Evaluation of the role of autophagy in fungal development and pathogenesis 3

136
APPENDIX
First author publication as listed on page xi.

CHAPTER TWENTY
Methods for Functional Analysis of
Macroautophagy in Filamentous Fungi
Yi Zhen Deng,* Marilou Ramos-Pamplona,* and Naweed I. Naqvi*
Contents
1. Introduction
296
1.1. Cellular functions of autophagy in filamentous fungi
296
2. Methods for the Functional Analysis of Autophagy
in Filamentous Fungi
297
2.1. Gene-deletion analyses to assess macroautophagy
in filamentous fungi
297
2.2. Use of chemical inhibitors to investigate autophagy in fungi
298
2.3. Microscopy methods to detect autophagy-associated
membrane structures
300
2.4. Monodansylcadaverine (MDC) staining of autophagic vesicles
302
2.5. LysoTracker-based visualization of vacuoles and
vesicular compartments
303
2.6. Analysis of glycogen sequestration and estimation
of glycogen content

304
2.7. Comparative proteomics for identifying the targets
of autophagic degradation
306
3. Concluding Remarks
307
Acknowledgments 307
References 308
Abstract
Autophagy is a bulk degradative process responsible for the turnover of mem-
branes, organelles, and proteins in eukaryotic cells. Genetic and molecular
regulation of autophagy has been independently elucidated in budding yeast
and mammalian cells. In filamentous fungi, autophagy is required for several
important physiological functions, such as asexual and sexual differentiation,
pathogenic development, starvation stress and programmed cell death during
heteroincompatibility. Here, we detail biochemical and microscopy methods
useful for measuring the rate of induction of autophagy in filamentous fungi,
Methods in Enzymology, Volume 451
#
2008 Elsevier Inc.
ISSN 0076-6879, DOI: 10.1016/S0076-6879(08)03220-5 All rights reserved.
* Fungal Patho-Biology Group, Temasek Life Sciences Laboratory and Department of Biological Sciences,
National University of Singapore, Singapore
295
and we summarize the methods that have been routinely used for monitoring
macroautophagy in both yeast and filamentous fungi. The role of autophagy in
carbohydrate catabolism and cell survival is discussed along with the specific
functions of macroautophagy in fungal development and pathogenesis.
1. Introduction
Autophagy is a highly conserved catabolic process in eukaryotes that is

responsible for organellar turnover, membrane recycling, and protein deg-
radation in vacuoles/lysosomes. Autophagy is induced in response to envi-
ronmental stress or developmental signals during cellular differentiation
(Besteiro et al., 2006; Noda and Ohsumi, 1998; Pinan-Lucarre et al.,
2003). Autophagy can act as a prosurvival signal or participate in pro-
grammed cell death, depending on the particular physiological conditions
(Codogno and Meijer, 2005).
There are three distinct classes of autophagy: macroautophagy, micro-
autophagy, and chaperone-mediated autophagy (CMA); the latter is selec-
tively used to degrade cytosolic proteins containing a specific pentapeptide
consensus motif (Majeski and Dice, 2004; Salvador et al., 2000). Macro-
autophagy and microautophagy (Reggiori and Klionsky, 2002) are consid-
ered nonselective and thus have more degradative capacity. The major
difference between macroautophagy and microautophagy is whether the
double-membrane vesicles, autophasogosomes, sequester cytoplasmic pro-
teins or organelles (macroautophagy) for delivery to the lysosome/vacuole
for degradation (Suzuki et al., 2001), or whether the cytoplasm is directly
engulfed into the vacuoles (microautophagy) (Mortimore et al., 1989).
Besides general autophagy, some specific types of autophagy exist, such as
crinophagy (the activity of lysosomes related to the secretory pathway and
endocrine functions; Glaumann, 1989), reticulophagy (degradation of ER;
Bernales et al., 2007; Bolender and Weibel, 1973) and pexophagy (degra-
dation of peroxisomes; Sakai and Subramani, 2000).
Thus far, 31 ATG genes (autophagy-related) have been characterized in
Saccharomyces cerevisiae, which has led to a better understanding of the genetic
and molecular regulation of autophagy (Kabeya et al., 2007; Klionsky et al.,
2003), particularly the formation of autophagy-associated vesicular compart-
ments, such as preautophagosomal structures (PAS), autophagosomes
(cytosolic), and autophagic bodies (vacuolar) (Suzuki et al.,2001).
1.1. Cellular functions of autophagy in filamentous fungi

Autophagy is reported to play a crucial role during differentiation of several
filamentous fungi such as Podospora, Aspergillus, Colletotrichum,andMagna-
porthe. Autophagy-deficient mutants of Magnaporthe oryzae are nonpathogenic
and show highly reduced asexual development (Liu et al., 2007; Veneault-
Fourrey et al., 2006). Loss of autophagy-assisted programmed cell death in
296 Yi Zhen Deng et al.
atg8D appressorium is proposed to be responsible for the failure of penetrating
the host cuticle ( Veneault-Fourrey et al., 2006). Furthermore, autophagy is
involved in lipid body turnover and thus is essential for turgor generation and
appressorium-mediated penetration (Liu et al., 2007). However, Colletotrichum
gloeosporioides, with a related infection strategy as M. oryzae, does not require
autophagic cell death for successful infection (Nesher et al.,2008). Surpris-
ingly, infection structures/appressoria from a CLK1-deletion (an ortholog of
MgATG1) mutant in Colletotrichum lindemuthianum, are unable to penetrate
the host cuticle (Dufresne et al.,1998), similar to the result from M. oryzae.In
S. cerevisiae, loss of autophagy leads to failure of sporulation, sensitivity to
nitrogen starvation, and increased pseudohyphal growth (Cutler et al., 2001;
Ma et al., 2007; Tsukada and Ohsumi, 1993). In A. oryzae, autophagy is
required for the differentiation of aerial hyphae and in conidial germination
(Kikuma et al.,2006). In contrast to its function in fungi mentioned previ-
ously, autophagy plays little or no role in the differentiation of the dimorphic
yeast Candida albicans within the host tissue (Palmer et al.,2007). The atg9D
mutant in C. albicans remains unaffected for yeast-hypha or chlamydospore
differentiation, though it shows specific defects in autophagy and the
cytoplasm-to-vacuole targeting (Cvt) pathway.
In P. anserina, autophagy is essential for sexual differentiation and cell
death by incompatibility. It remains controversial whether autophagy exe-
cutes a programmed cell death function or acts as a prosurvival response in
Podospora (Dementhon et al., 2003, 2004; Pinan-Lucarre et al., 2003; Pinan-
Lucarre et al., 2005). It was initially thought that autophagy acts as the cause

of cell death during incompatible interactions for it is induced when cells of
unlike genotypes fuse in P. anserina (Dementhon et al., 2004; Pinan-Lucarre
et al., 2003). A recent study suggests that autophagy serves a prosurvival role
during incompatibility, as loss of autophagy results in accelerated cell death
(Pinan-Lucarre et al., 2005).
In this chapter, we present a technical review of the most frequently used
methods to study autophagy in yeast and fungal species and focus on the
methods that are useful to monitor the induction and rate of autophagy
in filamentous fungi. The following protocols and methods have been
validated in the model fungus M. oryzae and can be easily adapted and
optimized for use in other filamentous fungi of interest.
2. Methods for the Functional Analysis of
Autophagy in Filamentous Fungi
2.1. Gene-deletion analyses to assess macroautophagy
in filamentous fungi
Generally, the one-step genedeletionstrategy (schematized in Fig. 20.1)isused
for disruption of requisite gene function in filamentous fungi. For gene
disruption of Magnaporthe ATG1 (MGG_06393.5), genomic DNA fragments
Analysis of Autophagy in Fungi 297
(about 1 kb each) representing the 5
0
and 3
0
flanks of the ATG1 open reading
frame were amplified by PCR, and ligated sequentially so as to flank the ILV1
cassette (confers resistance to sulfonyl urea) in pFGL385 to obtain plasmid
vector pFGLatg1KO. The pFGLatg1KO plasmid was transformed into wild-
type M. oryzae using Agrobacterium T-DNA-mediated transformation for
homology-dependent replacement of the ATG1 gene. Gene-disruption con-
structs can also be delivered into the fungal species of choice using electropo-

ration of spheroplasts (Talbot et al., 1993; Vollmer and Yanofsky, 1986;
Xoconostle-Cazares et al., 1996; Yelton et al.,1984). Transformants were
selected for resistance to chlorimuron ethyl (100 mg/mL; Cluzeau Labo,
France) and correct gene-replacement confirmed by PCR analysis and South-
ern blotting. The primers used for amplifying the 1-Kb region at the 5
0
-and
3
0
-flank of the ATG1 gene were as follows: ATG1-5F (5
0
- GAGTGA-
GAATTCGCGGGACTAAGCAGGCCCAGGA-3
0
), ATG1-5R (5
0
-GA
GTGA
GAATTCTGCACTTAGAAACACTCGGGCT-3
0
), ATG1-3F
(5
0
- GAGACTGTTCTGCAGCCTGGCAGTGGTTATCGGTTCG-3
0
),
and ATG1-3R (5
0
- GAGAGTGTTAAGCTTGGACGTACAGTAGG-
TAATTGGT-3

0
). The preceding protocol is validated for Magnaporthe but
can be easily optimized for other fungal species of choice, provided the
requisite sequence information is available. Other selectable marker cassettes
for transformation in fungi include hygromycin resistance (HPH1)or
ammonium-gluphosinate resistance (BAR). Gene targeting of the marker
cassette can also be achieved by providing homology within the coding
sequence per se and need not be restricted to the flanking sequences as
described previously.
2.2. Use of chemical inhibitors to investigate
autophagy in fungi
There are several chemical inhibitors of autophagy that are commercially
available and routinely used in mammalian cells. Although these inhibitors
block autophagy, their effects are not entirely specific. Wortmannin ( WM)
Figure 20.1 Schematic representation of a one-step gene replacement strategy for the
ATG1 locus (MGG_06393.5) in Magnaporthe. Solid bars and short open boxes represent
coding regions and introns, respectively, whereas grey bars indicate the genomic f la nks
used as regions of homology for gene targeting. Relevant restriction enzyme sites have
been depicted and ILV1 refers to the sulfonyl urea-resistance cass ette used to replace
ATG1 to create the atg1D strain.
298 Yi Zhen Deng et al.
is an inhibitor of PI3-kinase and blocks the induction of autophagy
(Blommaart et al., 1997; Petiot et al., 2000). 3-methlyadenine (3-MA) is
also a classical inhibitor of the autophagic pathway (Seglen and Gordon,
1982). N-ethylmaleimide (NEM) inhibits several vesicular transport events
and thus blocks the formation of autophagic vacuoles ( Woodman, 1997).
Such inhibitors can be potentially useful in studying autophagy in fungal
systems provided special caution is exercised in analyzing the results. WM-
treatment of vegetative mycelia (Fig. 20.2)ofMagnaporthe wild-type strain
mimics the phenotype of an atg8D mutant, in which starvation fails to

induce autophagy (Deng and Naqvi, unpublished data).
Figure 20.2 Epifluorescence microscopy-based assessment of autophagosomes. Myce-
lia from the wild type or atg8D mutant were stained with MDC and imaged using
epifluorescence microscopy. MDC-stained wild-type mycelia pretreated with the
autophagic-inhibitor Wortmannin (WM) prior to starvation, served as a negative
control. Scale bar denotes 5 mm. Micrographs were pseudo colored using Photoshop
Ve r s i o n 7.
Analysis of Autophagy in Fungi 299
Procedure
1. Small amount of vegetative mycelia (approximately 50 mg of wet
weight; scraped from a colony surface with inoculation loop) or conidia
(ca 4 Â 10
3
) of wild-type and atg8D strains are cultured in 20 ml of
complete medium (CM; yeast extract 0.6%, casein hydrolysate 0.6%,
sucrose 1%) for 48 h at 28

