198

Environmental Toxicology and Chemistry, Vol. 20, No. 1, pp. 198–204, 2001q 2001 SETAC

Printed in the USA0730-7268/01 $9.00 1 .00

POTENTIAL MULTIDRUG RESISTANCE GENE POHL: AN ECOLOGICALLYRELEVANT INDICATOR IN MARINE SPONGES

ANATOLI KRASKO,† BRANKO KURELEC,‡ RENATO BATEL,§ ISABEL M. MULLER,† and WERNER E.G. MULLER*††Institut fur Physiologische Chemie, Abteilung Angewandte Molekularbiologie, Universitat Mainz, Duesbergweg 6, D-55099 Mainz, Germany

‡Center for Marine Research, Institute Ruder Boskovic, 10000 Zagreb, Croatia§Center for Marine Research, Institute Ruder Boskovic, 52210 Rovinj, Croatia

(Received 6 October 1999; Accepted 30 May 2000)

Abstract—Sponges are sessile filter feeders found in all aquatic habitats from the tropics to the arctic. Against potential environmentalhazards, they are provided with efficient defense systems, e.g., protecting chaperones and/or the P-170/multidrug resistance pumpsystem. Here we report on a further multidrug resistance pathway that is related to the pad one homologue (POH1) mechanismrecently identified in humans. It is suggested that proteolysis is involved in the inactivation of xenobiotics by the POH1 system.Two cDNAs were cloned, one from the demosponge Geodia cydonium and a second from the hexactinellid sponge Aphrocallistesvastus. The cDNA from G. cydonium, termed GCPOHL, encodes a deduced polypeptide with a size of 34,591 Da and that fromA. vastus, AVPOHL, a protein of a calculated Mr of 34,282. The two sponge cDNAs are highly similar to each other as well as tothe known sequences from fungi (Schizosaccharomyces pombe and Saccharomyces cerevisiae) and other Metazoa (from Schistosomamansoni to humans). Under controlled laboratory conditions, the expression of the potential multidrug resistance gene POHL is,in G. cydonium, strongly upregulated in response to the toxins staurosporin (20 mM) or taxol (50 mM); the first detectable transcriptsappear after 1 d and reach a maximum after 3 to 5 d of incubation. The relevance of the expression pattern of the G. cydoniumgene POHL for the assessment of pollution in the field was determined at differently polluted sites in the area around Rovinj(Croatia; Mediterranean Sea, Adriatic Sea). The load of the selected sites was assessed by measuring the potency of XAD-7concentrates of water samples taken from those places to induce the level of benzo[a]pyrene monooxygenase (BaPMO) in fish andto impair the multidrug resistance (MDR)/P-170 extrusion pump in clams. These field experiments revealed that the levels ofinducible BaPMO activity in fish and of the MDR potential by the water concentrates are highly correlated with the level ofexpression of the potential multidrug resistance gene POHL in G. cydonium. This report demonstrates that the detoxification POHpathway, here mediated by the G. cydonium GCPOHL gene, is an additional marker for the assessment of the environmental loadin a given marine area.

Keywords—Multidrug resistance Northern Adriatic POHL gene expression Chemosensitizers Bioindicator

INTRODUCTION

Sponges (Porifera) represent the phylogenetically oldest,still extant metazoan taxon, which comprises approximately10,000 species [1]. Molecular biological studies revealed thatsponges form, with the other Metazoa, a monophyletic king-dom [2]. These animals, which are primarily marine organisms,are found in all aquatic habitats from the tropics to the arctics;they are the most abundant multicellular animals (macrofauna)found in the marine hard-substrate benthos both with respectto the number of species and in biomass [3]. As sessile filterfeeders [4], some of them filter every day ;24,000 L/kg ofwater [5] or pump a volume of water equal to their own volumeevery 5 s [6]. Therefore, these animals must have developedpotent strategies to protect themselves against the continuousexposure to environmental stress (e.g., dissolved organic car-bon) [7] as well as against foreign invaders.

Sponges are provided with at least two general types ofdefense mechanisms against environmental stress and attack-ing organisms, namely cellular defense systems, e.g., heat-shock proteins, targeting chaperones like the 14-3-3 protein(s),multidrug resistance, or thioredoxin, and organismic protec-tion, including a well-developed innate immune system andbioactive secondary metabolites [8–10]. In the past, our groupsestablished that sponge cells are provided with a multidrug

* To whom correspondence may be addressed(wmueller@mail.uni-mainz.de).

resistance (MDR) mechanism that appears to be identical withthe MDR-associated protein P-170 pump system found in ver-tebrate tumor cells [11]. The MDR was demonstrated for thesponges Tethya lyncurium [12], Suberites domuncula [13],Geodia cydonium, and Verongia aerophoba [14]. It was dem-onstrated that verapamil, a drug known to inhibit the P-170extrusion pump [15], causes accumulation of xenobiotics, like2-acetylaminofluorene, benzo[a]pyrene, daunomycin, vincris-tine, calcein-AM, or rhodamine B [8]. For the clam Corbiculafluminea, it was shown that verapamil causes an accumulationof 2-acetylaminofluorene, resulting in an augmentation of thegenotoxicity of this model carcinogen [16].

