Antibacterial activity in four marine crustacean decapods

1. Fish & Shellfish Immunology (2002) 12, 371–385
doi:10.1006/fsim.2001.0378
Available online at http://www.idealibrary.com on
Antibacterial activity in four marine crustacean decapods
TOR HAUG1*, ANITA K. KJUUL1, KLARA STENSVAG1, ERLING SANDSDALEN2 AND
r
OLAF B. STYRVOLD1
1
Institute of Marine Biotechnology, The Norwegian College of Fishery
Science, University of Tromsø, Breivika, N-9037 Tromsø, Norway and
2
Norwegian Institute of Fisheries and Aquaculture, N-9291 Tromsø, Norway
(Received 28 June 2001, accepted 31 August 2001)
A search for antibacterial activity in di#erent body-parts of Pandalus borealis
(northern shrimp), Pagurus bernhardus (hermit crab), Hyas araneus (spider
crab) and Paralithodes camtschatica (king crab) was conducted. Dried samples
were extracted with 60% (v/v) acetonitrile, containing 0·1% (v/v) trifluoro-
acetic acid, and further extracted and concentrated on C18 cartridges. Eluates
from the solid phase extraction were tested for antibacterial, lysozyme and
haemolytic activity. Antibacterial activity against Escherichia coli, Vibrio
anguillarum, Corynebacterium glutamicum and Staphylococcus aureus was
detected in extracts from several tissues in all species tested, but mainly in the
haemolymph and haemocyte extracts. V. anguillarum and C. glutamicum were
generally the most sensitive micro-organisms. In P. borealis and P. bernhardus
most of the active fractions were not a#ected by proteinase K treatment, while
in H. araneus and P. camtschatica most fractions were sensitive to proteinase
K treatment, indicating antibacterial factors of proteinaceous nature. In
P. bernhardus the active fractions were generally heat labile, whereas in
H. araneus the activities were resistant to heat. Di#erences between active
extracts regarding hydrophobicity and sensitivity for heat and proteinase
K treatment indicate that several compounds are responsible for the anti-
bacterial activities detected. Lysozyme-like activity could be detected in some
fractions and haemolytic activity against human red blood cells could be
detected in haemolymph/haemocyte and exoskeleton extracts from all species
tested. 2002 Elsevier Science Ltd. All rights reserved.
Key words: marine bioprospecting, natural products, antibacterial
activity, invertebrate immunology, Pandalus borealis, Pagarus
bernhardus, Hyas araneus, Paralithodes camtschatica.
I. Introduction
Marine invertebrates are constantly exposed to high concentrations of micro-
organisms. In crustaceans, the defence system against microbes rests largely
on cellular activities performed by haemocytes such as adhesion, phago-
cytosis, encapsulation, nodule formation, and melanisation. The multimeric
coagulation and phenoloxidase systems are also considered to be important
defences in these organisms. Other factors described as part of the immune
*Corresponding author: E-mail: torh@nfh.uit.no
371
1050–4648/02/$-see front matter 2002 Elsevier Science Ltd. All rights reserved.

2. 372 T. HAUG ET AL.
system include agglutinins, haemolysins, lysozyme and antimicrobial
factors [1–3]. Antimicrobial activity has been detected in several decapod
crustaceans, including lobster, crabs, shrimps, and freshwater crayfish [4–7].
However, little is known about the nature of the antimicrobial factors
in crustaceans, and only a few compounds have been fully characterised.
Chattopadhyay et al. [8] isolated an antimicrobial lectin, named scyllin, from
the mud crab (Scylla serrata). Antimicrobial peptides have been isolated and
characterised from the crabs Carcinus maenas [9] and Callinectes sapidus [10],
and from the shrimp Penaeus vannamei [11]. Such peptide antibiotics seem to
be important defence molecules in all living organisms, including bacteria,
plants, invertebrates and vertebrates [12]. In most of the crustacean species
studied, the antimicrobial activity has been located in the haemolymph and/or
in the haemocytes. However, potent antimicrobial activity has also been
detected in other organs/tissues [13–15].
The evolution of antibiotic-resistant pathogenic bacteria has stimulated the
search for alternative antimicrobial agents from natural sources. Many
antimicrobial peptides show a high specificity for prokaryotes and a low
toxicity for eukaryotic cells, and their mode of action (destroying mem-
branes) is considered unlikely to lead to development of resistance. These
properties have favoured their investigation and exploitation as potential new
antibiotics [16, 17].
The aim of the present work was to detect and compare antibacterial
activity in di#erent parts of the body of Pandalus borealis Krøyer (northern
shrimp), Pagurus bernhardus L. (hermit crab), Hyas araneus L. (spider crab)
and Paralithodes camtschatica Tilesius (king crab). They are all members of
the order Decapoda, with king crab and hermit crab belonging to the
infra-order Anomura, and the shrimp and the spider crab belonging to the
infra-orders Caridea and Brachyura, respectively.
II. Materials and Methods
EXPERIMENTAL ANIMALS AND SAMPLE COLLECTION
Live specimens of the shrimp P. borealis (220 specimens, average weight 8 g),
the crabs P. bernhardus (62 specimens, average weight 14 g) and H. araneus
(44 specimens, average weight 200 g) were obtained o# the coast of Tromsø,
Norway. P. camtschatica (10 specimens, average weight 1500 g, all male) were
collected from Varangerfjord, Finnmark, Norway. All animals were caught in
the period from March–May 2000, and kept in separate tanks with circulating
seawater until haemolymph and tissue collection. In H. aureus and P.
camtschatica haemolymph was collected by entering the unsclerotised mem-
brane at the base of the chelipeds and pereiopods with a 21 gauge needle
attached to a syringe. In P. borealis and P. bernhardus haemolymph was
withdrawn from the heart with a 25 gauge needle. In the large crabs (H.
araneus and P. camtschatica) the haemolymph was immediately centrifuged at
800 g at 4 C for 20 min to separate the haemocytes from the plasma (cell-free
haemolymph). After the animals were bled to death, di#erent tissues and parts
of the body were sampled, pooled and kept on ice. P. borealis and P.

3. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 373
bernhardus were divided into eggs, cephalothorax (including internal organs),
exoskeleton and muscle (abdominal part). From H. araneus; gills, exoskeleton,
internal organs, and eggs were collected. P. camtschatica were separated into
gills, exoskeleton, muscle, and internal organs. All samples were lyophilised
and frozen separately at 20 C until extraction.
EXTRACTION AND SEPARATION OF ANTIBACTERIAL FACTORS
Freeze-dried samples (3–50 g) were extracted with 10 volumes (v/w) of 60%
(v/v) acetonitrile (ACN; HPLC-grade, SDS, Peypin, France) containing 0·1%
trifluoroacetic acid (TFA; Fluka Chemie AG, Buchs, Switzerland) for 24 h at
4 C. The supernatant was collected, stored at 4 C, and the residue was
extracted once again under the same conditions. The combined supernatants
were incubated at 20 C for 1–2 h to allow two liquid phases, an acetonitrile-
rich phase and an aqueous-rich phase, to be formed. The two phases were
separated, lyophilised and kept frozen at 20 C until further extraction. The
remaining pellet was discarded. Because of small quantities and low water
solubility of several of the acetonitrile-rich fractions, these samples were
excluded from the study.
The aqueous phase fractions were solubilised in acidified water (pH 4·7) to a
concentration of 100 mg ml 1. Salt was removed from the aqueous phase by
solid phase extraction. The fractions (maximum 100 ml) were separately
loaded on to 35 cc Sep-Pak C18 Vac cartridges (Water Associates, MA, U.S.A.)
equilibrated in acidified water (0·05% TFA). After washing with acidified
water, three stepwise elutions were performed with 10, 40 and 80% (v/v) ACN
in acidified water, respectively. Non-bound material was discarded. The
di#erent eluates collected were lyophilised and reconstituted with Milli-Q
water (Millipore Corp., MA, U.S.A.) before testing for antibacterial activity.
BACTERIAL STRAINS AND GROWTH CONDITIONS
The Gram negative bacteria Vibrio anguillarum, serotype O2 (FT 1801, also
coded as AL 104/LFI 6004), originally isolated from Atlantic salmon by sta# of
the Norwegian Veterinary Institute (Oslo, Norway), Escherichia coli (ATCC
25922), and the Gram positive bacteria Staphylococci aureus (ATCC 9144) and
Corynebacterium glutamicum (ATCC 13032) were used as test organisms. V.
anguillarum and C. glutamicum were chosen because in preliminary studies
they were shown to have high sensitivity to various marine extracts. All
isolates were grown at room temperature in Mueller Hinton Broth (MHB;
Difco Laboratories, Detroit, U.S.A.).
ANTIBACTERIAL ACTIVITY TESTING
After solid phase extraction, each fraction was reconstituted in Milli-Q
water and the protein content was determined by the BCA protein assay
(Pierce, Rockford, IL, U.S.A.). All samples were diluted to a protein con-
centration of 250 g ml 1 and serial two-fold dilutions were made before
antibacterial activity testing. Aliquots of 50 l test fractions were incubated in

