Induction of (2¢)5¢)oligoadenylate synthetase in the marine
sponges
Suberites domuncula
and
Geodia cydonium
by the bacterial endotoxin lipopolysaccharide
Vladislav A. Grebenjuk
1
, Anne Kuusksalu
2
, Merike Kelve
2
, Joachim Schu¨ tze
1
, Heinz C. Schro¨ der
1
and Werner E. G. Mu¨ ller
1
1
Institut fu
¨
r Physiologische Chemie, Abteilung fu
¨
r Angewandte Molekularbiologie, Johannes Gutenberg-Universita
¨
t, Mainz, Germany;
2
Institute of Chemical Physics and Biophysics, Tallinn, Estonia
Recent studies have shown that the Porifera, with the
examples of the d emosponges Suberites domuncula and
Geodia cydonium, comprise a series of pathways found also

anisms to protect themselves against unfavorable co nditions,
e.g. environmental stress (ultraviolet exposure or xenobiot-
ics) [4,5]. Because sponges have the capacity to process their
own volume of water every 5 s in order to extract edible
material [6] they are exposed to a huge amount of bacteria
and also viruses that are present in the seawater [7,8]. To cope
with these threats, sponges have developed an efficient
chemical defense system [9] as well as humoral and cellular
defense mechanisms [10], that provided also t he basis for the
evolution to metazoan o rganisms [10].
One efficient protection against invading microorganisms
is the (2¢)5¢)oligoadenylate synthetase [(2–5)A synthetase]
system [11–13]. The (2–5)A synthetase(s) is activated by
certain classes of RNA, mainly double-stranded RNA [14].
In vertebrates the (2–5)A pathway is also induced by
interferons [15]. The major enzyme in this pathway, the
(2–5)A synthetase catalyzes the synthesis of a series of 2¢)5¢-
linked oligoadenylates, termed (2–5)A [ ¼ pppA(2¢p5¢A)
n
[p
n
A
n
], with chain lengths of 1 £ n £ 30] from ATP
[16,17]. (2–5)A acts as an allosteric activator of a latent
endoribonuclease, the RNase L, which degrades single-
stranded, viral or cellular RNA [18].
Only very rarely viruses have been observed in sponges
[19], while intracellular bacteria are frequently present [20].
Some of the bacteria (both Gram positive and negative)

(2–5)A synthetase form 1 , termed SD25A-1, a nd for the (2–5)A
synthetase form 2, termed SD25A-2, h ave been de posited in the
EMBL/GenBank database under accession numbers AJ 301652 and
AJ301653, re spectively.
(Received 2 6 September 2001, revised 7 January 2002, ac cepted
11 January 2002)
Eur. J. Biochem. 269, 1382–1392 (2002) Ó FEBS 2002
9-2 functions as a proapoptotic protein of the Bcl-2 family
[29].
Considering the fact that in sponges the molecules
involved in immune response are closer related to deutero-
stomian (vertebrate) a nimals than to Protosto mia (insects or
nematodes; reviewed in [10]), we postulated that also
elements of the (2–5)A system exist in sponges. The first
sponge species studied was Geodia cydonium (Demospong-
iae) which in fact showed high levels of (2–5)A oligoade-
nylate synthesis in comparison t o vertebrate c ells [30]. T he
reaction products were identified by thin-layer chromatog-
raphy, immunologically and by high-performance liquid
chromatography. The biological activity of (2–5)A oligo-
mers was verified by inhibition of the protein synthesis in
rabbit reticulocyte lysate [30]. The (2–5)A synthetase
reaction products were also confirmed by MALDI-MS
and by NMR analysis [31]. We succeeded in cloning the
sponge ( 2–5)A s ynthetase from G. cydonium [32]. A calcu-
lation based on the rates of amino-acid substitutions
revealed that the sponge enzyme branched off from a
common ancestor  520 million years ago.
In view of the finding that sponges do contain the (2–5)A
synthetase system like vertebrates, while this e nzyme is