C with gentle shaking (150 rpm) to obtain
sufficient biomass.
2. Wortmannin stock (1 mM, in DMSO) is added into the CM, to a final
concentration of 200 nM. The mycelia are treated with Wortmannin for
3 h at 28

C with gentle shaking.
3. Mycelia are harvested by filtration through sterile Miracloth (Calbio-
chem, USA) and washed twice by filtration with sterile distilled water.
4. A small amount of the freshly cultured mycelia is then inoculated into
20 ml of minimal medium lacking nitrogen [MM-N: 0.5 g/L KCl,
0.5 g/L MgSO
4

, 1.5 g/L KH
2
PO
4
, 0.1% (v/v) trace elements, 10 g/L
glucose, pH6.5; (Talbot et al., 1993)] containing 2 mM PMSF, and
grown for 16 h with gentle shaking at 28

C. Please note that this step
is carried out to induce autophagy and there may not be a visible change
in fungal biomass.
5. The pretreated and starved wild-type mycelia are stained with MDC and
observed using epifluorescence microscopy, as described subsequently.
2.3. Microscopy methods to detect autophagy-associated
membrane structures
Activation or induction of autophagy can be visualized by differential inter-
ference contrast (DIC; or Nomarski optics) microscopy. In S. cerevisiae (Lang
et al.,1998), C. albicans (Palmer et al.,2007), and P. anserina (Dementhon
et al., 2004), starvation stress leads to enlarged vacuoles, which can be
observed by DIC optics. Large punctuate perivacuolar structures or vesicles
inside the vacuolar lumen can also be visualized, indicating the formation of
autophagosomes or accumulation of autophagic bodies. In the Podospora
DpspA mutant, which lacks the PSPA vacuolar protease, the accumulation
of autophagic bodies in the vacuolar lumen is even more striking and easily
detectable by simple microscopy observations (Dementhon et al.,2003).
Fluorescence microscopy is another method to monitor the induction of
autophagy. N-terminal tagging of Atg8 with a fluorescent protein such as
GFP or RFP helps in epifluorescent detection of autophagosomes and has
been used in several filamentous fungi such as A. oryzae (Kikuma et al.,
2006), P. anserina (Pinan-Lucarre et al., 2005), and M. oryzae (Deng and

Naqvi, unpublished results). An advantage of utilizing GFP- or RFP-tagged
Atg8 is that the extent of autophagosome formation correlates well with the
amount of GFP/RFP-Atg8PE (Kabeya et al., 2000), so that the induction of
300 Yi Zhen Deng et al.
autophagy can be easily quantified by Western blotting using commercially
available anti-GFP/RFP antibodies.
Transmission electron microscopy (TEM) is the gold standard for ultra-
structural investigation of autophagy-associated membrane compartments
in filamentous fungi such as P. anserina (Pinan-Lucarre et al., 2003),
A. oryzae (Kikuma et al., 2006), and M. oryzae (Liu et al., 2007; Veneault-
Fourrey et al., 2006). Fig. 20.3 depicts the TEM analysis of the vacuolar
lumen in the wild type and an autophagy-deficient mutant (atg8D)of
Magnaporthe.
Procedure
1. Fresh conidia (4 Â 10
3
) or small amounts of mycelia (scraped from a
colony using inoculation loop; about 50 mg) from the wild-type or
atg8D strain in Magnaporthe are grown in 20 ml of CM for 48 h at 28

C
with gentle shaking (150 rpm).
2. Mycelia are harvested by filtration through Miracloth and washed
thoroughly using sterile distilled water.
3. Washed mycelia from individual strains are grown in liquid MM-N
medium [0.5 g/L KCl, 0.5 g/L MgSO
4
, 1.5 g/L KH
2
PO

4
, 0.1% (v/v)
trace elements, 10 g/L glucose, pH 6.5; Talbot et al., 1993] containing
2mM PMSF, for 16 h with gentle shaking, at 28

C.
4. A small amount of the fungal biomass harvested by filtration through
Miracloth, is placed in a microfuge tube and resuspended in 200 mlof
fixation reagent (2.5% glutaraldehyde in 0.1 M phosphate buffer, v/v,
pH 7.2) or sufficient to cover the mycelial sample. Initially the fixation
is carried out under vacuum for 15 min at room temperature and
subsequently at 4

C overnight.
Figure 20.3 Ultrastructural analysis of autophagy-related membrane compartments.
Wild-type or atg8D mycel ia grown in complete medium for 2 days and subjected to
nitrogen starvation for 16 h (in the presence of 2 mM PMSF) were processed for thin-
section transmission electron microscopy. Numerous autophagic bodies can be detected
in the vacuole of the wild-t ype strain. Scale bar ¼ 0.5 mm.
Analysis of Autophagy in Fungi 301
5. Fixed mycelia are washed 3 times (10 min each) with 0.1 M phosphate
buffer, pH 7.2.
6. The washed mycelial samples are postfixed for 3 h in 250 ml of osmium
tetraoxide (1%, w/v).
7. Fixed mycelia are again washed 3 times (10 min each) in 0.1 M phos-
phate buffer, pH 7.2.
8. Samples are dehydrated in a graded ethanol series (25%, 50%, 75%,
100%; 10 min in 500 ml each).
9. The samples are then washed twice, for 15 min each, in 250 mlof
propylene oxide.

10. Samples are infiltrated in 500 ml of propylene oxide-Spurr’s resin (1:1)
for 2 h, and then infiltrated overnight in 100% Spurr’s resin.
11. Next, samples are embedded in Spurr’s resin and polymerized overnight
at 70

C in EMS embedding molds (Electron Microscopy Sciences,
USA).
12. Ultrathin (80 nm) sections are generated using Leica Ultracut UCT and
mounted on 200 mesh copper grids.
13. Mounted sections are stained for 15 min at room temperature with a
mixture of 2% uranyl acetate and 2% Reynolds’ lead citrate (10 ml
for each grid) and examined using a JEM1230 transmission electron
microscope ( Jeol, Tokyo, Japan) at 120 kV.
2.4. Monodansylcadaverine (MDC) staining of
autophagic vesicles
MDC is an acidotropic dye that labels late stage autophagosomes or autop-
hagic vesicles (Niemann et al.,2001). MDC staining was successfully used for
monitoring the increased autophagic activity in nitrogen-starved Magnaporthe
mycelia (see Fig. 20.2), and the incorporation of MDC into late-stage
autophagosomes or autophagic vesicles was inhibited by pretreatment with
WM, the chemical inhibitor of autophagic sequestration. Furthermore,
MDC staining with the conidiating cultures of Magnaporthe at different stages
likely reflects the natural induction of autophagy (either basal levels or
developmentally-induced), without starvation. MDC-incorporated compart-
ments were copious in the conidiation-specific cell types, including aerial
hyphae, and both young and mature conidia. An important limitation is that
MDC staining fails to differentiate between late stage autophagosomes and
autophagic (acidified) vacuoles.
Procedure
1. Magnaporthe wild-type or atg8D strains are cultured on Prune agar

medium (PA; per liter: prune juice 40 ml, lactose 5 g, yeast extract 1 g,
agar 20 g) in the dark at 28

C for 2 days.
302 Yi Zhen Deng et al.
2. The wild-type and atg8D strains are then subjected to constant illumina-
tion (using overhead room lighting) to induce conidiation at room
temperature.
3. At 6, 12, and 48 h after photoinduction, the conidiating cultures of the
wild-type and atg8D strain are harvested by scraping with an inoculation
loop and stained with 0.05 mM MDC solution (stock solution: 5 mM in
normal phosphate buffered saline, pH 7.0) at 37

C for 15 min. The
MDC is then washed out with PBS before microscopic observation.
4. MDC-stained mycelia are observed using an epifluorescence microscope
equipped with the following filter sets: excitation wavelength 350 nm,
emission 320 to 520 nm.
2.5. LysoTracker-based visualization of vacuoles and
vesicular compartments
LysoTracker Green DND-26 and LysoTracker Red DND-99 (Invitrogen-
Molecular Probes, USA) are commonly used to stain and visualize acidic
compartments, including autophagic compartments (Liu et al., 2005; Scott
et al., 2004). LysoTracker dyes differ slightly from MDC and label acidified
autophagic vacuoles (Fig. 20.4) but fail to incorporate into late-stage
autophagosomes. In conidiating aerial hyphae of Magnaporthe, MDC stain-
ing was prominent, while very rare staining of LysoTracker Green DND-26
or LysoTracker Red DND-99 was detected. Both the LysoTracker- and
MDC-labeled spherical compartments were evident in conidia, indicating
that aerial hyphae are devoid of vacuoles that are mostly formed in conidia.

One major drawback of the use of LysoTrackers is the inability to perform
co-localization studies with RFP-Atg8 labeled vesicles (Deng and Naqvi,
unpublished data).
Figure 20.4 Mycelia from the wild-type B157 st rain of Magnaporthe was stained with
LysoTracker Green DND-26 or LysoTracker Red DN D99 and subjected to the requisite
epifluorescence microscopy to visualize acidified autophagic vacuoles. Bar ¼ 5 mm.
Different morphological variants of the fungal vacuoles (numerous small vesicles, top,
or big round vesicles, bottom) are detected through LysoTracke r DND staining.
Analysis of Autophagy in Fungi 303
Procedure
1. Magnaporthe wild-type and atg8D strains are grown on PA medium in the
dark at 28

C in an incubator for 2–3 days. The diameter of the resulting
fungal colonies is about 1–2 cm at this stage.
2. Conidiation is induced in the wild-type and atg8D colonies by subjecting
them to constant illumination at room temperature.
3. At 6, 12, and 48 h after photoinduction, the wild-type and atg8D con-
idiating cultures are harvested by scraping with an inoculation loop in
sterile distilled water and collected through filtration using Miracloth.
The harvested biomass is then stained with 50 nM LysoTracker Green
DND26 solution (Invitrogen-Molecular Probes, USA; the stock is 1 M
in DMSO) for 10 min at 37

