INVESTIGATION
Membrane Organization and Cell Fusion During
Mating in Fission Yeast Requires Multipass
Membrane Protein Prm1
M.-Ángeles Curto,* Mohammad Reza Sharifmoghadam,†,‡ Eduardo Calpena,* Nagore De León,*
Marta Hoya,* Cristina Doncel,* Janet Leatherwood,† and M.-Henar Valdivieso*,1
*Departamento de Microbiología y Genética/Instituto de Biología Funcional y Genómica, Universidad de Salamanca/Consejo
Superior de Investigaciones Científicas, 37007 Salamanca, Spain, †Department of Molecular Genetics and Microbiology,
Stony Brook University, New York 11790, and ‡Department of Biology, Faculty of Science, Ferdowsi University of Mashhad,
9177948974 Iran
ABSTRACT The involvement of Schizosaccharomyces pombe prm1+ in cell fusion during mating and its relationship with other genes
required for this process have been addressed. S. pombe prm1D mutant exhibits an almost complete blockade in cell fusion and an
abnormal distribution of the plasma membrane and cell wall in the area of cell–cell interaction. The distribution of cellular envelopes is
similar to that described for mutants devoid of the Fig1-related claudin-like Dni proteins; however, prm1+ and the dni+ genes act in
different subpathways. Time-lapse analyses show that in the wild-type S. pombe strain, the distribution of phosphatidylserine in the
cytoplasmic leaflet of the plasma membrane undergoes some modification before an opening is observed in the cross wall at the cell–
cell contact region. In the prm1D mutant, this membrane modification does not take place, and the cross wall between the mating
partners is not extensively degraded; plasma membrane forms invaginations and fingers that sometimes collapse/retract and that are
sometimes strengthened by the synthesis of cell-wall material. Neither prm1D nor prm1D dniD zygotes lyse after cell–cell contact in
medium containing and lacking calcium. Response to drugs that inhibit lipid synthesis or interfere with lipids is different in wild-type,
prm1D, and dni1D strains, suggesting that membrane structure/organization/dynamics is different in all these strains and that Prm1p
and the Dni proteins exert some functions required to guarantee correct membrane organization that are critical for cell fusion.
M
EMBRANE fusion is essential for several developmental processes. The characterization of mating in yeasts
represents a useful tool for understanding the mechanisms
involved in intercellular membrane merger and their regulation. In the fission yeast Schizosaccharomyces pombe,
heterothallic cells belong to one of two mating types, M
(h2 cells) or P (h+ cells) while h90 strains are homothallic
(Arcangioli and Thon 2004). In rich medium, h+ and h2 cells
can proliferate actively together without mating. When nitrogen becomes scarce the cAMP level decreases, which triggers
Copyright © 2014 by the Genetics Society of America
doi: 10.1534/genetics.113.159558
Manuscript received November 10, 2013; accepted for publication January 30, 2014;
published Early Online February 10, 2014.
Supporting information is available online at http://www.genetics.org/lookup/suppl/
doi:10.1534/genetics.113.159558/-/DC1.
This article is dedicated to Dr. Angel Duran on the occasion of his retirement for his
contributions to the field of the yeast cell wall and morphogenesis and for his efforts
to make the IBFG an outstanding research center.
1
Corresponding author: Lab. P1.1, Edificio IBFG, Calle Zacarias Gonzalez 2, 37007
Salamanca, Spain. E-mail: [email protected]
the expression of genes required for sexual differentiation. Cells
arrest in G1 and polarize, giving rise to specialized cells termed
shmoos (Yamamoto et al. 1997; Davey 1998; Nielsen 2004;
Yamamoto 2004). After agglutination, the cross wall separating
both parental cells is degraded, allowing cell fusion. Efficient
cell fusion requires the correct organization of the cytoskeleton
(Petersen et al. 1995, 1998a,b; Kurahashi et al. 2002; Doyle
et al. 2009). The S. cerevisiae FUS1 and the S. pombe fus1+
genes have the same name and both mutants exhibit cell fusion
defects during mating, leading to an accumulation of prezygotes (mating intermediates in which the mating partners have
established a stable contact but have not yet fused). However,
the corresponding proteins are not related; while Saccharomyces cerevisiae Fus1p is a membrane protein (Trueheart et al.
1987), S. pombe Fus1p is a formin required for the organization
of actin patches at the shmoo tip (Petersen et al. 1995, 1998b).
S. pombe diploid nuclei are unstable, such that meiosis occurs
immediately after karyogamy, and zygotes give rise to asci
containing four ascospores (Nielsen 2004; Yamamoto 2004).
Genetics, Vol. 196, 1059–1076 April 2014
1059
Genome-wide analyses performed using nitrogen starvation and pheromone treatments have been used to analyze
gene expression under mating conditions (Mata et al. 2002;
Mata and Bahler 2006). Study of some previously uncharacterized genes whose expression was induced in response
to mating conditions led to the characterization of the h+specific agglutinin map4+ (Yamamoto et al. 1997; Mata and
Bahler 2006; Sharifmoghadam et al. 2006), and dni1+ and
dni2+, two members of the Fig1-related family of fungal
claudin-like proteins (Clemente-Ramos et al. 2009). dni1D
and dni2D mutants exhibit a temperature-sensitive cell fusion defect due to a lack of coordination between membrane
organization and cell-wall remodeling at the cell–cell contact region (Clemente-Ramos et al. 2009). The SPAP7G5.03
gene, which codes for a protein with several potential transmembrane domains that is 22% identical to the S. cerevisiae
Prm1 (Heiman and Walter 2000), was not identified as
a mating-induced gene in the extensive analyses (Mata et al.
2002; Mata and Bahler 2006). In budding yeast, PRM1 is
highly induced in response to pheromones and its absence
leads to a 50% reduction in the efficiency of cell fusion
during mating. S. cerevisiae prm1D mating partners degrade
the cross wall separating both cells and their plasma membranes remain apposed but not fused (Heiman and Walter
2000). In the prm1D zygotes, intercellular bubbles and fingers are produced when the cytoplasm of one of the cells
invades part of the cytoplasm of the other mating partner
because the apposed membranes cannot support turgor
pressure in the absence of the cell wall (Heiman and Walter
2000; Jin et al. 2004). It has been proposed that Prm1p is
a fusion facilitator that might help to form a pore at the
contact area between mating partners (White and Rose
2001; Olmo and Grote 2010a). Neurospora crassa has
a PRM1 ortholog whose deletion leads to a 50% reduction
in cell fusion during vegetative and sexual cell fusion, the
appearance of intercellular bubbles/fingers with apposed
plasma membranes during vegetative cell fusion, and
defects in postmeiotic events that lead to complete sterility
(Fleissner et al. 2009). Intercellular bubbles/fingers have
also been observed in S. cerevisiae fig1D mutants, which
are defective in a claudin-like tetraspan protein that facilitates calcium influx and is required for cell polarization and
fusion during mating (Erdman et al. 1998; Muller et al.
2003; Aguilar et al. 2007).
Cell fusion is a challenging process that requires the
reorganization of cellular envelopes. S. cerevisiae prm1D
cells undergo contact-dependent lysis during mating (Jin
et al. 2004). Calcium depletion enhances lysis in prm1D
and fig1D mutants while deletion of FIG1 in a prm1D background enhances the cell-fusion defect but not zygote lysis
(Aguilar et al. 2007). It has been proposed that calcium
would promote a plasma membrane repair mechanism that
compensates the lack of Fig1p and Prm1p (Aguilar et al.
2007). Although there is a fairly large body of information
about S. cerevisiae Prm1p (Heiman and Walter 2000; Jin
et al. 2004; Aguilar et al. 2007; Engel et al. 2010; Olmo
1060
M.-Ángeles Curto et al.
and Grote 2010a,b), the nature of its function remains
unknown.
S. cerevisiae and S. pombe are distantly related yeasts
(Sipiczki 2000) and hence studies addressing mating in both
organisms might help to understand the conserved and divergent functions of proteins involved in this developmental
process (Merlini et al. 2013). In this work we have analyzed
the role of S. pombe prm1+ in mating and its relationship
with other genes involved in cell fusion. We have confirmed
that prm1+ is upregulated during mating and have found
that Prm1p localizes to the tip of the shmoos. In prm1D
mutants, the cross walls separating the mating partners
are not digested extensively and plasma membranes form
invaginations/fingers. In some prezygotes, new cell-wall
material is produced at the fusion area so that this structure
undergoes a similar disorganization to that observed in dniD
zygotes. However, genetic analyses revealed that the function of prm1+ and the dni+ genes is independent. The behavior of phosphatidylserine at the inner leaflet of plasma
membrane and response to drugs that interfere with lipids
indicate that the structure/organization/dynamics of the
plasma membrane is different during vegetative growth
and during mating and that it is different in the wild-type,
dni1D, and prm1D strains. These results suggest that the
Prm1p and the Dni proteins exert different functions required to maintain proper membrane organization and that
these functions are particularly relevant for cell fusion.
Materials and Methods
Strains and growth conditions
All general growth conditions and yeast manipulations have
been described previously (Moreno et al. 1991; http://
biotwiki.org/foswiki/bin/view/Pombe/NurseLabManual).
S. pombe strains (see Table S1) are derivatives of the 972
h2 and 975 h+ wild-type (WT) strains and were grown in yeastextracts with supplements (YES) or Edinburgh minimal medium (EMM). EMM–Ca was prepared by eliminating CaCl2
and replacing calcium pantothenate by sodium pantothenate. When desired, different concentrations of calcium,
BAPTA (Sigma), and EGTA (Sigma) were added to this medium. EMM–N was EMM without ammonium chloride and
was prepared with deionized water. Geneticin (G418, Formedium) and Nourseothricin (Werner BioAgents) were
used at 120 mm/ml and 50 m/ml, respectively. Miconazole
(Sigma; stock at 10 mM in DMSO), Myriocin (Sigma; stock
at 1 mg/ml in ethanol), filipin complex (Sigma; stock at
5 mg/ml in DMSO), Duramycin (Santa Cruz Biotechnology;
stock at 1 mM in methanol:H2O 1:1), and Papuamide B
(Flintbox; stock at 0.5 mg/ml in methanol:H2O 1:1) were
used at different concentrations. To assess sensitivity to filipin, actively growing cells were collected by centrifugation
and resuspended in YES medium containing different concentrations of the antibiotic; then, serial 1:4 dilutions were
performed on microtiter plates with the same medium so
Figure 1 prm1+ is expressed under mating conditions. (A) Semiquantitative reverse transcriptase
PCR of prm1+, dni1+, and dni2+
in strains bearing (WT) or lacking
(D) the corresponding gene. The
act1+ gene was used for normalization. Numbers indicate the relative level of expression. (B) Top:
phase contrast and fluorescence
micrographs of a homothallic
h90 strain expressing the indicated fusion proteins. Bottom:
phase contrast and fluorescence
micrographs of heterothallic strains
bearing the indicated fusion proteins and treated with pheromone.