MATERIALS AND METHODS
Materials
Restriction endonucleases and other enzymes for recombin-
ant DNA techniques and vectors were obtained from
Stratagene (La Jolla, CA, USA), Qiagen (Hilden, Germany),
Roche (Mannheim, Germany), USB (Cleveland, OH, USA),
Amersham (Buckinghamshire, UK) and Promega (Madi-
son, WI, USA). In addition, DIG (digoxigenin) DNA
labeling kit, DIG-11-dUTP, anti-DIG AP Fab fragments,
disodium 2-chloro-5-(4-methoxyspiro{1,2-dioxetane-3,2¢-
(5¢-chloro)-tricyclo[3.3.1.1
3,7
]decan}-4-yl)phenyl phosphate
(CDP) and positively charged nylon membrane (no.
1209272) were from Roche (Mannheim, Germany)
and Na-hexafluorosilicate from Aldrich (Deisenhofen,
Germany). Shrimp alkaline phosphatase was purchased
from USB Corporation (Cleveland, OH, USA), [
14
C]ATP
(542 mCiÆmmol
)1
) from Amersham International PLC
(Buckinghamshire, England), LPS from Escherichia coli
(L2880) a nd adenosine 5¢-triphosphate from Sigma Chemical
Co. (St Louis, MO, USA), and polyethyleneimine cellulose
TLC plates from Sc hleicher & Sc huell (Keene, NH, U SA).
Rotiquant reagent was purchased from Roth (Germany).
Sponge
Live specimens of S. domuncula [Porifera, Demospongiae,

. The incubation temperature was set to 17 °C. The
experiments were p erformed with six animals pe r assay, each.
Incubations
Sponges were cut into cubes with a side of approximately
0.5 c m. For each s et of experiments ( exposure to LPS and
control) samples from t he same sponge specimen were used.
Incubations were performed i n filtered, oxygenated sea-
water. The sponge cubes o r primmorphs were treated f or 0–
24 h in the absence or presence of 1 lgÆmL
)1
or 10 lgÆmL
)1
of LPS in seawater a t ambient temperature. In control
experiments the samples remained untreated for the
entire incubation period. Thereafter the sponge cubes/
primmorphs were immediately frozen ()80 °C).
Cell extracts
Frozen sponge cubes were ground in liquid nitrogen and an
equal a mount (v/w) of t he polymer ase a ssay buffer (PAB)
Ó FEBS 2002 (2¢)5¢)Oligoadenylate synthetase in sponges (Eur. J. Biochem. 269) 1383
[20 m
M
Tris/HCl, pH 7.5, containing 100 m
M
KCl, 5 m
M
MgCl
2
and 5% (v/v) glycerol] was added during homoge-
nization. The primmorphs were suspended in polymerase

was added [38].
The w ells were sealed tightly and the s ynthesis of (2–5)A was
allowed to occur usually for 4–12 h. For HPLC analysis
50 lL of the reaction buffer without radioactive t racer was
used to produce the oligomers and the reactions were
performed in microcentrifuge tubes.
Thin-layer chromatography
The reaction products were eluted with distilled water
and separated by TLC on polyethyleneimine cellulose using
0.4
M
Tris/HCl pH 8.6 , 30 m
M
MgCl
2
as the mobile phase
[30]. The TLC plates were exposed to a CS-imaging screen
and scanned w ith the GS-525 Molecular Imager (Bio-Rad;
Hercules, CA, USA). The amounts of ATP and (2–5)A
oligomers were quantified by the relative intensities of the
corresponding spot areas on the autoradiograms.
High-performance liquid chromatography
The 2¢-5¢ linked oligoadenylates produced in the assay in
their triphosphorylated forms were applied to the HPLC
column (Supelcosil LC-18, 30 cm · 4 mm, 5 lm; Supelco)
and separated in a 0.5–30% methanol gradient in 50 m
M
NH
4
H