Recently, the group of Dubiel [17] demonstrated that, invertebrates, besides the P-170 extrusion mechanism, a secondpleiotropic resistance system exists that appears to be a func-tional homologue of the PAD-mediated resistance pathway inSchizosaccharomyces pombe [18]; the corresponding gene inhumans was termed pad one homologue (POH1) [17]. Trans-fection studies revealed that POH1 expression confers resis-tance especially to the protein kinase inhibitor staurosporinand the spindle poison taxol [17]. It has been suggested thatPOH1 is a novel component of the 26S proteasome complex,which mediates resistance through AP-1 transcription factors,suggesting that proteolysis is involved in determining the sen-sitivity of human cells to toxic treatments [17].

In the present study, we demonstrate that the sponges Geo-dia cydonium (demosponge) and Aphrocallistes vastus (an

Multidrug resistance gene POHL Environ. Toxicol. Chem. 20, 2001 199

Fig. 1. Sampling sites for G. cydonium in the Rovinj area (northernAdriatic, Mediterranean Sea). The sites are termed S-1, Rovinj; S-2,Rovinj; S-3, San Giovanni; S-4, entrance to the Limski Kanal; andS-5, end of the Limski Kanal. The animals were collected and ribon-ucleic acid (RNA) immediately extracted. Northern blot experimentswere performed as described in Materials and Methods. The sampleswere hybridized with the GCPOHL probe. The Northern blot withthe references to the sites shown under A are shown in panel B. Therelative level of expression of the gene for GCPOHL is indicated inpanel A. The bars (and the added numbers) mark the level of themultidrug resistance gene POHL expression in the sponge tissue.

hexactinellid sponge) contain genes that display high sequencesimilarity to the human POH1 gene. In addition, it is shownthat the expression of the G. cydonium gene is upregulatedafter incubation with staurosporin and taxol in controlled lab-oratory experiments and also in the field at sites of high loadwith xenobiotics. As a functional test for the definition of thedegree of pollution at those sites from which the animals weretaken from the sea, the level of benzo[a]pyrene monooxygen-ase (BaPMO) activity in the liver of fish and the MDR-inhib-itory potential of XAD-7 concentrates were determined.

MATERIALS AND METHODS

Materials

Staurosporin (S4400) was obtained from Sigma (St. Louis,MO, USA), XAD-7 (Serdolit AD-7) resin from Serva (Hei-delberg, Germany), rhodamine B from Molecular Probes Eu-rope BV (Leiden, The Netherlands), and cyclosporin A fromSandoz (Nurnberg, Germany). Restriction endonucleases andother enzymes for recombinant DNA techniques and vectorswere obtained from Stratagene (La Jolla, CA, USA), QIAGEN(Hilden, Germany), Boehringer Mannheim (Mannheim, Ger-many), GibcoBRL (Grand Island, NY, USA), Amersham(Buckinghamshire, UK), USB (Cleveland, OH, USA), DU-PONT (Bad Homburg, Germany), Epicentre Technologies(Madison, WI, USA), and Promega (Madison, WI, USA). TaqDNA polymerase, DIG (digoxigenin) DNA labeling kit, DIG-11-dUTP, Fab fragment of anti-DIG alkaline phosphatase, andCDP (disodium 2-chloro-5-(4-methoxyspiro{1,2-dioxetane-3,29-(59-chloro)-tricyclo[3.3.1.13,7]decan}-4-yl)phenyl phos-phate) were from Boehringer Mannheim (Mannheim, Ger-many). Taxol (Paclitaxel) was a gift of Bristol-Myers Squibb(Munchen, Germany).

Sponge—Geodia cydonium

Live specimens of G. cydonium (Porifera, Demospongiae,Geodiidae) were collected by SCUBA diving near Rovinj (Cro-atia; Mediterranean Sea, Adriatic Sea) and were either im-mediately frozen in liquid nitrogen until analysis or immedi-ately exposed to the indicated compounds.

For the laboratory experiments, the animals were collectedfrom the nonpolluted site S-3 (Fig. 1). For studies to determinethe level of gene expression, G. cydonium specimens weretaken from five different sites (Fig. 1), which are in the closevicinity of fixed sampling sites used in earlier environmentalstudies [19], characterized by a distinct gradient of pollution.Site S-1 is a small bay at Rovinj (influence of a fish cannery);S-2 is at the rim of the harbor area of Rovinj (direct influenceof urban runoff and a tobacco factory); S-3 is near the islandof S. Giovanni in Pelago (uncontaminated); S-4 is the entranceto the Limski Kanal, placed close to the tourist camps Blesickain the north and Valalta in the south (tourist waste); and S-5is 2 km from the end of the Limski Kanal (uncontaminatedand only occasionally under the influence of a fish farm).