C. Staining with LysoTracker Red DND99
(1 M in DMSO; Invitrogen-Molecular Probes, USA) followed the same
protocol as described previously.
4. The stained mycelia are observed with epifluorescence microscopy using
the following filter sets: excitation wavelength 488 nm, emission 505 to
530 nm. LysoTracker Red DND99 staining is visualized using 543 nm

excitation and a 560 nm emission filter.
2.6. Analysis of glycogen sequestration and estimation
of glycogen content
In most eukaryotic cells, glycogen is stored as a carbohydrate reserve.
Autophagy is involved in glycogen catabolism, which occurs in response
to depletion of nutrients or to particular growth/differentiation conditions.
In S. cerevisiae, glycogen can be degraded in the cytoplasm or inside vacuolar
compartments. In the cytoplasm, Gph1 mediates glycogen breakdown
resulting in the release of glucose 1-phosphate (G1P) (Hwang et al.,
1989). In vacuoles, glycogen degradation is catalyzed by the vacuolar
glucoamylase (Colonna and Magee, 1978), which produces glucose
6-phosphate (G6P). The vacuolar degradation of glycogen is sporulation-
specific and probably relies on autophagy for sequestration and delivery of
glycogen (Fonzi et al., 1979; Francois and Parrou, 2001; Wang et al., 2001;
Yamashita and Fukui, 1985).
To monitor the sequestration of glycogen in budding yeast, KI-I
2
staining or iodine vapor exposure is widely used (Hwang et al., 1989;
Wang et al., 2001). In M. oryzae, detection of glycogen by KI-I
2
staining
in differentiating conidia (Park et al., 2004; Thines et al., 2000) or by PAS
(Periodic acid-Schiff ) staining with sectioned fungal materials (Clergeot
et al., 2001) has been reported. Total glycogen content in the yeast or fungal
cells can be determined enzymatically by hydrolyzing the extracted glyco-
gen with a-amylase and amyloglucosidase, and then measuring the released
glucose in a colorimetric assay containing glucose oxidase, peroxidase and
o-dianisidine (Gascon and Lampen, 1968; Lillie and Pringle, 1980). The kits
304 Yi Zhen Deng et al.
for total starch (Megazyme, UK) estimations are also commercially available

and have been successfully used in Magnaporthe samples (Deng and Naqvi,
unpublished data).
Procedures
Iodine vapor staining:
1. The fungal strain of interest is subcultured on PA medium and
allowed to grow in the dark at 28

C for 2 days.
2. The dark-grown cultures are then subjected to constant illumination
at room temperature to induce conidiation.
3. At 0, 2, and 4 d after photoinduction, the culture dishes containing
the colonies are inverted directly (with the lids removed) over a glass
beaker containing iodine crystals for approximately 15 min. Caution:
This step has to be carried out in a proper fume hood using appropri-
ate safety measures.
4. The iodine-stained colonies (Fig. 20.5) are photographed immediately.
Estimation of total glycogen in fungal tissue(s)
1. The fungal strains (wild type or the autophagy-deficient mutant of
interest) are grown on prune agar medium in the dark at 28

Cfor2days.
2. Conidiation is induced in the fungal strains by subjecting them to
constant illumination at room temperature.
3. At 0, 2, and 4 d after photoinduction, colonies from the respective
strains are harvested by scraping the colony surface with inoculation
loops in approximately 10–15 ml of sterile distilled water and then
filtered using sterile Miracloth (Calbiochem, USA).
4. The harvested biomass is ground to a powder in liquid nitrogen using
a mortar and pestle.
5. Thetotal glycogen content is estimated usingthe Megazyme Total Starch

Kit (Megazyme, UK) as instructed. Totalglycogen content isnormalized
to the wet weight of the fungal biomass used for the estimation.
Figure 20.5 Analysis of glycogen accumulation during Magnaporthe conidiation.
A wild-type strain grown in the dark for 2 days was subjected to constant illumination
for the specified ti me intervals, and finally ex posed to iodine vapor for 15 min and
quickly photographed.
Analysis of Autophagy in Fungi 305
2.7. Comparative proteomics for identifying the targets
of autophagic degradation
Because autophagy is a catabolic process responsible for lysosomal (vacuolar)
degradation of proteins, a block of the autophagy pathway will lead to
accumulation of proteins that are destined for autophagic degradation. To
identify proteins that are regulated by autophagic degradation during con-
idiogenesis, we performed an SDS-PAGE fractionation of total lysates from
4-day-old conidiating cultures of the wild-type, the atg8D mutant, and the
complemented strains. Mass spectrometry was then used to identify the
proteins that showed differential accumulation. A vacuolar a-mannosidase
(MGG_04464) in the Magnaporthe genome was present in the atg8D mutant
but absent in the wild-type or the complemented strains. In yeast, Ams1 is
delivered to the vacuoles via Cvt and autophagy pathway and acts as a
vacuolar hydrolase for free oligosaccharide degradation (Chantret et al.,
2003; Hutchins and Klionsky, 2001). Another protein identified in this
assay is Gph1 (MGG_01819), which is responsible for cytoplasmic glycogen
degradation (Hwang et al., 1989). The third protein that accumulates in
atg8D mutant but not the wild type was a methionine synthase
(MGG_06712). In S. cerevisiae, Pichia pastoris, Cryptococcus neoformans, and
C. albicans, the cobalamin-independent methionine synthase is required for
the growth of yeast or mycelial cells on methionine-free minimal media
(Burt et al., 1999; Csaikl and Csaikl, 1986; Huang et al., 2005; Pascon et al.,
2004). In the human pathogenic fungi C. neoformans and C. albicans,

methionine synthase acts as a virulence factor (Burt et al., 1999; Pascon
et al., 2004). An aconitase (MGG_03521) function showed reduced expres-
sion on loss of autophagy. All the identified proteins are involved in fungal
metabolism and potentially subjected to regulation by autophagy, thus
validating the importance of such a comparative proteomics approach in
identifying the potential targets of the autophagic degradation pathway for
proteins.
Procedure
1. Wild type and the atg8D mutant are cultured on PA medium in the dark
at 28

C for 2 days.
2. The dark-grown cultures are subjected to constant illumination to
induce conidiation.
3. At 4 d after photoinduction, the conidiating cultures are harvested by
scraping the colony with inoculation loops in 10–15 ml of sterile distilled
water. The fungal biomass is collected by filtration through sterile
Miracloth (Calbiochem, USA).
4. The harvested fungal biomass is ground to a fine powder in liquid
nitrogen, using an autoclaved mortar and pestle, and resuspended in
300 ml of extraction buffer (10 mM Na
2
HPO
4
, pH7.0, 0.5% SDS,
1mM DTT and 1 mM EDTA). Lysates were cleared by centrifugation
306 Yi Zhen Deng et al.
at 12,000 g for 30 min at 4

C. Protein concentration in the supernatant

fraction is determined by Nanodrop ND-1000 spectrophotometer
(Thermo Scientific, USA).
5. Normalized protein sample from each extract is fractionated by SDS-
PAGE (10%), mixing with equal volume of loading buffer (Bio-Rad,
USA).
6. Protein bands of interest (showing differential expression levels between
the wild-type and atg mutant) are excised from the gel.
7. In-gel digestion and peptide extraction are performed using the following
protocol: />52567430063b5d6/2265a5645f93475785256cbe00558a60/$FILE/P3650
5D.pdf.
8. Proteins of interest are identified by peptide sequencing or MALDI-
TOF mass spectrometry.
3. Concluding Remarks
A recent review (Klionsky et al.,2007) discusses methodologies to study
the dynamics of autophagy in yeast and mammalian cells. The methods
described previously measure the steady-state levels of autophagy, covering
the induction, vesicle formation, and vesicle fusion. Very few assays that
monitor the complete autophagic flux have been reported in filamentous
fungi. One such assay relates to measuring the total protein breakdown in
starved cells in wild-type versus an autophagy-deficient mutant in yeast
(Tsukada and Ohsumi, 1993). An alternate method requires quantification
of the total alkaline phosphatase activity in a yeast strain expressing a truncated
form of Pho8, Pho8D60, which exists as an inactive precursor in the cytosol
and is entirely dependent on autophagy to be delivered to the vacuole for its
activity during starvation. Pho8D60 gains its phosphatase activity on specific
processing by vacuolar enzymes so that the measurement of its activity can be
an indicator of the fusion of autophagosomes with the vacuole and the
subsequent processing therein (Kirisako et al., 1999; Wang et al.,2001). See
the chapters by Noda and Klionsky, Cabrera and Ungermann, and Mayer in
this volume for information on the Pho8D60 assay.

ACKNOWLEDGMENTS
We thank Yang Ming for generating the atg1 deletion mutant, and the Fungal Patho-biology
group for discussions and suggestions. We are grateful to S. Naqvi for critical reading of
the manuscript. Intramural research support from the Temasek Life Sciences Laboratory,
Singapore is gratefully acknowledged.
Analysis of Autophagy in Fungi 307
REFERENCES
Bernales, S., Schuck, S., and Walter, P. (2007). ER-phagy: Selective autophagy of the
endoplasmic reticulum. Autophagy 3, 285–287.
Besteiro, S., Williams, R. A., Morrison, L. S., Coombs, G. H., and Mottram, J. C. (2006).
Endosome sorting and autophagy are essential for differentiation and virulence of
Leishmania major. J. Biol. Chem. 281, 11384–11396.
Blommaart, E. F., Krause, U., Schellens, J. P., Vreeling-Sindelarova, H., and Meijer, A. J.
(1997). The phosphatidylinositol 3-kinase inhibitors wortmannin and LY294002 inhibit
autophagy in isolated rat hepatocytes. Eur. J. Biochem. 243, 240–246.
Bolender, R. P., and Weibel, E. R. (1973). A morphometric study of the removal of
phenobarbital-induced membranes from hepatocytes after cessation of threatment.
J. Cell Biol. 56, 746–761.
Burt, E. T., O’Connor, C., and Larsen, B. (1999). Isolation and identification of a 92-kDa
stress induced protein from Candida albicans. Mycopathologia 147, 13–20.
Chantret, I., Frenoy, J. P., and Moore, S. E. (2003). Free-oligosaccharide control in the yeast
Saccharomyces cerevisiae: Roles for peptide:N-glycanase (Png1p) and vacuolar mannosidase
(Ams1p). Biochem. J. 373, 901–908.
Clergeot, P. H., Gourgues, M., Cots, J., Laurans, F., Latorse, M. P., Pepin, R.,
Tharreau, D., Notteghem, J. L., and Lebrun, M. H. (2001). PLS1, a gene encoding a
tetraspanin-like protein, is required for penetration of rice leaf by the fungal pathogen
Magnaporthe grisea. Proc. Natl. Acad. Sci. USA 98, 6963–6968.
Codogno, P., and Meijer, A. J. (2005). Autophagy and signaling: Their role in cell survival
and cell death. Cell Death Differ. 12, 1509–1518.
Colonna, W. J., and Magee, P. T. (1978). Glycogenolytic enzymes in sporulating yeast.