obtained from two different c DNA libraries resulting in three
independent clones each. DNA sequencing was performed
with an automatic DNA sequenator (Li-Cor 4000S). Two
different complete sequences have been obtained; they
were termed SD25A-1 and SD25A-2. The corresponding
deduced proteins were named 25A-1_SD and 25A-2_SD.
Sequence comparisons
The sequences were analyzed using computer programs
BLAST
[42] and
FASTA
[43]. Multiple alignments were
performed with
CLUSTAL W
ver. 1.6 [44]. Phylogenetic trees
were constructed on the basis of amino-acid sequence
Fig. 1. S. domuncula: animals and prim-
morphs. (A) The siliceous sponge S. domun-
cula (red) has been kept for m o re than
6 months together with the second demo-
sponge D. avara (violet) in the a quarium
(·0.1). (B–D) Primmorph f ormation of
S. domuncula. (B) Dissociated single cells
(·200). (C) Primm orphs formed after 5 days;
(·5). (D) Cross section throu gh a primmorph,
which has been su bseq uently subjected to
incubation with antiserum r aised against
S. domuncula cells (·5).
1384 V. A. Grebenjuk et al. (Eur. J. Biochem. 269) Ó FEBS 2002
alignments by neighbour-joining, as implemented in the

electrophoresed through 1 % formaldehyde/agarose gel and
blotted onto Hybond N
+
membrane following the m anu-
facturer’s instructions (Amersham; Little Chalfont, Buck-
inghamshire, UK) [5]. Hybridization was performed with a
0.7-kb part of SD25A-1. The probe was labeled with the
PCR-DIG-Probe-Synthesis Kit according to the manufac-
turer’s instructions (Roche). In one series of experiments
poly(A)
+
-RNA was purified from sponge tissue with
Oligotex mRNA kit (Qiagen) and analysed. For the
quantification of the Northern b lot signals the chemilumi-
nescence procedure w as applied [50]; C DP-Star was used as
substrate. The screen was scanned with the GS-525
Molecular Imager (Bio-Rad).
Immunohistological analysis of primmorphs
Fresh tissue was fixed in paraformaldehyde, embedded in
Technovit 8100 and sectioned, essentially as described
previously [23]. The 2-lm thick slices reacted with anti-
serum, raised against S. domuncula cells.
The polyclonal antiserum against cells from S. domuncula
was raised in female rabbits (White New Zealand). An
amount of 3 · 10
6
cells [34] was injected at 4 -week inte rvals;
after three boosts, s erum was prepared [ 51]. The antiserum
obtained was termed anti-S. domuncula.
After fixation of the slices from S. domuncula prim-

Kuusksalu, A. Lopp, T. Reintamm and H. Kelve, unpub-
lished data). The product pattern of the enzyme from
S. domuncula differs under the same reaction conditions
from that of G. cydonium. The synthesis level is significantly
lower and the main synthesis product is p
3
A
2
as verified by
the c omigration with p
3
A
2
standard (TLC and HPLC) and
A
2
standard (HPLC) after dephosphorylation with shrimp
alkaline phosphatase (not shown). Recently we h ave shown
in G. cydonium as well as in S. domuncula crude extracts
that the e nzyme, catalyzing the formation of (2–5)A, does
not require dsRNA for activity (submitted). In the present
study we took advantage of this phenomenon and per-
formed the assays for (2–5)A synthesis with crude cell
extracts using positively charged membranes for partial
purification of the enzymes.
(2–5)A synthetase activity from field/aquarium animals
Interestingly, samples from S. domuncula kept in the
aquarium for 6 months had lower (2–5)A synthesizing
activities compare d to those which were cut into pieces and
frozen after only 2 days maintenance in an aquarium. The

