http://www.elsevier.com/locate/bba

Biochimica et Biophysica Act

D-3 phosphoinositides of the ciliate Tetrahymena: Characterization and

study of their regulatory role in lysosomal enzyme secretion

George Leondaritisa, Arno Tiedtkeb, Dia Galanopouloua,*

aLaboratory of Biochemistry, Department of Chemistry, University of Athens, 15771 Athens, GreecebInstitute for General Zoology and Genetics, University of Münster, D-48149 Münster, Germany

Received 28 February 2005; received in revised form 16 June 2005; accepted 20 June 2005

Available online 18 July 2005

Abstract

Phosphatidylinositol 3-phosphate, PtdIns(3)P, is a phosphoinositide which is implicated in regulating membrane trafficking in both

mammalian and yeast cells. It also serves as a precursor for the synthesis of phosphatidylinositol 3,5-bisphosphate, PtdIns(3,5)P2, a

phosphoinositide, the exact functions of which remain unknown. In this report, we show that these two phosphoinositides are constitutive

lipid components of the ciliate Tetrahymena. Using HPLC analysis, PtdIns(3)P and PtdIns(3,5)P2 were found to comprise 16% and 30–40%

of their relevant phosphoinositide pools, respectively. Treatment of Tetrahymena cells with wortmannin (0.1–10 AM) resulted in thedepletion of PtdIns(3)P and PtdIns(3,5)P2 without any effect on D-4 phosphoinositides. Wortmannin was further used for the investigation of

D-3 phosphoinositide involvement in the regulation of lysosomal vesicular trafficking. Incubation of Tetrahymena cells with wortmannin

resulted in enhanced secretion of two different lysosomal enzymes without any change in their total activities. Experiments performed with a

T. thermophila secretion mutant strain verified that the wortmannin-induced secretion is specific and it is not due to a diversion of lysosomal

enzymes to other secretory pathways. Moreover, experiments performed with a phagocytosis-deficient T. thermophila strain showed that a

substantial fraction of wortmannin-induced secretion was dependent on the presence of functional phagosomes/phagolysosomes.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Tetrahymena; Phosphoinositide 3-kinase; Wortmannin; LY294002; Lysosome; Trafficking

1. Introduction

Phosphatidylinositol (PtdIns) is a minor membrane

component of all eukaryotic cells studied and it serves as

the precursor of several mono- and poly-phosphorylated

analogues (phosphoinositides, PIs) which are involved in

0167-4889/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbamcr.2005.06.011

Abbreviations: PtdIns, phosphatidylinositol; PI, phosphoinositide;

PtdInsP, phosphatidylinositol phosphate; PtdInsP2, phosphatidylinositol

bisphosphate; PtdIns(3)P, phosphatidylinositol 3-phosphate; PtdIns(4)P,

phosphatidylinositol 4-phosphate; PtdIns(3,5)P2, phosphatidylinositol 3,5-

bisphosphate; PtdIns(4,5)P2, phosphatidylinositol 4,5-bisphosphate; Gro-

PInsP, glycerophosphoinositol phosphate; Vps, vacuolar protein sorting;

Fab, formation of aploid and binucleate; FYVE, Fab1-YOTB-Vac1-EEA1-

homology domain; PX, phagocyte oxidase-homology domain; EEA1, early

endosomal antigen1

* Corresponding author. Tel.: +30 210 7274471; fax: +30 210 7274476.

E-mail address: galanopoulou@chem.uoa.gr (D. Galanopoulou).

the regulation of important cellular functions including

signalling, cell growth and differentiation, actin cytoskeletal

arrangement and intracellular vesicular trafficking [1–4].

Recently, new information has been obtained from the study

of D-3 phosphorylated PIs (D-3 PIs) and the kinases

involved in their biosynthesis. PtdIns and D-4 PIs serve as

substrates to various members of the PI 3-kinase family that

catalyze the addition of a phosphate group to the D-3

position of the inositol ring of these lipids [2]. Of the four

distinct D-3 PIs identified so far in mammalian cells,

PtdIns(3)P and PtdIns(3,5)P2 are considered to be con-

stitutively present and not in general implicated in signalling

[2].

In mammalian cells, PtdIns(3)P is found in amounts

corresponding to 2–5% of total PtdInsP; the most part of

this PtdIns(3)P pool is likely to be produced by type III PI

3-kinase (yeast Vps34p homolog) [5,6], although type II PI

a 1745 (2005) 330 – 341

BBAMCR-15287; No. of pages: 12; 4C:

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341 331

3-kinases and other pathways such as dephosphorylation of

PtdIns(3,4)P2 might contribute under certain conditions [2].

In contrast to mammalian cells, S. cerevisiae PtdIns(3)P is

found in almost equimolar amounts to PtdIns(4)P and it is

produced by the sole PI 3-kinase identified so far in the

yeast genome, Vps34p [7,8]. Studies on PtdIns(3)P local-

ization with the use of FYVE- or PX-domain probes have

shown that PtdIns(3)P is restricted to endomembranes of the

early and late endosomal network and in the vacuole of

yeast cells [9] and, also, in phagosomal membranes in

macrophages [10,11]. This localization pattern is in accord-

ance with PtdIns(3)P proposed function in regulating early/

late endosomal network trafficking [3,4]. In addition,

PtdIns(3)P serves as a substrate for type III PtdInsP kinases,

two of which have been studied and cloned so far, PIKfyve

and Fab1p. The product of these kinases is PtdIns(3,5)P2, a

phosphoinositide which has been recently identified in

mammalian, plant and yeast cells [12,13].

Several studies have implicated PtdIns(3)P in the

regulation of membrane trafficking. Most of them have

expanded early observations in S. cerevisiae , where

inactivation of the VPS34 gene causes missorting of a

subset of newly synthesized vacuolar hydrolases to the

vacuole and, as a consequence, aberrant secretion of

vacuolar proenzymes to the medium [8]. A generally

accepted model for PtdIns(3)P role in trafficking pathways

is based on the highly specific interactions of PtdIns(3)P-

enriched membrane microdomains with FYVE- or PX-

domain-containing proteins which participate in membrane

budding or fusion reactions [3,4]. Few effectors or binding

proteins of PtdIns(3,5)P2 have been described so far

[14,15]. However, in vivo studies of PIKfyve and Fab1p

in mammalian and yeast cells, respectively, have shown that

PtdIns(3,5)P2 is implicated in vacuole/late endosome

homeostasis and multivesicular body formation [16,17].

Information on PtdIns(3)P (and PtdIns(3,5)P2) and its

role in trafficking pathways has largely relied on studies in

yeast and mammalian cells. One still unanswered question is

whether these PIs are present in lower eukaryotes, like

ciliates, and if they posses similar roles in membrane

trafficking. Studies in these systems may well reveal

additional roles or modes of action or simply facilitate

analysis of PI roles in mammalian-type trafficking pathways

which are present in ciliates but not in yeast, e.g.,

phagocytosis or regulated exocytosis.

In this paper, we provide evidence that PtdIns(3)P and a

putative PtdIns(3,5)P2 are constitutive membrane lipids of

the ciliated protozoan Tetrahymena and that their synthesis

in vivo is inhibited by the mammalian PI 3-kinase inhibitor

wortmannin. Based on this inhibition, we proceed in

establishing these PIs as putative regulators of lysosome

homeostasis in Tetrahymena by showing that their depletion

is accompanied by an enhancement of constitutive lysoso-

mal enzyme secretion, a pathway which is essential for the

extracellular digestion in this organism [18]. Moreover,

using two different mutant Tetrahymena strains, deficient in

enzyme secretion and phagosome formation, respectively,

we propose that the regulation takes place at the step of

phagolysosome formation.