J. Bacteriol. 134, 844–853.
Csaikl, U., and Csaikl, F. (1986). Molecular cloning and characterization of the MET6 gene
of Saccharomyces cerevisiae. Gene 46, 207–214.
Cutler, N. S., Pan, X., Heitman, J., and Cardenas, M. E. (2001). The TOR signal
transduction cascade controls cellular differentiation in response to nutrients. Mol. Biol.
Cell 12, 4103–4113.
Dementhon, K., Paoletti, M., Pinan-Lucarre, B., Loubradou-Bourges, N., Sabourin, M.,
Saupe, S. J., and Clave, C. (2003). Rapamycin mimics the incompatibility reaction in the
fungus Podospora anserina. Eukaryot. Cell 2, 238–246.
Dementhon, K., Saupe, S. J., and Clave, C. (2004). Characterization of IDI-4, a bZIP
transcription factor inducing autophagy and cell death in the fungus Podospora anserina.
Mol. Microbiol. 53, 1625–1640.
Dufresne, M., Bailey, J. A., Dron, M., and Langin, T. (1998). clk1, a serine/threonine
protein kinase-encoding gene, is involved in pathogenicity of Colletotrichum lindemuthia-
num on common bean. Mol. Plant Microbe Interact. 11, 99–108.
Fonzi, W. A., Shanley, M., and Opheim, D. J. (1979). Relationship of glycolytic inter-
mediates, glycolytic enzymes, and ammonia to glycogen metabolism during sporulation
in the yeast Saccharomyces cerevisiae. J. Bacteriol. 137, 285–294.
Francois, J., and Parrou, J. L. (2001). Reserve carbohydrates metabolism in the yeast
Saccharomyces cerevisiae. FEMS Microbiol. Rev. 25, 125–145.
Gascon, S., and Lampen, J. O. (1968). Purification of the internal invertase of yeast.
J. Biol.
Chem. 243, 1567–1572.
Glaumann,
H. (1989). Crinophagy as a means for degrading excess secretory proteins in rat
liver. Revis. Biol. Celular 20, 97–110.
Huang, L., Li, D. Y., Wang, S. X., Zhang, S. M., Chen, J. H., and Wu, X. F. (2005).
Cloning and identification of methionine synthase gene from Pichia pastoris. Acta Biochim.
Biophys. Sin. (Shanghai) 37, 371–378.
308 Yi Zhen Deng et al.

Hutchins, M. U., and Klionsky, D. J. (2001). Vacuolar localization of oligomeric alpha-
mannosidase requires the cytoplasm to vacuole targeting and autophagy pathway
components in Saccharomyces cerevisiae. J. Biol. Chem. 276, 20491–20498.
Hwang, P. K., Tugendreich, S., and Fletterick, R. J. (1989). Molecular analysis of GPH1,
the gene encoding glycogen phosphorylase in Saccharomyces cerevisiae. Mol. Cell. Biol. 9,
1659–1666.
Kabeya, Y., Mizushima, N., Ueno, T., Yamamoto, A., Kirisako, T., Noda, T., Kominami, E.,
Ohsumi, Y., and Yoshimori, T. (2000). LC3, a mammalian homologue of yeast Apg8p, is
localized in autophagosome membranes after processing. EMBO J. 19, 5720–5728.
Kabeya, Y., Kawamata, T., Suzuki, K., and Ohsumi, Y. (2007). Cis1/Atg31 is required for
autophagosome formation in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun.
356, 405–410.
Kikuma, T., Ohneda, M., Arioka, M., and Kitamoto, K. (2006). Functional analysis of the
ATG8 homologue Aoatg8 and role of autophagy in differentiation and germination in
Aspergillus oryzae. Eukaryot. Cell 5, 1328–1336.
Kirisako, T., Baba, M., Ishihara, N., Miyazawa, K., Ohsumi, M., Yoshimori, T., Noda, T.,
and Ohsumi, Y. (1999). Formation process of autophagosome is traced with Apg8/
Aut7p in yeast. J. Cell. Biol. 147, 435–446.
Klionsky, D. J., Cregg, J. M., Dunn, W. A., Jr., Emr, S. D., Sakai, Y., Sandoval, I. V.,
Sibirny, A., Subramani, S., Thumm, M., Veenhuis, M., and Ohsumi, Y. (2003).
A unified nomenclature for yeast autophagy-related genes. Dev. Cell 5, 539–545.
Klionsky, D. J., Cuervo, A. M., and Seglen, P. O. (2007). Methods for monitoring
autophagy from yeast to human. Autophagy 3, 181–206.
Lang, T., Schaeffeler, E., Bernreuther, D., Bredschneider, M., Wolf, D. H., and
Thumm, M. (1998). Aut2p and Aut7p, two novel microtubule-associated proteins are
essential for delivery of autophagic vesicles to the vacuole. EMBO J. 17, 3597–3607.
Lillie, S. H., and Pringle, J. R. (1980). Reserve carbohydrate metabolism in Saccharomyces
cerevisiae: Responses to nutrient limitation. J. Bacteriol. 143, 1384–1394.
Liu, X. H., Lu, J. P., Zhang, L., Dong, B., Min, H., and Lin, F. C. (2007). Involvement of a
Magnaporthe grisea serine/threonine kinase gene, MgATG1, in appressorium turgor and

pathogenesis. Eukaryot. Cell 6, 997–1005.
Liu, Y., Schiff, M., Czymmek, K., Talloczy, Z., Levine, B., and Dinesh-Kumar, S. P.
(2005). Autophagy regulates programmed cell death during the plant innate immune
response. Cell 121, 567–577.
Ma, J., Jin, R., Jia, X., Dobry, C. J., Wang, L., Reggiori, F., Zhu, J., and Kumar, A. (2007).
An interrelationship between autophagy and filamentous growth in budding yeast.
Genetics 177, 205–214.
Majeski, A. E., and Dice, J. F. (2004). Mechanisms of chaperone-mediated autophagy. Int.
J. Biochem. Cell Biol. 36, 2435–2444.
Mortimore, G. E., Poso, A. R., and Lardeux, B. R. (1989). Mechanism and regulation of
protein degradation in liver. Diabetes Metab. Rev. 5, 49–70.
Nesher, I., Barhoom, S., and Sharon, A. (2008). Cell cycle and cell death are not necessary
for appressorium formation and plant infection in the fungal plant pathogen Colletotrichum
gloeosporioides. BMC Biol. 6, 9.
Niemann, A., Baltes, J., and Elsasser, H. P. (2001). Fluorescence properties and staining
behavior of monodansylpentane, a structural homologue of the lysosomotropic agent
monodansylcadaverine. J. Histochem. Cytochem. 49, 177–185.
Noda, T., and Ohsumi, Y. (1998). Tor, a phosphatidylinositol kinase homologue, controls
autophagy
in yeast. J. Biol. Chem. 273, 3963–3966.
Palmer, G. E., Kelly, M. N., and Sturtevant, J. E. (2007). Autophagy in the pathogen
Candida albicans. Microbiol. 153, 51–58.
Park, G., Bruno, K. S., Staiger, C. J., Talbot, N. J., and Xu, J. R. (2004). Independent
genetic mechanisms mediate turgor generation and penetration peg formation during
plant infection in the rice blast fungus. Mol. Microbiol. 53, 1695–1707.
Analysis of Autophagy in Fungi 309
Pascon, R. C., Ganous, T. M., Kingsbury, J. M., Cox, G. M., and McCusker, J. H. (2004).
Cryptococcus neoformans methionine synthase: Expression analysis and requirement for
virulence. Microbiol. 150, 3013–3023.
Petiot, A., Ogier-Denis, E., Blommaart, E. F., Meijer, A. J., and Codogno, P. (2000).

Distinct classes of phosphatidylinositol 3
0
-kinases are involved in signaling pathways that
control macroautophagy in HT-29 cells. J. Biol. Chem. 275, 992–998.
Pinan-Lucarre, B., Paoletti, M., Dementhon, K., Coulary-Salin, B., and Clave, C. (2003).
Autophagy is induced during cell death by incompatibility and is essential for differentia-
tion in the filamentous fungus Podospora anserina. Mol. Microbiol. 47, 321–333.
Pinan-Lucarre, B., Balguerie, A., and Clave, C. (2005). Accelerated cell death in Podospora
autophagy mutants. Eukaryot. Cell 4, 1765–1774.
Reggiori, F., and Klionsky, D. J. (2002). Autophagy in the eukaryotic cell. Eukaryot. Cell 1,
11–21.
Sakai, Y., and Subramani, S. (2000). Environmental response of yeast peroxisomes: Aspects
of organelle assembly and degradation. Cell. Biochem. Biophys. 32, 51–61.
Salvador, N., Aguado, C., Horst, M., and Knecht, E. (2000). Import of a cytosolic protein
into lysosomes by chaperone-mediated autophagy depends on its folding state. J. Biol.
Chem. 275, 27447–27456.
Scott, R. C., Schuldiner, O., and Neufeld, T. P. (2004). Role and regulation of starvation-
induced autophagy in the Drosophila fat body. Dev. Cell 7, 167–178.
Seglen, P. O., and Gordon, P. B. (1982). 3-Methyladenine: specific inhibitor of autophagic/
lysosomal protein degradation in isolated rat hepatocytes. Proc. Natl. Acad. Sci. USA 79,
1889–1892.
Suzuki, K., Kirisako, T., Kamada, Y., Mizushima, N., Noda, T., and Ohsumi, Y. (2001).
The pre-autophagosomal structure organized by concerted functions of APG genes is
essential for autophagosome formation. EMBO J. 20, 5971–5981.
Talbot, N. J., Ebbole, D. J., and Hamer, J. E. (1993). Identification and characterization of
MPG1, a gene involved in pathogenicity from the rice blast fungus Magnaporthe grisea.
Plant Cell. 5, 1575–1590.
Thines, E., Weber, R. W., and Talbot, N. J. (2000). MAP kinase and protein kinase
A-dependent mobilization of triacylglycerol and glycogen during appressorium turgor
generation by Magnaporthe grisea. Plant Cell. 12, 1703–1718.