4. 374 T. HAUG ET AL.
96-well microtitre plates (Costar, No. 3599; Corning Inc., NY, U.S.A.) with
50 l of a suspension of a mid-logarithmic phase culture of bacteria at a
starting concentration of 5 103 cells per well in MHB. Incubations were
performed at room temperature with continuous shaking. Bacterial growth
was assayed by measurement of the optical density at 600 nm using a
microtitre plate reader (THERMOmax; Molecular Devices Corp., CA, U.S.A.)
every 6 h for 48 h. Negative controls included media plus sample, and media
plus milli-Q water. Cecropin P1 (0·5 g ml 1), made synthetically as described
by Kjuul et al. [18], was used as a positive control. Antibacterial activity was
determined when the optical density of the growth control (bacteria plus
media) reached an absorbance at 600 nm of 0·150–0·200. Fractions were
regarded as active when the optical density was less than 50% of the control,
and fractions that were active at protein concentrations of 31·25 g ml 1 or
lower were considered as fractions with high activity.
PROTEINASE K AND HEAT TREATMENT
Fractions showing antibacterial activity were tested for proteinase K
sensitivity. Proteinase K (Promega, Maddison, WI, U.S.A.) was dissolved in
50 mM Tris–HCl (pH 7·5) at a concentration of 2·5 mg ml 1. The fractions were
diluted in Milli-Q water to a protein concentration of 250 g ml 1. A volume
of 10 l proteinase K solution was added per 50 l sample. The mixture was
incubated at 42 C for 90 min for protein digestion. The temperature was then
elevated to 85 C for 15 min to inactivate the proteinase K. As a control (heat
treatment) 10 l 50 mM Tris–HCl (pH 7·5) was added to 50 l of the diluted
sample and subjected to the same treatment as the proteinase K sample.
Cecropin P1 (0·5 g ml 1) was used as control to ensure the activity of
proteinase K. The activity in the treated samples were determined in micro-
titre plates (Costar) as described above. Fractions that no longer showed
antibacterial activity after proteinase K/heat treatment were regarded as
sensitive.
LYSOZYME ASSAY
In order to test whether lysozyme-like compounds were present, all fractions
were tested for lysozyme activity according to the method of Nilsen et al. [19].
A standard suspension of Micrococcus luteus (Sigma Chemical Co., No.
M-3770, St. Louis, MO, U.S.A.) cell walls (0·2 mg ml 1) was prepared in 50 mM
sodium acetate (pH 5·2) and 50 mM NaCl (1:1). The test fractions were diluted
to a protein concentration of 500 g ml 1. Experimental conditions consisted
of 1 l sample added to 150 l of M. luteus suspension in microtitre plates
(Costar, No. 3599). The activity was determined spectrophotometrically
(SPECTRAmax Plus Microplate Spectrophotometer, Molecular Devices
Corp.) by recording the decrease in absorbance at 450 nm (due to cell wall
lysis) for 6 min at room temperature. Enzyme activity was expressed as units

5. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 375
where one unit represents a decrease in absorbance of 0·001 min 1. Lysozyme-
like activity was judged to be present when enzyme activity was more than
two units per g protein.
HAEMOLYTIC ASSAY
To test whether the animals contain factors which are toxic to eukaryotic
cells, the haemolytic activity in extracts from haemolymph/plasma, haemo-
cytes, eggs, muscle, and exoskeleton were determined using fresh human red
blood cells (RBC). Four millilitres of blood were collected from a healthy
person into a polycarbonate tube containing heparin to a final concentration
of 10 U ml 1. The erythrocytes were isolated by centrifugation at 450 g for
10 min and washed three times with phosphate-bu#ered saline (PBS;
320 mOsm, pH 7·4) in order to remove plasma and bu#y coat. The cell pellet
was resuspended in 4 ml of PBS. The haematocrit value (Hct) was determined
and the RBC suspension was further diluted to a Hct value of 10%.
The test samples were diluted to a protein concentration of 500 g ml 1 and
the test was performed in 96 well U-shaped microtitre plates (Nunclon
Surface, No. 163320; Nalge Nunc Int., Denmark). To each well was added 40 l
PBS, then 50 l test fraction and lastly 10 l of the RBC suspension. After
incubation in a shaker at 37 C for 1 h the plates were centrifuged at 200 g
for 5 min. The supernatants (60 l) were carefully transferred to new flat-
bottomed polycarbonate microtitre plates (Nunc No. 269620, Nalge Nunc
Int.) and the absorbance of the supernatant was measured at 550 nm. Baseline
haemolysis and 100% haemolysis were defined as the amount of haemoglobin
released in the presence of PBS and 0·1% Triton X-100 (Sigma), respectively.
III. Results
The results show that in vitro antibacterial activity is found in all crus-
tacean species tested. However, when the antibacterial activity in di#erent
SPE-eluates were compared, wide di#erences were found between eluates and
organs as well as between species (Table 1 and 2). Of the four di#erent bacteria
tested, C. glutamicum was the most sensitive, while E. coli was the least
sensitive.
In P. borealis the highest activity was found in haemolymph (all fractions),
but mainly against C. glutamicum. Some activity was also observed in eggs (all
fractions), cephalothorax (10 and 80% fraction), muscle (10 and 80% fraction),
and exoskeleton (all fractions). Figure 1 shows the time course study of the
antibacterial e#ect of the 10% haemolymph extract against C. glutamicum. A
discernible antibacterial e#ect was evident down to a protein concentration of
7·8 g ml 1, and no bacterial growth occurred at a protein concentration of
31·3 g ml 1. Activity against E. coli was not detected in any extract, and
none of the antibacterial fractions lost their activity after proteinase K
treatment.
In P. bernhardus high antibacterial activity was found in haemolymph,
cephalothorax and exoskeleton. Some activity was also detected in the eggs.
The highest antibacterial activity was mainly detected in the 40 and 80% ACN