The pictures were captured with
a conventional fluorescence microscope. Scale bar, 3 mm.
that cells were diluted but the antibiotic concentration was
the same in all the wells. Plates were incubated at 32° and
the OD600nm of each culture was measured after 2 days
of incubation using a microplate photometer (Multiskan
Ascent, Thermo).
by Isogen; final concentration 1.5 mg/ml from a stock prepared
at 5 mg/ml in methanol) as described (Stern and Nurse 1997;
Clemente-Ramos et al. 2009). cyr1D sxa2D cells are sensitive to
low concentrations of pheromones in nitrogen-containing media (Stern and Nurse 1997).
Mating analysis
Molecular and genetic manipulations
h90
Mating was induced in homothallic
cells in liquid EMM–
N as described previously (Sharifmoghadam and Valdivieso
2008); cells growing exponentially in EMM were washed
extensively with deionized water, inoculated at a cell concentration of about 107 cells/ml in EMM–N prepared with
deionized water, and incubated at 26° with gentle shaking
(70 rpm). Under these conditions, mating was evident about
3–4 hr after nitrogen starvation; unless stated otherwise,
samples were collected at this time. To perform mating analysis on plates, patches of cells that had been growing overnight on rich YES medium were replica plated onto EMM
plates and incubated at 32°. Under these conditions, cells
divide until complete nitrogen depletion and then initiate
sexual differentiation. In a WT h90 strain, a few prezygotes
were first observed after 12 hr of incubation; after 15 hr of
incubation, prezygotes and late zygotes (zygotes with wide
necks) were abundant, while mature asci were scarce. Mature asci (with four refringent spores) were observed after
24 hr of incubation. Samples were taken at different times
depending on the nature of the experiment. Sporulation
efficiency, estimated as the number of asci with respect to
the number of asci plus prezygotes and late zygotes, was
used as an indirect measurement of cell fusion. To analyze
the effect of calcium in mating, cells were induced to mate in
EMM–Ca or EMM–Ca supplemented with the desired
amounts of calcium or calcium chelators. Shmoo formation
was induced in strains devoid of the adenylate cyclase cyr1+
and the peptidase sxa2+ genes by adding P factor (synthesized
DNA fragments (1 kb) corresponding to the 59- and 39noncoding regions of prm1+ were amplified by PCR and cloned
into the pALKS+ vector. To recover a genomic copy of the
prm1+ ORF with the “gap repair” technique (Orr-Weaver
et al. 1983), this plasmid was linearized and used to transform a wild-type S. pombe strain. A prm1D deletion cassette
in which the KanMX6 gene (Bähler et al. 1998) had been
cloned between the 59- and 39-noncoding regions was used
to transform the strains of interest. Correct integration was
always assessed by PCR. Site-directed mutagenesis was used
to introduce a NotI restriction site immediately upstream of
the stop codon. The GFP and mCherry proteins were cloned
as a NotI/NotI DNA fragments; fusion proteins were functional, according to the complementation of the fusion defect.
All tagged proteins were integrated at the chromosome and
expressed under the control of their endogenous promoters.
A LactC2–GFP fusion protein was constructed by ligating
a PCR-amplified DNA fragment containing the C2 discoidinlike domain of bovine lactadherin fused to the GFP from the
LactC2–GFP–p416 plasmid (Addgene; Yeung et al. 2008) to
PCR-amplified act1+ 59- and 39-untranslated regions. The
construct was cloned into a KS+ vector carrying the NatMX6
gene, which produces resistance to nourseothricin (Sato et al.
2005) and was then integrated at the act1+ locus.
Semiquantitative reverse transcriptase PCR
Total RNA was prepared from h90 strains incubated in YES
medium (vegetative growth) and in EMM–N (mating conditions)
Fission Yeast Prm1p and Cell Fusion
1061
for 4 hr. RNA was extracted using the TRISure reagent (Bioline) according to the manufacturer’s protocol and treated
with DNase I (amplification grade, Invitrogen). First-strand
cDNA was synthesized from 15 mg total RNA using SuperScript first-strand synthesis system for RT–PCR (Invitrogen).
A total of 375 ng of the synthesized cDNA was used as templates in PCR. PCR was performed using Taq polymerase
(Bioline) as follows: denaturation at 94° for 30 sec, annealing
at 51° (for dni1+ and dni2+) or at 57° (for prm1+ and act1+)
for 30 sec, and extension at 72° for 70 sec for 18 cycles (for
prm1+ and act1+) or for 22 cycles (for dni1+ and dni2+).
Microscopy
For fluorescence microscopy, images were captured with either
a Leica DM RXA conventional microscope equipped with
a Photometrics Sensys CCD camera, using the Qfish 2.3
program, or with an inverted Olympus IX71 microscope
equipped with a personal DeltaVision system and a Photometrics CoolSnap HQ2monochrome camera. In the latter case,
stacks of z-series sections were acquired at 0.2-mm intervals.
Unless stated otherwise, fluorescence images are maximum
two-dimensional projections of the 6 z-series that corresponded to the middle of the cell and were analyzed using
deconvolution software from Applied Precision. Time-lapse
experiments were performed at 28° by acquiring epifluorescence cell images in single planes with the aforementioned
inverted microscope using a PlanApo 1003/1.40 IX70 objective; images were restored and corrected by 3D Deconvolution
(conservative ratio, 10 iterations, and medium noise filtering)
through softWoRx imaging software. Images were processed
with Adobe Photoshop and IMAGEJ (National Institutes of
Health). To assess zygote lysis, mating mixtures were exposed
to methylene blue (0.02% final concentration) and photographed immediately under a bright-field microscope; darkcolored zygotes were scored from the photographs. Staining
with Alexa-Fluor 568-Phalloidin (Invitrogen), filipin
(Sigma), Calcofluor (Blankophor; Bayern), and Hoechst
33258 (Sigma) were performed as described (Marks and
Hyams 1985; Wachtler et al. 2003; Sharifmoghadam et al.
2012). In the case of filipin, cells were stained with 5 mg/ml of
the complex and photographed immediately, such that cells
were exposed to the antibiotic for a total time of ,5 min. A
LactC2–GFP fusion protein, which binds to phosphatidylserine
(Yeung et al. 2008), was used to visualize the plasma membrane. Electron microscopy was performed on cells that had
been growing on EMM plates for 12 hr at 32°; cells were fixed
in glutaraldehyde (EM grade, Sigma) and stained with permanganate, as described (Neiman 1998). Ultrathin sections
were cut on a Jung Reichert microtome (Leica Mikroskopie
and System GmbH) and examined using a JEM1010 transmission electron microscope (Jeol) at 100 kV.
Statistical analyses
Data are presented as means 6 SD (standard deviation),
averaged for a minimum of three replicates of the experiments. Data were analyzed with Student’s t-test. A P-value
1062
M.-Ángeles Curto et al.
of ,0.05 was accepted as indicating a statistically significant
difference as compared with controls.
Results
prm1+ is induced during mating and Prm1p localizes
to the tip of mating cells
The timing of prm1+ expression was similar to that of fus1+,
map4+, dni1+, and dni2+, although the level of induction
was considerably lower (data obtained from Mata et al.
2002; Mata and Bahler 2006); this was probably the reason
why prm1+ was not identified in genome-wide analyses. To
confirm these results, semiquantitative reverse transciptase
PCR analyses were performed. As shown in Figure 1A,
prm1+ is expressed during vegetative growth and its expression increases under mating conditions (4 hr in EMM–N). In
agreement with previous results (Mata et al. 2002), our data
confirmed that prm1+ is induced to a lesser extent than
dni1+ and dni2+ in nitrogen-devoid medium. To investigate
whether Prm1p was produced under mating conditions, an
h90 prm1D mutant bearing the Prm1–GFP and Dni1–
mCherry fusion proteins was induced to mate in EMM–N
for 4 hr. Observation of zygotes by fluorescence microscopy
revealed that the Prm1p protein localized to the point of
contact between the mating partners (Figure 1B) and that
the protein colocalized with Dni1p; this colocalization did
not depend on cell–cell contact, since it was observed in
a heterothallic strain, bearing Prm1–mCherry and Dni1–
GFP, treated with the P factor pheromone (Figure 1B,
bottom).
prm1+ is required for cell fusion
The above results suggested that prm1+ might be involved
in mating. To confirm this hypothesis, h90 wild-type and
prm1D cells were induced to mate on EMM plates that were
incubated for 24 hr at 32°. Observation under a bright-field
microscope revealed that .90% of the zygotes had produced asci in the cultures from the control strain. In the
cultures from the mutant an accumulation of prezygotes
was observed, such that ,10% of the mating pairs had
formed asci (not shown). A similar result was obtained
when cells were induced to mate on sporulation agar
(SPA) or malt extract agar (MEA) plates (not shown). To
analyze the mating process in more detail, h90 wild-type and
prm1D strains bearing the Psy1 t-SNARE fused to the GFP
and the Hht1 histone fused to the mCherry RFP were induced to mate in liquid EMM–N. This strain allowed the
observation of the plasma membrane and nuclei in vivo.
The number of prezygotes, zygotes, and asci was scored
for both strains at different times. Figure 2A (top) shows
several examples of each cell type scored. Prezygotes
exhibited plasma membrane between mating partners, narrow mating bridges, and two nuclei that were apart from
each other. The criterion to score a cell as a zygote was the
lack of plasma membrane separating both cell bodies; some
zygotes had a narrow mating bridge and unfused nuclei,
while others had a wide bridge and fused nuclei. Cells with
four nuclei were scored as asci. The percentage of each cell
type was calculated with respect to the number of prezygotes, plus zygotes, plus asci. As shown in Figure 2A (bottom), in the control strain there was a progression along
time, such that the number of prezygotes decreased while
the number of zygotes increased; 22 hr after nitrogen depletion most of the mating cells were asci. In the prm1D
strain, the number of prezygotes remained high (.95%)
along the experiment. Twenty-two hours after nitrogen depletion, ,5% of the mating cells was asci.