2. Materials and methods

2.1. Materials

Yeast extract, ferrous sulphate/chelate solution, adenine

nucleotides, p-nitrophenyl substrates, bovine brain PIs and

wortmannin were obtained from Sigma. Proteose-peptone

and silica gel H were obtained from Merck, peptone from

Serva, methylamine (33% solution in ethanol) from Fluka

and LY294002 from Calbiochem and Alexis. D-myo

[2-3H]inositol (specific activity 21.0 Ci/mmol) was pur-

chased from ICN, [3H]PtdIns(4,5)P2 (specific activity 6.5

Ci/mmol) from ARC and [3H]Ins(1,4,5)P3 was from

Amersham (TRK1000). HPLC column was purchased from

MZ Analysentechnik and scintillation fluid Floscint IV from

Packard. All other chemicals and solvents were of analytical

grade.

2.2. Cells and cell culture

Tetrahymena pyriformis (strain W) was routinely

cultured in a 2% proteose-peptone, 0.5% glucose and

0.2% yeast extract standard medium. Tetrahymena ther-

mophila CU438.1 (wild type with respect to secretion and

phagocytosis: pmr1-1/pmr1-1; pm-s, IV) and secretion

mutant (sec�) MS-1 (chx1-1/chx1-1; cy-r, sec�, II) cells

were grown in the standard medium enriched with 1%

ferrous sulphate/chelate solution. Tetrahymena thermophila

phagocytosis mutant (phg�) A2 cells (pmr1-1/pmr1-1; pm-

r, phg�, IV) (A. Tiedtke, unpublished data) were grown in

a medium of Fe2+-enriched standard medium and supple-

mented synthetic medium at a ratio of 4:1, as described for

phagocytosis mutant strains [19]. Working cultures (10–25

mL) were incubated at 26 -C under constant shaking (75–100 rpm) for approximately 72 h (late-logarithmic phase of

growth). Final cell densities ranged between 0.6–1�106cells/mL for T. pyriformis and 1–2�106 cells/mL forFe2+-enriched medium grown cells. S. cerevisiae MUCL

27831 strain was cultured in a 2% peptone, 1% glucose

and 1% yeast extract medium under constant shaking (150

rpm). Washed rabbit platelets were prepared essentially

according to Pinckard et al. [20] and suspended in a Ca2+-

free Tyrode-gelatin buffer, pH 6.5, at a final concentration

of 0.5�1010 plts/mL.

2.3. Labelling with myo-[3H]inositol, lipid extraction and

TLC analysis

Tetrahymena cells were labelled in vivo with [3H]inositol

at a concentration of 1 ACi/mL during the logarithmic phaseof growth. [3H]Inositol was added to 24 h-cultures (cell

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341332

density 5–10�104 cells/mL) and, after additional 48 h ofincubation, cells were first chilled on ice and then harvested

by centrifugation (1000�g for 8 min) and lipids wereextracted using the Schacht method with previously

described modifications [21]. S. cerevisiae cultures (10–

20 mL) were labelled with [3H]inositol at 1 ACi/mL for 19 h(final cell density 0.9�107 cells/mL) and the lipids wereextracted in the presence of glass beads according to

Wurmser and Emr [22]. Washed rabbit platelets resuspended

in Ca2+-free Tyrode-gelatin buffer, pH 6.5, were labelled

with [3H]inositol (15 ACi/mL) for 2 h prior to lipidextraction.

Total [3H]inositol-labelled lipids from all sources were

separated by TLC on oxalate-impregnated silica gel plates,

prepared by mixing silica gel H and 1% potassium oxalate

at a ratio of 1:2.6. A basic solvent system, chloroform/

methanol/ammonium hydroxide/water (86:76:6:16), was

used for the separation [21]. In order to detect and purify

PIs from total lipids, bovine brain standards were added to

the lipid extract prior to chromatographic separation. For

further analysis, lipids were extracted from silica gel and the

radioactivity of [3H]inositol-labelled PIs was measured by

liquid scintillation counting after resuspension in 0.5 mL

methanol and addition of a toluene-based scintillation

cocktail [21].

2.4. Methylamine deacylation and HPLC analysis of

glycerophosphoinositol phosphates

TLC-purified, [3H]inositol-labelled Tetrahymena, yeast

or rabbit platelet PtdInsP and PtdInsP2, or total Tetrahy-

mena lipids, were deacylated using methylamine as

described previously [23]. One to 2 mL of freshly

prepared methylamine reagent (methylamine in 33%

ethanol/water/n-butanol, 10:3:1) was added to dried lipids,

the suspension was briefly sonicated in a sonication bath

and, after incubation at 53 -C for 50 min, samples werecooled on ice and dried under a stream of nitrogen without

heating. The water-soluble glycero-derivatives were resus-

pended in water, washed twice with a mixture of n-

butanol/petroleum ether (bp 40–60 -C)/ethyl formate(20:4:1) and the washed aqueous phase was recovered

and freeze-dried. Glycerophosphoinositol phosphates (Gro-

PInsPs) were finally dissolved in filtered HPLC-grade

water and the pH was adjusted to 7–7.5 with HEPES/

KOH 50 mM, pH 7.5.

The separation was performed using a Hewllet Packard

HPLC on a 250�4.0 mm Partisphere 5-SAX column.Separation of GroPInsP isomers was performed with a

gradient system based on buffers A (water) and B

(NH4H2PO4 1.4 M, pH 3.8 at 25 -C with H3PO4) at aconstant flow rate of 1 mL/min as follows: 0 min, 0% B; 5

min, 0% B; 39 min, 17% B; 41 min, 37% B; 51 min, 37% B;

61 min, 100% B; 71 min, 100% B; 72 min, 0% B. Separation

of GroPInsP2 isomers was performed with the following

gradient: 0 min, 10% B; 5 min, 10% B; 45 min, 30% B; 50

min, 100% B; 55 min, 100% B; 56 min, 10% B. When

total [3H]inositol-labelled deacylated lipids were analyzed

and in order to separate all PtdInsP and PtdInsP2 isomers

in the same run, the following system was used: 0 min, 0%

B; 5 min, 0% B; 55 min, 25% B; 60 min, 100% B; 65 min,

100% B; 66 min, 0% B. In all cases, samples (250 AL)were mixed with ATP, ADP and AMP (final concentration

0.03 mM each) and the pH was adjusted to 7–7.5 with

HEPES/KOH 50 mM, pH 7.5. Samples were centrifuged in

order to precipitate particulate matter and the supernatant

was injected into a 300-AL HPLC loop. Fractions werecollected at regular intervals (0.5 min) throughout the run

and the radioactivity was determined by liquid scintillation

counting after addition of 3–5 mL of Floscint scintillation

fluid. The gradient system was evaluated using AMP, ADP,

ATP, [3H]GroPIns derived from methylamine-deacylated

purified Tetrahymena [3H]PtdIns [21], [3H]GroPIns(4)P

derived from methylamine-deacylated rabbit platelet

PtdInsP (see Results), [3H]GroPIns(4,5)P2 derived from

methylamine-deacylated standard [3H]PtdIns(4,5)P2 and

standard [3H]Ins(1,4,5)P3. Adenine nucleotide elution was

monitored by UV detection at 254 nm. In order to

accurately determine the [3H]GroPInsP and [3H]GroPInsP2content of samples, fractions were counted for 5 min and

the radioactivity of base line fractions was subtracted from

the radioactivity of peak fractions. The amount of

[3H]GroPInsP and [3H]GroPInsP2 was corrected for

recovery and quenching and was referenced to the total

radioactivity of [3H]GroPIns in the same sample. Thus, the

amount of all PIs is expressed as the % of PtdIns in each

sample.