6. 376 T. HAUG ET AL.
Table 1. Antibacterial activity in extracts from Pandalus borealis and Pagurus
bernhardus against Vibrio anguillarum (V.a.), Escherichia coli (E.c.), Corynebacterium
glutamicum (C.g.) and Staphylococcus aureus (S.a.)
Antibacterial activity
Organ/tissue Fraction1 Pandalus borealis Pagurus bernhardus
(% ACN)
V.a. E.c. C.g. S.a. V.a. E.c. C.g. S.a.
10 ++ – +++ + + + + +
Haemolymph 40 – – +++ – + – +++ –
80 + – +++ – + – ++ ++
10 – – + – – – – –
Eggs 40 + – + – + – + –
80 – – +++ + + – + –
10 – – – – + – – –
Cephalothorax2 40 – – – – + – + –
80 ++ – ++ + + – +++ +++
10 + – + – – – – –
Muscle 40 – – – – – – + –
80 + – + + – – – –
10 + – + – + – + –
Exoskeleton 40 – – + – + – + –
80 – – + + – – ++ ++
Abbreviations:
– No antibacterial activity at a protein concentration of 125 µg m–1.
+ Antibacterial activity at a protein concentration of 125 µg m–1.
++ Antibacterial activity at a protein concentration of 62.5 µg m–1.
+++ Antibacterial activity at a protein concentration of 31.25 µg m–1.
Sensitivity to proteinase K (but not sensitive to heat).
Sensitivity to heat (but might be sensitive to proteinase K).
1 Eluates from solid phase extraction.
2 Including internal organs.
fractions, and Gram positive bacteria were most sensitive. Only two active
fractions (haemolymph—10% and eggs—40%) were sensitive to proteinase K
treatment, while most of the fractions were sensitive to heat treatment. One
fraction (haemolymph—10%) showed activity against E. coli. No or little
activity was measured in the muscle fractions.

7. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 377
Table 2. Antibacterial activity in extracts from Hyas araneus and Paralithodes
camtschatica against Vibrio anguillarum (V.a.), Escherichia coli (E.c.), Corynebacterium
glutamicum (C.g.) and Staphylococcus aureus (S.a.)
Antibacterial activity
Organ/tissue Fraction Hyas araneus Paralithodes camtschatica
(% ACN)
V.a. E.c. C.g. S.a. V.a. E.c. C.g. S.a.
10 +++ +++ +++ – – – – –
Plasma 40 + – + + – – + –
80 ++ – +++ – – – ++ +
10 ++ + ++ + + – – –
Haemocytes 40 +++ +++ +++ +++ + – + –
80 +++ +++ +++ – +++ +++ +++ +
10 – – – – – – – –
Gills 40 + ++ +++ – – – – –
80 – – ++ – + – – –
10 – – + – – – ++ –
Internal organs 40 – – – – – – – –
80 + + + ++ ++ – ++ –
10 + – – – – – + –
Exoskeleton 40 ++ ++ +++ – + – +++ –
80 – – + + – – + –
Abbreviations:
– No antibacterial activity at a protein concentration of 125 µg m–1.
+ Antibacterial activity at a protein concentration of 125 µg m–1.
++ Antibacterial activity at a protein concentration of 62.5 µg m–1.
+++ Antibacterial activity at a protein concentration of 31.25 µg m–1.
Sensitivity to proteinase K (but not sensitive to heat).
Sensitivity to heat (but might be sensitive to proteinase K).
1 Eluates from solid phase extraction.
2 Including internal organs.
High antibacterial activity (against all bacteria) was detected in the
haemocytes and plasma of H. araneus. Activity was also detected in the gills
(40 and 80% fraction), the internal organs (mainly 80% fraction), and the
exoskeleton (mainly 40 and 80% fraction). No activity was observed in the egg

8. 378 T. HAUG ET AL.
0.3
OD 600 nm
0.2
0.1
0 10 20 30 40
Time (h)
Fig. 1. Antibacterial activity of 10% haemolymph extract from P. borealis against
C. glutamicum grown in MHB. The optical density at 600 nm was measured in a
bacterial suspension of 5 103 cells per well containing bacteria alone (

9. ), or
bacteria plus extract adjusted to a protein concentration of 7·8 g ml 1 (
),
15·6 g ml 1 (), and 31·3 g ml 1 (), respectively.
0.3
OD 600 nm
0.2
0.1
0 10 20 30 40
Time (h)
Fig. 2. Antibacterial activity of 80% haemocyte extract from P. camtschatica against
C. glutamicum grown in MHB. The optical density at 600 nm was measured in a
bacterial suspension of 5 103 cells per well containing bacteria alone (

10. ), or
bacteria plus extract adjusted to a protein concentration of 7·8 g ml 1 (
),
15·6 g ml 1 (), and 31·3 g ml 1 (), respectively.
extracts (data not shown). In contrast to P. bernhardus and P. borealis, most
of the activities found in H. araneus were sensitive to proteinase K treatment.
The 10% fraction from plasma showed higher activity than the 10% fraction
from haemocytes, and the activity in both fractions were not a#ected by
proteinase K treatment.
In P. camtschatica, as in the other species, the highest antibacterial activity
was found in the haemocyte extracts (40 and 80% fractions). These fractions
were active against all bacteria tested. Figure 2 shows the time course study
of the antibacterial e#ect of the 80% haemocyte extract against C. gluta-
micum. The activity was higher with increasing protein concentration and no
bacterial growth occurred at protein concentrations higher than 31·3 g ml 1.
The activity in plasma was low compared to the haemocyte extracts. A
relatively high activity was also detected in the internal organs (10 and 40%
fraction) and the exoskeleton (mainly 40% fraction). No activity was
measured in the muscle (data not shown). All active 40% ACN fractions were
sensitive to proteinase K treatment, while none of the 10% fractions were
sensitive. Of the 80% fractions with antibacterial activity, only the haemocyte

11. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 379
Table 3. Haemolytic activity in extracts from Pandulus borealis, Pagurus bernhardus,
Hyas araneus and Paralithodes camtschatica against human erythrocytes (RBC)
Haemolytic activity (%)2
Fraction1
Organ/tissue Pandulus Pagurus Hyas Paralithodes
(% ACN)
borealis bernhardus araneus camtschatica
Haemolymph3 10 90 10 10 0
40 30 10 50 0
80 nt4 90 40 100
Haemocytes 10 nt nt 40 20
40 nt nt nt4 0
80 nt nt 30 20
Eggs 10 100 0 0 nt
40 0 0 20 nt
80 10 0 0 nt
Muscle 10 0 0 nt 0
40 20 10 nt 0
80 nt4 20 nt 0
Exoskeleton 10 0 0 20 10
40 50 20 50 50
80 100 100 100 100
Abbreviations: nt, not tested; 1, eluates from solid phase extraction; 2, the data are presented
as % haemolysis compared to control (Triton X-100); 3, including haemocytes in P. borealis and
P. bernhardus, without haemocytes in H. araneus and P. camtschatica; 4, not tested due to lack of
material.
fraction (and only the activity against V. anguillarum) was sensitive to
proteinase K.
Cecropin P1, an antimicrobial peptide originally isolated from pig intestine
[20], had a bactericidal e#ect on V. anguillarum and E. coli, but showed little
or no activity against C. glutamicum and S. aureus. The activity of cecropin
P1 was totally abolished after proteinase K treatment, and heat-treated
proteinase K showed no antibacterial activity.
Lysozyme-like activity was detected in some of the crustacean extracts (data
not shown). These were the 80% egg fractions of P. borealis, P. bernhardus and
H. araneus, and the 80% gill fraction of P. camtschatica. The activities
detected in the egg fractions ranged from 4–8 units per g protein and in
the gill fraction of P. camtschatica a value of 12 units per g protein was
recorded.
Several extracts from the crustaceans were lytic for human erythrocytes.
The most abundant activity was detected in the haemolymph/plasma extracts
and in the exoskeletal extracts (Table 3). In general, the highest activity was
detected in the 80% fractions. However, high haemolytic activity (more than
50% haemolysis compared to the control) was also detected in 10% fractions
from haemolymph and eggs of P. borealis. In addition, high activity could be
measured in internal organs from H. araneus (10 and 80% fraction) and P.
camtschatica (80% fraction), and in the gills (80% fraction) of P. camtschatica
(data not shown).

12. 380 T. HAUG ET AL.
IV. Discussion
A screening for antibacterial activity in di#erent organs and tissues of four
crustacean species was conducted. The results show that all species tested
possess antibacterial activity in vitro. Antibacterial activity has previously
been described in a wide range of crustacean species [7, 15, 21–24]. In most of
the species studied, only the haemolymph and/or the haemocytes have been
tested for activity. The present study demonstrated the presence of anti-
bacterial factors in several other tissues, such as gills, eggs, exoskeleton and
internal organs. Whether the same antibacterial factors are responsible for
the activity in all organs, is unknown. However, the major activity was mainly
located in the haemolymph and/or the haemocytes. In fact, in P. borealis, H.
araneus and P. camtschatica, some haemolymph/haemocyte fractions showed
activity with protein concentrations as low as 8 g/ml 1 (data not shown).
This is comparable to studies in the shore crab, C. maenas, where haemocyte
lysate supernatants had antibacterial activity at protein concentrations
of approximately 2 g ml 1 [6]. Some of the other tissues collected will
necessarily contain some haemolymph and haemocytes, but any dominant
e#ect on activity is not expected.
High antibacterial activity could be detected in the gills of H. araneus,
whereas little or no activity was detected in P. camtschatica. The process of
removal of bacteria from the crustacean circulation involves the formation of
a large number of haemocyte-bacteria aggregations in the gill lamellae [25].
The antibacterial activity in haemolymph/haemocytes of crustaceans can also
be inducible [5, 23]. Several authors have described aggregations of haemo-
cytes in the gills of crustaceans in which viruses, bacteria or fungal spores
have been injected [25–28]. Therefore, if the H. araneus in the present study
had an ongoing microbial infection, it is possible that the antibacterial
activity detected in the gills is of haemocyte origin.
In H. araneus and P. camtschatica, some antibacterial activity was found in
the internal organs. The cephalothorax from P. bernhardus and P. borealis,
which mainly consists of exoskeleton and internal organs, also showed
activity. Antibacterial activity has previously been found in internal
tissues, such as seminal plasma of S. serrata [15] and hepatopancreas of
Homarus americanus [13]. In the latter study some activity was also detected
in the haemolymph, but negligible activity was found in the haemocytes.
These findings indicate that antibacterial factors are also produced in
crustacean tissues other than in haemolymph/haemocytes. In insects, the fat
body (a functional equivalent of the mammalian liver) seems to be an
important organ for synthesis of antibacterial compounds, although anti-
bacterial factors are produced in granular haemocytes and other tissues as
well [12, 29]. In other arthropods, like myriapods and chelicerates, the
haemocytes seem to be the main tissue for production of antibacterial agents
[30, 31].
All animals in the present study were caught in their spawning period, and
the eggs sampled were ready to hatch. Some of the egg fractions from P.
borealis and P. bernhardus showed antibacterial activity, mainly against
marine bacteria. There was no detectable activity in the eggs from H. araneus.

13. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 381
Antimicrobial activity has previously been found in the eggs of various marine
invertebrates [32, 33]. The antibacterial activity might be due to factors of the
innate immune system. However, the activity might also be of microbial
origin. The antimicrobial activity detected in the eggs of the decapods
Palaemon macrodactylus and H. americanus was demonstrated to be due to
bacterial symbionts [34, 35]. Micro-organisms might in some cases also be
responsible for the antibacterial activity detected in the gills, exoskeleton and
even within the internal organs, although this is probably not likely for most
of the samples.
The exoskeleton showed antimicrobial activity in all species tested. The
exoskeleton of crustaceans is composed mainly of chitin, a polymer of
N-acetyl-glucosamine covalently bound to protein. Biologically, a deacetylase
transforms chitin to chitosan by hydrolysing the acetamido groups of
N-acetyl-glucosamine [36]. It has been reported that both chitin and chitosan
from crustaceans possess antimicrobial activity [36, 37]. However, chitin is
insoluble in water and acid while chitosan is relatively insoluble in water, but
soluble in acid [36]. Due to these properties, it is unlikely that chitin and/or
chitosan are responsible for the antibacterial activity detected in this study.
Several exoskeletal proteins have been characterised from crustaceans. Some
of these proteins are cationic, low molecular weight peptides [38, 39], which is
characteristic for antimicrobial peptides. However, none of these peptides
have been tested for antibacterial activity (Andersen, pers. comm.).
The detection of antibacterial activity in the exoskeleton suggests that this
activity is important in the defence against micro-organisms present in the
marine environment. It has been reported that most of the bacteria residing on
the carapace of C. sapidus are susceptible to the hosts antibacterial factors,
and crabs having depressed levels of this antibacterial activity were in higher
risk of developing shell disease [40].
The solid phase extraction method separates compounds according to their
hydrophobicity. As antibacterial activity was detected in 10, 40 and 80%
fractions, it is reasonable to assume that multiple factors are responsible for
the antibacterial activity. Some of the organic phase fractions (acetonitrile-
rich phase), containing fat, carotenoids and other lipophilic compounds
possessed antibacterial activities (data not shown). The stock solutions of the
aqueous phase fractions were based on protein concentration, and dilutions
were made from these stock solutions. Some fractions may contain a high
proportion of non-active proteins, and it is therefore a possibility that some
active components are missed due to non-inhibitory concentrations.
In H. araneus, most of the active fractions were sensitive to proteinase K
treatment. Since enzymatic digestion destroys the antibacterial activity, the
active molecules are most likely of proteinaceous nature. Antibacterial
peptides and proteins have previously been identified in C. sapidus [10],
C. maenas [9, 41], S. serrata [15], Penaeus setiferus [7], P. vannamei [11] and
Parachaeraps bicarinatus [4], and thereby seem to represent a common feature
in the defence system of crustaceans. It must be emphasised that the heat
labile fractions might also be sensitive to proteinase K treatment. The heat
treatment was included in the test to ensure that the antibacterial activity
detected was not caused by proteinase K itself.

14. 382 T. HAUG ET AL.
In P. borealis and P. bernhardus, on the other hand, most of the active
fractions were resistant to proteinase K treatment. The antibacterial
components in these fractions are therefore probably of non-proteinaceous
nature. In P. bernhardus, the active fractions were generally heat labile,
whereas in H. araneus the activities were resistant to heat. Heat labile
antibacterial factors have been detected in several crustaceans [5, 42, 43].
However, Relf et al. [41] isolated an antibacterial protein of 11·5 kDa which
was active even after heating for 10 min at 100 C. These results indicate
that crustaceans have developed a variety of defence molecules against
pathogenic micro-organisms and the kind dominating vary among the
di#erent species.
Since many fractions (especially from P. bernhardus) were sensitive to heat
treatment, the antibacterial activity might be of enzymatic character.
Lysozyme-like activity could be detected in all of the species tested in this
study, but only in the eggs from P. bernhardus, P. borealis and H. araneus, and
in the gills from P. camtschatica. Lysozyme(s) might therefore be responsible
for the antibacterial activity detected in these extracts. In contrast, no
antibacterial activity was detected in the eggs from H. araneus. Attempts to
demonstrate lysozyme activity in crustaceans have so far given equivocal
results, and no lysozyme has so far been purified and characterised. Lysozyme-
like activity has been detected in the cysts (dormant eggs) of the marine
branchiopod Artemia franciscana [14], and in the haemolymph of several
freshwater crayfish [43, 44]. A common feature of these factors is that the
highest activity is demonstrated at pH values of 6·2–8·0, and the activity is
heat labile. In the present study we examined for lysozyme activity at pH 5·2.
It is therefore possible that we, under other experimental conditions, would
have detected higher activities. On the other hand, no lysozyme-like
activity was detected in the haemolymph of P. setiferus [7], P. vannamei [24]
and C. sapidus [42]. Based on previous reports and our results, the eggs seem
to be a better target for lysozyme search than haemolymph in marine
crustaceans.
Some of the fractions, especially of the haemolymph/plasma and exoskel-
eton, contained high haemolytic activity. Cantacuzène [45] reported in 1912
the presence of haemolytic factors in serum of Eupagurus prideauxii, which is
closely related to P. bernhardus. Since then, haemolytic activity has been
detected in Maia squinado [46], Panulirus argus [47], Penaeus californiensis
[48], and in Nematopalaemon tenuipes [49]. In the latter study, the active
compound was shown to be a steroid. In the present study, several extracts
showed both antibacterial and haemolytic activity. Whether the same com-
pound(s) are responsible for both activities remains to be clarified. From a
pharmaceutical point of view it is advantageous that antibacterial drugs have
no side e#ects, such as haemolytic activity.
In conclusion, this study shows that a number of marine decapod crus-
taceans contain factors with antibacterial activity, particularly in the haemo-
lymph and/or the haemocytes. This property seems to be a common feature
throughout the order. Di#erences between active extracts regarding hydro-
phobicity and sensitivity to heat/proteinase K treatment indicate that several
compounds are responsible for the antibacterial activities. This study has also

15. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 383
shown the existence of haemolytic factors in the crustaceans tested. Whether
the same factors are responsible for both activities or not, remain to be
explored. Purified compounds are required to study their mechanism(s) of
action, their genetics, and their role in host defence. Although further
analyses are continuing, tentative findings suggest that the antibacterial
activities are caused by compounds with molecular weights ranging from
2–12 kDa.
This work was supported by the Norwegian Research Council (No. 111257/120) and
the Marine Biotechnology in Tromsø (MABIT) research programme (No. BS 0001). We
thank the crew of the research vessels F/F ‘Johan Ruud’ and F/F ‘Hyas’ for collecting
animals for this study. We also thank K. Øverbø for assistance with the lysozyme assay.
References
1 Smith, V. J. Chisholm, J. R. S. (1992). Non-cellular immunity in crustaceans. Fish
Shellfish Immunology 2, 1–31.
2 Söderhäll, K. Cerenius, L. (1992). Crustacean immunity. Annual Review of Fish
Diseases 2, 3–23.
3 Sritunyalucksana, K. Söderhäll, K. (2000). The proPO and clotting system in
crustaceans. Aquaculture 191, 53–69.
4 Schwab, G. E., Reeves, P. R. Turner, K. J. (1966). Bactericidal activity of serum of
the yabbie (Parachaeraps bicarinatus). British Journal of Experimental Pathology
47, 266–274.
5 Stewart, J. E. Zwicker, B. M. (1972). Natural and induced bactericidal activities
in the hemolymph of the lobster, Homarus americanus: products of hemocyte-plasma
interaction. Canadian Journal of Microbiology 18, 1499–1509.
6 Chisholm, J. R. S. Smith, V. J. (1992). Antibacterial activity in the haemocytes of
the shore crab, Carcinus maenas. Journal of the Marine Biological Association of the
United Kingdom 72, 529–542.
7 Noga, E. J., Arroll, T. A., Bullis, R. A. Khoo, L. (1996a). Antibacterial activity in
hemolymph of white shrimp, Penaeus setiferus. Journal of Marine Biotechnology 4,
181–184.
8 Chattopadhyay, T., Guha, A. K. Chatterjee, B. P. (1996). Novel antimicrobial
activity of scyllin, a haemolymph lectin of the edible crab Scylla serrata. Biomedical
Letters 53, 29–40.
9 Schnapp, D., Kemp, G. D. Smith, V. J. (1996). Purification and characterization
of a proline-rich antibacterial peptide, with sequence similarity to bactenecin-7,
from the haemocytes of the shore crab, Carinus maenas. European Journal of
Biochemistry 240, 532–539.
10 Khoo, L., Robinette, D. W. Noga, E. J. (1999). Callinectin, an antibacterial peptide
from blue crab, Callinectes sapidus, hemocytes. Marine Biotechnology 1, 44–51.
11 Destoumieux, D., Bulet, P., Loew, D., Dorsselaer, A. V., Rodriguez, J. Bachère, E.
(1997). Penaeidins, a new family of antimicrobial peptides isolated from the shrimp
Penaeus vannamei (Decapoda). Journal of Biological Chemistry 272, 28398–28406.
12 Boman, H. G. (1995). Peptide antibiotics and their role in innate immunity. Annual
Review of Immunology 13, 61–92.
13 Mori, K. Stewart, J. E. (1978). Natural and induced bactericidal activities of
the hepatopancreas of the American lobster, Homarus americanus. Journal of
Invertebrate Pathology 32, 171–176.
14 Stabili, L., Miglietta, A. M. Belmonte, G. (1999). Lysozyme-like and trypsin-like
activities in the cysts of Artemia franciscana Kellog, 1906. Is there a passive
immunity in a resting stage? Journal of Experimental Biology and Ecology 237,
291–303.