Tsukada, M., and Ohsumi, Y. (1993). Isolation and characterization of autophagy-defective
mutants of Saccharomyces cerevisiae. FEBS Lett. 333, 169–174.
Veneault-Fourrey, C., Barooah, M., Egan, M., Wakley, G., and Talbot, N. J. (2006).
Autophagic fungal cell death is necessary for infection by the rice blast fungus. Science
312, 580–583.
Vollmer, S. J., and Yanofsky, C. (1986). Efficient cloning of genes of Neurospora crassa. Proc.
Natl. Acad. Sci. USA 83, 4869–4873.
Wang, Z., Wilson, W. A., Fujino, M. A., and Roach, P. J. (2001). Antagonistic controls of
autophagy and glycogen accumulation by Snf1p, the yeast homolog of AMP-activated
protein kinase, and the cyclin-dependent kinase Pho85p. Mol. Cell. Biol. 21, 5742–5752.
Woodman, P. G. (1997). The roles of NSF, SNAPs and SNAREs during membrane fusion.
Biochim. Biophys. Acta 1357, 155–172.
Xoconostle-Cazares, B., Leon-Ramirez, C., and Ruiz-Herrera, J. (1996). Two chitin
synthase genes from Ustilago maydis
. Microbiol. 142, 377–387.
Yamashita,
I., and Fukui, S. (1985). Transcriptional control of the sporulation-specific
glucoamylase gene in the yeast Saccharomyces cerevisiae. Mol. Cell. Biol. 5, 3069–3073.
Yelton, M. M., Hamer, J. E., and Timberlake, W. E. (1984). Transformation of Aspergillus
nidulans by using a trpC plasmid. Proc. Natl. Acad. Sci. USA 81, 1470–1474.
310 Yi Zhen Deng et al.
©2009 Landes Bioscience. Do not distribute.
Autophagy, a conserved pathway for bulk cellular degrada-
tion and recycling in eukaryotes, regulates proper turnover of
organelles, membranes and certain proteins. Such regulated degra-
dation is important for cell growth and development, particularly
during environmental stress conditions, which act as key inducers
of autophagy. We found that autophagy and MoATG8 were
significantly induced during asexual development in Magnaporthe
oryzae. An RFP-tagged MoAtg8 showed specific localization and

enrichment in aerial hyphae, conidiophores and conidia. We
confirmed that loss of MoATG8 results in dramatically reduced
ability to form conidia, the asexual spores that propagate rice-
blast disease. Exogenous supply of glucose or sucrose significantly
suppressed the conidiation defects in a MoATG8-deletion mutant.
Comparative proteomics-based identification and characteriza-
tion of Gph1, a glycogen phosphorylase that catalyzes glycogen
breakdown, indicated that autophagy-assisted glycogen homeo-
stasis is likely important for proper aerial growth and conidiation
in Magnaporthe. Loss of Gph1, or addition of G6P, significantly
restored conidiation in the Moatg8Δ mutant. Overproduction of
Gph1 led to reduced conidiation in wild-type Magnaporthe strain.
We propose that glycogen autophagy actively responds to and regu-
lates carbon utilization required for cell growth and differentiation
during asexual development in Magnaporthe.
Introduction
Magnaporthe oryzae is a filamentous ascomycete that causes
blast disease, the most important fungal disease of rice and several
monocot species.
1
During its life cycle, M. oryzae undergoes an
asexual reproductive process known as conidiation.
2
The signaling
of the onset of conidiation is poorly understood. Thus far, the only
known environmental requirement for Magnaporthe conidiation is
exposure of the growing colony to light. Darkness is reported to be
necessary for the differentiation of mycelia into aerial hyphae (the
hyphae that grow perpendicular into the air from the vegetative
mycelial mat; Fig. 1A), which imparts a fluffy appearance to the

colony.
3
As depicted in Figure 1A, the conidiophore, or conidiog-
enous hypha, is composed of a stalk and distinguished through a
bulbous swelling at its tip. The bulbous outgrowth finally develops
into a spindle-shaped conidium, and eventually a septum delineates
the conidium from the stalk. After the first conidium is produced,
the next three or four conidia develop sequentially at the side of the
stalk in a sympodial manner (Fig. 1A). A mature conidium is pyri-
form and composed of three cells each containing a nucleus derived
from a common mother nucleus. Magnaporthe conidia are dispersed
by air and act as key determinants of the spread and severity of blast
disease. Rather surprisingly, very little is known about the genetic
control of conidia formation in Magnaporthe.
3-8
Macroautophagy (hereafter autophagy), a highly conserved
catabolic process in eukaryotes, is responsible for organellar turn-
over, membrane recycling and lysosomal (vacuolar) degradation of
proteins. Autophagy is induced during several biological processes,
including response to environmental stress or pathogen invasion,
and cellular remodeling during development and differentiation.
9-14

Thus far, 31 ATG genes (AuTophaGy) have been identified in
Saccharomyces cerevisiae and has led to a proper understanding of
the genetic and molecular regulation of autophagy.
15-19
In yeast, loss
of autophagy leads to a reduction in cell viability during nitrogen
starvation. Furthermore, all atg mutants show defects in sporulation,

indicating the importance of autophagy during sexual develop-
ment. Among these genes, ATG8 has been reported to be required
specifically for autophagosome formation in yeast.
20
Atg8p is a
membrane-bound ubiquitin-like protein that associates with pre-
autophagosomal structures (PAS), autophagosomes (in cytoplasm)
and autophagic bodies (in vacuolar compartment) and has therefore
been used extensively as a marker for such autophagy-associated
vesicular compartments.
21
The amount of Atg8-PE correlates well
with the extent of autophagosome formation.
22-25
MoATG8 is
Research Paper
Autophagy-assisted glycogen catabolism regulates asexual
differentiation in Magnaporthe oryzae
Yi Zhen Deng, Marilou Ramos-Pamplona and Naweed I. Naqvi*
Temasek Life Sciences Laboratory; and Department of Biological Sciences; 1 Research Link; National University of Singapore; Singapore
Abbreviations: Atg8, autophagy related gene 8; Cfu, colony forming unit; G6P, glucose 6-phosphate; G1P, glucose 1-phosphate; hpi, hours
post inoculation; LC3, microtubule-associated light chain 3; MDC, monodansyl cadavarine; ORF, open reading frame; PE, phosphatidyle-
thanolamine; PMSF, phenylmethylsulfonylfluoride; RFP, red fluorescent protein; RT-PCR, reverse transcription polymerase chain reaction;
TEM, transmission electron microscopy; TOR, target of rapamycin
Key words: autophagy, Atg8, LC3, Gph1, Magnaporthe, glycogen, asexual reproduction, conidiogenesis, rice-blast
[Autophagy 5:1, 33-43; 1 January 2009]; ©2009 Landes Bioscience
*Correspondence to: Naweed Naqvi; Temasek Life Sciences Laboratory; 1 Research
Link; National University of Singapore; Singapore 117604 Singapore; Tel.:
65.6872.7493; Fax: 65.6872.7007; Email:
Submitted: 03/29/08; Revised: 10/02/08; Accepted: 10/13/08

Previously published online as an Autophagy E-publication:
/>www.landesbioscience.com Autophagy 33
©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
34 Autophagy 2009; Vol. 5 Issue 1
then subjected to photo-induction to initiate asexual development/
conidiation. At 3–6 hpi, numerous aerial hyphae were evident in the
wild type, whereas aerial hyphae in the atg8Δ were stunted in growth
and highly diminished in numbers (Fig. 1B). At 9–12 hpi, a majority
of the WT aerial hyphae started to differentiate into conidiophores as
judged by the swollen tips and/or the inception of conidium initials
essential for autophagic cell death during infection-related morpho-
genesis and appressorium formation in Magnaporthe.
26
Conidiation
defects upon loss of autophagy have been reported recently in
Magnaporthe
27
and Aspergillus spps.
28,29
Autophagy was initially defined as a nonselective, bulk degradation
system within cells, until the first evidence for specific sequestration
of the smooth ER by autophagy was reported.
30
Recent reports on
glycogen autophagy, the breakdown of intracellular glycogen reserves
within autophagic vacuoles, also showed that autophagy does possess
some selectivity. In yeast, the vacuolar compartment is also implicated
as the degradation site for glycogen that is delivered by autophagy,
to cope with the nutrient limitation.

31-33
Glycogen autophagy has
been implicated in newborn animals as a strategy to cope with rapid
energy and metabolic requirements before gluconeogenesis can be
initiated.
34-37
Like conventional autophagy, glycogen autophagy can
be induced by rapamycin in newborn rat hepatocytes,
38
suggesting a
TOR-dependent regulation.
In the present study, we report that MoAtg8 (hereafter Atg8)
and autophagic activity are naturally induced at an early stage
during conidiation in Magnaporthe, and localized to the subcellular
structures that are highly correlated. We show that RFP-Atg8-PE
is recruited to autophagosomes in conidiation-related cell types.
We confirm that the loss of ATG8 causes pleiotropic defects during
asexual development in Magnaporthe.
26,27
Furthermore, we identify
proteins differentially regulated in the atg8Δ mutant and investigate
the role of autophagy and nutrient utilization during conidiation
and in planta growth in Magnaporthe. By genetic and biochemical
studies, we show that autophagy and glycogen homeostasis are
involved in tuning carbohydrate metabolism and utilization as an
important function during asexual differentiation in Magnaporthe.
Results
Autophagy regulates asexual differentiation and conidiation
in Magnaporthe. We created an atg8Δ strain in Magnaporthe by
replacing the entire MGG_01062 (NCBI accession XP_368182)

ORF with the hygromycin-resistance marker cassette (HPH1; Fig.
S1A). The full-length genomic copy of Magnaporthe ATG8 was
introduced into the atg8Δ mutant for genetic complementation
analysis. The replacement of ATG8 and the single-copy ectopic
integration of the genomic ATG8 locus were confirmed by Southern
blotting (Fig. S1B). Two independent strains in each instance were
then characterized for various aspects of vegetative growth, colony
formation and pathogenesis. Autophagy is normally induced in
response to nutrient depletion.
39
Consistent with a recent study,
26

the atg8Δ mutant failed to form autophagosomes in response to
nitrogen starvation (Suppl. data, Fig. S2A). Furthermore, numerous
MDC-positive autophagosomes were apparent in the wild type,
but were undetectable in atg8Δ mycelia or in the WT treated with
Wortmannin (Fig. S2B).
When utilizing lactose as the sole carbon source, the atg8Δ mutant
showed radial growth comparable to the wild-type (Fig. S1C), but
displayed a significant reduction in aerial hyphae imparting a flat
appearance to the colony, which was particularly evident in cross-
sectional view (Fig. 1B; top) and was confirmed by microscopic
analysis of the respective conidiating cultures (Fig. 1B). We quanti-
fied the aerial hyphal growth and differentiation, in a time-course
analysis on wild type and atg8Δ grown in the dark for two days and
Figure 1. Growth characteristics, aerial hyphal development, and conidia-
tion in the atg8Δ mutant in Magnaporthe. (A) Schematic representation of the
asexual cycle in Magnaporthe depicted as a time series (in hours) of devel-
opmental events induced in a vegetative mycelium upon exposure to light at