of gram-negative b acteria, may influence the activity of the
(2–5)A synthetase.
Effect of LPS on (2–5)A synthetase activity in tissues
from
S. domuncula
and
G. cydonium
Tissue samples from S. domuncula specimens, kept for
6 m onths in the aquarium are almost devoid of (2–5)A
synthetase activity, under the conditions used. Approxi-
mately 1% of the substrate was converted to (2–5)A
oligomers during 12 h synthesis. Tissue f rom these animals
was used to analyze if the endotoxin LPS has the capacity to
induce the enzyme. The data revealed that in the p resence of
1 lgÆmL
)1
of LPS the (2–5)A synthetase activity started to
increase; after 3–12 h incubation period 4% of the ATP
substratewasconvertedtop
3
A
2
(Table 1). This increase
was transient and during longer incubation periods (24 h)
the p roduct level dropped again. Higher concentrations of
LPS (10 lgÆmL
)1
) caused a lower effect on the (2–5)A
synthesizing activity in S. domuncula.
Despite the initially high (2–5)A synthesizing activity in

brightly with the antiserum (Fig. 1D). Control sections,
incubated with preimmune serum did not show any
reaction.
The experiments show again that after incubation of the
primmorphs with 1 lgÆmL
)1
of LPS for 3 h an increase of
(2–5)A synthetase activity can be measured (from 1.5%
(controls) to 3.3% of th e ATP substrate was converted to
p
3
A
n
after this period), Table 1. This amount does not
change significantly during a prolonged incubation for up to
24 h . The i dentity of the p
3
A
2
product synthesized by the
(2–5)A synthetase both in tissue and in primmorphs of
S. domuncula was verified by TLC and HPLC analysis as
triphosphorylated and/or core oligomers.
Two CDNAs encoding the putative
S. domuncula
(2–5)A
synthetase
Two cDNAs, named SD25A-1 (accession number
AJ301652) and SD25A-2 (AJ301653), have been isolated
which comprise 1175 and 1205 nucleotides. The longest

3
A
2
and p
3
A
3
)whichwere
runinparallelisshown.
1386 V. A. Grebenjuk et al. (Eur. J. Biochem. 269) Ó FEBS 2002
the 25A-1_SD sequence (Fig. 3A). The ATP-binding site
essential for enzyme activity [53,54] resides between amino
acid 273 a nd amino acid 284. The dsRNA b inding region of
(2–5)A synthetase has been narrowed down to the segment
within amino acid 104 and amino acid 158 of the murine
enzyme [53]; in S. do muncula a related stretch has been
found between amino acid 76 and amino acid 125. The
polyA-related domain, found in enzymes such as poly(A)
polymerase (2–5)A synthetase and topoisomerase 1 (acces-
sion number IPR001201 [55]), spans from amino acid 148
and amino acid 212.
Phylogenetic analysis of sponge (2–5)A synthetases
Based o n sequence similarity no sequence related to (2–5)A
synthetases from sponges or from vertebrates, is present in
the Protostomia Caenorhabditis elegans or Drosophila
melanogaster (Advanced
BLAST
available from http://
www.ncbi.nlm.nih.gov/blast/blast.cgi); a related enzyme is
also lacking in y easts (e.g. Saccharomyces cerevisiae)or

(%)
Product
(%)
S. domuncula tissue
Sea – – 83.84 16.26
Aquarium – – 98.88 1.12
Aquarium 1 3 95.34 4.66
Aquarium 1 12 96.61 3.39
Aquarium 1 24 97.78 2.22
Aquarium 10 3 98.78 1.22
Aquarium 10 12 97.80 2.20
Aquarium 10 24 98.90 1.10
S. domuncula primmorphs
Aquarium 0 – 98.48 1.52
Aquarium 1 3 96.71 3.29
Aquarium 1 12 96.44 3.56
Aquarium 1 24 96.28 3.72
Aquarium 10 3 98.41 1.59
Aquarium 10 12 97.93 2.07
Aquarium 10 24 98.38 1.62
Table 2. Determination of (2–5)A synthetase activity in tissue (from sea animals and aquarium animals) of G. cyd onium. Where indicated, incubation
with 1 or 10 lgÆmL
)1
of LPS was performed for 0–24 h. Extracts were prepared, reacted in the enzyme assay with ATP for 3.5 h, the products were
analysed by HPLC as described i n Materials and methods. The amount of 2¢)5¢ linked dimers (p
3
A
2
)aswellastrimers(p
3

