Genes involved in protein glycosylation determine the activity and cell

internalization of the antifungal peptide PAF26 in Saccharomyces

cerevisiae
Eleonora Harries
a,b
, Lourdes Carmona
a
, Alberto Muoz
b,1
, Jos I. Ibeas
c
, Nick D. Read
b,1
,
Mnica Ganda
a
, Jose F. Marcos
a,
a
Food Science Department, Instituto de Agroqumica y Tecnologa de Alimentos (IATA), Consejo Superior de Investigaciones Cientcas (CSIC), Avda Agustn Escardino 7, 46980
Paterna, Valencia, Spain
b
Fungal Cell Biology Group, Institute of Cell Biology, University of Edinburgh, Rutherford Building, Edinburgh EH9 3JH, UK
c
Centro Andaluz de Biologa del Desarrollo, Universidad Pablo de Olavide-Consejo Superior de Investigaciones Cientcas (CSIC), Carretera de Utrera km. 1, 41013 Sevilla, Spain
a r t i c l e i n f o
Article history:
Received 21 May 2013
Accepted 2 August 2013
Available online 11 August 2013
Keywords:
Antimicrobial peptide
Histatins
Melittin
Cell wall
Protein glycosylation
Cell-penetrating peptide
a b s t r a c t
We have previously characterized the synthetic hexapeptide PAF26 as a cell-penetrating and non-lytic
antifungal peptide that is active against Saccharomyces cerevisiae and lamentous fungi. Numerous cell
wall (CW) proteins are glycosylated in fungi and many of these play important roles in fungal pathogen-
esis. In this study, we screened a collection of S. cerevisiae deletion mutants for protein glycosylation
genes whose deletion altered the sensitivity to PAF26. Increased tolerance to PAF26 was observed in
mutants with the following disrupted genes: PMT1-6, EOS1, ALG5, MNN1, MNN4 and MNN5. Signicantly,
genes coding for protein O-mannosyltransferase 2 (Pmt2p), which is responsible for the addition of the
rst mannosyl residue of O-linked carbohydrates, and for Eos1p, an enzyme involved in N-linked glyco-
sylation of proteins, showed resistance to PAF26 and defects in CW integrity. Microscopic studies on the
S. cerevisiae Deos1 deletion mutant demonstrated a blockage of peptide internalization by cells. Protop-
lasts lacking CWs interacted with the peptide, but were more resistant to peptide killing than cells pos-
sessing CWs due to a blockage in PAF26 internalization. Interestingly, protoplasts obtained from Deos1
behaved similarly to those of the parental strain. Collectively, these observations demonstrate that the
CW is a positive factor that determines the internalization of the PAF26, and that Eos1p exerts its activity
through the glycosylation of specic protein(s) involved in peptide internalization.
2013 Elsevier Inc. All rights reserved.
1. Introduction
Antimicrobial peptides (AMPs) have been isolated from a vast
number of organisms, including bacteria, insects, plants and hu-
mans (Zasloff, 2002), and are considered a potential source of alter-
native compounds for the control of pathogenic microorganisms
(Hancock and Sahl, 2006; Marcos et al., 2008; Peters et al., 2010;
Duncan and ONeil, 2013). Peptides with antifungal activity have
not received as much attention as antibacterial peptides, probably
as consequence of the impact of bacterial infections on human
health. However, a number of AMPs have demonstrated selective
and promising activities against fungal pathogens. A detailed
understanding of the antimicrobial mechanism is of high priority
if peptides are to be considered as useful antimicrobial agents. It
also might facilitate the discovery of novel targets in the fungal cell
for antifungal therapy. Properties common to AMPs and small anti-
microbial proteins are: direct antimicrobial activity, abundance of
cationic and hydrophobic residues, and amphipathic conforma-
tions. In the past, the killing activity of cationic and amphipathic
peptides was considered to be mainly due to disruption of the plas-
ma membrane causing cell lysis. However, more recent evidence
clearly demonstrated that some AMPs are able to penetrate cells
and exert their activity via intracellular targets (Yeaman and You-
nt, 2003; Brogden, 2005; Henriques et al., 2006; Marcos and Gand-
a, 2009).
PAF26 is a synthetic cationic antifungal hexapeptide that was
identied by a combinatorial approach (Lpez-Garca et al.,
2002), and has been proposed as a suitable model to study the
mechanism of action of short, cell-penetrating antifungal peptides
(Muoz et al., 2013a). This peptide belongs to the class of cationic,
tryptophan-rich AMPs that includes peptides of natural origin such
as indolicidin and tritrpticin, peptides derived from proteins such
as lactoferricins, and de novo designed peptides of which PAF26
1087-1845/$ - see front matter 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.fgb.2013.08.004

Corresponding author. Fax: +34 963 636 301.

E-mail address: [email protected] (J.F. Marcos).
1
Current address: Manchester Fungal Infection Group, Institute of Inammation
and Repair, University of Manchester, Core Technology Facility, Grafton Street,
Manchester M13 9NT, UK.
Fungal Genetics and Biology 58-59 (2013) 105115
Contents lists available at ScienceDirect
Fungal Genetics and Biology
j our nal homepage: www. el sevi er . com/ l ocat e/ yf gbi
is an example. PAF26 is selective against lamentous fungi having
in vitro inhibitory activity in the low micromolar range, but it is
also active against the budding yeast at higher concentrations. In
different studies with Saccharomyces cerevisiae, Penicillium digita-
tum, Aspergillus fumigatus and Neurospora crassa, it has been shown
that PAF26 rst interacts with the outer layers of the fungal cell, it
is then translocated to the cell interior where it nally exerts its
intracellular killing mechanism (Muoz et al., 2006, 2012, 2013b).
A transcriptomic study conducted with S. cerevisiae demon-
strated that genes encoding proteins that are involved in reinforce-
ment of the cell wall (CW) participate in the response to
sub-inhibitory concentrations of PAF26 (Lpez-Garca et al.,
2010). The yeast CW is an essential and dynamic structure and
may account for up to one third of the dry weight of the cell. Major
components of the yeast CW are glycoproteins (mannoproteins),
b-1,3-linked glucans, branched b-1,6-glucans and chitin (Klis
et al., 2006). On the outer layer of the CW there are glycoproteins
that are extensively O- and N-mannosylated. Protein glycosylation
is the most universal and structurally diverse form of post-transla-
tional modication (Deshpande et al., 2008). It occurs by the
attachment of a glycan to a protein either at an asparagine residue
(N-glycosylation), or at hydroxylysine, hydroxyproline, serine, or
threonine residues (O-glycosylation). Protein glycosylation plays
diverse biological roles in the secretion, localization and function
of many proteins, as well as in CW integrity, and has been impli-
cated as being important in the morphogenesis and virulence of
fungal pathogens (Girrbach and Strahl, 2003; Munro et al., 2005;
Prill et al., 2005; Schirawski et al., 2005; Zhou et al., 2007; Fernn-
dez-lvarez et al., 2009). In addition, several protein glycosylation
genes have been shown to determine the sensitivity of fungi to
antifungal peptides and proteins (Lussier et al., 1995; Gentzsch
and Tanner, 1996; Kimura et al., 1999; Ibeas et al., 2000; Koo
et al., 2004; Bai et al., 2006; Harris et al., 2009; Szafranski-Schnei-
der et al., 2012).
Following these previous observations, we screened the public
collection of S. cerevisiae deletion strains for protein glycosylation
genes encoding proteins that alter the sensitivity to PAF26, in order
to gain further insights into the antifungal mechanism of this cell-
penetrating peptide. We demonstrated an increased tolerance to
PAF26 in mutants with the disrupted protein glycosylation genes
PMT1-6, EOS1, ALG5, MNN1, MNN4 and MNN5, revealing the impor-
tance of glycosylation pathways for the antifungal action of this
peptide. Evidence is presented that PAF26 requires the presence
of a functional CW to be internalized by yeast cells to exert its kill-
ing activity, and that Eos1p is particularly involved in PAF26
internalization.