This result showed that fission yeast Prm1p plays a
relevant role in mating and strongly suggested that cell
fusion was blocked in most prm1D zygotes. To confirm this
hypothesis, a soluble GFP expressed under the control of
map4+ was introduced in the wild-type and prm1D homothallic strains. The map4+ gene is mating-type specific, such
that it is expressed only in h+ cells (Mata and Bahler 2006;
Sharifmoghadam et al. 2006). Thus, if cell fusion had taken
place, the green fluorescence would have been observed
throughout the whole zygote body, whereas if fusion had
been defective, only half of the zygote would have exhibited
fluorescence. The strains were induced to mate on EMM
plates for 12 hr. In the control strain the whole zygote was
fluorescent in .90% of the zygotes (an example is shown in
Figure 2B). In the prm1D strain, only 4% of the zygotes
exhibited this pattern of fluorescence, while 96% of the
zygotes exhibited fluorescence in only one part of the zygote
body. This result confirmed a correlation between the lack of
sporulation and a defect in cell fusion in the prm1D mutant.
Out of the 96% of mutant prezygotes, an 85% exhibited
fluorescence in half of the zygote body (see cell a in Figure
2B), while 11% of the prezygotes exhibited fluorescence in
more or in less than half of the zygote body (see cells b–d in
Figure 2B; see also Supporting Information, Figure S1). This
observation suggested that intercellular bubbles/fingers,
like those described for the S. cerevisiae and N. crassa prm1D
mutants (Heiman and Walter 2000; Fleissner et al. 2009),
had been produced.
The distribution of cell wall and plasma membrane at
the cell–cell contact area of prm1D zygotes is abnormal
When the mating process was analyzed in strains bearing
fluorescent membrane and nuclei markers (Figure 2A), it
was found that most of the control and prm1D prezygotes
exhibited a flat plasma membrane separating the mating
partners (an example is denoted by an arrow in Figure
3A). However, in the mutant strain a significant number of
prezygotes exhibited an accumulation of membranous material (denoted by arrowheads in Figure 3A), membrane
bubbles/fingers (asterisks in Figure 3A), and cell compartments delimited by two membranes (denoted by a cross)
that were sometimes accompanied by a bubble. The number
of prezygotes with abnormal membranous structures at
the cell–cell contact area reached 40% of total prezygotes
4 hr after nitrogen depletion and decreased at later time
points.
The fact that the highest number of zygotes with an
abnormal membrane distribution correlated with the time at
which Prm1–GFP was observed at the tip of the shmoos is in
agreement with the hypothesis that Prm1p would be required for proper membrane organization and/or behavior
at the cell fusion region. To confirm that aberrant membrane
distribution was a consequence of the lack of Prm1p, and
not of an abnormal behavior of the GFP–Psy1 fusion protein,
a similar experiment was performed using a different membrane marker. The C2 discoidin-like domain from bovine
lactadherin fused to the GFP (LactC2–GFP; see Materials
and Methods) was used to visualize the plasma membrane.
This domain binds phosphatidylserine, a phospholipid
enriched in the cytoplasmic leaflet of plasma membrane. To
determine whether extensive cell-wall digestion was a prerequisite for the formation of membrane bubbles/fingers
h90, h90 prm1D, and h90 fus1D strains were induced to mate
in EMM–N and the cells were stained with Calcofluor and
photographed. Figure 3B (left) shows the typical morphology of a wild-type prezygote (with flat cell wall and plasma
membranes separating both cell bodies—cell a), and a wildtype zygote (in which the cell wall was digested and the
membranes fused—cell b). In all the prm1D prezygotes
(mating pairs with unfused plasma membranes) there was
cell-wall material at the intercellular contact area, and in
40% of the prezygotes there was an abnormal distribution
of the plasma membrane. Figure 3B (center) shows different
prm1D prezygotes; in prezygotes a and b one of the mating
partners exhibits a flat membrane (denoted by arrows)
while the other partner exhibits wide membrane invaginations into the cell body. Cell-wall material separates the
mating cells and partially surrounds the membrane invaginations (arrowheads). Prezygotes in which one mating partner exhibits a flat plasma membrane while the other exhibits
blebs and membrane invaginations have been described for
the S. cerevisiae kex2D prm1D double mutant (Heiman et al.
2007). Prezygotes c and d exhibit accumulations of membrane and cell-wall material at the intercellular contact area.
The formation of membrane invaginations and fingers and
the accumulation of cellular envelopes could not have been
an indirect consequence of the halt in cell fusion, since they
were never observed in the fus1D mutant (Figure 3B,
right).
We performed electron microscopy to further characterize the cell fusion defect in prm1D zygotes (see Figure 4).
h90 and h90 prm1D cells were induced to mate on EMM
plates for 12 hr. As it has been described (Calleja et al.
1977), in the h90 strain the mating partners established contact through their cell walls (photograph a in Figure 4, left);
later, the two-layer cross wall thinned so that plasma membranes from both mating cells approached each other (arrowhead in photograph b) and finally fused (photograph c).
In the h90 prm1D strain, we did not detect regions of intercellular contact in which the cell wall had been digested
Fission Yeast Prm1p and Cell Fusion
1063
Figure 2 prm1D cells have a defect in cell fusion. (A) h90 and h90 prm1D cells bearing GFP–Psy1 and Hht1–mCherry fusion proteins to label the plasma
membrane and nuclei, respectively, were induced to mate in liquid EMM–N medium for 4 hr and photographed at different times; prezygotes, zygotes,
and asci were quantified from the photographs. Top row: phase contrast (PC) and fluorescence micrographs of several examples of each cell type
scored. Bottom rows: percentage of prezygotes, zygotes, and asci in the h90 (WT, left) and h90 prm1D (right) strains. The mating efficiency, M(%),
calculated as the percentage of prezygotes plus zygotes and asci with respect to the number of prezygotes plus zygotes, asci, and cells, is shown for
each time point in the graph. (B) Phase contrast, fluorescence, and merged micrographs of h90 (top set of photographs) and h90 prm1D zygotes
expressing a soluble GFP from the map4+ promoter. Scale bars, 3 mm.
and the membranes were apposed (like those described for
the S. cerevisiae and Neurospora prm1D mutants. Heiman
and Walter 2000; Fleissner et al. 2009). In 30% of
S. pombe prm1D zygotes (Figure 4, right), there were regions
where the cross wall had been partially digested and regions
where it had thinned (denoted by arrowheads in the magnifications of photographs a and b). In the prm1D prezygote
shown in photograph a, the cell seemed to have repaired an
opening or a region were the cross wall had been partially
eroded. Plasma membrane outgrowths into the cytoplasm
(denoted by arrows in the magnifications of photographs
d, e, and f) and structures that seemed to be small plasma
membrane bubbles (denoted by arrows in the magnifications of photographs b, c, and d) were frequent in the proximity of the cell–cell contact point. The block arrow in
photograph d denotes structures that resembled the cellwall-adhered blebs described in the S. cerevisiae double
prm1D kex2D mutant (see Heiman et al. 2007, Figure 6F
and Figure 9), which also accumulated small membrane
1064
M.-Ángeles Curto et al.
bubbles at the intercellular contact area even when the cross
wall was present. In .80% of the prm1D zygotes, it was
possible to observe an abnormal cell-wall growth that surrounded portions of cytoplasm (photographs a, d, e, and f).
Since many of the enzymes required for cell-wall synthesis
are plasma membrane proteins, this abnormal growth of the
cell wall must have been preceded by an aberrant membrane growth.
Distribution of cell-wall enzymes in prm1D zygotes
As described above, in most S. pombe prm1D zygotes the
cross wall separating the mating partners was not extensively digested. To understand whether this phenotype
was due to a mislocalization of glucanases, h90 and h90
prm1D cells bearing the glucanase Agn1–GFP were induced
to mate in liquid medium. As shown in Figure 5A, this enzyme was present at the cell fusion point. Additionally, a defect in the action of glucanases cannot be invoked to explain
the lack in cell fusion in the S. pombe prm1D mutant, since
Figure 3 prm1D zygotes exhibit an abnormal cell wall and plasma membrane distribution at the cell–cell contact area. (A) Fluorescence micrographs of
h90 prm1D zygotes bearing the GFP–Psy1 and Hht1–mCherry fusion proteins. The arrow denotes a flat plasma membrane separating mating partners;
arrowheads denote accumulations of membranous material; asterisks denote membrane bubbles/fingers, and crosses denote cell compartments
delimited by two flat plasma membranes. (B) h90 (WT), h90 prm1D and h90 fus1D zygotes bearing the LactC2–GFP fusion protein to label internally
exposed phosphatidylserine in the plasma membrane were stained with Calcofluor to label the cell wall. Micrographs show the cell wall (magenta),
plasma membrane (green), and the superimposed images. For the wild-type strain (a) a prezygote and (b) a zygote are shown. For the prm1D strain, four
prezygotes are shown; arrows denote a flat plasma membrane at the cell–cell contact area of prezygotes exhibiting an invagination/finger, and
arrowheads denote cell-wall material around invaginations/fingers. Images were obtained with a DeltaVision Deconvolution system. For the fus1D
mutant, two prezygotes are shown. Scale bars, 3 mm.
such a defect would result in the presence of a flat, perhaps
thickened, cell wall between mating partners, but not in the
overgrowths of this structure into the cytoplasm observed in
the prm1D zygotes (cells a and b in Figure 3B and Figure 4).
By contrast, these overgrowths must have been produced by
the synthesis of new cell wall. In agreement with this notion,
the b(1,3)glucan synthase Bgs4p was observed around some
membrane inavginations/bubbles that had formed into the
cytoplasm of one of the mating partners (Figure 5B). The
synthesis of new cell wall from the membrane ingrowths
and membrane invagination/fingers would contribute to
the engulfment of portions of cytoplasm by cell-wall material, giving rise to some of the structures observed by electron microscopy (Figure 4).
Cellular envelopes show abnormal behavior during cell
fusion in prm1D zygotes
To obtain more detailed information about the chronology
of the events that led to the cell fusion defect in prm1D
zygotes, time-lapse experiments were performed using the
h90, h90 prm1D, and h90 fus1D strains bearing the LactC2–
GFP construct. Cells were induced to mate in liquid EMM–N,
stained with Calcofluor, and photographed every 4–5 min
(Figure 6, Figure S2, File S1, and File S2). In h90 wild-type
zygotes, the fluorescence corresponding to the phosphatidylserine in the inner leaflet of the plasma membrane lost intensity at the cell–cell contact point (denoted by an asterisk;
time point 359 in Figure 6A). This change in the appearance
of the plasma membrane was observed before a partially
eroded cross-wall region was observed (denoted by an arrow; time point 409). At later time points, an opening in the
cross wall was apparent and became wider with time, the
plasma membrane was no longer present between the mating partners, and the conjugation bridge grew (Figure 6A
and File S1).