Aquarium 1 24 56.80 32.86 10.50 43.20
Aquarium 10 3 60.70 30.81 8.49 39.30
Aquarium 10 12 57.19 32.21 10.60 42.81
Aquarium 10 24 67.11 26.77 6.12 32.89
Ó FEBS 2002 (2¢)5¢)Oligoadenylate synthetase in sponges (Eur. J. Biochem. 269) 1387
outgroup to root the tree. The anthocyanidin synthase is
known to be involved in the catalysis o f the colorless
leucoanthocyanidins to the colored anthocyanidins [56].
The phylogenetic relationship reveals that the t hree sponge
sequences form the basis of the tree from which the
vertebrate sequences bran ch off.
Fig. 3. The t wo putative sponge (2–5)A synthetases from S. domuncula. (A). Alignment o f the amino-acid sequ en ce of the t wo sponge sequences,
25A-1_SD and 25A-2_SD, deduced from the cDNAs SD25A-1 and SD25A-2, with the related proteins from the sponge G. cydonium
(25A_GEOCY, accession number Y1 8497), a s w ell a s f rom m ouse (25 A_MOUSE, P 11928) and from chicken (25A_CHICK, AB0025 86). T he
alignment was performed using the
CLUSTAL W
program. Residues of amino acids, similar a mong all sequences, are in inv erted type and residues
conserved in a t least three sequences are shaded. The characteristic signatures of the (2–5)A synthetase are indicated: the two conserved signatures
(| Sig-1 and | Sig-2), the potential ATP-binding region (|+ ATP), the dsRNA binding segment (|– Bdg: dsRNA) and the polyA-related domain
(|::: polyA-related domain). (B) The phylogenetic relationship of the five (2–5)A synthetase sequenc es. The tree was routed with the distantly related
sequence of anthoc yanidin synthase f rom the plant Dianthus caryophyllus (ANTO_DC, U82432). The numbers at the nodes are an indication o f the
level of confidenc e for the b ranches as d etermined b y bootstrap analysis (1000 b ootstrap r eplicates). The s cale bar ind icates an evolu tionary distance
of 0.1 amino-acid substitutions pe r position i n the se quence.
1388 V. A. Grebenjuk et al. (Eur. J. Biochem. 269) Ó FEBS 2002
Increase in the steady-state level of the (2–5)A
synthetase transcripts by LPS
The effect of LPS on the steady-state level of (2–5)A
synthetase transcripts, SD25A-1, was monitored by Nor-
thern blotting in a semiquantitative way both in t issue from
aquarium animals as well as in primmorphs obtained from

the 1.4-kb s ignal even increased if RNA was an alyzed from
primmorphs, incubated with LPS for 24 h (Fig. 4B, lane e).
Increase in the steady-state level of the (2–5)A
synthetase transcripts during incubation with
E. coli
In view of our earlier finding that S. domuncula cells
respond to exposure to heat-killed E. coli with a reduced cell
proliferation and cell viability [23], primmorphs were
exposed to dead bacteria under the conditions described.
The bacteria were added at a concentration of 10 lgof
nitrogen per mL to the primmorphs; 1 2 and 24 h later RNA
wasextractedandthenprobedwiththeSD25A-1 cDN A in
the Northern blotting experiment.
The results show that in the absence of the heat-killed
bacteria no transcripts, corresponding to 1.4 kb (2–5)A
synthetase mRNA, can be identified in the Northern blotting
approach (Fig. 4C, lanes a to c). In contrast, the steady-state
level of the transcripts increased strongly, even after the short
incubation period of 12 h (Fig. 4D, l ane b vs. lane a ; at time
0). A prolonged incubation for 24 h resulted inan even higher
level of the (2–5)A synthetase transcripts (Fig. 4C, lane c).
DISCUSSION
Inhibition of cell growth, apoptosis and inhibition of
protein synthesis are ways of protection of metazoan
organisms against death caused by microbes. The b acterial
endotoxin LPS causes cell growth inhibition [57] as well as
induction of apoptosis [27] in vertebrates very likely via a
(2–5)A synthetase-mediated pathway. Also in sponges LPS
inhibits cell proliferation and apoptosis [22,23]. Further-
more, LPS strongly inhibits protein s ynthesis in S. domun-