2. Materials and methods
2.1. S. cerevisiae strains and culture conditions
Experiments were performed with S. cerevisiae BY4741 parental
strain (MATa; his3D1; leu2D0; met15D0; ura3D0), corresponding
isogenic deletion strains from the EUROSCARF public collection
(http://web.unifrankfurt.de/fb15/mikro/euroscarf), and additional
strains obtained in this study, which are detailed in Table 1. S. cere-
visiae strains were grown on YPD (yeast peptone dextrose) plates
and liquid media. The yeast cells were grown overnight at 30 C
with shaking at 200 rpm. The culture cell concentration was ad-
justed to an optical density at 600 nm (OD
600
) of 0.2 and grown
at 30 C for 3 h to exponential phase (OD
600
0.4-0.5). Cells were
then collected and processed as described below, depending on
the experiment.
2.2. Peptides
PAF26, P113 (which is a derivative of histatin 5), and cecropin A
were synthesized and provided at >95% purity by Genscript Corpo-
ration (Piscataway, NJ, USA) (see peptide amino acid sequences in
Table 2). Melittin was purchased from SigmaAldrich (Reference
M2272). PAF26 labelled with tetramethyl-rhodamine (TMR-
PAF26) by covalent modication of its N-terminus was also
synthesized. Stock solutions of peptides were prepared in 10 mM
3-(N-morpholino)-propanesulfonic acid (MOPS) pH 7 buffer and
stored at 20 C. TMR-PAF26 was rst dissolved in a small volume
of dimethylsulfoxide (DMSO) and then diluted to 200 lM in
MOPS buffer, to maintain DMSO at a concentration of less than
1% in the stock solution. Peptide concentrations were determined
spectrophotometrically.
2.3. Antimicrobial activity assays
Fungicidal assays with PAF26 were carried out as described pre-
viously (Lpez-Garca et al., 2010). Briey, serial ve-fold dilutions
from exponential phase cells were treated either with 64 lM
PAF26 or with MOPS buffer (as the non-treated control) at 30 C
for 24 h, and then plated onto peptide-free YPD plates to determine
cell viability. Additionally, samples of non-treated cells were pla-
ted onto YPD plates amended with either the cell wall disrupting
uorophore calcouor white (CFW) (SigmaAldrich, F3543) or
the plasma membrane disrupting detergent sodium dodecyl sul-
phate (SDS) (SigmaAldrich, L4509).
S. cerevisiae BY4741 strain and selected deletion mutants Deos1,
Dalg5, Dmnn5 and Dpmt2 were treated with tunicamycin (TM)
(SigmaAldrich, T7765) and/or PAF26 as follows. Exponential
Table 1
S. cerevisiae strains used in this study.
Strain Genotype Source
BY4741 MATa his3D1 leu2D0 met15D0 uraD0 Euroscarf
Y07225 (Deos1) Same as BY4741 eos1D::KanMX4 Euroscarf
Y01065 (Dalg5) Same as BY4741 alg5D::KanMX4 Euroscarf
Y01239 (Dmnn5) Same as BY4741 mnn5D::KanMX4 Euroscarf
Y03792 (Dpmt1) Same as BY4741 pmt1D::KanMX4 Euroscarf
Y00385 (Dpmt2) Same as BY4741 pmt2D::KanMX4 Euroscarf
Y01750 (Darg1) Same as BY4741 arg1D::KanMX4 Euroscarf
SCLC01 (Darg1) Same as BY4741 arg1D:: HphR This study
SCLC02 (Deos1Darg1) Same as Y07225 arg1D:: HphR This study
SCLC03 (Dpmt2Darg1) Same as Y00385 arg1D:: HphR This study
SCLC04 (Dpmt2Dpmt1) Same as Y00385 pmt1D:: HphR This study
Table 2
Antifungal activity of peptides used in this work.
Peptide Amino acid sequence MIC (lM)
BY4741 Deos1 Dpmt2
PAF26 RKKWFW 32 64 64
P113 AKRHHGYKRKFH 128 256 256
Melittin GIGAVLKVLTTGLPALISWIKRKRQQ-NH2 32 32 16
Cecropin A KWKLFKKIEKVGQNIRDGIIKAGPAVAVVGQATQIAK-NH2 32 32 16
106 E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115
phase cells (OD
600
0.40.5) were adjusted to 4 10
5
cells/mL in
20% YPD containing 40 lg/mL chloramphenicol (to avoid bacterial
contamination) and subjected to no treatment (control) or treat-
ment with 4 lg/mL TM at 30 C for 24 h. Subsequently, serial
ve-fold dilutions were prepared in either 20% YPD (i), or 20%
YPD containing 4 lg/mL TM (ii), and subjected to 64 lM PAF26
treatment as described above. After incubation at 30 C for 24 h,
aliquots of 5 lL of each sample were dotted onto YPD agar plates
to determine cell viability. The agar plates were incubated at
30 C for two days to allow colony visualization.
Dose response curves of the effect of AMP on yeast strains were
carried out as described previously (Lpez-Garca et al., 2010). The
viability of cells was determined after treatment for 24 h with the
different peptides (PAF26, P113, cecropin A, and melittin) at con-
centrations ranging from 4 lM to 256 lM.
2.4. Flow cytometry
Quantication of peptide labeling of cells of S. cerevisiae strains
was carried out as described (Lpez-Garca et al., 2010) using TMR-
PAF26 (5 or 30 lM). Analysis was performed using a FACSCANTO
ow cytometer (Beckton Dickinson,USA) equipped with an ar-
gon-ion laser emitting a 488 nm beam at 15 mW. Data (10,000
cells/sample) were analyzed with the DIVA software included in
the ow cytometry system. Three independent experiments were
carried with triplicate samples in each.
2.5. Fluorescence microscopy analyses
Confocal laser scanning microscopy to analyze differences in
the localization patterns of the uorescently-labelled PAF26 pep-
tide in wild type and mutant strains was performed as described
previously (Muoz et al., 2012, 2013b). Yeast cells at 2.5 10
7
-
cells/mL were incubated in sterile water with 2.5 lM TMR-PAF26
for 1 h at 30 C in the dark, and subsequently transferred to and
imaged in 8-well slide culture chambers (Nalge Nunc International,
Rochester, NY). The confocal microscope used was a Radiance 2100
system (Bio-Rad Microscience, Hemel Hempstead, UK) mounted
onto a Nikon TE2000U Eclipse inverted microscope equipped with
blue diode, argon ion and HeNe lasers. Excitation at 543 nm and
RFP lters for emission at 580700 nm were used to visualize
TMR-PAF26 and propidium iodide (PI) uorescence. Simultaneous
brighteld images were captured with a transmitted light detector.
Confocal images were captured using Lasersharp software (v 5.1,
BioRad) and were subsequently processed using ImageJ software
(v. 1.44, MacBiophotonics).
2.6. Protoplast preparation
Protoplasts were obtained as described previously (Petit et al.,
1994) with modications. Yeast cells (1 10
8
cells/mL) were
washed with sterile water, suspended in pretreatment buffer (Cle-
lands reagent, 1 M sorbitol, 60 mM EDTA, 20 mM dithiothreitol)
and incubated for 30 min at 30 C. Next, cells were washed with
1 M sorbitol and protoplasting buffer (1 M sorbitol, 60 mM EDTA,
100 mM sodium citrate) containing 2 mg/mL of enzymatic solution
(Glucanex, SigmaAldrich L1412). Samples were incubated for
90 min at 30 C and then were washed with 1 M sorbitol. Protop-
lasts were suspended in 1 M sorbitol and treated for 1 h with
2.5 lM TMR-PAF26 and visualized by confocal microscopy as de-
scribed above. To determine the viability of protoplasts, samples
were treated rst with 2.5 lM PAF26 for 1 h and then with 2 lg/
mL of the cell death marker PI (Sigma-Aldrich P4864), followed
by confocal microscopy visualization.