2.5. Treatment of Tetrahymena with PI 3-kinase inhibitors in

vivo

T. pyriformis W cells were cultured and labelled in vivo

with [3H]inositol as described above. Cells were collected

by centrifugation at room temperature (650�g for 10 min),washed in an inorganic medium (Tris–HCl 10 mM, pH 7.0

or Na2HPO4 3.7 mM, KH2PO4 2.9 mM, MgSO4 1 mM, pH

7.0) and resuspended in the same medium at a final cell

density of 2–3�106 cells/mL. After further incubation for30 min, samples were treated with DMSO (0.1% final

concentration) or inhibitors (wortmannin or LY294002 at

0.1–10 AM and 50–100 AM final concentrations, respec-tively, from 1000-fold concentrated solutions in DMSO).

Samples were gently swirled during the incubation (10–20

min) after which, ice-cold perchloric acid was added (0.5 M

final concentration) and the samples were rapidly transferred

on ice. After 15–20 min on ice, samples were centrifuged at

2500�g for 5 min, the supernatant was aspirated and thepellet was suspended by addition of 3 mL of ice-cold

chloroform/methanol (1:2) and vigorous vortexing. After 20

min on ice, samples were brought to room temperature and

total lipids were extracted and deacylated as described

above.

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341 333

2.6. Lysosomal enzyme secretion assays

T. pyriformis or T. thermophila cells were grown to late-

logarithmic phase and then centrifuged at room temperature

as above and washed twice with Tris–HCl 10 mM, pH 7.0.

After incubation for 15 min, a 1-mL aliquot was withdrawn

for the determination of the time point 0, the rest of the

suspension was divided into flasks and DMSO (0.1% final

concentration) or inhibitor was added. Cells were incubated

at 26 -C with shaking and at regular time intervals (20–180min) 1-mL aliquots were withdrawn, incubated on ice for 5

min and centrifuged for 1 min in a microfuge. A 500-ALsample of the cell-free supernatant was recovered and the

rest of the supernatant was aspirated without damaging the

cell pellet. Cells were resuspended in 500 AL of ice-coldTris–HCl 10 mM, pH 7.0 and they were homogenized by

probe sonication on ice (3�30 s bursts with 30-s intervals).Acid phosphatase and h-hexosaminidase activities of

supernatants and cell homogenates were assayed as pre-

viously described, using the corresponding p-nitrophenyl

substrate [24]. Extracellular or cellular activity is expressed

as nmol/min/mL of initial cell suspension and the secretion

is calculated as the percentage of extracellular activity to the

total (sum of extracellular and cellular) activity. Data in

Figs. 4 and 5 were derived by subtraction of the % secretion

at time point 0; this value ranged between 1.1–4.9% for

acid phosphatase and 6.6–12% for h-hexosaminidase andwas due to incomplete removal of already secreted enzymes

during growth. In these experiments, lactate dehydrogenase

and alkaline phosphatase, which served as controls for non-

specific secretion or cell leakage, were assayed in the same

samples as described previously [25,26]. The extracellular

activity of these enzymes was negligible in both control and

wortmannin-treated samples.

3. Results

3.1. Identification of PtdIns(3)P and PtdIns(3,5)P2 in

Tetrahymena

Early studies have shown that phosphoinositides are

present in Tetrahymena [27] and subsequent experiments

from our group have partially characterized these phospho-

lipids and particularly PtdIns [21]. In addition, a recent study

provided in vitro evidence for the presence of a PI 3-kinase-

like activity in Tetrahymena [28]. However, our attempts in

demonstrating the presence of PtdIns(3)P in Tetrahymena

using a borate-based TLC solvent system yielded incon-

sistent results [21]. In order, therefore, to confirm or exclude

the presence of PtdIns(3)P in this organism, we followed the

approach of analyzing [3H]inositol-labelled deacylation

products of T. pyriformis PIs by HPLC chromatography on

an anion-exchange column. A gradient system for the

resolution of GroPInsP isomers was developed based on

earlier studies [5,7,12]. For PtdInsP isomer analysis, total

[3H]inositol-labelled Tetrahymena phospholipids were sep-

arated by TLC and the PtdInsP spot was scrapped off the

plate, extracted from the silica gel and deacylated by

methylamine. The water-soluble [3H]inositol-labelled deacy-

lation products were chromatographed on a Partisphere 5-

SAX column under the conditions depicted in Fig. 1. This

resulted in the resolution of two [3H]inositol-labelled

compounds with retention times expected for GroPIns(3)P

and GroPIns(4)P (Fig. 1A). These two compounds were

identified by comparison to parallel HPLC chromatograms

of TLC-purified, methylamine-deacylated, [3H]inositol-

labelled PtdInsP from S. cerevisiae cells and washed rabbit

platelets (Fig. 1B). As shown in Fig. 1B, deacylated yeast

PtdInsP was resolved into GroPIns(3)P and GroPIns(4)P as

expected [7], while platelet PtdInsP consisted mainly of

PtdIns(4)P. In 5 independent isolations of T. pyriformis

phospholipids, PtdIns(3)P was found to represent

16.5T2.5% of the total PtdInsP pool of the cell, a valuehigher than in any other cell type or organism examined so

far except yeast. These experiments were performed using

cells which were labelled during the logarithmic phase of

growth. Similar results concerning the relative abundance of

Tetrahymena PtdIns(3)P have been obtained by using

alternative conditions, namely labelling of cells during

starvation in inorganic buffer (Tris 10 mM, pH 7.4) or

during lag phase of growth (data not shown). Therefore, we

conclude that, under basal conditions, PtdIns(3)P is a major

PI in Tetrahymena comprising approximately 16% of the

amount of total PtdInsP. It is worth-mentioning that this is

the first report on D-3 PI presence in ciliated protozoa.

We followed the same approach in order to analyze the

PtdInsP2 isomers in Tetrahymena. The gradient system was,

therefore, optimized for the resolution of these isomers as

described in Materials and methods. HPLC analysis of

[3H]inositol-labelled, TLC-purified, methylamine-deacy-

lated PtdInsP2 from T. pyriformis cultures is shown in

Fig. 1C. Tetrahymena [3H]GroPInsP2 is resolved into two

labelled compounds one of which is co-chromatographed

with standard methylamine-deacylated [3H]PtdIns(4,5)P2.

The other compound, which constituted approximately 30%

of total GroPInsP2, was identified as a putative Gro-

PIns(3,5)P2 by comparison to parallel HPLC chromato-

grams of S. cerevisiae GroPInsP2. In all HPLC analyses,

PtdIns(3,5)P2 accounted for 30–40% of total PtdInsP2levels. Such a high relative abundance of this PI has never

been reported for any mammalian cell type or unicellular

organism studied so far.

3.2. Wortmannin but not LY294002 inhibits PtdIns(3)P and

PtdIns(3,5)P2 synthesis in Tetrahymena

Wortmannin has been widely used as a specific inhibitor

of PI 3-kinases especially in mammalian cells [2], although

at high (micromolar) concentrations it has been shown to

inhibit also PtdIns 4-kinases [29,30]. T. pyriformis cultures

were labelled with [3H]inositol and the cells were pelleted,

Fig. 1. Analysis of Tetrahymena PtdInsP and PtdInsP2 isomers by HPLC. (A) Tetrahymena [3H]PtdInsP and (B) [3H]PtdInsP from S. cerevisiae (open

diamonds) and rabbit platelets (filled diamonds) were isolated as described in the text, deacylated with methylamine and water-soluble products were mixed

with adenine nucleotide standards and chromatographed on a 250�4.0 mm column using a gradient system based on water and NH4H2PO4 1.4 M, pH 3.8(shown only in B). Fractions (0.5 mL) were collected and radioactivity was measured by liquid scintillation counting, while adenine nucleotide elution was

monitored by UV detection at 254 nm. Chromatograms are representative of approximately 20 individual runs. Platelet [3H]GroPIns3P is readily detected when

larger amounts of radioactivity are analyzed (not shown). (C) Tetrahymena [3H]PtdInsP2 was isolated as described in the text, deacylated with methylamine

and water-soluble products were chromatographed using a modified gradient system based on water and NH4H2PO4 1.4 M, pH 3.8. The elution positions of

[3H]GroPIns(3,5)P2 and [3H]GroPIns(4,5)P2 from S. cerevisiae and [

3H]GroPIns(4,5)P2 standard chromatographed in parallel are shown (arrows).