incubated f or 0 h (B; lane c) , 12 h (lane d) o r 24 h (lane e). M , m arker
RNAs, which were run in parallel. In (B; lane a) poly(A)
+
-RNA,
isolated f rom tissue of an aquarium anim als, which was tre ated with
1 lgÆmL
)1
of LPS for 12 h, was analysed. Determination of the effect
of heat-killed E. coli on the steady-state level of (2–5)A synthetase
mRNA in p rimmorphs. Primmorphs were incubated in the absence
(C) or presence of of the heat-killed bacteria (D) for 0–24 h . At the
indicated times primmorphs were taken, RNA was extracted and
subjected at the same concentrations (5 lg) to Northern blotting
experiments using the SD25A-1 cDNA as a probe.
Ó FEBS 2002 (2¢)5¢)Oligoadenylate synthetase in sponges (Eur. J. Biochem. 269) 1389
for a longer period, more than 6 months, in t he aquarium
(aquarium animals) almost no enzyme activity was observed
(Fig. 2 ; Tables 1 and 2). One reason for this effe ct is th e f act
that the bacterial load, with respect to the number a s well as
the species diversity of bacteria, is reduced under the
controlled aquarium conditions (clo sed c ircuit). The reduc-
tion of the bacterial flora in specimens kept in the aquarium
has been recently documented [22].
To test the assumption that bacterial load of sponge
tissue is causatively connected with (2–5)A synthetase
activity, the endotoxin LPS f rom t he outer bacterial cell
wall was used as a substitution/model component. Incuba-
tion studies with tissue f rom aquarium animals (S. domun-
cula, G. cydonium) revealed that LPS causes a significan t
and rapid stimulation of the synthetase activity. The extent

stimulation of (2–5)A synthetase activity by a hitherto
unknown signal transduction pathway. I n a previous study
it had been shown that the mitogen-activated protein k inase
pathway is involved in the cell response to LPS [22]. Until
now a potential involvement o f this pathway in the (2–5)A
synthetase system has not been reported. Nonetheless,
the fast response of the cells to LPS argues in favor of a
post-translational/allosterical activation of t he (2–5)A
synthetase.
TheeffectofLPSonthesteady-statelevelofthe
S. domuncula (2–5)A synthetase transcripts was analyzed
in tissue and primmorphs, incubated with LPS and heat-
killed bacteria. The results revealed that the steady-state
level of the transcripts i s strongly up-regulated after an at
least 12-h incubation period. This finding supports the
view that LPS causes not only a post-translational/
allosteric activation of the (2–5)A synthetase activity in
cells and tissue but also an increased transcript level. The
potency of LPS to modulate gene expression in vertebrate
cells is well established [58]; nevertheless, the involvement
of the toxin in the ( 2–5)A synthetase pathway in these
systems has not yet been described. However, the partici-
pation of LPS in apoptosis has been documented as
reviewed recently [59].
Even though the documentation of virus infection/
presence in sponges is very poor in contrast to that of
bacterial association/infection, which is very a bundant in
Demospongiae, the data presented show that the activity o f
the enzyme as well a s the steady-state level of t he transcripts
of the respective gene increases in cells after LPS/bacteria

Furthermore, seq uence data show that genes encoding a
putative (2–5)A synthetase are present in different sponge
species. This adds further support for the view that the
immune system in sponges is closer related to the deutero-
stomian, vertebrate, t axa than to the protostomian systems
[60], which are lacking not only a series of characteristic
cytokines [61] but also t he (2–5)A synthetase system. Future
transfection studies must show if the genes encoding the
putative (2–5)A synthetases from S. domunc ula are indeed
responsible for the (2–5)A synthetase activity measured in
cells from S. domuncula.
ACKNOWLEDGEMENTS
This work was supported b y grants from the Deutsche Forschungs-
gemeinschaft (Mu
¨
348/14-1), the European Commission (project:
SPONGE), the Bundesministerium fu
¨
r Bildung und Forschung
(project: C enter of Competence BIOTEC-MARIN), the I nternational
Human Frontier Science Program (RG-333/96-M) and the Estonian
Science Foundation.
REFERENCES
1. Mu
¨
ller, W.E.G. (1995) Molecular phylogeny of Metazoa (ani-
mals): monophyletic origin. Naturwiss. 82, 321–329.
2. Mehl, D., M u
¨
ller, I. & Mu