3. Results
3.1. Genes involved in protein glycosylation determine the sensitivity
of S. cerevisiae to PAF26
We have used the EUROSCARF collection of S. cerevisiae deletion
mutants to determine the sensitivity of protein glycosylation mu-
tants to PAF26 treatment. A total of 37 mutants were selected, with
23 corresponding to genes annotated as being involved in protein
N-glycosylation, 10 in O-glycosylation, and 4 that participate in
both pathways (Supplemental Table 1 and Fig. 1) (Saccharomyces
genome database, http://www.yeastgenome.org/). They were ana-
lyzed for their sensitivity to PAF26 in a fungicidal assay (Supple-
mental Table 1 and Fig. 2). For those mutants showing a
phenotype of altered sensitivity to PAF26, their sensitivities to
the CW disrupting agent CFW and the plasma membrane disrupt-
ing detergent SDS were also determined. Only minor increased
sensitivity to PAF26 was observed in some strains (i.e., a 1 score
in Supplemental Table 1 indicates a one dilution of difference in
growth). However, several mutants showed a strong increase in
resistance towards PAF26, including: (1) mutants in ALG5 that
codes for the UDP-glucose:dolichyl-phosphate glucosyltransferase
that transfers glucose dolichyl-phosphate (Dol-P) to elongate the
dolichol-linked core oligosaccharide precursor of N-glycans in the
endoplasmic reticulum (ER) (Heesen et al., 1994) (Supplemental
Fig. 1A), (2) MNN1 and MNN5 that encode mannosyltransferases
that decorate N- and O-linked protein glycans with mannose resi-
dues (Rayner and Munro, 1998) (Supplemental Fig. 1B), (3) PMT
genes that encode enzymes that transfer the rst mannose residue
in the rst step of O-linked protein glycosylation in the ER (Supple-
mental Fig. 1C) (Strahl-Bolsinger et al., 1999), and (4) EOS1 which is
a gene involved in N-glycosylation (Nakamura et al., 2007) (also
see below). Fig. 1 shows representative data on the sensitivities
of four of the most signicant of these deletion mutants to
PAF26, CFW and SDS. Signicantly, the Dpmt2 and Deos1 mutants
were resistant to PAF26 although they have a defective CW as
Fig. 1. Sensitivity to PAF26, CFW and SDS of S. cerevisiae deletion mutants defective in N- and O-glycosylation genes. Serial ve-fold dilutions of cells of single deletion
mutants Deos1, Dalg5, Dmnn5, Dpmt2 and the parental strain BY4741 were treated with either buffer (control treatment, A, C and D) or with 64 lM PAF26 (B) for 24 h, and
then plated onto YPD peptide-free plates. Same dilutions of control treatments were applied onto YPD plates amended with CFW (C) or with SDS (D).
E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115 107
demonstrated by their increased sensitivity to CFW and SDS,
respectively.
3.2. The N-glycosylation mutant Deos1 shows tolerance to PAF26 and
Tunicamycin
Tunicamycin (TM) is an antibiotic inhibitor of N-glycosylation
in eukaryotes and is frequently used in the characterization of
deletion strains with defects in protein glycosylation pathways (El-
bein, 1987). It is known that deletion of EOS1 results in tolerance to
TM (Nakamura et al., 2007), an observation conrmed in our assays
of TM treatment (4 lg/mL) (Fig. 2). This is signicantly different to
the near wild-type sensitivity to TM of the other glycosylation mu-
tants analyzed (Fig. 2). Therefore, the sensitivity prole of the
strains analyzed is qualitatively different for PAF26 and TM. The
Deos1 strain was the only one that showed tolerance to both
agents. We therefore questioned whether TM and PAF26 might
have overlapping activities. Cells were treated rst with TM and la-
ter with PAF26 and then plated onto plates free of these antifungal
agents (Fig. 2). At the tested concentrations of either PAF26 or TM,
there was no loss of CFU recovery in the Deos1 mutant; however
when both agents were combined, signicant loss of CFU was ob-
served. Therefore, our data suggest synergistic effects of both anti-
fungal agents acting on the Deos1 mutant, a behaviour also shared
by the other mutants analyzed with deletions of glycosylation-
associated genes (Fig. 2).
3.3. The fungicidal activity of PAF26 requires different cellular
processes
Recently, we presented evidence that arginine metabolismis in-
volved in the fungicidal action of PAF26 (Lpez-Garca et al., 2010;
Carmona et al., 2012). The S. cerevisiae deletion mutant of ARG1,
which encodes the cytosolic enzyme arginino-succinate synthase
involved in arginine metabolism, is also resistant to PAF26
(Lpez-Garca et al., 2010) (see also as control in Fig. 3). We
decided to analyze the inuence on peptide sensitivity of blocking
the arginine and protein glycosylation pathways together. As de-
scribed in Supplemental Fig. 3, we generated the double deletion
mutants of ARG1, with either PMT1 or EOS1. The double deletion
mutants SCLC02 (Deos1-Darg1) and SCLC03 (Dpmt2-Darg1)
showed a weakened growth phenotype. However, none of them
showed an enhancement of resistance to the peptide and displayed
sensitivity levels similar to those of strains carrying a single gene
deletion (Fig. 3). These results indicate that each gene targets dif-
ferent steps of the same killing mechanism.
Similarly, we generated a double mutant of PMT2 and PMT1
(SCLC04) that also had impaired growth (Fig. 3) and an extremely
weakened CW determined by its sensitivity to CW disrupting
agents (data not shown) (Gentzsch and Tanner, 1996). It was also
resistant to PAF26 and showed no increased resistance as com-
pared with the respective single mutants (Fig. 3).
3.4. PMT2 and EOS1 also play roles in the mode of action of another
cell-penetrating antifungal peptide
We investigated whether PMT2 and EOS1 also mediate the sen-
sitivity to other AMPs with different modes of action (Table 2).
P113 is a 12-amino acid fragment from the human histatin 5,
which is also described as a cell-penetrating antifungal peptide
(Jang et al., 2008). Cecropin A is a natural peptide which has bacte-
rial membrane disrupting activity and is also antifungal (Steiner
et al., 1981; Brogden, 2005). Melittin is a non-specic antimicrobial
peptide with a cytolytic mode of action (Terwillinger and Eisen-
berg, 1982).
Doseresponse experiments were carried out for the four pep-
tides against strains BY4741, Deos1 and Dpmt2 (Fig. 4). These as-
says conrmed the increased resistance to PAF26 in both
mutants (Fig. 4A and Table 2). Increased resistance was also ob-
served in the case of P113 although this peptide had lower anti-
fungal activity than PAF26 under our assay conditions, and the
increased resistance of the mutants was not so obvious in the
doseresponse curves (Fig. 4C and Table 2). On the contrary, for
the membrane targeting peptides cecropin A and melittin, the
mutations did not result in increased resistance (Table 2) but
rather in increased sensitivity. This was more evident in the case
of Dpmt2 and melittin (Fig. 4B), probably as consequence of their
weakened CWs. These results provide evidence that PMT2 and
EOS1, besides playing a role in the mechanism of action of PAF26,
Fig. 2. Analysis of sensitivity to PAF26 and tunicamycin of S. cerevisiae deletion mutants. Cells were rst exposed to either buffer (A and B) or tunicamycin (TM) (C and D) for
24 h. Subsequently, serial ve-fold dilutions of cells were either control-treated (A and C) or treated with PAF26 (B and D), as described in Fig. 1, and dotted onto YPD plates.
Fig. 3. Sensitivity to PAF26 of S. cerevisiae single and double deletion mutants in
arginine metabolism and glycosylation pathways. Serial ve-fold dilutions of cells
of Darg1, Deos1 and Deos1Darg1 (arginine metabolism, top panels) and Dpmt1,
Dpmt2, Dpmt2Dpmt1 and Dpmt2Darg1 (protein glycosylation, bottom panels) with
the corresponding parental strain BY4741, were plated onto YPD peptide-free plates
after treatment with control buffer (A) or with 64 lM PAF26 (B) for 24 h.
108 E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115
are also important for the antifungal activity of another cell-pene-
trating antifungal peptide (P113).