0 h. (B) Loss of ATG8 leads to a huge reduction in aerial hyphae formation
and conidiation. Cross-section from 5-day old wild type or atg8Δ colony
grown under constant illumination was photographed (top). Wild type or
atg8Δ colonies were stained with acid fuchsin and analyzed by bright field
microscopy at the indicated time points post photo-induction. Scale bar
represents 10 μm. (C) Loss of autophagy leads to pleiotropic abnormalities
in conidiation. Conidiation was induced in wild type or atg8Δ strain and
samples stained with calcofluor white. Epifluorescence microscopic images
shown are representative (n = 300 each) of the developmental stage (or
defects) depicted by the majority of aerial hyphae or conidiophores in the
respective samples. Conidia started forming and maturing properly in the
wild-type strain, whereas the atg8Δ mutant showed abnormalities in all
aspects of conidia development. The experiment was repeated three times
using a total of 10 colonies per strain in each instance. Scale bar equals
10 micron.
©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
www.landesbioscience.com Autophagy 35
from a likely cleavage at Gly116) fused to RFP. It is possible that
this peptide is degraded rapidly in the cytosol, since the Atg8-RFP
could be detected as a faint signal in micrographs (Fig. 2B) and in
immunoblots with RFP antiserum (Fig. 2C). Having ascertained
the specificity of the antiRFP antisera (Fig. S3), we then assessed
the various predicted Atg8 modifications by western analyses with
antiRFP antisera on total lysates from RFP-ATG8 or ATG8-RFP
strain grown either in CM or MM - N medium. Based on relative
mobility and predicted size, the two bands detected by the antiRFP
antibody in ATG8-RFP lysates (Fig. 2C; MM - N) were judged to
be the full-length Atg8-RFP and likely the DLFEEVE peptide fused
to RFP (asterisk; although unconfirmed but likely generated upon

cleavage at glycine 116) respectively. The stronger Atg8p-RFP band
also indicated the enhanced expression of ATG8 under nitrogen
limiting conditions. In MM - N cultured RFP-ATG8 strain, proteins
likely representing RFP-Atg8p, RFP-Atg8p-PE and RFP alone were
detected (Fig. 2C). Compared to the CM grown mycelia, there was
an obvious and expected increase in the amount of the predicted
RFP-Atg8-PE (Fig. 2C) along with the other modified forms.
The mRFP moiety in an mRFP-LC3 fusion is resistant to the
lysosomal acidic and degradative conditions.
40
The intense 26-kDa
band likely corresponds to such free vacuolar RFP in the RFP-ATG8
strain, and indirectly indicates enhanced autophagic response during
nitrogen starvation in Magnaporthe. These results helped us to deduce
that Atg8p is induced during nitrogen starvation and likely under-
goes posttranslational processing such as endoproteolytic cleavage
(likely at glycine 116 as reported in some organisms, although we
do not rule out an alternate cleavage site in Magnaporthe Atg8) and
conjugation to PE. Furthermore, we infer that the lipidated form of
RFP-Atg8 is likely enriched in autophagosomes and could serve as an
appropriate tool to detect and depict autophagy in Magnaporthe.
Induction and subcellular localization of RFP-Atg8 in
Magnaporthe. Using the RFP-ATG8 strain and epifluorescence
microscopy, we visualized the temporal and spatial distribution of
RFP-Atg8 protein (and autophagosomes/autophagic bodies) during
asexual differentiation in Magnaporthe. Calcofluor white was used as
a co-stain to delineate the outline of the analyzed fungal structures.
After growth in prune agar medium (PA; non-induced medium for
autophagy) in the dark for 2 days, the RFP-ATG8 cultures were
subjected to constant illumination to induce conidiation. During

growth in the dark, profuse aerial hyphae were formed, which
upon light induction showed tip swelling and differentiated into
conidiophores and finally produced mature conidia. At early stages,
RFP-Atg8p was prominently seen localizing as discrete puncta in the
aerial hyphae (Fig. 2D). Later on, the RFP-Atg8 signals appeared
as punctate structures as well as showed vacuolar localization in
the stalks and the swollen tips of conidiophores (Fig. 2D). Upon
initiation of the first incipient conidium, the RFP-Atg8 puncta were
located prominently in the developing conidium, and remained
highly enriched in these structures until proper differentiation of
asexual spores (Fig. 2D). The induction of autophagy in aerial struc-
tures during asexual differentiation was distinct and significanty
stronger (p < 0.0001) as compared to the starvation or age-related
induction of autophagy in mycelia (Suppl. Fig. S4A and B). In a
parallel experiment, Lysotracker Green DND-26 staining confirmed
copious autophagic vacuoles or late-stage autophagosomes in several
conidiation specific cell types (Fig. 2E). We conclude that RFP-Atg8
therein (Fig. 1B, lower). In contrast the atg8Δ showed a significant
decrease in aerial hyphae formation and conidiophore differentia-
tion (Fig. 1B). At 24 hpi, 34.0 ± 4.5% of wild-type aerial hyphae
produced conidiophores, as opposed to only 2.0 ± 1.0% in the
atg8Δ mutant (p = 0.001). Such decreased efficiency of conidiophore
formation in the atg8Δ remained unchanged even at 48 hpi (data
not shown). Proper conidiophore differentiation (24.0 ± 4.8%) was
restored in the complemented atg8Δ strain.
Conidiation defects upon loss of autophagy have recently been
reported in Magnaporthe.
26,27
We performed a detailed quantifica-
tion and confirmed that compared to the wild type (98.7 ± 10.9

x 10
2
conidia/cm
2
), the atg8Δ (1.4 ± 0.03 x 10
2
conidia/cm
2
)
showed severely reduced (~98%) conidiation and that this defect
was suppressed in the complemented atg8Δ strain (97.5 ± 9.1 x 10
2

conidia/cm
2
). Furthermore, viability of the atg8Δ conidia (51.2 ±
1.2% cfu) was significantly lower than the wild-type conidia (77.5
± 2.5% cfu). To assess the overall morphology of the resultant struc-
tures, we stained the conidiating cultures of wild type and atg8Δ
strains with calcofluor white at various time points during asexual
growth. As shown in Figure 1C (lower), the atg8Δ conidiophores and
conidia were heterogeneous with a vast majority remaining undif-
ferentiated and/or showing abnormal morphology. Two kinds of
abnormalities were easily discernible: failure to produce and elongate
the conidiophore stalk and cessation of growth at the conidium initi-
ation stage (Fig. 1C). Such persistent defects in the atg8Δ (Fig. 1C)
were likely due to gross abnormalities observed in all the preceding
stages of conidiation (aerial hyphae, aerial hyphal tips, stalks, conid-
iophores). Normal and mature conidia were rare in the atg8Δ strain
(Fig. 1C, last) and in a majority of the cases only a single aberrant

conidium was produced per conidiophore (Fig. 1C). At the corre-
sponding time points, wild-type hyphae differentiated proper aerial
structures that developed appropriately to initiate conidiophores to
finally produce the tricellular conidia in a sympodial manner (Fig.
1C). Taken together, these results reaffirm that Magnaporthe atg8Δ
is defective for autophagy, and establish that Atg8 (and by inference
autophagy) plays a critical role during aerial hyphal development and
conidia formation in Magnaporthe.
Post-translational processing and Atg8p targeting to autophago-
somes in Magnaporthe. In Atg8p/LC3/GATE-16, a specific cleavage
by Atg4 protease exposes the carboxyl-terminal glycine residue (Fig.
2A) that is essential for a ubiquitylation-like conjugation reaction
22

and subsequent amidation to PE. Such a glycine residue was found
to be well conserved in Magnaporthe Atg8 (Fig. 2A). To assess post-
translational processing and subcellular localization of Atg8p, we
generated Magnaporthe strains expressing RFP-Atg8 or Atg8-RFP
fusion proteins under native regulation and as the sole copy of ATG8
in each instance. These two strains were cultured in liquid complete
medium (CM or MM + N) for 2 days, and then inoculated into
liquid MM - N (deprived of nitrogen) to induce autophagy in the
presence of PMSF. During growth in CM, the RFP signals were
undetectable or weak in RFP-ATG8 or ATG8-RFP strains (Fig. 2B).
However, upon growth in MM - N for about 16 hours, predomi-
nant RFP-Atg8 puncta or rings representing autophagosomes could
be easily observed. Such punctate structures were absent in the
ATG8-RFP strain, even under starvation conditions. Instead, faint
RFP signals were seen distributed uniformly in the cytosol (Fig. 2B)
possibly representing the 7 amino acid peptide (DLFEEVE, resulting

©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
36 Autophagy 2009; Vol. 5 Issue 1
predominantly localizes to autophagosomes
and autophagic vacuoles during important
steps of asexual development and further
construe that autophagy is naturally induced
during Magnaporthe conidiation.
Suppression of conidiation defects in
atg8Δ by alternate carbohydrate sources.
Rather serendipitously, we found that the
atg8Δ gained a fluffy appearance, suggesting
restoration of aerial hyphal growth, when
supplemented with glucose or sucrose as
a carbon source. In experiments described
earlier (Figs. 1B and S1C), loss of aerial
growth in the atg8Δ was highly pronounced
on media containing lactose as the sole source
of carbon. Therefore, we tested the conidi-
ation capability of atg8Δ in the presence of
a variety of alternate carbohydrate sources
mostly sugars. Interestingly, the addition
of sucrose or glucose significantly restored
conidiation in atg8Δ even in the presence
of the nonrepressible disaccharide, lactose
(p = 0.001; Fig. 3A). Although profuse
(Fig. 3B) the suppression of atg8Δ conidi-
ation was lower in glucose (about 64 x 10
2

conidia/cm
2
; Fig. 3A) when compared to that
induced by sucrose (average 99 x 10
2
conidia/
cm
2
; p < 0.01). Not surprisingly, galactose,
which is not utilized well by Magnaporthe,
failed to cause a significant repression of
conidiation defects in atg8Δ mutant (Fig.
3A). We conclude that the availability of a
readily utilizable sugar source such as glucose
or sucrose circumvents the requirement of
autophagy during asexual development in
Magnaporthe. Furthermore, we infer that
the nutrient status and proper regulation
of carbohydrate metabolism are important
during conidiation and influence both the
quantity and quality of conidia production in
Magnaporthe.
Gph1 is involved in glycogen metabo-
lism during Magnaporthe conidiogenesis.
To identify proteins that are regulated by
autophagic degradation during conidiogen-
esis, we performed SDS-PAGE fractionation
of total protein extracts from four-day old
conidiating cultures of the wild type, the
atg8Δ and the complemented strain. Mass

spectrometry was then used to identify the
proteins that showed differential accumula-
tion (Fig. 4A). Among these was a glycogen
phosphorylase of 98 kDa, encoded by
MGG_01819, an ortholog of yeast GPH1.
41

In S. cerevisiae, GPH1 is transcriptionally
induced at late exponential growth phase,
concomitant with the onset of intracellular
Figure 2. Posttranslational modification and subcellular localization of RFP-tagged Atg8p. (A)
ClustalW
57
assisted sequence comparison between Magnaporthe Atg8 protein and its orthologs.
MgAtg8 protein (XM_368182) was compared with GATE-16 from Mus musculus (NM_026693),
ScAtg8p from S. cerevisiae (YBL078C), GABARAP from Homo sapiens (NM_007278), and HsLC3
from Homo sapiens (BC041874). Residues that are conserved or similar in at least three out of the five
sequences are boxed in black or in gray respectively. Arrowhead indicates the site of intramolecular
cleavage to expose the conserved glycine residue for lipidation with PE. (B) ATG8-RFP or RFP-ATG8
strain cultured in Complete medium for 2–3 days was subjected to nitrogen starvation for 16 h and
imaged using laser-scanning confocal microscope to detect the RFP signals. Scale Bar = 10 μm. (C)
Total protein lysates from the indicated strains were fractionated using SDS-PAGE and analyzed by
immunoblotting with anti-RFP antibodies (upper). The immunoblot was subsequently reprobed with
anti-Porin antibody to serve as a loading control (lower). Asterisk indicates the predicted DLFEEVE-RFP
peptide resulting from the proteolytic cleavage at glycine 116. CM refers to nitrogen replete whereas
MM - N to nitrogen limiting growth conditions. (D) Subcellular localization of RFP-Atg8p during conidi-
ation. Conidiating cultures of the RFP-ATG8 strain were stained with calcofluor white and analyzed by
epifluorescence microscopy to detect the RFP signal at different stages of asexual development (aerial
hyphae, conidiophore and conidia). Bar = 5 μm. (E) Wild-type strain was stained with Lysotracker
Green DND-26 to visualize acidified autophagic vacuoles at the same developmental stages as shown