In the prm1D zygote shown in Figure 6B and File S2, the
fluorescence of the LactC2 probe in the membrane separating the mating partners (denoted by an asterisk in the time
frame 359) was strong throughout the time course; at 409,
a partially eroded cross-wall region was observed (denoted
by an arrow). At 459, an opening in the cross wall became
evident due to a thickening of the cell wall surrounding it,
but it did not become wider at later time points. At this time
point (459), the plasma membrane invaginated into the
Fission Yeast Prm1p and Cell Fusion
1065
Figure 4 Electron microscopy of cellular envelopes at the cell fusion area. Transmission electron microscopy of h90 zygotes (WT; three different zygotes
are shown) and h90 prm1D zygotes (six different zygotes are shown). The micrographs show the intercellular contact regions. For the prm1D strain,
regions of interest (delineated with dotted lines) have been magnified. Arrowheads denote areas of partial cell-wall erosion and arrows denote
abnormal membrane structures. Scale bar, 0.75 mm.
cytoplasm of one of the mating partners giving rise to a structure that collapsed or retracted at 509. After this time, the
membrane separating the mating partners resembled that of
the initial time point. The cell repaired the eroded cross-wall
region 559 after the initial time point, after which the cross
wall thickened. Although in this zygote the membrane invagination was produced soon after the cross wall had undergone partial erosion, in other zygotes there was a delay
between the time of cross-wall erosion and the appearance
of the membrane invagination. Thus, in the zygote shown in
Figure S2 (top), the cross wall partially eroded region was
already present at the beginning of the experiment (time
point 09; denoted by an arrow) and the membrane invagination was not observed until the 129 time point. In the
zygote shown in Figure S2 (bottom), the cross-wall erosion
was produced before the 09 time point (denoted by an arrow) and was closed at 249 (denoted by an arrowhead); the
membrane invagination was observed at 489, preceded by
an outgrowth of the cell wall at the mating bridge (denoted
1066
M.-Ángeles Curto et al.
by a square). Thus, the formation of the membrane invaginations was apparently not an immediate consequence of
the partial erosion of the cross wall. The membrane invaginations observed in the zygotes shown in Figure 6B and
Figure S2 collapsed/retracted soon after their appearance,
indicating that these structures were transient and probably
fragile. In 20% of the prm1D zygotes the membrane invaginations remained for longer times and cell-wall material was
synthesized from them (File S3). In all the prm1D zygotes
observed, the intensity of the fluorescence corresponding to
the phosphatidylserine in the cytoplasmic side of the plasma
membrane at the intercellular contact region was strong
throughout the experiments, and the fluorescence corresponding to the cross wall thickened after the closure of
the opening at that region.
In sum, according to the observation of 6 zygotes from
the wild-type and 10 zygotes from the prm1D mutant by
time-lapse microscopy, it could be established that the initial
difference between both strains that could be observed was
temperature sensitive (Clemente-Ramos et al. 2009) and
hence the influence of temperature on the cell fusion efficiency of h90, h90 dni1D dni2D, and h90 prm1D strains was
compared. Sporulation efficiency (percentage of asci with
respect to prezygotes plus asci) was used as an indirect
measurement of cell fusion for the following reasons:
Figure 5 Distribution of cell-wall enzymes in zygotes. (A) Agn1–GFP
localization in the indicated strains that had been induced to mate in
liquid EMM–N. Arrows denote Agn1p accumulation at the cell–cell contact area of zygotes. Images were captured with a conventional fluorescence microscope. (B) Localization of LactC2–GFP and mCherry–Bgs4p in
prm1D zygotes. Arrows denote an intracellular bubble/finger. Images are
maximal projections of z-stacks obtained with a DeltaVision Deconvolution system. Scale bars, 3 mm.
the behavior of the phosphatidylserine marker associated to
the cytoplasmic leaflet of the plasma membrane at the cell–
cell contact point. In the fus1D mutant, the mating partners
remained adhered but unfused and the mating bridge became longer with time. In this strain, the fluorescence corresponding to the LactC2–GFP marker weakened some minutes
after cell–cell contact (asterisk in the time point 359) but no
erosion in the cell wall or membrane invaginations were
observed along the time course (Figure 6C and data not
shown).
To confirm that a change in the distribution of phosphatidylserine preceded cell fusion in the control strain, h90
wild-type cells bearing the LactC2–GFP marker and
mCherry–Bgs4 (with 16 transmembrane domains) were induced to mate in liquid medium and early zygotes were
photographed. Serial z-sections were analyzed; as shown
in Figure 6D, the fluorescence corresponding to the
LactC2–GFP probe was excluded from a point at the cell–
cell contact area, while that of the mCherry–Bgs4 was not. A
similar result was obtained when mCherry–Psy1 (with one
transmembrane domain) was used (Figure 6D). Exclusion of
the LactC2–GFP fluorescence at the fusion area was never
observed in prm1D prezygotes (not shown).
prm1+ and dni+ genes act in different subpathways
The defects in the distribution of the plasma membrane and
cell wall in the area of cell–cell fusion in prm1D zygotes
were reminiscent of those observed in the dni1D mutant
(Clemente-Ramos et al. 2009). Accordingly, the functional
relationship between prm1+ and the dni+ genes was analyzed. The cell fusion defect in dni1D and dni2D mutants is
1. In S. pombe, sporulation proceeds immediately after cell
fusion; in the wild-type strain, the number of fused
zygotes that had not sporulated was negligible 22 hr after
nitrogen depletion (Figure 2A).
2. We had observed that dni1D and prm1D zygotes had
a defect in fusion (Clemente-Ramos et al. 2009, Figure
2B, and Figure S1).
3. In the dni1D and prm1D mutants, fused zygotes that had
not sporulated were almost never detected (Figure 2A
and data not shown), such that the percentage of
asci and the percentage of prezygotes were always
complementary.
Cells were induced to mate on EMM plates and were
incubated for 3–5 days at different temperatures; long incubation times were used to ensure that all asci were mature
so that they would be easily recognized under the brightfield microscope because of spore refringency. As shown in
Figure 7A, sporulation in the double dni1D dni2D mutant
was a thermosensitive process; sporulation efficiency in this
strain was fairly similar to that of the wild type at 20° and
was very low at 32°. In contrast the sporulation efficiency of
the h90 prm1D mutant was very low at all temperatures
tested, suggesting that the nature of the fusion defect in
prm1D and dniD mutants might be different.
The sporulation efficiency in a triple dni1D dni2D prm1D
mutant could not be compared with that of the single prm1D
mutant to investigate whether the prm1+ and the dni+
genes acted in the same or in different subpathways because
the defect in the single prm1D mutant was already close to
100%. Accordingly, several experiments were undertaken to
clarify this issue. First, sporulation was analyzed in prm1D
cells that had been transformed with an empty plasmid or
with plasmids overexpressing prm1+, dni1+, or dni2+. As
shown in Figure 7B, sporulation was restored only by the
plasmid overexpressing prm1+. Similarly, prm1+ overexpression did not restore sporulation in dni1D, dni2D, and
dni1D dni2D mutants (not shown). These results suggested
either that prm1+ and the dni+ genes acted in different
subpathways or that these genes acted at the same point
of the same subpathway so that they regulated each other.
To investigate whether Prm1p and the Dni proteins formed
a complex, the dependence of these proteins on each other
for their localization to the cell–cell contact area was analyzed. As shown in Figure 7C, Prm1–GFP localized to the
intercellular contact region in a dni1D dni2D mutant. Similarly, Dni1–GFP and Dni2–GFP were localized to the intercellular contact region of early prm1D zygotes (zygotes with
a short, narrow mating bridge; Figure 7C). In late zygotes
(zygotes with a long, wide, mating bridge), the fluorescent
Fission Yeast Prm1p and Cell Fusion
1067
Figure 6 Plasma membrane and cell-wall behavior during cell fusion. (A) h90 WT, (B) h90 prm1D, and(C) h90 fus1D zygotes bearing the LactC2–GFP
marker to label internally exposed phosphatidylserine in the plasma membrane were stained with Calcofluor to label the cell wall, induced to mate in
liquid EMM–N, and time-lapse-imaged every 5 min. The 09 time indicates the moment when the first photograph was taken. Since there was no
temporal marker to adjust the first image, the pictures from different zygotes labeled with the same time (in minutes) do not necessarily correspond to
the same moment after the initial cell–cell contact had taken place. Asterisks denote a change in the intensity of the LactC2–GFP fluorescence in the
fusion area, with respect to time 09, and arrows denote openings or eroded cross-wall areas. The insets are magnifications of the cell–cell fusion area. (D)
h90 WT prezygotes bearing the LactC2–GFP phosphatidylserine probe and mCherry-tagged Bgs4 (left) or Psy1 (right) plasma membrane proteins. Images
are medial sections obtained with a DeltaVision Deconvolution system. Scale bars, 3 mm.
signal of both proteins was broader, due to the expansion of
the cell–cell contact area after cell fusion failure in the mutant. Thus, Prm1p and the Dni proteins are not interdependent for their localization. Finally, the sporulation efficiency
in unilateral and bilateral crosses was estimated. Sporulation in unilateral WT 3 prm1D crosses was more efficient
1068
M.-Ángeles Curto et al.
than in bilateral prm1D 3 prm1D crosses (Figure 7D), showing that the presence of Prm1p in one mating partner was
sufficient to promote cell fusion at almost wild-type levels.
The result was similar when the prm1D strain was h+ and h2.
The same kind of behavior was found for the dni1D, dni2D,
and fus1D mutants (Petersen et al. 1995; Clemente-Ramos
Figure 7 prm1+ and the dni+ genes act in different subpathways. (A) Percentage of sporulation (percentage of asci with respect to prezygotes plus asci)
in the indicated homothallic strains incubated on EMM plates at different temperatures. (B) Bright-field micrographs of mating mixtures of h90 prm1D
cells bearing an empty vector and plasmids overexpressing the indicated genes. Arrows denote asci with mature spores. (C) Localization of the indicated
fusion proteins in zygotes of homothallic strains bearing the indicated gene deletions. The insets corresponding to the prm1D mutant show early zygotes
while the main photographs show late zygotes with long, wide, mating bridges. (D) Percentage of sporulation in mating mixtures involving heterothallic
h+ and h2 strains bearing the indicated gene deletions. Mating was performed on EMM plates incubated at 32° for 3 days. The experiments were
performed a minimum of three times with duplicates. The mean, standard deviation, and statistical significance of the relevant differences are shown
(*, P , 0.05; #, P , 0.01). Scale bars, 3 mm.
et al. 2009; Figure 7D, and data not shown). The sporulation
efficiency of the unilateral WT 3 dni2D prm1D cross was significantly lower than that of the unilateral WT 3 dni2D and
WT 3 prm1D crosses (Figure 7D), suggesting that prm1+ and
dni2+ acted in different genetic subpathways for cell fusion.