Sponge—Aphrocallistes vastus

The specimens of A. vastus (Porifera, Hexactinellida,Aphrocallistidae) were collected from Saanich Inlet and Bar-kley Sound, British Columbia (Canada), by SCUBA diving.They were a gift of Sally P. Leys. The material was imme-diately frozen in liquid nitrogen until use.

Exposure of Geodia cydonium to staurosporin and taxol

Tissue samples of ;10 g of G. cydonium remained eitheruntreated or were exposed to 20 mM of staurosporin or to 50mM of taxol for 0 to 5 d in 10-L aquaria at 178C under con-tinuous aeration; the water was changed once at day 2. Aliquotsof the respective sponge samples were immediately frozen inliquid nitrogen and stored at 2808C.

Construction of cDNA libraries

The cDNA libraries from both G. cydonium and A. vastuswere constructed as described previously [20,21]. Briefly, totalribonucleic acid (RNA) was extracted from sponge tissue andpolyadenylated mRNA was isolated. cDNA was prepared fromG. cydonium by using the ZAP Express cDNA synthesis kit(Stratagene). The cDNA library of A. vastus was prepared inHybriZAP (Stratagene) and packaged in vitro using MaxPlaxPackaging Extract (Epicentre Technologies). The resultingnumber of independent plaque-forming units of the primarylibrary was approximately 2.4 3 106; the amplified library wasstored at 48C.

Polymerase chain reaction cloning of Geodia cydoniumand Aphrocallistes vastus multidrug resistance cDNAPOHL

The sponge cDNAs, encoding the POH1-like molecules,termed POHL, were isolated from the G. cydonium and A.vastus cDNA libraries by polymerase chain reaction [22]. Thedegenerate forward primer, directed against the conserved aasegments found in the sequences from the human POH1 (ac-

200 Environ. Toxicol. Chem. 20, 2001 A. Krasko et al.

cession number NPp005796 [17]; aa50–aa57) and the Schisto-soma mansoni Pad1 homologue (AAC02298; aa53–aa60) 59-CCC/AATGGAA/GGTGATGGGGCTGATG-39, was used inconjunction with the 39-end vector-specific primer. The poly-merase chain reaction was carried out at an initial denaturationat 958C for 3 min, then 30 amplification cycles at 958C for 30s, 638C for 45 s, 748C for 1.5 min, and a final extension stepat 748C for 10 min. The reaction mixture was as describedearlier [23]. Fragments of ;800 bp for both cDNAs wereobtained. The two sequences were completed by primer walk-ing as described [22]. The longest inserts obtained had a sizeof 1,010 nucleotides (nt) (excluding the poly(A) tail) for theG. cydonium sequence GCPOHL and 1,067 nt for the A. vastuscDNA AVPOHL. The clones were sequenced using an auto-matic DNA sequencer (Li-Cor 4200, MWG Biotech, Ebers-berg, Germany).

Sequence comparisons

The sequences were analyzed using computer programsBLAST [24] and FASTA [25]. Multiple alignments were per-formed with CLUSTAL W (Version 1.6) [26].

Northern blotting

The RNA was extracted from liquid-nitrogen pulverizedsponge tissue with TRIzol Reagent (GibcoBRL) as recom-mended by the manufacturer. A total of 4 mg of RNA waselectrophoresed through a formaldehyde/agarose gel and blot-ted onto a Hybond N1 membrane following the manufacturer’sinstructions (Amersham). Hybridization experiments were per-formed with ;800 bp fragments of GCPOHL from G. cy-donium and the G. cydonium b-tubulin (accession numberY17002) GCBTUB (;800 bp). These probes were labeled withDIG-11-dUTP by the DIG DNA labeling kit. Hybridizationwas performed with the antisense DIG-labeled probes at 428C,following the instructions of the manufacturer (Boehringer)and as recently described [27]. After washing, DIG-labelednucleic acid was detected with anti-DIG Fab fragments andvisualized by the chemiluminescence technique using CDP-Star. The signals of the Northern blots were quantitated by thechemiluminescence procedure [27]. The screen was scannedwith the GS-525 Molecular Imager (Biorad, Richmond, CA,USA). The relative values for the expressions of the GCPOHLand GCBTUB genes in G. cydonium tissue were correlatedwith the intensities of the bands measured for the expressionof the tubulin gene.

Preparation of XAD-7 water concentrates

Water samples of 25 L were passed through an XAD-7column (particle size of 0.3–1.0 mm; 5 g) at a flow rate of 80ml/min. The adsorbed material was eluted from the resin with30 ml of acetone; this material was evaporated in a rotaryevaporator and the dry residue was dissolved in corn oil (fordetermination of BaPMO activity) or dimethyl sulfoxide forthe experiments to determine the effect on the P-170 transportprotein (ratio, 25 L:120 ml water:corn oil or dimethyl sulf-oxide).