8. Kennish, M.J. (199 4) Practical Handbook o f Marine Science.CRC
Press, Boca Raton, FL.
9. Proksch, P. (1994) Defensive role for secondary metabolites from
marine sponges and sponge-feeding n udib ranch s. Toxicon 32,
639–655.
10. Mu
¨
ller, W.E.G., Blumbach, B. & Mu
¨
ller, I.M. (1999) Evolution of
the innate and adaptive immune systems: relationships between
potential immune molecules in the lowest metazoan phylum
[Porifera] and those in vertebrates. Transplantation 68, 1215–1227.
11. Lengyel, P.A.R.B. (1982) Biochemistry of interferons and their
action. Annu. Rev. Biochem. 51, 251–282.
12. Sen, G.C. & Lengyel, P. (1992) The interferon system: a bird’s eye
view of its b ioch emistry. J. Biol. C hem. 267, 5017–5022.
13. Rebouillat, D., Hovnanian, A., Marie, I. & Hovanessian, A.G.
(1999) The 100-kDa 2¢,5¢-oligoadenylate A synthetase catalyzing
preferentially the synthesis of dimeric ppp2¢pA5¢ molecules is
composed of th ree homologous domains. J. Biol. Chem. 274,
1557–1565.
14. Hartmann, R., Noerby , P.L., Martensen, P.M., Joergensen, P.,
James, M.C., Jacobson, C., Moestrup, S.K., Clemens, M.J. &
Justesen, J. (1998) Activation of 2–5 oligoadenylate synthetase by
single-stranded and double-stranded RNA aptamers. J. Biol.
Chem. 273, 3 236–3246.
15. Pestka, S., Langer, J.A., Zoon, K.C. & Samuel, C.E. (1987)
Interferons and their actions. Annu. Rev. Biochem. 56, 727–777.
16. Chebath, J., Benech, P., H ovanessian, A.G., Galabru, J. , Robert,

response of t he sp onge Suberites domuncula to bacterial infection.
Mar. Biol. 139, 1037–1045.
23. Wagner, C., Steffen, R., Koziol, C., Batel, R., Lacorn, M.,
Steinhart, H., Simat, T. & Mu
¨
ller, W.E.G . (1998) Apoptosis in
marine spo nges: a bioma rker for environme ntal stress (cadmium
and bacteria). Mar. Biol. 131, 411–421.
24. Wiens, M., Krasko, A., Mu
¨
ller, C.I. & Mu
¨
ller, W.E.G. (2000)
Molecular evo lution o f ap optotic pathways: cloning of key
domains from sponges (Bcl-2 homology d omains and death
domains) and their phylogenetic relationships. J. Mol. Evol. 20,
520–531.
25. Wiens, M., Krasko, A., Mu
¨
ller, I.M. & Mu
¨
ller, W.E.G. (2000)
Increased expression of the potential proapopto tic m olecule DD 2
and increased synthesis of leukotriene B
4
during allograft r ejection
in a marine sponge. Cell Death Diff. 7, 461–469.
26. Rysiecki, G., Gewert, D.R. & Williams, B.R. (1989) Constitutive
expression of 2¢,5¢-oligoadenylate synthe tase cDNA resu lts in
increased antiviral activity and g rowth suppression. J. Interferon