3.5. Subcellular location of PAF26 within S. cerevisiae cells
demonstrates an inhibition of peptide internalization in the EOS1
mutant
Flow cytometry has been used to correlate the amount of uo-
rescently labelled peptides that bind yeast cells of selected resis-
tant mutants (Harris et al., 2009; Lpez-Garca et al., 2010). We
used the same approach to study the Deos1 and Dpmt2 mutants
exposed to two different peptide concentrations (see also
Fig. 4A): a sub-inhibitory concentration of 5 lM and an inhibitory
concentration of 30 lM (Fig. 5). At low concentrations both mu-
tants had a statistically signicant higher amount of peptide that
labelled the cells (p < 0.05), which was somehow unexpected due
to the higher resistance shown by these strains (Figs. 1 and 4A).
At high peptide concentration (Fig. 5 and Supplemental Fig. 4)
more than 90% of BY4741 cells were labelled with peptide, an
amount not signicantly different from Dpmt2, while around 75%
of Deos1 cells were labelled.
Previously, we reported that FITC-labelled PAF26 is internalized
into yeast cells at sub-inhibitory concentrations (Lpez-Garca
et al., 2010). Recently, we have described three sequential steps
during the interaction/internalization of this antifungal peptide
in S. cerevisiae using TMR-labelled PAF26 and confocal microscopy
(Marcos et al., 2012; Muoz et al., 2013b). BY4741 cells treated
with a sub-inhibitory concentration (2.5 lM) of TMR-PAF26 for
1 h (e.g. see Fig. 6A) showed three patterns of subcellular localiza-
tion within the cell population. A small proportion of cells had the
labelled peptide located at the cell surface (pattern 1 in Fig. 6B);
most of the cells had the peptide internalized in intracellular vac-
uolar-like organelles (pattern 2); and a small proportion showed
TMR-PAF26 labelling throughout the whole cell (pattern 3). These
sequential steps and subcellular location patterns are similar to
those described in the lamentous model fungus N. crassa (Muoz
et al., 2012).
We investigated whether the mutation of genes involved in pro-
tein glycosylation affected not only the amount of peptide bound
(Fig. 5) but also the distribution and subcellular location of peptide
within cells. There was no signicant variation in the subcellular
location of PAF26 in any of the mutants disrupted in protein
mannosyltransferase genes (PMT) as compared with the parental
strain (Supplemental Fig. 5), including the Dpmt2 deletion strain
that is resistant to PAF26. These ndings might be partially ex-
plained by the functional redundancy of the PMT gene family in
S. cerevisiae (Gentzsch and Tanner, 1997; Strahl-Bolsinger et al.,
1999; Girrbach and Strahl, 2003). Interestingly, the Deos1 deletion
strain demonstrated an inhibition of TMR-PAF26 internalization
into the cells (Fig. 6A), showing the labelled PAF26 predominately
located at the cell envelope in the majority of cells (pattern 1,
Fig. 6B). These results are consistent with the idea that the cell
envelope is critical for peptide interaction and internalization,
and also suggest that Eos1p exerts its activity through the glycosyl-
ation of cell envelope protein(s) required for the internalization of
PAF26.
Fig. 4. Doseresponse curves of fungicidal activity of different peptides against
specic S. cerevisiae deletion mutants. Cell survival after treatment with PAF26 (A),
melittin (B) and P113 (C) are shown. Cells of BY4741 (black circles), Dpmt2 (white
triangles) and Deos1 (white squares) were exposed to different concentrations of
each peptide for 24 h. Cell survival (CFU/mL) was determined by dilution and
plating.
Fig. 5. Differential interaction of S. cerevisiae deletion mutants with TMR-PAF26.
Flow cytometry measurements of TMR-PAF26 signal from cells of the deletion
mutants and parental strain. The graph shows the percentage of uorescently
labelled cells after exposure of 10,000 cells to either 5 (up) or 30 lM (bottom) TMR-
PAF26. Mean standard deviation from two replicas in each of two independent
experiments are shown for each strain.
E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115 109
3.6. A functional cell wall is required for internalization and activity of
PAF26
The next logical step was to explore the role of the CW in the
interaction and internalization of PAF26 by yeast cells. We found
that BY4741 protoplasts exhibited limited internalization of the
peptide (Fig. 6A) and, in fact, mimicked to a large extent the peptide
localization phenotype of walled Deos1 cells (see also Fig. 6B). Fur-
ther characterization of the differential uptake of uorescent pep-
tide was carried out on protoplasts of the Deos1 mutant. Here the
amount of uorescence detected within protoplasts was not signif-
icantly different to that detected within BY4741 protoplasts or
Deos1 walled cells (Fig. 6A and B). Therefore, protoplasting changed
the distribution of peptide signal in BY4741 but not in Deos1 cells.
It is expected for an AMP, whose proposed mechanism of anti-
fungal action is by membrane permeabilization, to be much more
active against cells lacking a CW. In N. crassa it has been shown that
peptide internalization is necessary for cell killing (Muoz et al.,
2012). Consistent with this and the above data, we hypothesized
that inhibition of peptide internalization by removal of the CW
would render cells less sensitive to PAF26. To test this hypothesis,
we further compared cell viability in protoplasts and whole cells
of both BY4741 and Deos1 strains after exposure to PAF26 by using
the PI cell viability assay (Fig. 7). Walled cells of the Deos1 mutant
exhibited a lower percentage of cell death (PI staining) as compared
with the parental strain, in agreement with the resistant phenotype
identied in this study. On the other hand, protoplasts from both
the BY4741 and Deos1 strains exhibited a similar percentage of cell
death that was lower than those of the corresponding intact walled
cells. This is clearly shown in Fig. 7A in which a smaller proportion
of the protoplasts in both the BY4741 and Deos1 strains have taken
up PI compared with the walled cells of these strains (Fig. 7). These
results are thus consistent with a functional CW being important
for PAF26 internalization and antifungal activity.
4. Discussion
It has long been recognized that differences in CW composition
and structure can inuence the selective toxicity of many AMPs
Fig. 6. Differential localization of TMR-PAF26 in S. cerevisiae BY4741 and Deos1 strains. (A) Representative bright eld and confocal microscopy images from walled yeast
cells (top row) and protoplasts (bottom row) for each strain after treatment with 2.5 lM TMR-PAF26 for 1 h. (B) Graph showing the percentages of walled cells (CW) or
protoplasts (Prot.) for each strain (as labelled on the x-axis) that show TMR-PAF26 uorescence with the three different patterns of localization detailed in the text and
illustrated on top as pattern 1 (white bars), pattern 2 (light grey bars), or pattern 3 (black bars). Results represent the means standard deviations of three experiments, each
performed in triplicate. Bar = 10 lm.
110 E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115
against microbes (Yeaman and Yount, 2003). The synthetic hexa-
peptide PAF26 has been proposed as a simple model to character-
ize the mode of action of cationic cell-penetrating antifungal
peptides (Muoz et al., 2012, 2013a). We have recently described
the PAF26 kinetics of interaction and killing in two model fungi,
N. crassa and S. cerevisiae (Marcos et al., 2012; Muoz et al.,
2012, 2013b). At a relatively low concentration, the peptide inter-
acts with the cell envelope and then diffuses to the plasma mem-
brane. The peptide is then internalized and reaches the
intracellular vacuolar system. In the case of N. crassa, it has been
shown that the internalization at low peptide concentration is en-
ergy dependent and involves endocytosis. After reaching a critical
concentration in the vacuolar network, active peptide transport
from vacuoles to the cytoplasm is coincident with the peptide ll-
ing the cell interior and killing the cell. These three steps of peptide
interaction, internalization and killing are reected in the three
subcellular peptide localization patterns shown in Fig. 6. Similar
observations have been reported in Candida albicans for the human
peptide histatin 5 (Mochon and Liu, 2008; Jang et al., 2010) and the
dermaseptin DsS3(1-16) (Harris et al., 2009).