Determination of BaPMO activity

The level of BaPMO activity in the liver of the fish commongrey-mullet, Mugil cephalus, was determined either after in-jection of 100 ml corn oil alone (controls) or of the oil togetherwith the XAD-7 water concentrates as described [28]. Twodays after injection, the BaPMO activity was determined in

the postmitochondrial fraction. The enzyme activity is givenin picomoles of benzo[a]pyrene hydroxide (BaPOH) formedper milligram of protein 3 minutes.

Bioaccumulation activity

Bioaccumulation activity on the P-170 pump system, as ameasure for the level of toxic compounds in the water con-centrates present, was determined in living gill cells from Cor-bicula fluminea, as described originally by Cornwall et al.[29]. The principle of the assay is based on the extrusion ofthe fluorescent dye rhodamine B from the cells mediated bythe multixenobiotic P-170 transport protein [30]. A high ex-trusion (efflux) of rhodamine B reflects a high activity of theP-170 pump; in contrast, a low extrusion rate indicates a com-petitive inhibition of the transporter.

The clams (density ø 4 clams/50 ml) were exposed to 10mM of rhodamine B in dechlorinated tap water alone or to-gether with XAD-7 water extracts (equivalent of 200 ml watersample) or with verapamil (1–30 mM) for 1 h as described[31]. After washing the animals, they were kept for 5 min indye-free water to allow rhodamine B to be released from themantle cavity. Subsequently, the amount of rhodamine B ingills from individual specimens was determined by the fluo-rescence procedure after homogenization of the tissue [31].The inhibitory potential of the extracts is expressed as verap-amil-equivalents per extract, as described [31,32].

Other methods

For statistics, the results were analyzed by paired Student’st tests [33]. The protein content was determined using theLowry method [34].

RESULTS

Cloning of the multidrug resistance cDNAs GCPOHL fromGeodia cydonium and AVPOHL from Aphrocallistes vastus

The cDNAs, encoding the putative POHL-like polypeptidesfrom G. cydonium (GCPOHL) and A. vastus (AVPOHL) are1,010 nt (G. cydonium) and 1,067 nt (A. vastus) long. The G.cydonium sequence has a potential open reading frame fromthe AUG initiation codon at nt 20 to 22 to the stop codon at950 to 952 that encodes a 310-aa-long polypeptide (Fig. 2).The deduced aa sequence of the putative multidrug resistanceprotein from G. cydonium, POHLpGEOCY, has a putative size(Mr) of 34,591 and an isoelectric point of 6.22 [35]. Northernblot analysis performed with the sponge GCPOHL clone as aprobe yielded one band of 1.2 kb, confirming that the fulllength cDNA was isolated (Fig. 3).

The open reading frame (from nt41/43 [start codon] to nt968/970

[stop codon]) of the A. vastus sequence AVPOHL is 924 ntlong and encodes the 308-aa-long polypeptide POHLpAPHVA(Fig. 2). The calculated Mr is 34,282, and the estimated iso-electric point is 5.98 [35]. One transcript size of 1.3 kb wasidentified, again indicating that the complete sequence wasobtained (data not shown).

The two polypeptides POHLpGEOCY and POHLpAPHVAare related to each other and share 76% of identical aa and85% of similar aa (with respect to the amino acid charge). Adatabank search with the G. cydonium polypeptidePOHLpGEOCY revealed a high identity (similarity) to the hu-man 26S proteasome-associated pad1 homologue [17] with81% (86%), the Pad1 homologue from Schistosoma mansoni(accession AAC02298) with 74% (84%), the proteasome reg-ulatory subunit from Caenorhabditis elegans (accession

Multidrug resistance gene POHL Environ. Toxicol. Chem. 20, 2001 201

Fig. 2. Putative sponge multidrug resistance proteins POHL. Comparison of the two sponge deduced polypeptides encoding the putative multidrugresistance proteins POHLpGEOCY and POHLpAPHVA, with the related human 26S proteasome-associated pad1 homologue (pPOH1pHUMA,accession number NPp005796) and the Pad1 homologue from Schistosoma mansoni (PAD1HpSCHM, AAC02298). Amino acids identical in allsequences are given in white on black (and as capitals above the sequences) and those in at least three sequences are in small capitals (abovethe sequences). The location of the primer at the nt level is indicated by a line.

Fig. 3. Effect of staurosporin and taxol on the expression of the po-tential multidrug resistance gene POHL in G. cydonium. Tissue sam-ples from G. cydonium were incubated for 0 d (Con [control], lanea), 1 d (lane b), 2 d (lane c), 3 d (lane d), or 5 d (lane e) in thepresence of staurosporin (20 mM) (A) or taxol (50 mM) (B). ThenNorthern blot analyses to estimate the level of expression of the geneswere performed using the probe for the potential multidrug resistancegene POHL from G. cydonium, GCPOHL. After blot, transfer hy-bridization was performed with the probe GCPOHL. In parallel, blotsfrom staurosporin-treated tissue were hybridized with a probe fromG. cydonium, encoding b-tubulin (C). The intensities of the transcriptsfor GCPOHL are correlated with the expression of b-tubulin.