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341334

washed and resuspended as described in Materials and

methods. Aliquots of the cell suspension were treated with

wortmannin 1 AM in DMSO or DMSO alone (0.1% finalconcentration), total [3H]inositol-labelled lipids were

extracted and deacylated and glycerophosphoinositols were

chromatographed on the HPLC column. Representative

HPLC profiles of deacylated PIs from wortmannin-treated

and control samples are illustrated in Fig. 2. Wortmannin

caused a substantial reduction of Tetrahymena PtdIns(3)P

levels without any effect on PtdIns(4)P (Fig. 2A). This

suggests that the target of this compound in Tetrahymena is

a PtdIns 3-kinase. The reduction of PtdIns(3)P levels was

already maximal at 10 min, as further incubation with

wortmannin for 20 min did not cause any additional

lowering of PtdIns(3)P levels (data not shown). Such a

rapid wortmannin-induced depletion of PtdIns(3)P levels in

vivo has been observed in all mammalian cells studied

[6,31]. A titration curve (Fig. 2B) showed that the reduction

of PtdIns(3)P levels proceeds with an IC50 value of

approximately 200 nM. This value is substantially higher

than the values in mammalian cells, which are in the low

nanomolar range [6,12,31]. However, no in vivo data are

available from studies with unicellular organisms. More-

over, considering the uniqueness of Tetrahymena, it should

be added that the apparent differences in the efficacy of both

wortmannin and LY294002 (see below) could be due to a

reduced availability of inhibitors at specific locations/

compartments within the cell. Tetrahymena cells are

characterized by an extensive and complex pellicle which

permits direct communication of plasma membrane with

intracellular space only at specific sites in sharp contrast to

the case in most mammalian cells studied. This would

explain why highest inhibitor concentrations are required for

efficient in vivo inhibition in contrast to what is expected

from in vitro experiments.

As illustrated in Fig. 2C, wortmannin treatment reduced

also [3H]PtdIns(3,5)P2 levels to undetectable values (

Fig. 2. Effects of wortmannin in vivo treatment on D-3 and D-4 phosphoinositides. (A) [3H]inositol-labelled T. pyriformis cells were incubated for 10 min in

the presence of wortmannin 1 AM (filled circles) or DMSO (open circles) and, after quenching with perchloric acid, total lipids were extracted, deacylated andchromatographed on a HPLC column as shown in Fig. 1. Representative samples from more than 10 independent experiments and only the appropriate parts of

the chromatograms are shown. The recovered [3H]GroPIns of the samples was 90,000–100,000 cpm. (B) [3H]inositol-labelled T. pyriformis cells were

incubated for 10 min in the presence of different wortmannin concentrations or DMSO (control) and deacylated [3H]inositol-labelled PIs were

chromatographed as in Fig. 1. The [3H]GroPIns(3)P content was referenced to the total radioactivity of [3H]GroPIns. Results are expressed as % of the relative

abundance of PtdIns(3)P in DMSO-treated samples (0.7–0.9% of PtdIns) and they are the meanTS.D. of 3 independent experiments. (C) [3H]inositol-labelled

T. pyriformis cells were incubated for 10 min in the presence of DMSO (open circles) or wortmannin 10 AM (filled circles) and total lipids were extracted,deacylated and chromatographed using a modified gradient as described in Materials and methods, which permitted the separation of all PI isomers in the same

run. Only the appropriate parts of the chromatograms are shown. The recovered [3H]GroPIns of samples was 230,000 (wortmannin) and 260,000 cpm

(DMSO).

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341 335

[3H]PtdIns(4)P that was sensitive to LY294002 was 25–

30%. Although we cannot exclude that this reduction might

be due to a non-specific effect because of the high

concentrations used, the above results indicate that the

metabolism of PIs in Tetrahymena bears rather unique

specificities to two mammalian PI 3-kinase inhibitors:

wortmannin inhibits a PtdIns 3-kinase, while LY294002

apparently inhibits a PtdIns 4-kinase (see also Discussion).

It should be added that all experiments described so far have

been reproduced in another Tetrahymena species, T.

thermophila CU438.1 with similar results (D. Deli, G.

Leondaritis and D. Galanopoulou, unpublished results).

3.3. Wortmannin enhances the secretion of lysosomal

enzymes

Having established the occurrence of D-3 PIs in

Tetrahymena, we proceeded in investigating their possible

implication in the regulation of lysosomal vesicular

trafficking. Tetrahymena research, however, lacks a

clean-cut tool for the dissecting Golgi to lysosome traffic

in contrast to S. cerevisiae and mammalian cells where

pulse-chase experiments of newly-synthesized carboxypep-

tidase Y and cathepsin D, respectively, have provided the

first evidence for a function of PtdIns(3)P in this process

[8,33]. We decided, therefore, to study the effect of

wortmannin on the constitutive lysosomal enzyme secre-

tion pathway that has been well characterized in this cell

[18].

T. pyriformis cells were grown and treated as described

in previous sections and the activities of two lysosomal acid

hydrolases with different secretion rates, acid phosphatase

and h-hexosaminidase, were assayed in both supernatantsand cell homogenates. Acid phosphatase secretion in T.

pyriformis cells occurred at a rate of 3.9–4.8% of total

activity/h (Fig. 4A), while h-hexosaminidase secretionoccurred at a rate of 14–20% (Fig. 4B), both consistent

with previously published data [25]. When T. pyriformis

cells were treated with 10 AM wortmannin, an enhancementof the secretion of both acid phosphatase and h-hexosami-nidase was observed (Fig. 4A–B). This increased secretion

was evident at 20 min (the earliest time point examined, not

shown), peaked at 1 h after addition of wortmannin and

accounted for 65–88% and 67–71% increases over control

values for acid phosphatase and h-hexosaminidase, respec-tively. The increased secretion of acid phosphatase and h-

Fig. 3. Effects of LY294002 in vivo treatment on D-3 and D-4 phosphoinositides. (A) [3H]inositol-labelled T. pyriformis cells were incubated for 10 min in the

presence of LY294002 50 AM (filled circles) or DMSO (open circles) and deacylated [3H]inositol-labelled PIs were chromatographed as in Fig. 1. Only theappropriate parts of the chromatograms are shown. The recovered [3H]GroPIns of the samples was 90,000–97,000 cpm. (B) Samples were treated with 50 or

100 AM LY294002 or DMSO and the [3H]GroPIns(3)P (open bars) and [3H]GroPIns(4)P (filled bars) content was referenced to the total radioactivity of[3H]GroPIns. Results are expressed as % of the relative abundances of the lipids in DMSO-treated samples and they are the meanTS.D. of 3 independent

experiments. (C) [3H]inositol-labelled T. pyriformis cells were treated as in (A) and deacylated [3H]inositol-labelled PIs were chromatographed as in Fig. 2C.

Only the appropriate parts of the chromatograms are shown. The recovered [3H]GroPIns of samples was 230,000 (LY294002) and 260,000 cpm (DMSO).

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341336

hexosaminidase was progressively eliminated and, after a 3-

h incubation with wortmannin, secretion returned to control

rates (Fig. 4A–B). As shown in Fig. 4C, total activity of

acid phosphatase in control and wortmannin-treated cells

was very similar during the time course of these experi-

ments, a similar behavior was observed for h-hexosamini-dase as well.