Interestingly, enzymatic digestion that removed the CW almost
completely blocked the internalization of TMR-PAF26 into protop-
lasts of BY4741 (Fig. 6), and resulted in lower percentages of cell
death (Fig. 7). The amount of TMR-PAF26 bound to the cell enve-
lopes was not signicantly affected in BY4741 protoplasts as com-
pared to walled cells (Fig. 6A). These data indicate that the
presence of a functional CW determines the internalization of
PAF26 into S. cerevisiae cells and therefore its antifungal activity,
and also further supports an intracellular killing mechanism for
PAF26. Recently, an inactive PAF26 sequence variant was shown
to be retained at the cell envelope without internalization or toxic-
ity (Muoz et al., 2013b). The fungal CW is also required for the
antifungal activity of the plant defensins Pn-AMP1 against yeast
cells (Koo et al., 2004) and NaD1 against the lamentous fungus
Fusarium oxysporum (van der Weerden et al., 2010). FITC-conju-
gated Pn-AMP1 localized to the surface of the S. cerevisiae BWG7a
cells but, in contrast to what we have shown with PAF26, removal
of the CW impeded the binding of the defensin to the cell and pro-
tected the fungus from being killed (Koo et al., 2004). In contrast,
yeast protoplasts were more sensitive than walled cells to the anti-
fungal tobacco protein osmotin and this higher sensitivity was also
observed for protoplasts of the resistant MNN4 mutant (Ibeas et al.,
2000). Altogether these data indicate different roles for the fungal
CW in the mechanism of antifungal action of PAF26, plant defen-
sins and osmotin.
We had previously demonstrated that exposure of S. cerevisiae
to PAF26 or to the cytolytic melittin results in induction of genes
that code for structural CW proteins, and null mutations of specic
CW-related genes such as Ecm33p resulted in increased sensitivity
to both peptides (Lpez-Garca et al., 2010). Similarly, deletion of
CW-related genes resulted in an increase in the sensitivity of yeast
to other AMPs (Morton et al., 2007). These observations suggest
that the CW broadly also acts as a defence mechanism against
the effects of AMPs. We therefore postulate a dual role for the
Fig. 7. Killing activity of PAF26 on walled cells and protoplasts of S. cerevisiae BY4741 and Deos1 strains. (A) Representative brighteld and confocal microscopy images from
walled yeast cells (top row) and protoplasts (bottom row) from each strain after treatment with 2.5 lM PAF26 for 1 h and staining with 2 lg/mL of the death cell marker PI.
(B) Graph showing the percentage cell death determined by PI uptake from images as shown in (A). Results represent the mean standard deviation of three experiments,
each performed in triplicate. The white bars represent walled cells and black bars protoplasts, indicating statistical signicance (p value) for each comparison. Bar = 10 lm.
E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115 111
fungal CW in the interaction and killing mechanism of PAF26 and
related peptides. The CW is the rst fungal cell component that
the peptide interacts with and, as discussed above, a positive
determinant of peptide action since the CW is required for peptide
internalization. However, it is also a defence barrier and CW
strengthening protects the cell from the detrimental effects of
the peptide. It is likely that these two roles are carried out by dif-
ferent molecular determinants of the CW and future studies will
aim to reveal them.
Most fungal CW proteins, and membrane proteins exposed to
the CW, are glycosylated. Our results indicate that the correct gly-
cosylation of proteins is required for the full susceptibility of bud-
ding yeast to the fungicidal action of PAF26 (Supplemental Table 1
and Figs. 1 and 2). Increased resistance to PAF26 was observed in
null mutants of protein glycosylation genes, including the O-glyco-
sylation genes PMT1-6, the N-glycosylation EOS1 and ALG5, as well
as MNN1, MNN4 and MNN5 (Supplemental. Table 1 and Fig. 8).
These genes (marked in red in Fig. 8) are involved in the addition
of sugars at two distinct locations of the glycan structure: either
at the linkage with the amino acid residues (i.e., PMT genes or
EOS1, see below) or at the outer mannose layers that decorate
the N- and O-protein glycans. As an aside, it should be mentioned
that we could not observe any signicant changes in the expres-
sion of these genes in yeast cells exposed to low concentrations
of PAF26 (data not shown).
It is well known that lamentous fungal and yeast mutants
defective in protein glycosylation show increased sensitivity to
various antifungal agents, slow growth rates and CW-related phe-
notypes that indicate a compromised CW integrity (Gentzsch and
Tanner, 1997; Strahl-Bolsinger et al., 1999; Mouyna et al., 2010).
Previous reports have found increased tolerance to small AMPs
and proteins among protein glycosylation mutants. However, un-
iquely the genes MNN1, MNN4 and MNN5 identied in our study
have been discussed extensively in the context of AMP resistance
(Ibeas et al., 2000; Bai et al., 2006; Harris et al., 2009).
S. cerevisiae mutants in genes responsible for the branching of
O-linked mannans confer resistance to plant antifungal proteins,
as is the case of MNN10, MNN11 and HOC1 to Pn-AMP1 (Koo
et al., 2004), and MNN2, MNN4 or MNN6 to tobacco osmotin (Ibeas
et al., 2000). In this latter study, MNN2, MNN4 or MNN6 were
required for osmotin binding to the fungal CW, while mutation of
MNN1 that exposed negatively charged phosphate groups of phos-
phomannans exhibited enhanced binding and sensitivity (Ibeas
et al., 2000). In the pathogenic yeast C. albicans, deletion of
MNN4 also resulted in enhanced resistance and reduced binding
to the synthetic peptide DsS3(1-16) (Harris et al., 2009), and
MNN5 was required for the antifungal activity of the bovine lacto-
ferrin (Bai et al., 2006). Resistance to yeast killer toxins has been
found in S. cerevisiae mutants in the N-glycosylation genes STT3
and ALG3 (Kimura et al., 1999), and in the PMT genes (Lussier
et al., 1995; Gentzsch and Tanner, 1996). Therefore, proper protein
glycosylation is required for the activity of a signicant number of
antifungal peptides and proteins, including PAF26.
All the above studies imply that there must be glycosylated CW
protein(s) to which the AMPs actually bind, although their identi-
cation has not been achieved in most cases, including the present
study. The CW protein Ssa2 from C. albicans is a chaperone-like
protein that binds histatin 5 (Sun et al., 2008) and has potential
glycosylation sites. Msb2p, that is a transmembrane mucin-like
protein localized at the CW and which is glycosylated by Pmt1p,
has also been associated with the antifungal activity of histatin 5
and LL-37 against C. albicans (Szafranski-Schneider et al., 2012).
In this example, null mutants of Pmt1p were more susceptible to
LL-37. The released glycosylated extracellular domain of Msb2 pro-
tected cells from AMP killing action by sequestering peptides and
preventing their binding to the cells. In a parallel study, we have
conrmed that or strains of S. cerevisiae, which naturally form
biolms on the surface of wine, can bind PAF26 and that a mutant
lacking the hyper-glycosylated CW mucin-like Flo11p lost the abil-
ity to bind the peptide (Bou Zeidan et al., in press). However, we
have not conrmed these ndings with laboratory strains of
S. cerevisiae such as the one used in the present study (data not
shown).
An alternative hypothesis to explain the involvement of protein
glycosylation in the mechanism of antifungal peptides is based on
the glycosylated nature of sensor proteins of fungal mitogen-acti-
vated protein kinase (MAPK) signaling pathways (de Nadal et al.,
2007; Lien et al., 2013). Protein glycosylation is required for the
proper function and signal transduction in the MAPK cascades. This
is the case of the above-mentioned Msb2p, which is a hyper-gly-
cosylated sensor protein of the osmotic stress and lamentous
growth pathways and whose defective glycosylation activates
them constitutively (Tatebayashi et al., 2007; Yang et al., 2009).
Previous studies have shown that yeasts adapt to some AMPs by
activating the Hog1 signaling pathway (Gamberi et al., 2007; Vylk-
ova et al., 2007). The possibility that protein glycosylation has a
role in the transduction pathways that signal the perception of
PAF26 by fungal cells will be examined in the near future.
We focused our analyses on the characterization of Dpmt2 and
Deos1 mutants because they are less susceptible to the cell-pene-
trating peptides PAF26 and P113 (a synthetic derivative of histatin
5), yet they show hypersensitivity to the CW- and plasma mem-
brane-disrupting compounds CFW or SDS, respectively. Moreover,
they are also more sensitive to membrane lytic peptides such as
melittin or cecropin A (Figs. 1 and 4, and Table 2). All our current
and previous analyses indicate that these two mutants have a
weakened CW that, presumably, should make them broadly more
sensitive to most antifungal agents. Therefore, their PAF26 toler-
ance phenotype is noteworthy and specic to a sub-class of AMPs
that penetrate cells by mechanisms (e.g. endocytosis) other than
plasma membrane permeabilization.