AAC26287) with 71% (82%), the PAD1 protein from Schi-zosaccharomyces pombe [18,36] with 65% (78%), and theSaccharomyces cerevisiae MPR1 protein (accession numberP43588) with 63% (76%) (Fig. 2).

Definition of the degree of pollution at the sampling sitesof Geodia cydonium

Two approaches were chosen to estimate the degree of en-vironmental load at the sampling sites. Both are based onfunctional investigations.

The level of BaPMO activity is indicative of the load ofthe marine environment with compounds such as polynucleararomatic hydrocarbons or some polychlorinated biphenyls[37]. Water samples from the five different sites chosen tocollect G. cydonium specimens were concentrated followingthe XAD-7 procedure, as described under Materials and Meth-ods. Aliquots were injected into the fish Mugil cephalus. Twodays later, the BaPMO activity was determined in the liver.The quantitation of the experiments revealed that the enzymeactivity, expressed in picomoles of BaPOH/mg of protein 3minutes, is lowest at site S-3, with 1.9 6 0.6 pmol/mg ofprotein 3 minutes, followed by sites S-5 (2.3 6 0.9) and S-4 (7.2 6 1.6); significantly higher levels (p , 0.001; n 5 10)were determined at S-1 (32.7 6 3.4) and S-2 (36.0 6 4.4)(Table 1). In control experiments, it was established that cornoil alone did not induce BaPMO activity; the level was ,0.56 0.2.

To determine the effect of pollutants potentially present inXAD-7 extracts, bioaccumulation experiments with gills fromC. fluminea were performed. Animals were loaded with rho-damine B in the absence or presence of XAD-7 water extractsas described in Materials and Methods. To estimate the inhib-itory potential of the concentrates, the fluorescence values ob-

202 Environ. Toxicol. Chem. 20, 2001 A. Krasko et al.

Table 1. Levels of benzo[a]pyrene monooxygenase (BaPMO) activity in liver of the fish Mugil cephalus 2 d after injection of XAD-7 waterconcentrates; the activity is given in picomoles of BaPOH/mg protein 3 minute; n 5 10. The multidrug resistance (MDR) potential in the waterconcentrates was measured in living gill cells from Corbicula fluminea. The inhibitory effect of the concentrates are indicated as verapamil-equivalents (mM of Ver-equ); n 5 5. Results are means (6 standard deviation) of n individual experiments. The expression level of the potentialmultidrug resistance gene POHL in specimens collected from the respective areas was determined by Northern blotting; the lowest expression

at site S-3 was set arbitrarily to one-fold

Site Description

BaPMOactivity

(pmol/mg·min)MDR potential

(mM of Ver-equ)POHL

expression

S-1S-2S-3S-4S-5

Fish canneryUrban runoff; tobacco factoryIsland (uncontaminated)Tourist campsLimski Kanal (uncontaminated)

32.7 6 3.436.0 6 4.4

1.9 6 0.67.2 6 1.62.3 6 0.9

24.6 6 2.929.1 6 2.8

1.4 6 0.84.8 6 2.22.0 6 1.1

23.7-fold18.3-fold

1-fold9.4-fold8.5-fold

tained in the experiments with the XAD-7 water extracts werecompared with those obtained after addition of verapamil, aknown inhibitor of the P-170 extrusion pump [15]. The effectsof the XAD-7 water extracts on the accumulation of rhodamineB are given in verapamil-equivalents. The experiments re-vealed again that the water extracts from site S-3 show thelowest effect on the P-170 extrusion pump, with an MDR-inhibitory potential of 1.4 6 0.8 verapamil-equivalents, fol-lowed by the extracts from sites S-5 (2.0 6 1.1) and S-4 (4.86 2.2) (Table 1). Significantly higher inhibitory potential (p, 0.001; n 5 5) was measured at sites S-2 (29.1 6 2.8) andS-1 (24.6 6 2.9) (Table 1). Preliminary experiments performedin parallel using P-170 preparations from G. cydonium re-vealed that, in this species, this system is not inducible (datanot shown).

Expression of the potential multidrug resistance genePOHL in response to toxins

Tissue from G. cydonium was incubated in the presence ofstaurosporin (20 mM) or taxol (50 mM) for 0 to 5 d undercontrolled laboratory conditions. After the respective time pe-riod, RNA was extracted from the tissue and the level ofGCPOHL expression was determined semiquantitatively byNorthern blotting (Fig. 3). In the absence of staurosporin ortaxol, no expression of the gene encoding the potential mul-tidrug resistance gene POHL can be detected (Fig. 3A and B).However, after an incubation of only 1 d in the presence ofeither of the two drugs, the expression of GCPOHL increasedsignificantly (;0.2-fold) and reached a maximum after 3 to 5d (.onefold [with respect to the expression of b-tubulin]). Inparallel blots, the level of expression of b-tubulin RNA fromstaurosporin-treated tissue was determined; no significant dif-ferences were seen, confirming that the same amount of RNAwas applied (Fig. 3C).