The comparison of the titration curves of wortmannin-

induced PtdIns(3)P depletion (Fig. 2B) and acid phospha-

tase secretion (Fig. 4D) illustrates that both effects occur at

approximately the same EC50 of wortmannin. Moreover, as

expected from the ineffectiveness of LY294002 in the

inhibition of PtdIns(3)P synthesis, LY294002 did not

increase the initial rate of acid phosphatase or h-hexosami-nidase secretion in parallel experiments (data not shown).

3.4. Wortmannin effect in secretion is different in two T.

thermophila strains defective in distinct steps of the

phagolysosomal trafficking pathway

Lysosomal enzyme trafficking pathway in Tetrahymena

consists of two branches. Mature lysosomes or vesicles

containing mature lysosomal enzymes are either targeted to

plasma membrane for secretion or they fuse to early

acidified phagosomes to produce phagolysosomes [18,24].

In fact, a substantial fraction of intracellular acid phospha-

tase and h-hexosaminidase in growing Tetrahymena cells isfound associated with phagosomes [24].

We found it intriguing to examine wortmannin-induced

secretion of lysosomal hydrolases in two T. thermophila

mutant strains that exhibit distinct defects in the phagolyso-

somal trafficking pathway, namely the secretion mutant T.

thermophila MS-1 and the phagocytosis mutant T. thermo-

phila A2. In T. thermophila MS-1 cells, biosynthesis and

processing of proenzymes is unaltered compared to wild type

cells, but the secretion of mature enzymes is blocked,

probably at late steps of the secretory pathway [18,34–36];

importantly, T. thermophila MS-1 cells display a normal

phagocytic pathway [24,34] and are apparently wild type

concerning other Tetrahymena secretory pathways like

regulated exocytosis of dense-core granules [34]. In T.

thermophila A2 cells, phagocytosis is blocked and therefore

no phagosomes are formed when cells are incubated with

Indian Ink or iron dextran particles under different growth

conditions (our unpublished results). Previous studies of

several independent phagocytosis mutant strains have shown

that the basal secretion of lysosomal enzymes does not rely on

the presence of a functional phagocytic pathway [18].

Wortmannin treatment of wild type T. thermophila

CU438.1 cells induced depletion of PtdIns(3)P and

PtdIns(3,5)P2 levels similar to this of T. pyriformis and

enhancement of lysosomal enzyme secretion, although at

Fig. 4. Effects of wortmannin in vivo treatment on lysosomal enzyme secretion in Tetrahymena. T. pyriformis cells were treated with DMSO (open bars) or

wortmannin 10 AM (filled bars) as described in Materials and methods and, at regular time intervals, aliquots were withdrawn and centrifuged and acidphosphatase (A) or h-hexosaminidase (B) activity was assayed in supernatants and cell homogenates. Secreted enzyme activity is expressed as % of the totalactivity of each sample. Results are meansTS.D. of duplicate samples from 2 independent experiments. (C) Cells were treated with DMSO (open bars) or

wortmannin 10 AM (filled bars) and total acid phosphatase activity (sum of supernatant and homogenate activity) is expressed as % of the initial value (448nmol/min/106 cells). Results are meansTS.D. of duplicate samples from 2 independent experiments. (D) T. pyriformis cells were treated with different

concentrations of wortmannin or DMSO for 1 h and the secreted acid phosphatase activity was calculated as in (A) and was referenced to the value obtained

with 10 AM wortmannin which was taken as 100%. Results are the meansTS.D. of 3–4 experiments.

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341 337

lower (10-fold) wortmannin concentrations. Interestingly,

treatment of T. thermophila CU438.1 with 1 AM wortman-nin caused hypersecretion of both acid phosphatase and h-hexosaminidase (Fig. 5B for acid phosphatase), which

persisted even after 3 h; this permitted us to use 1 AMwortmannin which would allow for even a slight increase in

secretion, particularly in the case of T. thermophila MS-1

cells, to be detected and also after prolonged incubation.

As shown in Fig. 5A, T. thermophila MS-1 cells exhibit a

defect in the secretion of acid phosphatase, as was originally

described by Hünseler et al. [34], while T. thermophila A2

cells exhibit a very similar basal secretion of acid phosphatase

when compared to T. thermophila CU438.1 cells. One

noticeable difference, however, was that the total activity of

acid phosphatase of this mutant was lower compared to that

of wild type cells (approximately 248 and 481 nmol/min/106

cells, respectively). The result of wortmannin treatment on

the secretion of all strains is shown in Fig. 5B and Table 1.

The secretion of acid phosphatase induced by wortmannin in

T. thermophila MS-1 cells was negligible even when higher

concentrations (10 AM) were used (data not shown). At thesame time, in T. thermophila A2 cells, wortmannin induced

an increase in secretion which, after 1 h of treatment,

accounted for 50–60% of the increase observed in T.

thermophila CU438.1 cells. The difference was more

obvious after prolonged treatment since, after 3 h, wortman-

nin-induced secretion in T. thermophila A2 was 15–20%

compared to that of T. thermophila CU438.1 cells (Fig. 5B

and Table 1).

4. Discussion

Despite the establishment of D-3 PIs as important

regulators of vesicular trafficking in mammalian cells, a

similar role in unicellular organisms has been studied in

detail only in the yeast S. cerevisiae. We decided to proceed

in the characterization of putative D-3 PI isomers and

investigation of their involvement in Tetrahymena traffick-

ing pathways since the ciliated protozoan Tetrahymena

provides an excellent and rare opportunity for identifying

potential sites of action of PIs in trafficking: it possesses

three well-characterized pathways, regulated exocytosis of

dense-core granules, secretion of lysosomal enzymes and

phagocytosis, all in a genetically tractable and amenable

cellular context [37].

PtdIns(3)P and PtdIns(3,5)P2 were identified in T.

pyriformis cells by means of in vivo [3H]inositol labelling,

methylamine deacylation and HPLC analysis of the corre-

sponding glycero-derivatives. PtdIns(3)P was found to

represent approximately a 16% fraction of total PtdInsP

pool, while PtdIns(3,5)P2 was found to represent 30–40% of

Table 1

Wortmannin-induced acid phosphatase secretion is different in T. thermo-

phila mutant strains

T. thermophila strain Wortmannin-induced acid phosphatase

secretion (% of total)

60 min 180 min

CU438.1 (wild type) 2.8T0.2 12.0T2.4MS-1 (sec�) 0.4T0.2 1.0T0.4

A2 (phg�) 1.5T1.1 2.2T0.5

T. thermophila strains were treated as in Fig. 5B and net secretion induced

by wortmannin was calculated by subtraction of the values of DMSO-

induced secretion. Thus, values in the table represent the fraction of total

acid phosphatase activity that is mobilized for secretion by wortmannin at

each time point and they are meansTS.D. of duplicate samples from 3

independent experiments.

Fig. 5. Effect of wortmannin on lysosomal enzyme secretion in T.

thermophila MS-1 and A2 cells. (A) T. thermophila CU438.1 (diamonds),

MS-1 (triangles) and A2 (circles) cells were resuspended in inorganic

medium and, at regular time intervals, aliquots were withdrawn and acid

phosphatase activity was assayed. Secreted acid phosphatase activity is

expressed as in Fig. 4. Results are meansTS.D. of duplicate samples from 2

to 3 independent experiments. Total acid phosphatase activity was 481

nmol/min/106 cells (CU438.1), 470 nmol/min/106 cells (MS-1) and 248

nmol/min/106 (A2 cells). (B) T. thermophila CU438.1, MS-1 and A2 cells

were treated with DMSO (open bars) or wortmannin 1 AM (filled bars) for60 or 180 min and the secreted acid phosphatase activity was calculated as

described above. Results are meansTS.D. of duplicate samples from 3

independent experiments.