This hypothesis was confirmed by analyzing the sporulation
efficiency in unilateral dni2D 3 prm1D crosses. On comparing
sporulation efficiency in the unilateral dni2D 3 prm1D cross
with that of the bilateral WT 3 WT, dni2D 3 dni2D, and
prm1D 3 prm1D crosses and with that of the unilateral
dni2D 3 WT and prm1D 3 WT crosses, we found that it was
similar to that of the unilateral crosses (Figure 7D). This result
confirmed that dni2+ and prm1+ act in parallel subpathways,
since if both genes acted at different points of the same subpathway, the result of the unilateral dni2D 3 prm1D cross would
have been similar to that of the bilateral prm1D 3 prm1D cross.
Similar results were obtained when dni1D 3 prm1D and
fus1D 3 prm1D unilateral crosses were analyzed (not shown).
In S. pombe, two subpathways have been shown to participate in the functional pathway required for cell fusion
during mating. One of the subpathways includes the formin Fus1p and the Fig1-related Dni proteins (ClementeRamos et al. 2009). As shown in Figure 7, prm1+ and the
dni+ genes acted in different subpathways. Similarly, analyses of unilateral and bilateral crosses, of actin distribution
at the tip of the shmoos, and of the distribution of Prm1–
GFP in a fus1D mutant confirmed that fus1+ and prm1+
acted in different subpathways (Figure S3 and data not
shown). For the other subpathway, only the Golgi protein
Cfr1p has been identified; in cfr1D mutants the cross wall
between mating partners is not digested, but no abnormal
outgrowth of the cell wall was detected (Cartagena-Lirola
et al. 2006). Unilateral and bilateral crosses directed toward investigating whether prm1+ and cfr1+ belonged to
the same subpathway were performed. The results were
not significant because the fusion defect was very strong
in the prm1D mutant and mild in the cfr1D mutant. In any
case, microscopy analyses revealed that Prm1–GFP reached
the cell surface efficiently in the cfr1D mutant (Figure S3).
These results suggest that prm1+ might define a third
Fission Yeast Prm1p and Cell Fusion
1069
functional subpathway for the cell fusion process in fission
yeast.
prm1D and dni1D mutants are slightly sensitive to drugs
that inhibit lipid synthesis and drugs that induce cell
death after interfering with lipids
The cell fusion defect in the prm1D mutants was concomitant with an abnormal behavior and distribution of the
plasma membrane at the intercellular contact area. This result suggested that Prm1p might play some role in the organization/structure/dynamics of the plasma membrane.
prm1+ was expressed during vegetative growth (Figure
1A); thus, Prm1p function might be also performed under
these conditions. To gain information about the nature of
Prm1p function in S. pombe, h90 cells were treated with
different lipid-disturbing drugs during vegetative growth
and during mating. The dni1D mutant was included in the
analyses.
When h90 wild-type, prm1D, and dni1D cells from logarithmic cultures were spotted onto YES plates supplemented with miconazole, a drug that inhibits the synthesis
of sterols, it was found that prm1D cells were partially
sensitive to the drug (Figure 8A); the dni1D mutant exhibited a sensitivity that was intermediate between that
of the control and the prm1D strain. The filipin complex
is a macrolide polyene antibiotic that binds ergosterol, disturbs the structure of the sterol-rich membrane domains
termed lipid rafts, and destabilizes plasma membrane proteins (Wachtler et al. 2003). Although filipin seems to bind
the sphingolipid-free pool of ergosterol (Jin et al. 2008),
the hydrophobic environment provided by sphingolipids is
required for filipin binding to sterols (Drabikowski et al.
1973); thus, sensitivity to this antibiotic might reflect
defects in the amount and organization of lipids at the
plasma membrane. When h90 wild-type, prm1D, and dni1D
cells growing in logarithmic phase were incubated in the
presence of filipin in liquid medium (see Materials and
Methods), it was found that dni1D vegetative cells were
partially sensitive to the antibiotic and that the sensitivity
of the prm1D cells was intermediate between that of the
control and the dni1D cells (Figure 8B). These results suggest that the amount or the organization of sterols is defective in the prm1D and dni1D strains and that the defects
in these mutants are different. During vegetative growth
neither of the mutants was sensitive to myriocin, an inhibitor of sphingolipid biosynthesis.
h90 cells were also treated with Duramycin and papuamide B, which are cytotoxic peptides that respectively bind
extracellularly exposed phosphatidyl ethanolamine and
phosphatidylserine and induce cell lysis (Marki et al. 1991;
Parsons et al. 2006). Normally, flippases maintain these
phospholipid components on the inner leaflet of the plasma
membrane; thus, an increased sensitivity to Duramycin and
to papuamide B would reflect a defective phospholipid organization in the plasma membrane. Cells that had been
growing actively in EMM and cells that had been induced
1070
M.-Ángeles Curto et al.
to mate in liquid EMM–N for 1 hr were exposed to the
solvent or the peptides for an additional 5 hr and then
plated onto YES medium, to analyze cell survival. In the
mock experiments (0 mg/ml papuamide B), all the strains
grew at the same rate on the plates (Figure 8C), suggesting
that under these conditions lysis was negligible in the
mutants. dni1D cells were slightly sensitive to papuamide
B during vegetative growth; sensitivity was enhanced when
the peptide was added to cells undergoing mating (Figure 8
C). This result suggested that the plasma membrane has
more externally exposed phosphatidylserine in the dni1D
mutant than in the wild-type strain and that this difference
is enhanced in cells undergoing mating. No reproducible
differences were found between the survival of dni1D and
control cells treated with Duramycin (not shown) and between the survival of prm1D and control cells treated with
either peptide (Figure 8C and data not shown).
When h90 strains were induced to mate in liquid EMM–N
and then plated onto EMM plates containing myriocin so
that cells that had initiated mating (and presumably had
undergone a change in their membranes) were exposed to
myriocin, it was found that the prm1D cells were more sensitive to this antibiotic than wild-type and h90 dni1D cells
(Figure 8D). Under mating conditions, no sensitivity to
miconazole or filipin was detected in any of the mutants.
To gain information about the extent of the membrane
defects in the prm1D and dni1D strains, heterothallic cyr1D
sxa2D cells were treated with pheromone and stained with
filipin to observe the distribution of sterols. cyr1D sxa2D
strains bearing the LactC2–GFP construct were treated
with pheromone to observe the phosphatidylserine on the
cytoplasmic leaflet of the plasma membrane. The filipin
and LactC2–GFP fluorescent signals were stronger along
the shmoo projections than in the cell bodies in both
strains. No apparent difference was observed between the
control, dni1D, and prm1D strains (Figure 8E and data not
shown). These results indicate that the plasma membrane
defects in the mutant strains are mild. Additionally, no
abnormal membranous structures or intracellular expansion of the plasma membrane were observed at the tip of
the shmoos in the mutant strains, showing that the abnormalities detected in the membrane of the mutant zygotes
were produced after cell–cell contact. The patterns of filipin staining and the LactC2–GFP signal were similar in
wild-type, prm1D, and dni1D strains during vegetative
growth (not shown).
All these results suggest that the composition/structure
of the plasma membrane is different during vegetative
growth than during mating and that it is different in the
wild type, in the prm1D mutant, and in the dni1D mutant.
The mild abnormalities in the plasma membrane of the mutant strains unveiled using lipid-disturbing drugs do not
seem to produce alterations during vegetative growth and
shmoo differentiation, but might be responsible for, or at
least contribute to, the failure in cell fusion during mating
in the prm1D and the dni1D mutants.
Figure 8 prm1D and dni1D mutants are sensitive to compounds that interfere with lipids. (A) 3 3 104 cells and serial 1:4 dilutions from the indicated
strains were spotted onto YES plates and YES plates supplemented with the indicated amount of miconazole and incubated at 32° for 2 days. (B) 3 3
104 cells and serial 1:4 dilutions from the indicated strains were incubated in liquid YES medium containing the indicated amounts of filipin for 2 days.
The y-axis represents the OD600nm of the cultures from the cell dilutions indicated on the x-axis. (C) Cells from the indicated strains were exposed to the
indicated amounts of papuamide B for 5 hr when they were growing actively in EMM (vegetative) or undergoing mating in EMM–N. After this time, 3 3
104 cells and serial 1:4 dilutions were spotted onto YES plates and incubated at 32° for 2 days to assess cell survival. (D) 3 3 104 cells and serial 1:4
dilutions from the indicated strains that had been induced to mate in liquid EMM–N were spotted onto EMM plates and EMM plates supplemented with
the indicated amount of myriocin and incubated at 32° for 2 days. The experiments in A–D were performed a minimum of three times with duplicates;
a representative example is shown for each experiment. (E) Left: cyr1D sxa2D control and prm1D haploid cells were treated with pheromone, stained
with filipin complex (5 mg/ml) and photographed immediately with a conventional fluorescence microscope. Right: cyr1D sxa2D control and prm1D
haploid cells bearing a LactC2–GFP construct were photographed with a DeltaVision Deconvolution system; the images correspond to medial sections.
Scale bar, 3 mm.
prm1D zygotes do not lyse after cell–cell contact
Membrane merger and cell-wall remodeling must be correctly cross-regulated during fusion to avoid cell lysis. A
significant number of S. cerevisiae prm1D zygotes undergo
a contact-dependent lysis that is enhanced after decreasing
the external calcium concentration (Jin et al. 2004; Aguilar
et al. 2007). It has been described that decreasing the external calcium concentration does not affect either cell fusion efficiency or zygote integrity in S. pombe wild-type and
dni1D strains (Clemente-Ramos et al. 2009). To gain further
information about the influence of external calcium concentrations in cell fusion and zygote integrity in the S. pombe
prm1D mutant, homothallic wild-type, dni2D, prm1D, dni2D
prm1D, and fus1D mutants were induced to mate in liquid
EMM–N medium (about 100 mM calcium), EMM–N without
calcium (traces of calcium), and EMM–N without calcium
supplemented with BAPTA. After 4 hr of incubation, zygotes
were exposed to methylene blue and photographed immediately; dark-colored zygotes were scored from the photographs. In all the strains and conditions tested, zygote lysis
was ,10% (data not shown). The same result was obtained
when the cells were incubated under those conditions for
24 hr. Thus, neither calcium depletion nor the elimination
of a protein from the Fus1–Dni subpathway produced irreversible damage to the membrane of the prm1D cells during
cell fusion.