Expression of the potential multidrug resistance genePOHL in the field

After having assessed the level of pollution, using the in-duction of the BaPMO system and the effect of the concen-trates on the function of the P-170 extrusion pump as param-eters, specimens were collected at the same sites in the Rovinjarea to determine the steady-state level of GCPOHL transcripts(Fig. 1). The lowest expression was found at site S-3 (S. Gio-vanni in Pelago); this expression was set arbitrarily to onefold.The analysis to determine the relative abundance of GCPOHLtranscripts in the same amount of RNA, obtained from tissuesamples from animals collected at the four other sites, revealedthe increasing order of expression as S-5 (end of the Limski

Kanal [8.5-fold]) , S-4 (entrance to the Limski Kanal [9.4-fold]) , S-2 Rovinj (harbor area [18.3-fold]) , S-1 Rovinj(fish cannery [23.7-fold]).

In Table 1, the levels of inducible BaPMO activity in fishliver and of the MDR potential by XAD-7 water concentratesare correlated with the level of expression of the potentialmultidrug resistance gene POHL of G. cydonium. The datashow a strong correlation between all three parameters of pol-lution with a gradient from S-3 (S. Giovanni in Pelago; lowestload) to the sites of highest level of pollution at S-2 Rovinj(harbor area) and at S-1 Rovinj (fish cannery [23.7-fold]).

DISCUSSION

Marine sponges show a wide-spread distribution, from theArctic to Antarctic, and live on all substrata, from a rockyenvironment (like G. cydonium at S-4 and S-5) to a sandy-muddy bottom (at site S-1) [30]. It is especially the marinebottom that accumulates adverse organic and inorganic ma-terial [38]. Therefore, sponges have had to develop powerfuldefense mechanisms against toxic compounds. It is now wellestablished that these animals have an efficient P-170/MDRextrusion pump [8], which eliminates xenobiotics that had beentaken up by the cells. Poisoning of the pump by toxins resultsin rapid cell death [13,39].

In the present study, it is demonstrated that sponges, withthe demosponge G. cydonium and hexactinellid sponge A. vas-tus as examples, are provided with a second type of multidrugresistance that is mediated by one component of the 26S pro-teasome complex, i.e., the polypeptide synthesized by the POHgene. The cDNAs from both animals were isolated; the sim-ilarity of the deduced sponge polypeptides to both fungal (S.pombe and S. cerevisiae) and metazoan sequences (from S.mansoni to humans) is unusually high, with .75% with re-spect to physico–chemical-related aa. This finding already sup-ports the notion that this POH pathway is of high functionalimportance. The mechanism by which the POH pathway con-fers resistance is unknown; it is suggested that proteolysis isinvolved in the inactivation of xenobiotics [17].

In the first set of the functional studies, the established drugsfound in earlier experiments with S. pombe [36] and humans[17] to be detoxified by the POH pathway, i.e., the proteinkinase inhibitor staurosporin and the spindle poison taxol, wereapplied. After an incubation period of only 1 d in the presenceof these two drugs, the expression of the G. cydonium POH-like gene, GCPOHL, is strongly upregulated. While in control(untreated) samples the transcript is not detectable, theGCPOHL transcripts are abundant after day 3.

In the next series of experiments, it was determined whether

Multidrug resistance gene POHL Environ. Toxicol. Chem. 20, 2001 203

the expression of the GCPOHL gene is different in the field,depending on the load of environmental pollution. As a firstparameter for the degree of pollution, extracts from differentsites around the Rovinj area (Croatia) in the Northern AdriaticSea were tested for their potency to induce the enzymeBaPMO. In a recent study, established that enzymes involvedin the BaPMO system, members of the cytochrome P450 su-perfamily, are present in sponges [40]. The experiments re-vealed that XAD-7 concentrates caused a range of up to 19-fold increased induction from the nonpolluted site (the islandof S. Giovanni) to the highly polluted location (at high urbanrunoff and the outlets of a tobacco factory) in fish, dependingon the location chosen. As the second parameter, the effect ofthe concentrates on the bioaccumulation activity was deter-mined. The assay system suitable to determine the efficacy ofthe P-170 pump system was applied. The measurementsshowed again the same order of pollution load from the islandof S. Giovanni to the highly polluted location at the high urbanrunoff and the outlets of a tobacco factory.