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341338

the PtdInsP2 pool. Concerning PtdIns(3,5)P2, Tetrahymena

seems to represent an interesting case: this lipid has been

found in minor amounts ranging from 0.2–0.5% of the

PtdInsP2 pool in mammalian cells [12] to 5% in S.

cerevisiae [13]. In any case, the calculated PtdIns(3)P/

PtdIns(3,5)P2 ratio in Tetrahymena (approximately 18:1)

has a value close to that of S. cerevisiae (20–30:1, in the

absence of hyperosmotic shock) [13]. However,

PtdIns(3,5)P2 concentration in S. cerevisiae increases

dramatically upon hyperosmotic stress reaching levels

similar to those of PtdIns(4,5)P2 [13]. We have been

unable so far to characterize an osmodependent activation

of PtdIns(3,5)P2 synthesis in Tetrahymena cells (D. Deli,

G. Leondaritis, and D. Galanopoulou, unpublished

results). Nevertheless, it is still an open question whether

this PtdIns(3,5)P2-based signalling pathway represents a

more widespread osmosensing pathway in eukaryotic

cells; apart from yeasts, only Chlamydomonas and some

plant cells have been reported to exhibit robust hyper-

osmotic activation of PtdIns(3,5)P2 synthesis [38].

In order to connect the presence of PtdIns(3)P and

PtdIns(3,5)P2 to functions in Tetrahymena, we decided to

use wortmannin and LY294002, two compounds that have

served as specific PI 3-kinase inhibitors in vivo and in vitro

in mammalian cell studies [2,32]; in fact, much of the work

that connected PtdIns(3)P to membrane trafficking has been

obtained using these compounds [4,33].

The data presented in Figs. 2 and 3 illuminate several

aspects of D-3 and D-4 PI metabolism in Tetrahymena.

Wortmannin causes a rapid, concentration-dependent,

reduction of PtdIns(3)P levels, quite similar to that in

mammalian cells, except the higher IC50 wortmannin

concentration which characterizes the effect in Tetrahy-

mena. Moreover, wortmannin treatment reduced the levels

of PtdIns(3,5)P2 as well. The above results predict the

presence of a PtdIns 3-kinase which is wortmannin-sensitive

and provide evidence for the synthesis of PtdIns(3,5)P2 via

a PtdIns(3)P 5-kinase activity in Tetrahymena. Whether

these enzymes are the Tetrahymena homologs of class III PI

3-kinases and type III PtdInsP kinases awaits future studies.

Interestingly, putative genes that show extended homology

to Vps34-PI3K III and Fab1/PIKfyve-type III PtdInsP

kinases have already been identified in the ongoing

Tetrahymena sequencing project (Suppl. Fig. 1).

Wortmannin has been shown to inhibit PtdIns 4-kinases

as well. In mammalian cells, the PtdIns 4-kinase that

confers wortmannin-sensitivity is the type III PtdIns 4-

kinase [29]. Additionally, in S. cerevisiae cells, the main

target of wortmannin in vivo and in vitro is a PtdIns 4-

kinase, the product of STT4 gene and not the Vps34p

PtdIns 3-kinase [39]. We believe that, apart from an

apparent absence of a wortmannin-sensitive PtdIns 4-

kinase in Tetrahymena, our results provide a strong basis

for the use of wortmannin in experiments aimed at

identifying specific roles of PtdIns(3)P in Tetrahymena

cellular functions. At the same time, LY294002 was

ineffective in reducing PtdIns(3)P and PtdIns(3,5)P2 levels

in Tetrahymena; instead PtdIns(4)P levels were slightly but

reproducibly reduced. Therefore, our results raised the

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341 339

question whether the preferred target of LY294002 in

Tetrahymena is a wortmannin-insensitive PtdIns 4-kinase.

This idea might not be excluded since these two

compounds use slightly different mechanisms for inhibition

of class I PI 3-kinases: although they both interfere with

the ATP-binding site, wortmannin is covalently bound and

effects a distortion in the catalytic domain while LY294002

makes extensive contacts with both the adenine and ribose

binding sites of the site [40]. These observations correlate

well with a recent study which showed that wortmannin

and LY294002 have different efficiencies in inhibiting

Tetrahymena PI kinase activity in vitro and in vivo [28].

We postulate, however, that the main target of LY294002

in vivo in Tetrahymena might not be a PtdIns 4-kinase but

possibly another protein or lipid kinase (see below).

Tetrahymena PtdIns 3-kinase seems to be rather different

from mammalian-type enzymes concerning sensitivity to

wortmannin and LY294002, at least in vivo. Therefore, we

sought to establish whether this discrepancy could be

explained at the molecular level. Towards this end, we

screened the ongoing Tetrahymena sequencing project for

genes homologues to mammalian-type PI 3-kinases using as

a bait the sequence of PI3Kg for which the crystal structure ofenzyme–inhibitor complexes has been resolved [40]. Inter-

estingly, four putative genes were recovered from this screen

and apart from a putative Vps34-type PI3K III (see above),

three genes showed extended similarity to class I PI3Ks

(Suppl. Fig. 1). A detailed comparative analysis of residues

implicated in wortmannin and LY294002 binding to human

PI3Kg showed that small differences do exist (Suppl. Fig. 2);however, these cannot clearly explain the differential

sensitivity to these two inhibitors and future studies will

hopefully clarify this issue. Different PI3Ks (with different

sensitivity to inhibitors) may as well account for the differ-

ential and concentration-dependent effects of wortmannin

and LY294002 on T. vorax differentiation [41] and Tetrahy-

mena phagocytosis (our unpublished results and [28]).

In mammalian and yeast cells, PtdIns(3)P serves as a

signal for membrane recruitment of FYVE- and PX-domain

containing proteins that regulate vesicular trafficking steps

in various aspects of endosomal trafficking [3,4]. Few

studies, however, have been focused on the lysosomes per

se: incubation of mammalian cells with wortmannin results

in a slight shift of lysosomal markers (including lysosomal

enzymes) towards less dense fractions in fractionation

gradients [42,43]. It has been proposed that, in this case,

wortmannin inhibits dense lysosome reformation after the

late endosome–lysosome fusion into a proteolytically active

hybrid organelle [43,44]. Several recent studies have also

highlighted the involvement of PtdIns(3)P in the maturation

of early phagosomes to phagolysosomes [11,45]. Incubation

of macrophages with wortmannin results in the inhibition of

PtdIns(3)P production on phagosomal membranes [10,11]

and in a block in the acquisition of early and late endosomal

markers (i.e., EEA1 and lyso-bisphosphatidic acid, respec-

tively) by nascent phagosomes [11,45].

Interestingly, Tetrahymena lysosomes serve at least two

functions: first, they fuse to – as yet unidentified – sites of

the plasma membrane and release their contents to the

extracellular space and second, they fuse to maturing

phagosomes which results in the formation of phagolyso-

somes [18,24,37]. The phagosomal maturation pathway in

ciliates is similar to that of mammalian macrophages in

several aspects: newly-formed phagosomes are rapidly

acidified by fusion with endosomal type organelles and

subsequently they become competent for fusion with

lysosomes or lysosomal vesicles containing active hydro-

lases [46]. Not surprisingly, pathogens, which escape

degradation by reprogramming the phagosomal maturation

pathway in mammalian cells [45], seem to do the same in

Tetrahymena as well [47].

The results of wortmannin-induced secretion of the two

T. thermophila mutant strains clearly implicate phagosomes/

phagolysosomes as putative target compartments of

PtdIns(3)P (and/or PtdIns(3,5)P2) function in Tetrahymena.

If the wortmannin-induced secretion involved a mechanism

of action that bypasses the normal secretory pathway, this

compound would induce the secretion of a fraction of total

lysosomal hydrolases in T. thermophila MS-1 cells com-

parable to that of wild type cells. As shown in Fig. 5B, this

is not the case. On the other hand, wortmannin treatment of

a strain devoid of phagosomes shows a decrease in

lysosomal enzyme secretion as high as 85% compared to

the secretion of wild type cells (Fig. 5B and Table 1).