Discussion
S. pombe Prm1p and cell fusion
The S. pombe prm1D mutant exhibits an almost complete
blockade in cell fusion, showing that its function is more
Fission Yeast Prm1p and Cell Fusion
1071
critical for this process than that of S. cerevisiae and N. crassa
PRM1 (Heiman and Walter 2000; Jin et al. 2004; Fleissner
et al. 2009). Analysis of zygotes bearing a soluble GFP suggested that the S. pombe prm1D mutants might produce
plasma membrane bubbles/fingers; however, electron microscopy revealed that the nature of the structures observed
in S. pombe might be different from that described in other
organisms. In ,5% of the S. pombe prm1D zygotes the cell
wall had been digested completely and cells had fused; in
the rest of zygotes a cross wall separated the mating pair. In
some zygotes, a localized erosion of the cross wall was
detected that sometimes seemed to have been repaired;
time-lapse analyses confirmed this observation.
It seems plausible that some kind of communication
exists between plasma membrane and cell wall. If this
communication is altered in the prm1D mutants, cells might
respond abnormally to some signal produced after the initial
cross-wall erosion. In this scenario, the outcomes produced
by the failure in membrane merger in different organisms
would be explained by the specific nature of these signals
and by the mechanism triggered after the initial cell–cell
contact to coordinate the remodeling of both cellular envelopes. S. pombe might trigger a signal to enhance cell-wall
synthesis to repair the erosion while S. cerevisiae and N.
crassa might favor the action of glucanases that would continue degrading the cross wall, and hence the areas of apposed membranes would become wider with time. These
different outcomes could help to explain the penetrance of
the cell fusion defect. In S. cerevisiae and N. crassa, after the
cell wall is degraded the membranes are apposed and
pushed one against the other by the turgor pressure; thus,
the prolonged contact between both membrane bilayers
would facilitate their merging after a time. In most S. pombe
zygotes, the small openings in the cross wall would be
repaired and the cell wall would provide a physical barrier
preventing membrane fusion; also, the small size and the
rapid repairing of the cell-wall openings might contribute to
the absence of zygote lysis.
Abnormal membrane and cell-wall structures at the
contact site of S. pombe prm1D prezygotes
Time-lapse experiments showed that large membrane invaginations/fingers formed at the area of intercellular contact
in S. pombe prm1D mating pairs; in most cases these structures collapsed/retracted a few minutes after their appearance. Electron microscopy revealed the presence of areas of
membrane growth toward the inside of the cells, small bubble-like structures, and portions of cytoplasm surrounded by
cell-wall material. Figure 9 illustrates some hypothetical
mechanisms that would explain the origin of these abnormal
structures:
1. After the formation of an opening in the cross wall, the
failure in membrane merger would allow turgor pressure
from the cytoplasm of one of the mating cells pushing
both membrane bilayers toward the inside of the cyto-
1072
M.-Ángeles Curto et al.
plasm of the mating partner with the weakest pressure,
giving rise to a finger (Figure 9A). This process would be
similar to that described for the S. cerevisiae prm1D mutant (Heiman and Walter 2000; Jin et al. 2004). The
pressure between both cells would become balanced,
the membranes would recover their natural position,
the cell-wall opening would be repaired, and the cell wall
at the contact point would thicken with time.
2. After the appearance of the finger, growth of the cell wall
on the same line as the old wall (directed from the membranes proximal to the edges of the gap) would repair the
cross-wall opening before the finger retracts, trapping
a portion of plasma membrane and originating a small
plasma membrane bubble in the cytoplasm (Figure 9B).
3. The synthesis of new cell wall on the same line as the old
wall together with cell-wall synthesis from the finger’s
plasma membrane would reinforce these structures.
These processes would generate portions of cytoplasm
engulfed by cell wall (Figure 9C, File S3, and Figure 4,
prm1D cells A, D, E, and F). Cell-wall regeneration at the
finger membrane has also been observed in the S. cerevisiae prm1D and erg6D mutants (Jin et al. 2008).
4. Although nothing is known about how the fungal cell
wall and plasma membrane are attached, it seems reasonable to think that some interaction exists between
cellular envelopes. Plasma membrane detachment from
the cell wall in one of the mating cells would produce
a membrane invagination into the cytoplasm, while
membrane would remain flat in the other cell (Figure
9D). Later, the membrane invagination might retract,
originate small bubbles, or direct the synthesis of new
cell wall.
The membrane detachment hypothesis is supported by
the facts that in several zygotes, membrane invaginations
were observed after the erosion in the cell wall seemed to
have been repaired (Figure S2, bottom, and data not shown)
and that in some zygotes, invaginations were present in one
mating partner while membrane was flat and proximal to
the cross wall in the other partner (Figure 3B, cells a and b,
Figure 5B, Figure S2, and data not shown). Serial sections
analyses confirmed that plasma membrane invaginated in
only one of the mating partners (Figure 9E and data not
shown). Additionally, while invaginations were observed in
40% of prm1D prezygotes, cytoplasm fingers (viewed using
the soluble GFP) were detected only in 11% of the prezygotes, strongly suggesting that cell-wall openings and membrane invaginations did not always coexist. Probably, several
forces/tensions arise at the cell–cell fusion point and these
forces undergo some change when the erosion of the cross
wall starts. If the membrane is not properly attached to the
cell wall, tension might be released through the invagination of plasma membrane into the cytoplasm of one of the
cells. This hypothesis requires that these forces are stronger
than turgor pressure, which pushes the membrane toward
the cell wall. If plasma membrane flexibility is altered in the
Figure 9 Possible models for the origin of the abnormal cell wall and membrane structures observed in S. pombe prm1D prezygotes. The drawings
depict the cell–cell contact area of prezygotes. (A) Erosion of the cross wall is followed by the appearance of a cytoplasmic finger into the opposite
mating cell; then, the finger retracts and the opening in the cell wall is repaired. (B) The finger is formed as in A, but the cell wall at the opening closes
trapping a portion of plasma membrane that originates a small bubble. (C) The finger is formed as in A, but it is reinforced by the synthesis of cell-wall
material and does not retract. (D) Partial erosion or the opening of a small gap in the cell wall results in the detachment of plasma membrane from the
cell wall in one of the mating partners; this detachment originates a membrane invagination that retracts, closes forming a small bubble, or is reinforced
by cell-wall synthesis. (E) Serial sections of a prm1D zygote labeled with the LactC2–GFP probe. Top: bright-field and fluorescence images of the whole
prezygote; the square delimits the area that has been enlarged (bottom). Scale bar, 3 mm. The asterisk depicts the cell whose membrane invaginated
and the arrow depicts the cell whose membrane remained flat. Numbers indicate z-sections. Images were captured with a DeltaVision system.
prm1D mutant (see below), it might be possible that it bends
abruptly toward the inside of the cytoplasm to release these
forces; later, turgor pressure would push it back toward the
cell wall. This phenomenon would be similar to that described in cell cultures when cells are abruptly detached
from a surface; in these circumstances, the plasma membrane forms transient blebs (Norman et al. 2010). While
in mammalian cells the membrane blebs toward the outside
of the cell, in the S. pombe prm1D zygotes invaginations
would be produced toward the inside of the cells because
of the cell-wall barrier.
Prm1, Fig1-related proteins, and
membrane organization
The composition and distribution of lipids in the plasma
membrane is important for cell fusion in S. cerevisiae; in
particular, the amount of sterols as well as their distribution/organization in the plasma membrane is relevant for
this process (Bagnat and Simons 2002; Proszynski et al.
2006; Jin et al. 2008; Aguilar et al. 2010; Grote 2010).
Prm1 and Fig1-related proteins are integral membrane
proteins and concentrate at the shmoo tip. The contactdependent lysis in the S. cerevisiae prm1D mutant is not
remedied by sorbitol (Jin et al. 2004) and is influenced by
calcium (Aguilar et al. 2007). In S. pombe, the cell-fusion
defect in the dniD mutants is influenced by temperature
(Clemente-Ramos et al. 2009). Since calcium and temperature affect membrane fluidity (Gordon et al. 1978), it is possible that different aspects of plasma membrane structure/
organization/dynamics might be regulated by the Prm1 and
Fig1-Dni proteins.
Experiments using the LactC2–GFP marker strongly suggested that in the wild-type phosphatidylserine is excluded
from the inner leaflet of plasma membrane at the cell–cell
contact sites before cell fusion. This result, together with the
fact that LactC2–GFP labels some parts of the shmoo differently than others (Figure 8E) suggests that normal cell
fusion involves dynamic changes in the amount and/or
Fission Yeast Prm1p and Cell Fusion
1073
distribution of this phospholipid by lateral segregation or
localized flipflop to the outer leaflet. This reorganization
of lipids seems to be altered in the S. pombe prm1D mutant.
Additionally, the sensitivity of the prm1D and dni1D mutants
to the filipin complex, which binds the sphingolipid-free
pool of ergosterol, and the sensitivity of the dni1D mutant
to papuamide B, which binds the externally exposed phosphatidylserine, suggest that the specific distribution/
topology of lipids is defective in prm1D and dni1D membranes; the lipid imbalance produced by a reduction in the
synthesis of ergosterol and sphingolipids by miconazole and
myriocin would enhance these defects resulting in sensitivity to those compounds. Loss of Prm1p and Dni1p function
has minor effects during vegetative growth but dramatic
effects during mating; this is probably the reason why cells
induce the expression of the prm1+ and dni1+ genes under
mating conditions.
The intracellular growth of the plasma membrane and
some of the structures observed in the intercellular
contact area in the S. pombe prm1D mutant were reminiscent of those observed at the mother-bud neck of Candida
albicans yeasts lacking Sur7p, a protein that localizes to
membrane microdomains and mediates the spatial regulation of plasma membrane organization (Alvarez et al.