Tunicamycin (TM) is an antifungal agent that inhibits protein N-
glycosylation (Nakamura et al., 2007). PAF26 and TM antifungal
activities are synergistic (Fig. 2). This suggests that although pro-
tein glycosylation determines sensitivity to AMPs, the antifungal
mechanism of PAF26 does not target protein glycosylation itself.
Previous data are consistent with this view. For instance, pharma-
cological or genetic inhibition of protein glycosylation resulted in
the activation of genes from the CW integrity signaling pathway
(Cantero et al., 2007; Arroyo et al., 2011; Cantero and Ernst,
2011), and the inhibition of N-glycosylation by TM triggers the
activation of PMT genes. This suggests a compensatory mechanism
to alleviate defects in protein glycosylation. However, the tran-
scriptomic analyses of PAF26-treated S. cerevisiae did not identify
any process involved in the induction of the CW integrity pathway
or protein glycosylation genes including PMTs (Lpez-Garca et al.,
2010), as would be expected if PAF26 affected protein glycosyla-
tion. On the other hand, recent studies demonstrated the increased
sensitivity of all S. cerevisiae Dpmt1-6 mutants to a specic O-gly-
cosylation inhibitor (OGT2468) (Arroyo et al., 2011), while in our
study the same mutants are more resistant to PAF26.
Previously, we have shown that disruption of ARG genes in-
volved in arginine metabolism also resulted in increased resistance
to PAF26 (Lpez-Garca et al., 2010). The mutant Darg1 does not
produce the intracellular nitric oxide (NO) associated with expo-
sure to PAF26, and an inhibitor of arginine-derived NO synthase
partially protects yeast cells from PAF26 action (Carmona et al.,
2012). Moreover, Darg1 showed a block in the transport of PAF26
from the vacuole to the cytosol that accompanies cell death
(Muoz et al., 2013b). Therefore ARG1 is likely involved in the
intracellular killing phase of the mode of action of PAF26. Conse-
quently, none of the generated double deletion strains that
112 E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115
targeted arginine metabolism or protein glycosylation resulted in
enhanced resistance to PAF26 (Fig. 3). These results are consistent
with the involvement of both biological processes (protein glyco-
sylation and arginine metabolism) in different steps of the same
pathway that results in cell killing by PAF26.
Most of the aforementioned previous studies identied manno-
syltransferase genes involved in decorating the outer branches of
the core N-glycan structure, some of which have been also identi-
ed in our study (Fig. 8). Few studies have reported roles for the
EOS1 or PMT genes in this context. Signicantly, EOS1 (YLN080C)
was included in the list of genes that when deleted showed en-
hanced resistance to the cyclic lipopeptide caspofungin in two
independent genome-wide studies (Lesage et al., 2004; Markovich
et al., 2004), although the involvement of EOS1 in AMP action was
unexplored and not discussed at that time. The Deos1 mutant
shows a pleiotropic phenotype with hypersensitivity to oxidative
and osmotic stress, as well as altered metal homeostasis (Nakam-
ura et al., 2007, 2010). One of the most remarkable phenotypes of
Deos1 is its high tolerance to TM (Fig. 2 and also Nakamura et al.,
2007), which together with the altered glycosylation pattern of cell
proteins and location of Eos1p at the ER justied its annotation as
involved in protein N-glycosylation (Nakamura et al., 2007). TM
inhibits the transfer of N-acetylglucosamine-P to Dol-P in the ER
in the rst step of the dolichol pathway of protein N-glycosylation,
a reaction catalyzed by Alg7p (Fig. 8). The null mutation of ALG7 is
lethal, and therefore could not be assayed in our study. We specu-
late that Eos1p participates in the N-glycosylation of a subset of
cell proteins, by interaction with Alg7p in the rst step of the dol-
ichol pathway (Fig. 8 and Supplemental Fig. 1A).
Compared with the BY4741 cells, we observed a remarkably dif-
ferent localization of TMR-PAF26 in Deos1 cells, in which the la-
belled peptide remained at the cell envelope and was not
internalized (Fig. 6A). This was coincident with a lower percentage
of cell death (Fig. 7), conrming the resistant phenotype found in
fungicidal assays (Fig. 1). These observations demonstrate that
EOS1 is necessary for the internalization of this antifungal hexa-
peptide. PAF26 localization (Fig. 6) and killing activity (Fig. 7)
was similar in protoplasts of BY4741 and Deos1, contrarily to
BY4741 in which protoplasting affected PAF26 localization and
activity. Therefore the phenotypic difference between the eos1 mu-
tant and the parental strain is lost after removal of the CW. A plau-
sible explanation is that EOS1 functions at the level of the CW,
through involvement in the glycosylation of a protein(s) required
for cell-penetration by PAF26. The killing activity of histatin 5 to
C. albicans is determined by the intracellular translocation of the
peptide mediated by the CW located protein Ssa2 (Sun et al.,
2008; Jang et al., 2010). In our case, an alternative hypothesis could
be that compensatory changes in the CW of the eos1 mutant in-
crease the content of (glyco)proteins and/or glycans that sequester
the peptide at the CW layer and avoid its internalization. Finally, it
cannot be ruled out that the pleiotropic effect of the EOS1 null
mutation (Nakamura et al., 2010) affects additional (maybe even
overlapping) cell processes that might explain the blockage of pep-
tide cell penetration and resistance of this mutant.
5. Conclusions
In conclusion, our data further conrm the involvement of pro-
tein glycosylation genes in the activity of short antifungal peptides
and expand the repertoire of glycosylation genes known in this
context. The results are consistent with a mechanism for which
PAF26 internalization is required for killing fungal cells (Muoz
et al., 2013a). It is shown that the CW and the N-glycosylation gene
EOS1 are required for internalization of PAF26. The EOS1 null muta-
tion results in the peptide being located at the cell surface and in-
creased tolerance of cells to its antifungal action. Our present
working hypothesis is that Eos1p participates in the glycosylation
of yet unidentied CW protein(s) required for peptide cell penetra-
tion. The identication of CW proteins that participate in the
Fig. 8. Schematic representation of the structure of representative S. cerevisiae N-linked (left) and O-linked (right) glycans. Diagram adapted from (Lehle et al., 2006). Squares
indicate N-acetylglucosamine residues, and circles indicate mannose residues. Phosphate residues are indicated by red circles. Open symbols indicate residues added to the
glycan at the cytosolic side of the ER membrane, lled symbols indicate residues added at the lumen side of the ER, and striped symbols indicate residues added in the Golgi
apparatus. The proteins involved in the addition of each sugar residue are labelled; and those proteins whose genes when deleted are lethal are shown in grey. The proteins
whose genes when deleted resulted in increased tolerance to PAF26 are highlighted in red.
E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115 113
internalization and antifungal activity of PAF26 is currently being
explored.
Acknowledgments
EHs visit to the University of Edinburgh was funded by the JAE-
PREDOC Program of CSIC (Spain) and the European FEDER fund.
The study was supported by research grants BIO2009-12919 (MIC-
INN, Spain) and ACOMP/2012/018 (Generalitat Valenciana, Spain).
This article is dedicated to the dearest memory of our friend and
colleague Enrique (Quique) Prez-Pay, with whom we had a fruit-
ful scientic collaboration in the eld of antimicrobial peptides
over the last two decades.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.fgb.2013.08.004.
References
Arroyo, J., Hutzler, J., Bermejo, C., Ragni, E., Garcia-Cantalejo, J., Botias, P., Piberger,
H., Schott, A., Sanz, A.B., Strahl, S., 2011. Functional and genomic analyses of
blocked protein O-mannosylation in bakers yeast. Mol. Microbiol. 79, 1529
1546.