Therefore, the presence of functional phagosomes/phagoly-

sosomes is necessary for the wortmannin-induced secretion

as, a significant part of it, is dependent on their presence. A

possible interpretation would be that PtdIns(3)P (and/or

PtdIns(3,5)P2) regulate the proper secretion at the level of

phagosome–lysosome fusion such that, in the absence of

PtdIns(3)P, lysosomal enzymes destined for delivery to

phagosomes could be diverted instead to the secretory

pathway thus resulting in a hypersecretion effect. This idea

is further suggested by the fact that wortmannin at the same

time causes an arrest of phagosomal maturation in early

stages (our unpublished results). The notion that blocking of

lysosome–phagosome fusion can result in a hypersecretion

effect of mature lysosomal enzymes has been suggested in

studies with chloroquine- and bafilomycin A1-treated J774

macrophages [48]. To our best knowledge, a similar

question has not been addressed in studies of wortmannin-

inhibition of phagosomal maturation in macrophages.

In conclusion, we have demonstrated the presence of D-3

PIs in the lower eukaryote Tetrahymena. Our data indicate a

role for these lipids in lysosomal trafficking and provide the

basis for future studies on their exact targets, which

probably lie in the phagolysosomal pathway of the cell.

Furthermore, the different effects of wortmannin and

LY294002 on Tetrahymena PtdInsP synthesis show that

this organism could be used for dissecting wortmannin and

LY294002 cellular targets which, until now, are considered

as identical in mammalian cells studies.

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341340

Acknowledgements

We wish to thank Dr. M. Typas (Department of Biology,

University of Athens, Greece) for the gift of the S.

cerevisiae MUCL 27831 strain and Dr A. Lykidis (Institute

for Biomedical Research of the Academy of Athens,

Greece) for critically reading of the manuscript. This work

was partially supported by a University of Athens grant (KA

70/4/2507).

Appendix A. Supplementary data

Supplementary data associated with this article can be

found, in the online version, at doi:10.1016/j.bbamcr.

2005.06.011.

References

[1] M.J. Berridge, Inositol trisphosphate and calcium signalling, Nature

361 (1993) 315–325.

[2] B. Vanhaesebroeck, S.L. Leevers, K. Ahmadi, J. Timms, R. Katso,

P.C. Driscoll, R. Woscholski, P.J. Parker, M.D. Waterfield, Synthesis

and function of 3-phosphorylated inositol lipids, Annu. Rev. Biochem.

70 (2001) 535–602.

[3] A. Simonsen, A.E. Wurmser, S.D. Emr, H. Stenmark, The role of

phosphoinositides in membrane transport, Curr. Opin. Cell Biol. 13

(2001) 485–492.

[4] S. Corvera, Phosphatidylinositol 3-kinase and the control of endosome

dynamics: new players defined by structural motifs, Traffic 2 (2001)

859–866.

[5] L. Stephens, P.T. Hawkins, C.P. Downes, Metabolic and structural

evidence for the existence of a third species of polyphosphoinositide in

cells: d-phosphatidyl-myo-inositol 3-phosphate, Biochem. J. 259

(1989) 267–276.

[6] L. Stephens, F.T. Cooke, R. Walters, T. Jackson, S. Volinia, I. Gout,

M.D. Waterfield, P.T. Hawkins, Characterization of a phosphatidyl-

inositol-specific phosphoinositide 3-kinase from mammalian cells,

Curr. Biol. 4 (1994) 203–214.

[7] P.T. Hawkins, L.R. Stephens, J.R. Piggott, Analysis of inositol

metabolites produced by Saccharomyces cerevisiae in response to

glucose stimulation, J. Biol. Chem. 268 (1993) 3374–3383.

[8] P.V. Schu, K. Takegawa, M.J. Fry, J.H. Stack, M.D. Waterfield, S.D.

Emr, Phosphatidylinositol 3-kinase encoded by yeast VPS34 gene

essential for protein sorting, Science 260 (1993) 88–91.

[9] D.J. Gillooly, I.C. Morrow, M. Linsday, R. Gould, N.J. Bryant, J.-M.

Gaullier, R.G. Parton, H. Stenmark, Localization of phosphatidylino-

sitol 3-phosphate in yeast and mammalian cells, EMBO J. 19 (2000)

4577–4588.

[10] C.D. Ellson, K.E. Anderson, G. Morgan, E.R. Chilvers, P. Lipp, L.R.

Stephens, P.T. Hawkins, Phosphatidylinositol 3-phosphate is generated

in phagosomal membranes, Curr. Biol. 11 (2001) 1631–1635.

[11] O.V. Vieira, R.J. Botelho, L. Rameh, S.M. Brachmann, T. Matsuo,

H.W. Davidson, A. Schreiber, J.M. Backer, L.C. Cantley, S. Grinstein,

Distinct roles of class I and class III phosphatidylinositol 3-kinases in

phagosome formation and maturation, J. Cell Biol. 155 (2001) 19–25.

[12] C.C. Whiteford, C.A. Brearly, E.T. Ulug, Phosphatidylinositol 3,5-

bisphosphate defines a novel PI 3-kinase pathway in resting mouse

fibroblasts, Biochem. J. 323 (1997) 597–601.

[13] S.K. Dove, F.T. Cooke, M.R. Douglas, L.G. Sayers, P.J. Parker, R.H.

Michell, Osmotic stress activates phosphatidylinositol-3,5-bisphos-

phate synthesis, Nature 390 (1997) 187–192.

[14] K. Tsujita, T. Itoh, T. Ijuin, A. Yamamoto, A. Shisheva, J. Laporte, T.

Takenawa, Myotubularin regulates the function of the late endosome

through the gram domain-phosphatidylinositol 3,5-bisphosphate inter-

action, J. Biol. Chem. 279 (2004) 13817–13824.

[15] S. Friant, E.I. Pecheur, A. Eugster, F. Michel, Y. Lefkir, D.

Nourrisson, F. Letourneur, Ent3p is a PtdIns(3,5)P2 effector required

for protein sorting to the multivesicular body, Dev. Cell. 5 (2003)

499–511.

[16] G. Odorizzi, M. Babst, S.D. Emr, Fab1p PtdIns(3)P 5-kinase function

essential for protein sorting in the multivesicular body, Cell 95 (1998)

847–858.

[17] O.C. Ikonomov, D. Sbrissa, A. Shisheva, Mammalian cell morphology

and endocytic membrane homeostasis require enzymatically active

phosphoinositide 5-kinase PIKfyve, J. Biol. Chem. 276 (2001)

26141–26147.

[18] A. Tiedtke, R. Rasmussen, Constitutive secretion of acid hydrolases in

Tetrahymena thermophila, J. Protozool. 36 (1989) 378–382.

[19] O. Orias, E.P. Hamilton, J.D. Orias, Tetrahymena as a laboratory

organism: useful strains, cell culture and cell line maintenance, in: D.J.

Asai, J.D. Forney (Eds.), Tetrahymena thermophila, Methods Cell

Biol., 62, 2000, pp. 28–103.

[20] R.N. Pinckard, R.S. Farr, D.J. Hanahan, Physicochemical and func-

tional identity of rabbit platelet-activating factor (PAF) released in

vivo during IgE anaphylaxis with PAF released in vitro from IgE

sensitized basophils, J. Immunol. 123 (1979) 1847–1857.

[21] G. Leondaritis, D. Galanopoulou, Characterization of inositol

phospholipids and identification of a mastoparan-induced polyphos-

phoinositide response in Tetrahymena pyriformis, Lipids 35 (2000)

525–532.