2008; Douglas et al. 2012). Thus, it is possible that Prm1p
is involved in the organization of a specialized membrane
microdomain at the tip of shmoos such that it would ensure the correct organization of lipids at the cell–cell contact area. The amino acid sequence of Prm1p does not
predict any enzymatic activity; accordingly, it is possible
that this protein plays a structural role at the shmoo tip
required to localize/compartmentalize a class of lipids to
provide the proper fluidity/curvature required for membrane merger. In agreement with the notion that fluidity
of membrane microdomains affects fusion, it has been
proposed that increasing the fluidity of cholesterol-rich
membranes inhibits hepatitis C virus entry (ChamounEmanuelli et al. 2013). Membrane curvature is also required for merger (Kozlov et al. 2010); if phosphatidylserine
is flipped to the outer leaflet at a small area in the control
strain, the surface of this leaflet would change, inducing
membrane curvature. prm1D defects in phosphatidylserine
lateral exclusion or flipping would produce alterations in
membrane curvature/fluidity, affecting membrane merger.
Lipids and transmembrane proteins modulate each other
(Marsh 2008); thus, Prm1p might have a role organizing
lipids at the cell–cell fusion area whose disorganization
might in turn alter the function/activity of transmembrane
proteins directly involved in fusion, membrane–cell-wall attachment and/or membrane–cell-wall communication. Also,
Prm1p might act as a scaffold required for the recruitment of
other fusion factors.
In summary, the results obtained on studying the
functions of the Prm1- and Fig1-families of proteins in
different organisms suggest that these proteins have
functions related to the organization of the plasma
1074
M.-Ángeles Curto et al.
membrane that have been conserved along evolution
and are relevant for cell fusion, acting as a safeguard
mechanism that ensures proper cell fusion and zygote
integrity.
Acknowledgments
We thank E. Grote for helpful discussions. N. Skinner is
acknowledged for language revision and Susan Van Horn
and the Central Microscopy Imaging Center at Stony Brook
University for assistance with electron microscopy. We are
indebted to J. Cooper, O. Nielsen, C. Shimoda, and the Yeast
Genetic Resource Center (YGRC, Japan) for strains and
plasmids. Financial support to H.V.M. from the Comisión
Interministerial de Ciencia Y Tecnología (Spain)/European
Union Fondo Europeo de DEsarrollo Regional program
(grant BFU2010-15085) made this work possible. M.A.C.
and N.d.L were supported by Formación de Personal Universitario fellowships from the Ministry of Science, and M.H.
was supported by a Junta de Ampliación de Estudios predoctoral fellowship from the Consejo Superior de Investigaciones Científicas, Spain. The authors have no conflict of
interest to declare.
Literature cited
Aguilar, P. S., A. Engel, and P. Walter, 2007 The plasma membrane proteins Prm1 and Fig1 ascertain fidelity of membrane
fusion during yeast mating. Mol. Biol. Cell 18: 547–556.
Aguilar, P. S., M. G. Heiman, T. C. Walther, A. Engel, D. Schwudke
et al., 2010 Structure of sterol aliphatic chains affects yeast
cell shape and cell fusion during mating. Proc. Natl. Acad. Sci.
USA 107: 4170–4175.
Alvarez, F. J., L. M. Douglas, A. Rosebrock, and J. B. Konopka,
2008 The Sur7 protein regulates plasma membrane organization and prevents intracellular cell wall growth in Candida albicans. Mol. Biol. Cell 19: 5214–5225.
Arcangioli, B., and G. Thon, 2004 Mating-type cassettes: structure, switchin and silencing, pp. 9–147 in The Molecular Biology
of Schizosaccharomyces pombe, edited by R. Egel. Springer,
Heidelberg, Germany.
Bagnat, M., and K. Simons, 2002 Cell surface polarization during
yeast mating. Proc. Natl. Acad. Sci. USA 99: 14183–14188.
Bähler, J., J. Q. Wu, M. S. Longtine, N. G. Shah, A. McKenzie, 3rd,
et al., 1998 Heterologous modules for efficient and versatile
PCR-based gene targeting in Schizosaccharomyces pombe. Yeast
14: 943–951.
Calleja, G., Y. Bong, and B. Johnson, 1977 Fusion and erosion of
cell walls during conjugation in the fission yeast Schizosaccharomyces pombe. J. Cell Sci. 25: 139–155.
Cartagena-Lirola, H., A. Duran, and M. H. Valdivieso, 2006 The
Schizosaccharomyces pombe cfr1+ gene participates in mating
through a new pathway that is independent of fus1+. Yeast
23: 375–388.
Chamoun-Emanuelli, A. M., E. I. Pecheur, R. L. Simeon, D. Huang,
P. S. Cremer et al., 2013 Phenothiazines inhibit hepatitis C
virus entry, likely by increasing the fluidity of cholesterol-rich
membranes. Antimicrob. Agents Chemother. 57: 2571–2581.
Clemente-Ramos, J. A., R. Martin-Garcia, M. R. Sharifmoghadam,
M. Konomi, M. Osumi et al., 2009 The tetraspan protein Dni1p
is required for correct membrane organization and cell wall
remodelling during mating in Schizosaccharomyces pombe. Mol.
Microbiol. 73: 695–709.
Davey, J., 1998 Fusion of a fission yeast. Yeast 14: 1529–1566.
Douglas, L. M., H. X. Wang, S. Keppler-Ross, N. Dean,, and J. B.
Konopka, 2012 Sur7 promotes plasma membrane organization and is needed for resistance to stressful conditions and to
the invasive growth and virulence of Candida albicans. MBio 3.
Doyle, A., R. Martin-Garcia, A. T. Coulton, S. Bagley, and D. P.
Mulvihill, 2009 Fission yeast Myo51 is a meiotic spindle pole
body component with discrete roles during cell fusion and spore
formation. J. Cell Sci. 122: 4330–4340.
Drabikowski, W., E. Lagwinska, and M. G. Sarzala, 1973 Filipin as
a fluorescent probe for the location of cholesterol in the membranes of fragmented sarcoplasmic reticulum. Biochim. Biophys.
Acta 291: 61–70.
Engel, A., P. S. Aguilar, and P. Walter, 2010 The yeast cell fusion
protein Prm1p requires covalent dimerization to promote membrane fusion. PLoS ONE 5: e10593.
Erdman, S., L. Lin, M. Malczynski, and M. Snyder, 1998 Pheromoneregulated genes required for yeast mating differentiation. J. Cell
Biol. 140: 461–483.
Fleissner, A., S. Diamond, and N. L. Glass, 2009 The Saccharomyces cerevisiae PRM1 homolog in Neurospora crassa is involved in
vegetative and sexual cell fusion events but also has postfertilization functions. Genetics 181: 497–510.
Gordon, L. M., R. D. Sauerheber, and J. A. Esgate, 1978 Spin label
studies on rat liver and heart plasma membranes: effects of
temperature, calcium, and lanthanum on membrane fluidity. J.
Supramol. Struct. 9: 299–326.
Grote, E., 2010 Secretion is required for late events in the cellfusion pathway of mating yeast. J. Cell Sci. 123: 1902–1912.
Heiman, M. G., and P. Walter, 2000 Prm1p, a pheromone-regulated
multispanning membrane protein, facilitates plasma membrane
fusion during yeast mating. J. Cell Biol. 151: 719–730.
Heiman, M. G., A. Engel, and P. Walter, 2007 The Golgi-resident
protease Kex2 acts in conjunction with Prm1 to facilitate cell
fusion during yeast mating. J. Cell Biol. 176: 209–222.
Jin, H., C. Carlile, S. Nolan, and E. Grote, 2004 Prm1 prevents
contact-dependent lysis of yeast mating pairs. Eukaryot. Cell 3:
1664–1673.
Jin, H., J. M. McCaffery, and E. Grote, 2008 Ergosterol promotes
pheromone signaling and plasma membrane fusion in mating
yeast. J. Cell Biol. 180: 813–826.
Kozlov, M. M., H. T. McMahon, and L. V. Chernomordik,
2010 Protein-driven membrane stresses in fusion and fission.
Trends Biochem. Sci. 35: 699–706.
Kurahashi, H., Y. Imai, and M. Yamamoto, 2002 Tropomyosin is
required for the cell fusion process during conjugation in fission
yeast. Genes Cells 7: 375–384.
Marki, F., E. Hanni, A. Fredenhagen, and J. Van Oostrum,
1991 Mode of action of the lanthionine-containing peptide
antibiotics duramycin, duramycin B and C, and cinnamycin as
indirect inhibitors of phospholipase A2. Biochem. Pharmacol.
42: 2027–2035.
Marks, J., and J. S. Hyams, 1985 Localisation of F-actin through
the cell division cycle of Schizosaccharomyces pombe. Eur. J. Cell
Biol. 39: 27–32.
Marsh, D., 2008 Protein modulation of lipids, and vice-versa, in
membranes. Biochim. Biophys. Acta 1778: 1545–1575.
Mata, J., and J. Bahler, 2006 Global roles of Ste11p, cell type, and
pheromone in the control of gene expression during early sexual
differentiation in fission yeast. Proc. Natl. Acad. Sci. USA 103:
15517–15522.
Mata, J., R. Lyne, G. Burns, and J. Bähler, 2002 The transcriptional program of meiosis and sporulation in fission yeast. Nat.
Genet. 32: 143–147.
Merlini, L., O. Dudin, and S. G. Martin, 2013 Mate and fuse: how
yeast cells do it. Open Biol. 3: 130008.
Moreno, S., A. Klar, and P. Nurse, 1991 Molecular genetic analysis
of fission yeast Schizosaccharomyces pombe. Methods Enzymol.
194: 795–823.
Muller, E. M., N. A. Mackin, S. E. Erdman, and K. W. Cunningham,
2003 Fig1p facilitates Ca2+ influx and cell fusion during mating of Saccharomyces cerevisiae. J. Biol. Chem. 278: 38461–
38469.
Neiman, A. M., 1998 Prospore membrane formation defines a developmentally regulated branch of the secretory pathway in
yeast. J. Cell. Biol. 140: 29–37.
Nielsen, O., 2004 Mating-type control and differentiation, pp.
281–296 in The Molecular Biology of Schizosaccharomyces
pombe, edited by R. Egel. Springer, Heidelberg, Germany.
Norman, L. L., J. Brugues, K. Sengupta, P. Sens, and H. ArandaEspinoza, 2010 Cell blebbing and membrane area homeostasis
in spreading and retracting cells. Biophys. J. 99: 1726–1733.
Olmo, V. N., and E. Grote, 2010a Prm1 functions as a disulfidelinked complex in yeast mating. J. Biol. Chem. 285: 2274–2283.