Geodia cydonium specimens taken from these selected siteswere analyzed for the expression of the GCPOHL gene. TheNorthern blot experiments clearly demonstrate that the ex-pression of this gene parallels the determined increasing levelof BaPMO-inducible activity in fish and the degree of poi-soning of the MDR/P-170 pump system in gills of the clamscaused by the XAD-7 concentrates taken from the identicalsites. Our preliminary experiments suggest that, in G. cydon-ium, the P-170 system is not inducible (data not shown).

Taken together, this report demonstrates for the first timethat the detoxification POH pathway, here mediated by the G.cydonium GCPOHL gene, is an additional marker for the as-sessment of the environmental load in a given marine area. Inaddition, these data indicate that both systems, the MDR/P-170 pump mechanisms and the proteasome-associated POHpathway, are reliable methods for measuring a great varietyof xenobiotics/chemosensitizers.

Acknowledgement—This work was supported by grants from the Eu-ropean Commission, Marine Science and Technology ProgrammeMast—MAS3-CT97-0118, and the International Human Frontier Sci-ence Program, RG-333/96-M. The sequences reported here are beingdeposited in the European Molecular Biology Laboratory ([EMBL],Heidelberg, Germany) data base (Geodia cydonium potential multi-drug resistance gene POHL, GCPOHL [accession AJ249917] and thecorresponding gene from Aphrocallistes vastus, AVPOHL[AJ245813]).

REFERENCES

1. Muller WEG. 1998. Origin of Metazoa: Sponges as living fossils.Naturwissenschaften 85:11–25.

2. Muller WEG. 1995. Molecular phylogeny of Metazoa (animals):Monophyletic origin. Naturwissenschaften 82:321–329.

3. Sara M, Vacelet J. 1973. Ecologie des Demosponges. In GrassePP, ed, Traite de Zoologie: Spongiaires. Tome III (1). Masson,Paris, pp 462–576.

4. Simpson TL. 1984. The Cell Biology of Sponges. Springer-Verlag,New York, NY, USA.

5. Vogel S. 1977. Current-induced flow through living sponges innature. Proc Natl Acad Sci USA 74:2069–2071.

6. Reiswig HM. 1974. Water transport, respiration, and energeticsof three tropical marine sponges. J Exp Mar Biol Ecol 14:231–249.

7. Clark RB. 1997. Marine Pollution. Clarendon Press, Oxford, UK.8. Kurelec B. 1997. A new type of hazardous chemical: The che-

mosensitizers of multixenobiotic resistance. Environ Health Per-spect 105:855–860.

9. Muller WEG, Blumbach B, Muller IM. 1999. Evolution of theinnate and adaptive immune systems: Relationships between po-

tential immune molecules in the lowest metazoan phylum [Por-ifera] and those in vertebrates. Transplantation 68:1215–1227.

10. Sarma AS, Daum T, Muller WEG. 1993. Secondary Metabolitesfrom Marine Sponges. Akademie gemeinnutziger Wissenschaftenzu Erfurt, Ullstein-Mosby Verlag, Berlin, Germany.

11. Bellamy WT. 1996. P-glycoproteins and multidrug resistance.Annu Rev Pharmacol Toxicol 36:161–183.

12. Kurelec B, Pivcevic B. 1992. The multidrug resistance mechanismin the marine sponge Tethya aurantium. Mar Environ Res 34:249–253.

13. Muller WEG, Steffen R, Rinkevich B, Kurelec B. 1996. Multix-enobiotic resistance mechanism in the marine sponge Suberitesdomuncula: Its potential application in assessment of environ-mental pollution. Mar Biol 125:165–170.

14. Kurelec B, Krca S, Pivcevic B, Ugarkovic D, Bachmann M, Im-siecke G, Muller WEG. 1992. Expression of P-glycoprotein genein marine sponges. Identification and characterization of the 125-kDa drug-binding glycoprotein. Carcinogenesis 13:69–76.

15. Gottesman MM, Pastan I. 1993. Biochemistry of multidrug re-sistance mediated by the multidrug transporter. Annu Rev Bioch-em 62:385–427.

16. Waldmann P, Pivcevic B, Muller WEG, Zahn RK, Kurelec B.1995. Increased genotoxicity of aminoanthracene by modulatorsof multixenobiotic resistance mechanism: Studies with the freshwater clam Corbicula fluminea. Mutat Res 342:113–123.

17. Spataro V, Toda T, Craig R, Seeger M, Dubiel W, Harris AL,Norbury C. 1997. Resistance to diverse drugs and ultraviolet lightconferred by overexpression of a novel human 26 S proteasomesubunit. J Biol Chem 272:30470–30475.

18. Shimanuki M, Saka Y, Yanagida M, Toda T. 1995. A novel es-sential fission yeast gene pad11 positively regulates pap1(1)-dependent transcription and is implicated in the maintenance ofchromosome structure. J Cell Sci 108:569–579.

19. Bihari N, Batel R. 1994. Alkaline elution of mussel DNA as atool for determination of environmental contamination by gen-otoxins. In Muller WEG, ed, Use of Aquatic Invertebrates asTools for Monitoring of Environmental Hazards. Gustav Fischer-Verlag, Stuttgart, Germany, pp 35–48.