in (D). Bar = 5 μm.
©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
www.landesbioscience.com Autophagy 37
Fig. 5A). However, exogenous G1P was unable to suppress the conid-
iation defects in the atg8Δ strain (Fig. 5A). Lastly, conidia production
in a GPH1 OP-2 strain (which indirectly simulates the increased
levels of Gph1 observed in the atg8Δ) was 42.4 ± 1.5 (x100/cm
2
)
showing a nearly 60% reduction compared to the wild type (Fig.
5A). This suggests that the decrease in conidia formation in the atg8Δ
mutant is partly due to the increased Gph1 levels observed therein.
The reduced conidiation defect in the GPH1 OP-2 strain could be
significantly suppressed with the addition of G6P or glucose (Fig.
5B; p < 0.001) to the growth medium. We conclude that the loss of
Gph1 function, either through deletion of GPH1 or by inhibition
of the enzyme activity using G6P, partially represses the conidiation
defects associated with the loss of autophagy in Magnaporthe. Taken
together, we concur that glycogen homeostasis, likely mediated
through Gph1 and autophagy, plays an important role during asexual
development in Magnaporthe.
Glycogen homeostasis is regulated by autophagy and Gph1
during conidiation in Magnaporthe. Since gph1Δ atg8Δ strain
showed a partial suppression of conidiation defects, and Gph1 is
indicated to be involved in glycogen catabolism, we performed
quantitative (Fig. 5C) and semi-quantitative assessment (Fig. S7)
of the steady-state levels of glycogen in the conidiating colonies of
the wild-type, atg8Δ, gph1Δ, gph1Δ atg8Δ or GPH1 OP-2 strain on
PA medium (lactose). We estimated glycogen at three critical time

points: 0d (immediately before photo-induction), 2d (early stage)
and 4d (late stage) during the conidiation cycle in Magnaporthe
(Figs. 5C and S7). The wild type showed a steady increase in
glycogen as conidiation proceeded (Figs. 5C and S7; p = 0.001).
However, compared to the wild type at 2d (1.2 ± 0.01%; Fig. 5C),
the atg8Δ showed strikingly high (2.32 ± 0.25%; p = 0.001) glycogen
levels, which were significantly reduced upon loss of GPH1 func-
tion (gph1Δatg8Δ; 1.23 ± 0.01%; p = 0.001). Glycogen level in the
gph1Δ, was comparatively lower at 0d (0.63 ± 0.01% versus 0.99
± 0.08% in wild type) but increased steadily at the early (1.26 ±
0.11%; p < 0.01) and late stage (2.2 ± 0.3%) of conidiation. Prior
to conidiation, however, glycogen levels were significantly high and
low (respectively; p < 0.001) in the gph1Δ atg8Δ and the gph1Δ,
when compared to the wild type or the atg8Δ strain (Fig. 5C). At the
late stage of conidiogenesis, the atg8Δ and the gph1Δ atg8Δ mutant
showed a decrease in glycogen (Fig. 5C) compared to the early stage.
Overproduction of Gph1 (GPH1 OP-2) caused only a marginal
increase in glycogen at 4d into the conidiation cycle (1.56 ± 0.07%;
Fig. 5C). Taken together, we infer that glycogen accumulation and
breakdown occurs during conidia formation in Magnaporthe and is
tightly regulated by autophagy and Gph1. We further construe that
glycogen catabolism at the onset of conidiation, likely regulated by
autophagy and Gph1, is required for proper asexual development in
Magnaporthe.
Glycogen autophagy and Magnaporthe pathogenesis. An earlier
study showed that Magnaporthe atg8Δ strain is incapable of host
penetration and is completely non-pathogenic.
26
Therefore, we
tested whether loss of Gph1 also suppresses the pathogenicity defects

associated with atg8Δ. The gph1Δ atg8Δ double mutant appeared
to be incapable of infecting the host through the appressorial route
(Fig. 6A). Microscopic analysis revealed that the gph1Δ atg8Δ mutant
failed to produce any invasive hyphae at 48 hpi (Fig. 6B) or later
stages postinoculation. The gph1Δ mutant did not differ significantly
glycogen accumulation.
41,42
Gph1 acts in the cytoplasm to release
G1P from glycogen.
We created a gph1Δ strain in Magnaporthe by replacing
MGG_01819 (NCBI accession XP_363893) with the bialaphos-
resistance marker (Bar) in the wild type, as well as in the atg8Δ
background. Correct gene replacements were confirmed by Southern
blotting (Fig. S5). Loss of GPH1 did not affect vegetative growth,
asexual development or pathogenesis in Magnaporthe. Deletion of
GPH1 in atg8Δ strain did not alter the colony morphology (Fig.
4B). However, the gph1Δ atg8Δ strain showed a partial but signifi-
cant restoration of conidiation (Fig. 4C; p = 0.001). The overall
conidiation of gph1Δ atg8Δ mutant was quantified to be 17.0 ± 0.8
[conidia (x100/cm
2
)], which was a significant increase compared to
atg8Δ mutant. However, there was no significant increase in conidial
viability in the gph1Δ atg8Δ compared to the atg8Δ mutant. At
12–30 hpi, calcofluor white staining of conidiating gph1Δ atg8Δ
cultures, revealed abundant healthy aerial hyphae, some already with
the characteristic swollen tips indicating the induction of conidio-
genesis. Subsequently, mature conidia were also observed (Fig. 4D)
together with multiple conidia growing sympodially at the tips of the
stalks. Compared with atg8Δ strain (Fig. 1C for comparison), the

gph1Δ atg8Δ produced normal aerial hyphae and less abnormal tips,
suggesting a likely reduction and a partial suppression of anomalies
associated with the loss of autophagy during asexual development in
Magnaporthe.
To further confirm the direct involvement of Gph1p-regulated
glycogen metabolism in Magnaporthe conidiogenesis, we assessed
the effect of G6P and G1P on conidiation in atg8Δ strain. G6P is an
efficient inhibitor of the active form of Gph1.
41
The addition of 0.5
mM G6P to the growth media caused a remarkable suppression of
conidiation defects in the atg8Δ strain (53.2 ± 4.4 x10
2
conidia/cm
2
,
Figure 3. Suppression of conidiation defects by alternate carbon sources in
atg8 deletion mutant. (A) Bar chart depicting quantitatively assessed conidia-
tion in the atg8Δ grown on prune-agar medium containing 0.5% lactose and
the indicated sugar (final concentration 1%). Mean values (±SE) presented
as percentage points were derived from three independent experiments (n
= 30 colonies for each sample). Assessments were performed 4 days post
induction. (B) Photomicrographs depicting the extent of conidia formation in
atg8Δ in the presence of the indicated carbon source(s). Images were taken
4 days post induction of conidiation. Scale bar = 10 micron.
©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
38 Autophagy 2009; Vol. 5 Issue 1
from the wild type in terms of its capacity
to breach the host surface and cause disease

(Fig. 6A and B). Interestingly, and unlike
the atg8Δ, the gph1Δ atg8Δ strain elicited
a slight induction of hypersensitive reaction
(HR; inferred from tiny brown speckles
at the infection site) in the host (Fig. 6A)
suggestive of at least some successful host
penetration attempts by the gph1Δ atg8Δ
strain, although further spread and infection
was not successful. Aniline blue staining
was therefore carried out to detect papillary
callose deposits in order to quantify the host
penetration events by atg8Δ or gph1Δ atg8Δ
strains in comparison to the wild type. As
depicted in Figure 6C, the ability to form
penetration pegs on barley leaves was about
77% (±5.6; n = 450) in the wild-type strain
at the 48 hr time-point, whereas it was 2.6%
(±0.3; n = 300) in the atg8Δ mutant. Even
upon extended incubation (72 h), a vast
majority of the atg8Δ appressoria (~95%)
failed to penetrate the host surface and elicit
callose deposits. The gph1Δ atg8Δ strain
showed only a marginal increase in the level
of papillary callose deposits (Fig. 6C) to
13% (±1.5) (n = 300; p = 0.01) respectively.
We conclude that the glycogen catabolism
function of autophagy plays only a minor
role during the host penetration as well as
the in planta growth stages of Magnaporthe
pathogenesis.

Discussion
Autophagy is essential for survival during
nutrient starvation and represents a cellular
degradation mechanism that uses vacuolar
proteases for turnover of certain proteins,
organelles and membranes.
43,44
In this
study, through gene deletion analysis and
genetic complementation, we confirmed that
loss of ATG8 leads to a loss of autophagy
and results in a huge reduction in asexual
sporulation in Magnaporthe. The primary
conidiation defect in Magnaporthe atg8Δ
mutant is the failure to produce sufficient
aerial hyphae and a subsequent inability
to differentiate into proper conidiophores.
The present study adds to the limited
but growing list of Magnaporthe mutants
that show reduced and aberrant conidia-
tion.
6,7,26,27
Apart from an essential role
in sexual sporulation in yeast,
45
autophagy
has recently been implicated in conidiation
and conidial germination in Aspergillus and
Magnaporthe.
26-28

Our results add to these
findings and suggest a functional role for
Figure 4. Identification and functional analysis of Gph1 in Magnaporthe. (A) Identification of proteins
differentially regulated in the atg8Δ mutant. Silver-stained profile of the SDS-PAGE gel depicting total
proteins extracted from conidiating mycelial cultures from wild type (WT), atg8Δ or the atg8Δ comple-
mented strain. Proteins differentially regulated and identified by mass spectrometry have been listed
together with their predicted molecular masses. Western blotting with anti-Porin antisera served as a
loading control. Mw refers to the molecular mass markers in kilodalton. (B) Colony and growth char-
acteristics of the wild type, atg8Δ, gph1Δ and gph1Δ atg8Δ double mutant. Colonies of the indicated
strains were grown on prune agar medium with lactose as the carbon source for 5 days in the dark and
photograghed. Scale bar = 1 cm. (C) Bar chart depicting quantification of conidiation in the wild type
(WT), atg8Δ, gph1Δ or gph1Δ atg8Δ strain grown on prune-agar medium containing 0.5% lactose as
the sole source of carbohydrate. Mean values (±SE) presented as percentage points were derived from
three independent experiments (n = 30 colonies for each strain). Assessments were performed 4 days
post induction. (D) Partial restoration (qualitative and quantitative) of conidiation in the atg8Δ mutant
upon loss of GPH1 function. Calcoflour-white stained gph1Δatg8Δ strain was analyzed by epifluores-
cence microscopy to assess the indicated developmental stages during conidiation. Lower panels depict
the representative incidence of aberrant conidiation-specific defects that still persist in the mutant. The
experiment was repeated three times using a total of 10 colonies. Bar = 10 μm.
©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
www.landesbioscience.com Autophagy 39
glucose) conditions, which efficiently suppress autophagy. However,
the essential requirement for autophagic cell death during appres-
sorium-mediated entry of Magnaporthe
26
into the host could not
be bypassed by the use of alternate carbon source(s). We observed
that atg8Δ conidia harvested from sucrose or glucose-containing
medium were unable to penetrate the host leaf surface and could not

proliferate as well as the wild type even when introduced through
wounded host tissue (data not shown). Overall, these results point to
an important role for autophagy in Magnaporthe response to nutri-
tional stress, and a possible connection between carbon homeostasis
and autophagy during conidiation as well as in invasive growth inside
the host tissue. Our data also suggest a key role for autophagy in
the breakdown of carbohydrate stores at the onset of Magnaporthe
conidiogenesis in order to produce sufficient intracellular glucose.
This mirrors a well-defined role for glycogen autophagy in newborn
mammals in the breakdown of intracellular glycogen reserves as a
strategy to cope with a sudden demand for ample energy substrates
to confront metabolic requirements.
34,35
In yeast, there are two distinct intracellular pools of glycogen:
cytosolic and vacuolar. Gph1 catalyzes the release of G1P from
glycogen and mediates glycogen breakdown in the cytosol.
41