[22] A.E. Wurmser, S.D. Emr, Phosphoinositide signalling and turnover:

PtdIns(3)P, a regulator of membrane traffic, is transported to the

vacuole and degraded by a process that requires lumenal vacuolar

hydrolase activities, EMBO J. 17 (1998) 4930–4942.

[23] N.G. Clarke, R.M. Dawson, Alkaline OYN-transacylation. A new

method for the quantitative deacylation of phospholipids, Biochem. J.

195 (1981) 301–306.

[24] T. Kiy, C. Vosskhuhler, L. Rasmussen, A. Tiedtke, Three pools of

lysosomal enzymes in Tetrahymena thermophila, Exp. Cell Res. 205

(1993) 286–292.

[25] Y. Banno, N. Sasaki, Y. Nozawa, Secretion heterogeneity of

lysosomal enzymes in Tetrahymena pyriformis, Exp. Cell Res. 170

(1987) 259–268.

[26] K.E. Howell, G.E. Palade, Hepatic Golgi fractions resolved into

membrane and content subfractions, J. Cell Biol. 92 (1982) 822–832.

[27] P. Kovacs, G. Csaba, Involvement of phosphatidylinositol (PI) system

in the mechanism of hormonal imprinting, Biochem. Biophys. Res.

Commun. 170 (1990) 119–126.

[28] P. Kovacs, E. Pallinger, Phosphatidylinositol 3-kinase-like activity in

Tetrahymena. Effects of wortmannin and LY294002, Acta Protozool.

42 (2003) 277–285.

[29] T. Balla, G.J. Downing, H. Jaffe, S. Kim, A. Zolyomi, K.J. Catt,

Isolation and molecular cloning of wortmannin-sensitive bovine type

III phosphatidylinositol 4-kinases, J. Biol. Chem. 272 (1997)

18358–18366.

[30] G.B. Willars, S.R. Nahorski, R.A.J. Chaliss, Differential regulation of

muscarinic acetylcholine receptor-sensitive polyphosphoinositide

pools and consequences for signaling in human neuroblastoma cells,

J. Biol. Chem. 273 (1998) 5037–5046.

[31] C.C. Whiteford, C. Best, A. Kazlauskas, E.T. Ulug, D-3 phosphoino-

sitide metabolism in cells treated with platelet-derived growth factor,

Biochem. J. 319 (1996) 851–860.

[32] C.J. Vlahos, W.F. Matter, K.Y. Hui, R.F. Brown, A specific inhibitor of

phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-ben-

zopyran-4-one (LY294002), J. Biol. Chem. 269 (1994) 5241–5249.

[33] H.W. Davidson, Wortmannin causes mistargeting of procathepsin D.

Evidence for the involvement of a phosphatidylinositol 3-kinase in

vesicular transport to lysosomes, J. Cell Biol. 130 (1995) 797–805.

http:\\dx.doi.org\doi:10.1016\j.bbamcr.2005.06.011

G. Leondaritis et al. / Biochimica et Biophysica Acta 1745 (2005) 330–341 341

[34] P. Hünseler, G. Scheidgen-Kleyboldt, A. Tiedtke, Isolation and

characterization of a mutant of Tetrahymena thermophila blocked in

secretion of lysosomal enzymes, J. Cell Sci. 88 (1987) 47–55.

[35] P. Hünseler, A. Tiedtke, K. von Figura, Biosynthesis of secreted h-hexosaminidase in Tetrahymena thermophila, Biochem. J. 252 (1988)

837–842.

[36] Y. Banno, Y. Okano, K. Furukawa, A. Tiedtke, A. Kobata, Y. Nozawa,

Processing and secretion of lysosomal acid a-glucosidase in Tetrahy-mena wild type and secretion-deficient mutant cells, J. Eukaryot.

Microbiol. 40 (1993) 515–520.

[37] J. Frankel, Cell biology of Tetrahymena thernophila, in: D.J. Asai,

J.D. Forney (Eds.), Tetrahymena thermophila, Methods Cell Biol., 62,

2000, pp. 28–103.

[38] H.J.G. Meijer, N. Divecha, H. van der Ende, A. Musgrave, T. Munnik,

Hyperosmotic stress induces rapid synthesis of phosphatidyl-d-

inositol 3,5-bisphosphate in plant cells, Planta 208 (1999) 294–298.

[39] M.S. Cutler, J. Heitman, M.E. Cardenas, STT4 is an essential

phosphatidylinositol 4-kinase that is a target of wortmannin in

Saccharomyces cerevisiae, J. Biol. Chem. 272 (1997) 27671–27677.

[40] S. Djordjevic, P.C. Driscoll, Structural insight into substrate specificity

and regulatory mechanisms of phosphoinositide 3-kinases, Trends

Biochem. Sci. 27 (2002) 426–432.

[41] P.E. Ryals, S. Bae, C.E. Patterson, Evidence for early signalling events

in stomatin-induced differentiation of Tetrahymena vorax, J. Eukaryot.

Microbiol. 46 (1999) 77–82.

[42] B.J. Reaves, N.A. Bright, B.M. Mullock, J.P. Luzio, The effect of

wortmannin on the localisation of lysosomal type I integral membrane

glycoproteins suggests a role for phosphoinositide 3-kinase activity in

regulating membrane traffic late in the endocytic pathway, J. Cell Sci.

109 (1996) 749–762.

[43] S.A. Mousavi, A. Brech, T. Berg, R. Kjeken, Phosphoinositide 3-

kinase regulates maturation of lysosomes in rat hepatocytes, Biochem.

J. 372 (2003) 861–869.

[44] N.A. Bright, B.J. Reaves, B.M. Mullock, J.P. Luzio, Dense core

lysosomes can fuse with late endosomes and re-formed from the

resultant hybrid organelles, J. Cell Sci. 110 (1997) 2027–2040.

[45] R.A. Fratti, J.M. Backer, J. Gruenberg, S. Corvera, V. Deretic, Role of

phosphatidylinositol 3-kinase and Rab5 effectors in phagosomal

biogenesis and mycobacterial phagosome maturation arrest, J. Cell

Biol. 154 (2001) 631–644.

[46] J.R. Nilsson, Phagotrophy in Tetrahymena, in: M. Levandowsky, S.H.

Hutner (Eds.), Biochemistry and Physiology of Protozoa, 2nd ed., vol.

2, Academic Press, 1979, pp. 339–373.

[47] E. Weidner, L.D. Sibley, Phagocytized intracellular microsporidian

blocks phagosome acidification and phagosome– lysosome fusion,

J. Protozool. 32 (1985) 311–317.

[48] V. Claus, A. Jahraus, T. Tjelle, T. Berg, H. Kirschke, H. Faulstich, G.

Griffiths, Lysosomal enzyme trafficking between phagosomes, endo-

somes and lysosomes in J774 macrophages, J. Biol. Chem. 273 (1998)

9842–9851.

D-3 phosphoinositides of the ciliate Tetrahymena: Characterization and study of their regulatory role in lysosomal enzyme secretionIntroductionMaterials and methodsMaterialsCells and cell cultureLabelling with myo-[3H]inositol, lipid extraction and TLC analysisMethylamine deacylation and HPLC analysis of glycerophosphoinositol phosphatesTreatment of Tetrahymena with PI 3-kinase inhibitors in vivoLysosomal enzyme secretion assays

ResultsIdentification of PtdIns(3)P and PtdIns(3,5)P2 in TetrahymenaWortmannin but not LY294002 inhibits PtdIns(3)P and PtdIns(3,5)P2 synthesis in TetrahymenaWortmannin enhances the secretion of lysosomal enzymesWortmannin effect in secretion is different in two T. thermophila strains defective in distinct steps of the phagolysosomal trafficking pathway

DiscussionAcknowledgementsSupplementary dataReferences