Bai, C., Xu, X.L., Chan, F.Y., Lee, R.T.H., Wang, Y., 2006. MNN5 encodes an iron-
regulated a-1,2-mannosyltransferase important for protein glycosylation, cell
wall integrity, morphogenesis and virulence in Candida albicans. Eukaryot. Cell
5, 238247.
Bou Zeidan, M., Carmona, L., Zara, S., Marcos, J.F., in press. FLO11 gene is involved in
the interaction of or strains of Saccharomyces cerevisiae with a biolm-
promoting synthetic hexapeptide. Appl. Environ. Microbiol. 79, 2013,
doi:10.1128/AEM.01647-13.
Brogden, K.A., 2005. Antimicrobial peptides: pore formers or metabolic inhibitors in
bacteria? Nat. Rev. Microbiol. 3, 238250.
Cantero, P.D., Ernst, J.F., 2011. Damage to the glycoshield activates PMT-directed O-
mannosylation via the Msb2-Cek1 pathway in Candida albicans. Mol. Microbiol.
80, 715725.
Cantero, P.D., Lengsfeld, C., Prill, S.K.H., Subanovic, M., Roman, E., Pla, J., Ernst, J.F.,
2007. Transcriptional and physiological adaptation to defective protein-O-
mannosylation in Candida albicans. Mol. Microbiol. 64, 11151128.
Carmona, L., Ganda, M., Lpez-Garca, B., Marcos, J.F., 2012. Sensitivity of
Saccharomyces cerevisiae to the cell-penetrating antifungal peptide PAF26
correlates with endogenous nitric oxide (NO) production. Biochem. Biophys.
Res. Commun. 417, 5661.
de Nadal, E., Real, F.X., Posas, F., 2007. Mucins, osmosensors in eukaryotic cells?
Trends Cell Biol. 17, 571574.
Deshpande, N., Wilkins, M.R., Packer, N., Nevalainen, H., 2008. Protein glycosylation
pathways in lamentous fungi. Glycobiology 18, 626637.
Duncan, V.M.S., ONeil, D.A., 2013. Commercialization of antifungal peptides. Fungal
Biol. Rev. 26, 156165.
Elbein, A.D., 1987. Inhibitors of the biosynthesis and processing of N-linked
oligosaccharide chains. Annu. Rev. Biochem. 56, 497534.
Fernndez-lvarez, A., Elas-Villalobos, A., Ibeas, J.I., 2009. The O-
mannosyltransferase PMT4 is essential for normal appressorium formation
and penetration in Ustilago maydis. Plant Cell. 21, 33973412.
Gamberi, T., Cavalieri, D., Magherini, F., Mangoni, L., De Filippo, C., Borro, M., Gentile,
G., Simmaco, M., Modesti, A., 2007. An integrated analysis of the effects of
Esculentin 1-21 on Saccharomyces cerevisiae. Biochim. Biophys. Acta Prot.
Proteom. 1774, 688700.
Gentzsch, M., Tanner, W., 1996. The PMT gene family: protein O-glycosylation in
Saccharomyces cerevisiae is vital. EMBO J. 15, 57525759.
Gentzsch, M., Tanner, W., 1997. Protein-O-glycosylation in yeast: protein-specic
mannosyltransferases. Glycobiology 7, 481486.
Girrbach, V., Strahl, S., 2003. Members of the evolutionarily conserved PMT family
of protein O-mannosyltransferases form distinct protein complexes among
themselves. J. Biol. Chem. 278, 1255412562.
Hancock, R.E.W., Sahl, H.G., 2006. Antimicrobial and host-defense peptides as new
anti-infective therapeutic strategies. Nat. Biotechnol. 24, 15511557.
Harris, M., Mora-Montes, H.M., Gow, N.A.R., Coote, P.J., 2009. Loss of
mannosylphosphate from Candida albicans cell wall proteins results in
enhanced resistance to the inhibitory effect of a cationic antimicrobial
peptide via reduced peptide binding to the cell surface. Microbiol. SGM 155,
10581070.
Heesen, S.t., Lehle, L., Weissmann, A., Aebi, M., 1994. Isolation of the ALG5 locus
encoding the UDP-glucose:dolichyl-phosphate glucosyltransferase from
Saccharomyces cerevisiae. Eur. J. Biochem. 224, 7179.
Henriques, S.T., Melo, M.N., Castanho, M.A.R.B., 2006. Cell-penetrating peptides and
antimicrobial peptides: how different are they? Biochem. J. 399, 17.
Ibeas, J.I., Lee, H., Damsz, B., Prasad, D.T., Pardo, J.M., Hasegawa, P.M., Bressan, R.A.,
Narasimhan, M.L., 2000. Fungal cell wall phosphomannans facilitate the toxic
activity of a plant PR-5 protein. Plant J. 23, 375383.
Jang, W.S., Li, X.W.S., Sun, J.N.N., Edgerton, M., 2008. The P-113 fragment of Histatin
5 requires a specic peptide sequence for intracellular translocation in Candida
albicans, which is independent of cell wall binding. Antimicrob. Agents
Chemother. 52, 497504.
Jang, W.S., Bajwa, J.S., Sun, J.N., Edgerton, M., 2010. Salivary histatin 5
internalization by translocation, but not endocytosis, is required for fungicidal
activity in Candida albicans. Mol. Microbiol. 77, 354370.
Kimura, T., Komiyama, T., Furuichi, Y., Iimura, Y., Karita, S., Sakka, K., Ohmiya, K.,
1999. N-glycosylation is involved in the sensitivity of Saccharomyces cerevisiae
to HM-1 killer toxin secreted from Hansenula mrakii IFO 0895. Appl. Microbiol.
Biotechnol. 51, 176184.
Klis, F.M., Boorsma, A., de Groot, P.W.J., 2006. Cell wall construction in
Saccharomyces cerevisiae. Yeast 23, 185202.
Koo, J.C., Lee, B., Young, M.E., Koo, S.C., Cooper, J.A., Baek, D., Lim, C.O., Lee, S.Y., Yun,
D.J., Cho, M.J., 2004. Pn-AMP1, a plant defense protein, induces actin
depolarization in yeasts. Plant Cell Physiol. 45, 16691680.
Lehle, L., Strahl, S., Tanner, W., 2006. Protein glycosylation, conserved from yeast to
man: a model organism helps elucidate congenital human diseases. Angew.
Chem. Int. Ed. 45, 68026818.
Lesage, G., Sdicu, A.M., Menard, P., Shapiro, J., Hussein, S., Bussey, H., 2004. Analysis
of beta-1,3-glucan assembly in Saccharomyces cerevisiae using a synthetic
interaction network and altered sensitivity to caspofungin. Genetics 167, 35
49.
Lien, E.C., Nagiec, M.J., Dohlman, H.G., 2013. Proper protein glycosylation promotes
mitogen-activated protein kinase signal delity. Biochemistry 52, 115124.
Lpez-Garca, B., Prez-Pay, E., Marcos, J.F., 2002. Identication of novel
hexapeptides bioactive against phytopathogenic fungi through screening of a
synthetic peptide combinatorial library. Appl. Environ. Microbiol. 68, 2453
2460.
Lpez-Garca, B., Ganda, M., Muoz, A., Carmona, L., Marcos, J.F., 2010. A genomic
approach highlights common and diverse effects and determinants of
susceptibility on the yeast Saccharomyces cerevisiae exposed to distinct
antimicrobial peptides. BMC Microbiol. 10, 289.
Lussier, M., Gentzsch, M., Sdicu, A.M., Bussey, H., Tanner, W., 1995. Protein O-
glycosylation in yeast. The PMT2 gene species a second protein O-
mannosyltransferase that functions in addition to the PMT1-encoded activity.
J. Biol. Chem. 270, 27702775.
Marcos, J.F., Ganda, M., 2009. Antimicrobial peptides: to membranes and beyond.
Expert Opin. Drug Discov. 4, 659671.
Marcos, J.F., Muoz, A., Prez-Pay, E., Misra, S., Lpez-Garca, B., 2008.
Identication and rational design of novel antimicrobial peptides for plant
protection. Annu. Rev. Phytopathol. 46, 273301.