Olmo, V. N., and E. Grote, 2010b Prm1 targeting to contact sites
enhances fusion during mating in Saccharomyces cerevisiae. Eukaryot. Cell 9: 1538–1548.
Orr-Weaver, T. L., J. W. Szostak, and R. J. Rothstein, 1983 Genetic
applications of yeast transformation with linear and gapped plasmids. Methods Enzymol. 101: 228–245.
Parsons, A. B., A. Lopez, I. E. Givoni, D. E. Williams, C. A. Gray
et al., 2006 Exploring the mode-of-action of bioactive compounds by chemical-genetic profiling in yeast. Cell 126: 611–
625.
Petersen, J., M. J. Heitz, and I. M. Hagan, 1998a Conjugation in S.
pombe: identification of a microtubule-organising centre, a requirement for microtubules and a role for Mad2. Curr. Biol. 8:
963–966.
Petersen, J., O. Nielsen, R. Egel, and I. M. Hagan, 1998b F-actin
distribution and function during sexual differentiation in Schizosaccharomyces pombe. J. Cell Sci. 111: 867–876.
Petersen, J., D. Weilguny, R. Egel, and O. Nielsen, 1995 Characterization
of fus1 of Schizosaccharomyces pombe: a developmentally controlled function needed for conjugation. Mol. Cell. Biol. 15:
3697–3707.
Proszynski, T. J., R. Klemm, M. Bagnat, K. Gaus, and K. Simons,
2006 Plasma membrane polarization during mating in yeast
cells. J. Cell Biol. 173: 861–866.
Sato, M., S. Dhut, and T. Toda, 2005 New drug-resistant cassettes
for gene disruption and epitope tagging in Schizosaccharomyces
pombe. Yeast 22: 583–591.
Sharifmoghadam, M. R., and M. H. Valdivieso, 2008 The Schizosaccharomyces pombe Map4 adhesin is a glycoprotein that can be
extracted from the cell wall with alkali but not with beta-glucanases
and requires the C-terminal DIPSY domain for function. Mol.
Microbiol. 69: 1476–1490.
Sharifmoghadam, M. R., P. Bustos-Sanmamed, and M. H. Valdivieso,
2006 The fission yeast Map4 protein is a novel adhesin required
for mating. FEBS Lett. 580: 4457–4462.
Sharifmoghadam, M. R., M. A. Curto, M. Hoya, N. De Leon, R.
Martin-Garcia et al., 2012 The integrity of the cytokinesis machinery under stress conditions requires the glucan synthase
Bgs1p and its regulator Cfh3p. PLoS ONE 7: e42726.
Sipiczki, M., 2000 Where does fission yeast sit on the tree of life?
Genome Biol. 1(2). Reviews 1011.
Stern, B., and P. Nurse, 1997 Fission yeast pheromone blocks Sphase by inhibiting the G1 cyclinB-p34cdc2 kinase. EMBO J. 16:
534.
Trueheart, J., J. D. Boeke, and G. R. Fink, 1987 Two genes required for cell fusion during yeast conjugation: evidence for
Fission Yeast Prm1p and Cell Fusion
1075
a pheromone-induced surface protein. Mol. Cell. Biol. 7: 2316–
2328.
Wachtler, V., S. Rajagopalan, and M. K. Balasubramanian,
2003 Sterol-rich plasma membrane domains in the fission
yeast Schizosaccharomyces pombe. J. Cell Sci. 116: 867–874.
White, J. M., and M. D. Rose, 2001 Yeast mating: getting close to
membrane merger. Curr. Biol. 11: R16–R20.
Yamamoto, M., 2004 Initiation of meiosis, pp. 297–310 in The
Molecular Biology of Schizosaccharomyces pombe, edited by
R. Egel. Springer, Heidelberg, Germany.
1076
M.-Ángeles Curto et al.
Yamamoto, M., Y. Imai, and Y. Watanabe, 1997 Mating and sporulation in Schizosaccharomyces pombe, pp. 1037–1106 in The
Molecular and Cellular Biology of the Yeast Saccharomyces, edited
by J. R. Pringle, J. R. Broach, and E. W. Jones. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, New York.
Yeung, T., G. E. Gilbert, J. Shi, J. Silvius, A. Kapus et al.,
2008 Membrane phosphatidylserine regulates surface charge
and protein localization. Science 319: 210–213.
Communicating editor: M. D. Rose
GENETICS
Supporting Information
http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.113.159558/-/DC1
Membrane Organization and Cell Fusion During
Mating in Fission Yeast Requires Multipass
Membrane Protein Prm1
M.-Ángeles Curto, Mohammad Reza Sharifmoghadam, Eduardo Calpena, Nagore De León,
Marta Hoya, Cristina Doncel, Janet Leatherwood, and M.-Henar Valdivieso
Copyright © 2014 by the Genetics Society of America
DOI: 10.1534/genetics.113.159558
2 SI
M.-Á. Curto et al.
M.-Á. Curto et al.
3 SI
4 SI
M.-Á. Curto et al.
Files S1-S3
Available for download as AVI files at
http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.113.159558/-/DC1
File S1 Cell fusion in a S. pombe wild-type zygote. Cell wall (Calcofluor staining) is shown in magenta and membrane (LactC2GFP probe) is shown in green.
File S2 Cell fusion in a S. pombe prm1Δ zygote. Cell wall (Calcofluor staining) is shown in magenta and membrane (LactC2-GFP
probe) is shown in green.
File S3 Cell fusion in a S. pombe prm1Δ zygote. Cell wall (Calcofluor staining) is shown in magenta and membrane (LactC2-GFP
probe) is shown in green.
M.-Á. Curto et al.
5 SI
Table S1 Source of the strains used in this work
Strain
Genotype
Source
HVP30
leu1-32 his3-1 ura418 ade6-M210 h-
Lab. stock
HVP117
leu1-32 his3-1 ura418 ade6-M216 h+
Lab stock
HVP281
leu1-32 ura418 ade6-M216 h90
Lab stock
HVP289
cyr1::ura4+ sxa2::ura4+ leu1-32 ura4Δ18 ade6-M704
P. Nurse
HVP423
fus1-HÁ:ura leu1-32 his3-1 ura418 ade6-M210 h-
Lab stock
HVP424
fus1-HÁ leu1-32 his3-1 ura418 ade6-M216 h+
Lab stock
HVP484
fus1::LEU2 ura4Δ18 h90
O. Nielsen
HVP781
dni1-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
Lab stock
HVP866
cyr1::ura4+ sxa2::ura4+dni1 ::KAN leu1-32 ura4Δ18 ade6-M704
Lab stock
HVP1077
dni1::KAN dni2 ::ura4+ leu1-32 ura418 ade6-M216 h90
Lab stock
HVP1085
dni1::KAN leu1-32 his3-1 his31 ade6 h-
Lab stock
HVP1086
dni1::KAN leu1-32 his3-1 his31 ade6 h+
Lab stock
HVP1113
dni2::ura4+ leu1-32 ura418 ade6-M216 h90
Lab stock
HVP1697
Pmap4+:GFP:Tmap4+:leu1+ leu1-32 ura418 ade6-M216 h90
Lab stock
HVP2185
prm1::KAN leu1-32 ura418 ade6-M216 h90
This work
HVP2258
dni1::KAN dni2::ura4+ prm1-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP2259
cfr1::his3+ prm1-GFP:leu1+ leu1-32 his3-1 ura418 ade6-M210 h-
This work
HVP2260
prm1::KAN prm1-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP2264
prm1::KAN Pmap4+:GFP:Tmap4+:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP2265
fus1::ura4+ prm1-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP2268
prm1::KAN dni1-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP2281
prm1::KAN leu1-32 ura418 ade6-M216 h90 with pREP3Xdni1+
This work
HVP2290
prm1::KAN leu1-32 ura418 ade6-M216 h90 with pREP3Xdni2+
This work
HVP2294
prm1::KAN leu1-32 ura418 ade6-M216 h90 with pREP3X
This work
HVP2297
prm1::KAN dni2-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP2421
prm1::KAN leu1-32 ura418 ade6-M216 h-
This work
HVP2449
GFP-Psy1:leu Hht1-mRFP:KAN leu1-32 ura418 ade6 h90
Lab stock
HVP2516
prm1::KAN dni2::ura4+ leu1-32 ura418 ade6-M216 h90
This work
HVP2591
prm1::KAN leu1-32 ura418 his3Δ1 ade6-M216 h+
This work
6 SI
M.-Á. Curto et al.
HVP2684
prm1::KAN leu1-32 ura418 ade6-M216 h90 with pREP3Xprm1+
This work
HVP2691
prm1::KAN cyr1::ura4+ sxa2::ura4+ leu1-32 ura4Δ18 ade6-M704 h-
This work
HVP2842
dni1-mCherry:leu1+ prm1-GFP:ura4+ leu1-32 ura418 ade6-M216 h90
This work
HVP3064
LactC2-GFP:NAT leu1-32 ura418 ade6-M216 h90
This work
HVP3074
prm1::KAN LactC2-GFP:NAT leu1-32 ura418 ade6-M216 h90
This work
HVP3092
cyr1::ura4+ sxa2::ura4+ dni1-GFP:KAN prm1-mCherry leu1-32 ura4Δ18 h-
This work
HVP3168
LactC2-GFP:NAT mCherry-Bgs4:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP3170
prm1::KAN LactC2-GFP:NAT mCherry-Bgs4:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP3285
fus1::LEU2 prm1::KAN ura4Δ18 h90
This work
HVP3299
dni2::ura4+ dni2-GFP:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP3412
fus1::ura4+ LactC2-GFP:NAT ura4Δ18 h90
This work
HVP3415
cyr1::ura4+ sxa2::ura4+ LactC2-GFP:NAT :ura4+ leu1-32 ura4Δ18 ade6-M704 h-
This work
HVP3416
cyr1::ura4+ sxa2::ura4+ prm1::KAN LactC2-GFP:NAT leu1-32 ura4Δ18 ade6-M704 h-
This work
HVP3423
fus1::LEU2 prm1::KAN dni2::ura4+ ura4Δ18 h90
This work
HVP3432
LactC2-GFP:NAT mCherry-Psy1:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP3434
prm1::KAN LactC2-GFP:NAT mCherry-Psy1:leu1+ leu1-32 ura418 ade6-M216 h90
This work
HVP3552
GFP-Psy1:leu Hht1-mRFP:KAN prm1::KAN leu1-32 ura418 ade6 h90
This work
M.-Á. Curto et al.
7 SI