However, during sexual sporulation, glycogen degradation releases
glucose and is catalyzed by a vacuolar glucoamylase.
42
In S. cerevisiae,
inhibition of the vacuolar glycogen degradation, prevents the onset of
sporulation,
31,46-48
indicating that autophagy likely serves to deliver
cytoplasmic glycogen into vacuoles for degradation, to produce suffi-
cient energy and/or appropriate intermediates (e.g., G6P or glucose)
for cellular differentiation.
Loss of autophagy led to increased accumulation of Gph1, and

rather surprisingly a significantly high steady-state level of glycogen
at the early stages of asexual differentiation in Magnaporthe. Such
concomitant increase in glycogen as well as Gph1 in atg8Δ is likely
due to growth under glucose-limiting conditions, which is known to
upregulate genes required for glycogen biosynthesis and breakdown
in yeast.
41
Deletion of GPH1 in atg8Δ mutant reduced the glycogen
levels to those observed in the wild type during early stages of
conidiation. However, glycogen levels are higher prior to induction
of conidiation in the gph1Δ atg8Δ mutant. We infer that glycogen
levels are in a state of flux during conidiation and speculate that a
strict temporal and spatial regulation of glycogen homeostasis, medi-
ated by Gph1 and autophagy, is critical for asexual differentiation in
Magnaporthe. Future studies would certainly aim at distinguishing
between the cytoplasmic and vacuolar stores of glycogen, and to
uncover the mechanisms that regulate glycogen metabolism during
asexual development in Magnaporthe.
The gph1Δ atg8Δ mutant showed an overall improvement in
the morphology of conidiation-specific structures suggesting that
sufficient glucose produced by glycogen autophagy is likely necessary
for proper growth and differentiation during asexual development.
Furthermore, G6P-based suppression of atg8Δ defects was stronger
than that achieved through deletion of GPH1, suggesting that G6P
is likely involved in regulation of other relevant processes such
as glycolysis and gluconeogenesis, which might be important in
regulating proper carbon homeostasis during asexual development
in Magnaporthe. However, deletion of GPH1 in atg8Δ background
did not increase the number of conidiophores per se, implying that
factor(s) other than carbohydrate metabolism are also necessary

autophagy in the recycling of nutrient resources, in particular carbon
(this study), nitrogen
32
and metal ions,
32
for extensive cellular
remodeling required during asexual differentiation in filamentous
fungi. Autophagy thus attains importance during nutrient-limiting
condition, which is a well-established trigger for asexual sporulation
in several fungi including Magnaporthe. Further studies are needed
to dissect the mechanistic regulation of conidiation, and to elucidate
the exact physiological role of autophagy in the process of asexual
development in Magnaporthe.
A key finding was the remarkable suppression of atg8Δ conidia-
tion defects by external supplementation of carbohydrates (such as
sucrose or glucose). This suggests that autophagy becomes essential
for aerial hyphal growth and conidiophore differentiation during
carbon-limiting conditions as well as in the presence of some nonre-
pressible sugars. Alternately, autophagy could be dispensable for
conidiation during growth under carbohydrate-replete (exogenous
Figure 5. Functional requirement of glycogen catabolism during Magnaporthe
conidiation. (A) Bar chart representing the quantitative analysis of conidia-
tion in the wild type, atg8Δ (treated with G6P or G1P) or the GPH1 OP-2
strains grown on PA medium. Results (mean values ± SE) represent three
independent experiments involving a total of 20 colonies per strain. (B)
Quantitative analysis of conidiation in the wild type (gray bar) or the GPH1
OP-2 strain (black bar) grown on PA medium (control) or on PA medium
supplemented with G6P or glucose. Results (mean values ± SE) represent
three independent experiments involving a total of 20 colonies per strain.
(C) Quantitative analysis of total internal glycogen concentration in the wild

type, atg8Δ, gph1Δ, gph1Δ atg8Δ and GPH1 OP-2 strains. The indicated
strains were initially grown in the dark for 2 days and then subjected to
constant illumination to induce conidiation. At the specified time-points post
induction, the fungal biomass was subjected to glycogen extraction and
estimation. The concentration of internal glycogen was presented as percent-
age of the total wet weight of the fungal biomass. Mean values (±SE) were
derived from three independent experiments (n = 15 colonies per strain).
©2009 Landes Bioscience. Do not distribute.
Glycogen autophagy and Magnaporthe development
40 Autophagy 2009; Vol. 5 Issue 1
suspension were inoculated on barley leaf explants and incubated
under humid conditions at 23°C for up to 96 hours.
52
Methanol
treatment and staining with 0.1% acid fuchsin (Sigma-Aldrich,
USA) was carried out as described
53
for observation of penetration
and infection hyphae.
Nucleic acid and protein-related manipulation. Standard molec-
ular manipulations were performed as described.
54
Fungal genomic
DNA was extracted using a modified potassium acetate method.
55

Plasmid DNA was isolated with QiaPrep plasmid miniprep kit
(Qiagen, Valencia, California) and nucleotide sequencing performed
using the ABI Prism Big Dye terminator method (PE-Applied
Biosystems, California). Homology searches of DNA/protein

sequences were performed using the BLAST programs
56
and multiple
sequence alignments carried out with ClustalW
57
and Boxshade
( /interfaces/boxshade.html). Total
RNA was isolated with RNeasy Plant Mini kit (QIAGEN, USA),
and cDNA synthesis conducted using AMV Reverse Transcriptase
(Roche Diagnostics, Germany). Following primers were used to
amplify the 5' and 3' UTR of the ATG8 gene: ATG8-5F (5'-GAG
AGT GAA CTC GAG GCT ATA ACC TGA GGG TAG-3'),
ATG8-5R (5'-GAG AGT GAG GAT CCC GGT TGA TTG AGA
CTT GT-3'), ATG8-3F (5'-GAG AGT GTT CTG CAG CGA GTG
AGC TTG CTC ACC-3'), and ATG8-3R (5'-GAG AGT GTT AAG
CTT CAC GTC CTC CCA-3'). The full-length genomic copy of
for Magnaporthe conidiogenesis, and are probably subjected to
autophagy-dependent regulation. Future experiments will focus
on the identification of such positive regulators as well as their
functional relationship with the autophagy pathway.
We established RFP-Atg8p-PE as a reliable marker for autophago-
somes and autophagic vacuoles in aerial hyphae and conidiophores in
Magnaporthe. As expected, RFP-Atg8 required the critical posttrans-
lational modifications (predicted endoproteolytic cleavage most likely
at G116 and lipidation to PE) prior to its recruitment to preautopha-
gosomal structures. Very little (if any) RFP-Atg8p was discernible in
vegetative mycelia when the strain was grown in the dark, whereas
RFP-Atg8p-PE was clearly visible in autophagosomes in the aerial
hyphae. Whether RFP-Atg8p expressing aerial hyphae subsequently
differentiate into conidiophores is an important question and is pres-

ently hindered by a lack of suitable conidiophore-specific markers.
We observed that autophagy is induced naturally during conidio-
genesis and that autophagosomes are distributed prominently in
conidiation-related structures such as aerial hyphae, conidiophores
and conidia. Recent studies on ATG8
26
and ATG1 function
27
show
that autophagy plays a key role in cell death, lipid mobilization, and
turgor generation during the pathogenic phase in Magnaporthe.
Future studies would certainly aim at identifying the specific targets
of the autophagosomal-vacuolar degradation machinery in asexual
and pathogenic growth phases of Magnaporthe.
Materials and Methods
Fungal strains and growth conditions. Magnaporthe wild-type
strain B157 (Field isolate, mat1-2) was obtained from the Directorate
of Rice Research (Hyderabad, India). Magnaporthe strains were
propagated on Prune-agar (PA) medium or complete medium (CM)
as described.
49,50
Carbohydrate-supplemented Prune-agar medium
for assessing conidiation in atg8Δ contained lactose (5 g/L) and
one of the following sugars: sucrose, glucose, lactose, galactose or
maltose at 10 g/L. The composition of MM and MM - N (used for
nitrogen starvation) was as reported earlier.
51
Two-day old liquid
CM-grown mycelia were ground in liquid nitrogen for the isolation
of nucleic acids. To assess the growth and colony characteristics,

Magnaporthe isolates were cultivated on CM agar or PA medium,
at 28°C for one week. Mycelia used for total protein extraction were
obtained by growing the relevant strains in liquid CM for 2–3 days,
with gentle shaking, followed by inoculation in MM or MM - N
for about 16 hours.
For quantitative analysis of conidiation, colonies were culti-
vated on PA medium in the dark for 2 days, followed by a 4-day
growth cycle under constant illumination at room temperature.
The surface of the colonies was then scraped with inoculation loops
in the presence of water and the fungal biomass was harvested in
Falcon
TM
Conical Tubes (BD Biosciences, San Jose, California). The
suspension was vortexed thoroughly to ensure maximum detach-
ment of conidia from mycelia, and then filtered through two layers
of Miracloth (Calbiochem, San Diego, California). Conidia thus
collected were washed twice with and finally resuspended in sterile
water. Conidia production in a given colony was quantified using a
hemocytometer and reported as the total number of conidia present
per unit area of the colony [conidia (x100/cm
2
)]. To test the pathoge-
nicity, conidia harvested from 6-day-old cultures were resuspended to
1 x 10
5
conidia per mL in sterile water. Droplets (20 μl) of conidial
Figure 6. Glycogen catabolism and Magnaporthe pathogenesis. (A) Host
penetration is partially restored in the gph1Δ atg8Δ mutant. Barley leaf
explants were inoculated with conidia from the wild type, gph1Δ atg8Δ or
gph1Δ atg8Δ Magnaporthe strains and disease symptoms assessed after 5

days. (B) The gph1Δ atg8Δ mutant is incapable of invasive growth in the
host. Equal number of conidia from the gph1Δ or the gph1Δ atg8Δ were
inoculated on barley leaf explants and allowed to proceed through infection-
related development for 48 hours. Asterisk denotes the infection hyphae in
planta. Scale bar represents 10 μm. (C) Partial restoration of host penetration
in the gph1Δ atg8Δ mutant. Bar chart representing papillary callose deposits
(as percentage of total appressoria) in barley leaf explants inoculated with
conidia from the indicated strains grown on PA medium. Quantification
was performed 48 hours post inoculation and represents mean ± SE values
derived from three independent experiments.