tases: cloning of the (2¢-5¢) oligoadenylate synthetase from the
marine sponge Geodia c ydonium. FEBS Lett. 462, 1 2–18.
33. Custodio, M.R., Prokic, I., Steffen, R., Koziol, C., Borojevic, R.,
Bru
¨
mmer, F., Nickel, M . & Mu
¨
ller, W.E.G. (1998) Primmorphs
generated from d issociated cells of the sponge Suberites domun-
cula: a model system for st udie s of cell proliferation and cell d eath.
Mech. Ageing Dev. 105, 45–59.
34. Mu
¨
ller, W.E.G., Wiens, M., Batel, R., Steffen, R., Borojevic, R. &
Custodio, M.R. (1999) Establishment of a primary cell culture
from a sponge: primmorphs from Suberites d omuncula. Mar. Ecol.
Progr. S er. 178, 205 –219.
35. Ulevitch, R.J. & Tobias, P.S. (1994) Recognition of endotoxin by
cells leading t o t ransmembrane signaling. Curr. Opin. Immunol. 6,
125–130.
36. Rottmann, M., Schro
¨
der,H.C.,Gramzow,M.,Renneisen,K.,
Kurelec, B., Dorn, A., Frie se, U. & M u
¨
ller, W.E.G. (1987) Specific
phosphorylation of proteins in pore complex-laminae from the
sponge Geodia cydonium by the homologous aggregation factor
and phorbol ester. R ole of p rotein kinase C in the phosphorylation
of DNA topoisomerase II. EMBO J. 6, 3939–3944.

42. BLAST (1997) http://www.ncbi.nlm.nih.gov/blast/blast.cgi
43. FASTA (1997) http://www.ncbi.nlm.nih.gov/BLAST/fasta.html
44. Thompson, J.D., Higgins, D.G. & G ibson, T.J. (1994) CLUSTAL
W: improving t he sensitivity of progressive multiple sequence
alignment t hrough sequence w eighting, positions-specific gap
penalties and weight matrix choice. N ucleic Acids Re s. 22, 4673–
4680.
45. Felsenstein, J. (1993) PHYLIP, Version 3.5. University of
Washington, Seattle, WA.
46. Dayhoff, M.O., Schwartz, R.M. & Orcutt, B.C. (1978) A model of
evolutionary change in protein. In Atlas of Protein Sequence a nd
Structure (Dayhoff, M.O., ed.), pp. 345–352. Nat. Biomed. Res.
Foundation, Washington, DC.
47. Nicholas, K.B. & Nicholas, H.B. Jr (1997) Genedoc: a Tool f or
Editing and Annot ating Multiple S equence A lignments, Version
1.1.004. Available a t http://www.psc.edu/biomed/genedoc/ebimet.
html
48. Kyte, J. & Doolittle, R.F. (1982) A simple method for displaying
the hydrophobic character of a protein. J. Mol. Bio l. 157, 105–132.
49. PC/GENE (1995) Data Banks CD-ROM; Release 14.0. Intelli-
Genetics, Inc., Mountain View, CA.
50. Stanley, P.E. & Kricka, L.J. (1990) Bio lumin escence and Chemi-
luminescence: Current Status. John Wiley, Sons New York.
51. Harlow, E . & Lane, D. ( 1988) Antibodies: A Laboratory Manual.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New
York.
52. Schu
¨
tze, J., Krasko, A ., Diehl-Seifert, B. & Mu
¨

¨
der, H.C., Skorokhod, A., Bu
¨
nz, C.,
Mu
¨
ller, I.M. & Grebenjuk, V.A. (2001) Con tribution of sponge
genes to unravel the genome of the hypothetical ancestor of
Metazoa (Urmetazoa). Gene 276, 161 –173.
61. Kruse,M.,Steffen,R.,Batel,R.,Mu
¨
ller, I .M . & M u
¨
ller, W.E.G.
(1999) Differential e xpression of allograft inflammatory factor 1
and o f glutathione pe roxidase during auto- and allograft response
in marine sponges. J. Cell. Sci. 112, 4305–4313.
1392 V. A. Grebenjuk et al. (Eur. J. Biochem. 269) Ó FEBS 2002