20. Pfeifer K, Haasemann M, Gamulin V, Bretting H, Fahrenholz F,Muller WEG. 1993. S-type lectins occur also in invertebrates:High conservation of the carbohydrate recognition domain in thelectin genes from the marine sponge Geodia cydonium. Glyco-biology 3:179–184.

21. Skorokhod A, Gamulin V, Gundacker D, Kavsan V, Muller IM,Muller WEG. 1999. Origin of insulin receptor tyrosine kinase:Cloning of the cDNAs from marine sponges. Biol Bull 197:198–206.

22. Ausubel FM, Brent R, Kingston RE, Moore DD, Smith JA, Seid-mann JG, Struhl K. 1995. Current Protocols in Molecular Bi-ology. John Wiley & Sons, New York, NY, USA.

23. Wiens M, Koziol C, Hassanein HMA, Batel R, Muller WEG.1998. Expression of the chaperones 14-3-3 and HSP70 inducedby PCB 118 (2,39,4,49,5-pentachlorobiphenyl) in the marinesponge Geodia cydonium. Mar Ecol Prog Ser 165:247–257.

24. Pearson WR, Lipman DJ. 1988. Improved tools for biologicalsequence comparison. Proc Natl Acad Sci USA 85:2444–2448.

25. Lipman DJ, Pearson WR. 1985. Rapid and sensitive protein sim-ilarity searches. Science 277:1435–1441.

26. Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W:Improving the sensitivity of progressive multiple sequence align-ment through sequence weighting, positions-specific gap penaltiesand weight matrix choice. Nucleic Acids Res 22:4673–4680.

27. Krasko A, et al. 1999. Ethylene modulates gene expression incells of the marine sponge Suberites domuncula and reduces thedegree of apoptosis. J Biol Chem 274:31524–31530.

28. Kurelec B, Protic M, Britvic S, Kezic N, Rijavec M, Zahn RK.1981. Toxic effects in fish and mutagenicity capacity of waterfrom the Sava River. Bull Environ Contam Toxicol 26:179–187.

29. Cornwall R, Toomey BH, Bard S, Bacon C, Jarman W, Epel D.1995. Characterization of multixenobiotic/multidrug transport inthe gills of the mussel Mytilus californianus and identificationof environmental substrates. Aquat Toxicol 31:277–296.

30. Neyfakh AA. 1988. Use of fluorescent dyes as molecular probesfor the study of multidrug resistance. Exp Cell Res 174:168–176.

31. Smital T, Kurelec B. 1997. The concentrations of inhibitors ofmultixenobiotic resistance mechanism in natural waters: The di-

204 Environ. Toxicol. Chem. 20, 2001 A. Krasko et al.

rect in vivo demonstration of their effect. Environ Toxicol Chem16:2164–2170.

32. Muller WEG, Steffen R, Kurelec B, Smodlaka N, Puskaric S,Jagic B, Muller-Niklas G, Queric NV. 1998. Chemosensitizers ofthe multixenobiotic resistance in the amorphous aggregates (ma-rine snow): Etiology of mass killing on the benthos in the North-ern Adriatic? Environ Toxicol Pharmacol 6:229–238.

33. Sachs L. 1984. Angewandte Statistik. Springer-Verlag, Berlin,Germany.

34. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Proteinmeasurement with the folin phenol reagent. J Biol Chem 193:265–275.

35. IntelliGenetics. 1995. PC/GENE Data Banks CD-ROM [EMBL,Swiss Prot], Release 14.0. Mountain View, CA, USA.

36. Usui T, Yoshida M, Honda A, Beppu T, Horinouchi S. 1995. 5AK-252a-resistance gene, sks11, encodes a protein similar to the

Caenorhabditis elegans F37 A4.5 gene product and confers mul-tidrug resistance in Schizosaccharomyces pombe. Gene 161:93–96.

37. Kurelec B, Britvic S, Rijavec M, Muller WEG, Zahn RK. 1977.Benzo[a]pyrene monooxygenase induction in marine fish—Mo-lecular response to oil pollution. Mar Biol 44:211–216.

38. Rand GM. 1995. Fundamentals of Aquatic Toxicology. Taylor &Francis, Washington, DC, USA.

39. Schroder HC, Badria FA, Ayyad SN, Batel R, Wiens M, Has-sanein HMA, Kurelec B, Muller WEG. 1998. Inhibitory effectsof toxin from the marine algae Caulerpa taxifolia and Caulerparacemosa on multixenobiotic resistance in the marine spongeGeodia cydonium. Environ Toxicol Pharmacol 5:119–126.

40. Muller WEG, Wiens M, Batel R, Steffen R, Borojevic R, CustodioMR. 1999. Establishment of a primary cell culture from a sponge:Primmorphs from Suberites domuncula. Mar Ecol Prog Ser 178:205–219.