Marcos, J.F., Ganda, M., Harries, E., Carmona, L., Muoz, A., 2012. Antifungal
peptides: exploiting non-lytic mechanisms and cell penetration properties. In:
Rajasekaran, K., Cary, J.W., Jaynes, J., Montesinos, E. (Eds.), Small Wonders:
Peptides for Disease Control. American Chemical Society, pp. 337357.
Markovich, S., Yekutiel, A., Shalit, I., Shadkchan, Y., Osherov, N., 2004. Genomic
approach to identication of mutations affecting caspofungin susceptibility in
Saccharomyces cerevisiae. Antimicrob. Agents Chemother. 48, 38713876.
Mochon, A.B., Liu, H.P., 2008. The antimicrobial peptide Histatin-5 causes a spatially
restricted disruption on the Candida albicans surface, allowing rapid entry of the
peptide into the cytoplasm. PLoS Pathog. 4, e1000190.
Morton, C.O., Hayes, A., Wilson, M., Rash, B.M., Oliver, S.G., Coote, P., 2007. Global
phenotype screening and transcript analysis outlines the inhibitory mode(s) of
action of two amphibian-derived, a-helical, cationic peptides on Saccharomyces
cerevisiae. Antimicrob. Agents Chemother. 51, 39483959.
Mouyna, I., Kniemeyer, O., Jank, T., Loussert, C., Mellado, E., Aimanianda, V.,
Beauvais, A., Wartenberg, D., Sarfati, J., Bayry, J., Prevost, M.C., Brakhage, A.A.,
Strahl, S., Huerre, M., Latge, J.P., 2010. Members of protein O-
mannosyltransferase family in Aspergillus fumigatus differentially affect
growth, morphogenesis and viability. Mol. Microbiol. 76, 12051221.
Muoz, A., Lpez-Garca, B., Marcos, J.F., 2006. Studies on the mode of action of the
antifungal hexapeptide PAF26. Antimicrob. Agents Chemother. 50, 38473855.
Muoz, A., Marcos, J.F., Read, N.D., 2012. Concentration-dependent mechanisms of
cell penetration and killing by the de novo designed antifungal hexapeptide
PAF26. Mol. Microbiol. 85, 89106.
Muoz, A., Ganda, M., Harries, E., Carmona, L., Read, N.D., Marcos, J.F., 2013a.
Understanding the mechanism of action of cell-penetrating antifungal peptides
using the rationally designed hexapeptide PAF26 as a model. Fungal Biol. Rev.
26, 146155.
Muoz, A., Harries, E., Contreras-Valenzuela, A., Carmona, L., Read, N.D., Marcos, J.F.,
2013b. Two functional motifs dene the interaction, internalization and toxicity
of the cell-penetrating antifungal peptide PAF26 on fungal cells. PLoS ONE 8,
e54813.
Munro, C.A., Bates, S., Buurman, E.T., Hughes, H.B., MacCallum, D.M., Bertram, G.,
Atrih, A., Ferguson, M.A.J., Bain, J.M., Brand, A., Hamilton, S., Westwater, C.,
Thomson, L.M., Brown, A.J.P., Odds, F.C., Gow, N.A.R., 2005. Mnt1p and Mnt2p of
Candida albicans are partially redundant a-1,2-mannosyltransferases that
participate in O-linked mannosylation and are required for adhesion and
virulence. J. Biol. Chem. 280, 10511060.
Nakamura, T., Ando, A., Takagi, H., Shima, J., 2007. EOS1, whose deletion confers
sensitivity to oxidative stress, is involved in N-glycosylation in Saccharomyces
cerevisiae Biochem. Biophys. Res. Commun. 353, 293298.
114 E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115
Nakamura, T., Takahashi, S., Takagi, H., Shima, J., 2010. Multicopy suppression of
oxidant-sensitive eos1 mutation by IZH2 in Saccharomyces cerevisiae and the
involvement of Eos1in zinc homeostasis. FEMS Yeast Res. 10, 259269.
Peters, B.M., Shirtliff, M.E., Jabra-Rizk, M.A., 2010. Antimicrobial peptides: primeval
molecules or future drugs? PLoS Pathog. 6, e1001067.
Petit, J., Boisseau, P., Arveiler, B., 1994. Glucanex: a cost-effective yeast lytic enzyme.
Trends Genet. 10, 45.
Prill, S.K.H., Klinkert, B., Timpel, C., Gale, C.A., Schroppel, K., Ernst, J.F., 2005. PMT
family of Candida albicans: ve protein mannosyltransferase isoforms affect
growth, morphogenesis and antifungal resistance. Mol. Microbiol. 55, 546560.
Rayner, J.C., Munro, S., 1998. Identication of the MNN2 and MNN5
mannosyltransferases required for forming and extending the mannose
branches of the outer chain mannans of Saccharomyces cerevisiae. J. Biol.
Chem. 273, 2683626843.
Schirawski, J., Bhnert, H.U., Steinberg, G., Snetselaar, K., Adamikowa, L., Kahmann,
R., 2005. Endoplasmic reticulum glucosidase II is required for pathogenicity of
Ustilago maydis. Plant Cell. 17, 35323543.
Steiner, H., Hultmark, D., Engstrom, A., Bennich, H., Boman, H.G., 1981. Sequence
and specicity of 2 anti-bacterial proteins involved in insect immunity. Nature
292, 246248.
Strahl-Bolsinger, S., Gentzsch, M., Tanner, W., 1999. Protein O-mannosylation.
Biochim. Biophys. Acta 1426, 297307.
Sun, J.N.N., Li, W.S., Jang, W.S., Nayyar, N., Sutton, M.D., Edgerton, M., 2008. Uptake
of the antifungal cationic peptide Histatin 5 by Candida albicans Ssa2p requires
binding to non-conventional sites within the ATPase domain. Mol. Microbiol.
70, 12461260.
Szafranski-Schneider, E., Swidergall, M., Cottier, F., Tielker, D., Roman, E., Pla, J.,
Ernst, J.F., 2012. Msb2 shedding protects Candida albicans against antimicrobial
peptides. PLoS Pathog. 8, e1002501.
Tatebayashi, K., Tanaka, K., Yang, H.Y., Yamamoto, K., Matsushita, Y., Tomida, T.,
Imai, M., Saito, H., 2007. Transmembrane mucins Hkr1 and Msb2 are putative
osmosensors in the SHO1 branch of yeast HOG pathway. EMBO J. 26, 3521
3533.
Terwillinger, T.C., Eisenberg, D., 1982. The structure of melittin. II. Interpretation of
the structure. J. Biol. Chem. 257, 60166022.
van der Weerden, N.L., Hancock, R.E.W., Anderson, M.A., 2010. Permeabilization of
fungal hyphae by the plant defensin NaD1 occurs through a cell wall-dependent
process. J. Biol. Chem. 285, 3751337520.
Vylkova, S., Jang, W.S., Li, W.S., Nayyar, N., Edgerton, M., 2007. Histatin 5 initiates
osmotic stress response in Candida albicans via activation of the Hog1 mitogen-
activated protein kinase pathway. Eukaryot. Cell 6, 18761888.
Yang, H.Y., Tatebayashi, K., Yamamoto, K., Saito, H., 2009. Glycosylation defects
activate lamentous growth Kss1 MAPK and inhibit osmoregulatory Hog1
MAPK. EMBO J. 28, 13801391.
Yeaman, M.R., Yount, N.Y., 2003. Mechanisms of antimicrobial peptide action and
resistance. Pharmacol. Rev. 55, 2755.
Zasloff, M., 2002. Antimicrobial peptides of multicellular organisms. Nature 415,
389395.
Zhou, H., Hu, H.Y., Zhang, L.J., Li, R.Y., Ouyang, H.M., Ming, J., Jin, C., 2007. O-
mannosyltransferase 1 in Aspergillus fumigatus (AfPmt1p) is crucial for cell wall
integrity and conidium morphology, especially at an elevated temperature.
Eukaryot. Cell 6, 22602268.
E. Harries et al. / Fungal Genetics and Biology 58-59 (2013) 105115 115