16. 384 T. HAUG ET AL.
15 Jayasankar, V. Subramoniam, T. (1999). Antibacterial activity of seminal plasma
of the mud crab Scylla serrata (Forskal). Journal of Experimental Marine Biology
and Ecology 236, 253–259.
16 Bax, R., Mullan, N. Verhoef, J. (2000). The millennium bugs—the need for and
development of new antibacterials. International Journal of Antimicrobial Agents
16, 51–59.
17 Mor, A. (2000). Peptide-based antibiotics: A potential answer to raging
antimicrobial resistance. Drug Development Research 50, 440–447.
18 Kjuul, A. K., Büllesbach, E. E., Espelid, S., Dunham, R., Jørgensen, T. Ø., Warr, G.
W. Styrvold, O. B. (1999). E#ects of cecropin peptides on bacteria pathogenic to
fish. Journal of Fish Diseases 22, 387–394.
19 Nilsen, I. W., Øverbø, K., Sandsdalen, E., Sandaker, E., Sletten, K. Myrnes, B.
(1999). Protein purification and gene isolation of chlamysin, a cold-active lysozyme-
like enzyme with antibacterial activity. FEBS Letters 464, 153–158.
20 Lee, J. Y., Boman, A., Chuanxin, S., Andersson, M., Jörnvall, H., Mutt, V. Boman,
H. G. (1989). Antibacterial peptides from pig intestine: isolation of a mammalian
cecropin. Proceedings of the National Academy of Sciences of the United States of
America 86, 9159–9162.
21 Chisholm, J. R. S. Smith, V. J. (1995). Comparison of antibacterial activity in
the hemocytes of di#erent crustacean species. Comparative Biochemistry and
Physiology A 110, 39–45.
22 Ueda, R., Sugita, H. Deguchi, Y. (1996). Determination of the naturally occurring
bactericidal activity of sera from Japanese coastal crustaceans. Fisheries Science
62, 325–326.
23 Sritunyalucksana, K., Sithisarn, P., Withayachumnarnkul, B. Flegel, T. W.
(1999). Activation of prophenoloxidase, agglutinin and antibacterial activity in
haemolymph of the black tiger prawn, Penaeus monodon, by immunostimulants.
Fish Shellfish Immunology 9, 21–30.
24 Alabi, A. O., Latchford, J. W. Jones, D. A. (2000). Demonstration of residual
antibacterial activity in plasma of vaccinated Penaeus vannamei. Aquaculture 187,
15–34.
25 Smith, V. J. Ratcli#e, N. A. (1980). Defensive reactions of the shore crab, Carcinus
maenas. In vivo haemocytic and histopathological responses to injected bacteria.
Journal of Invertebrate Pathology 35, 65–74.
26 Solangi, M. A. Lightner, D. V. (1976). Cellular inflammatory response of Penaeus
aztecus and P. setiferus to the pathogenic fungus, Fusarium sp., isolated from the
California brown shrimp, P. californiensis. Journal of Invertebrate Pathology 27,
77–86.
27 McCumber, L. J. Clem, L. W. (1977). Recognition of viruses and xenogeneic
proteins by the blue crab, Callinectes sapidus. I. Clearance and organ
concentration. Developmental and Comparative Immunology 1, 5–14.
28 White, K. N., Ratcli#e, N. A. Rossa, M. (1985). The antibacterial activity of
haemocyte clumps in the gills of the shore crab, Carcinus maenas. Journal of the
Marine Biological Association of the United Kingdom 65, 857–870.
29 Ho#mann, J. A. (1995). Innate immunity of insects. Current Opinion in Immunology
7, 4–10.
30 Xylander, W. E. R. Nevermann, L. (1990). Antibacterial activity in the hemolymph
of myriapods (Arthropoda). Journal of Invertebrate Pathology 56, 206–214.
31 Iwanaga, S., Kawabata, S.-I. Muta, T. (1998). New types of clotting factors and
defense molecules found in horseshoe crab hemolymph: their structures and
functions. Journal of Biochemistry 123, 1–15.
32 Stabili, L. Pagliara, P. (1994). Antibacterial protection in Marthasterias glacialis
eggs—characterization of lysozyme-like activity. Comparative Biochemistry and
Physiology B 109, 709–713.
33 Benkendor#, K., Bremner, J. B. Davis, A. R. (2000). Tyrian purple precursors in
the egg masses of the Australian muricid, Dicathais orbita: A possible defensive
role. Journal of Chemical Ecology 26, 1037–1050.

17. ANTIBACTERIAL ACTIVITY IN FOUR MARINE CRUSTACEAN DECAPODS 385
34 Gil-Turnes, M. S., Hay, M. E. Fenical, W. (1989). Symbiotic marine bacteria
chemically defend crustacean embryos from a pathogenic fungus. Science 246,
116–118.
35 Gil-Turnes, M. S. Fenical, W. (1992). Embryos of Homarus americanus are
protected by epibiotic bacteria. Biological Bulletin 182, 105–108.
36 Koide, S. S. (1998). Chitin-chitosan: Properties, benefits and risks. Nutrition
Research 18, 1091–1101.
37 Tsai, G.-J. Su, W.-H. (1999). Antibacterial activity of shrimp chitosan against
Escherichia coli. Journal of Food Protection 62, 239–243.
38 Nousiainen, M., Rafn, K., Skou, L., Roepstor#, P. Andersen, S. O. (1998).
Characterization of exoskeletal proteins from the American lobster, Homarus
americanus. Comparative Biochemistry and Physiology B 119, 189–199.
39 Andersen, S. O. (1999). Exoskeletal proteins from the crab, Cancer pagurus.
Comparative Biochemistry and Physiology A 123, 203–211.
40 Noga, E. J., Engel, D. P., Arroll, T. W., McKenna, S. Davidian, M. (1994). Low
serum antibacterial activity coincides with increased prevalence of shell disease in
blue crabs Callinectes sapidus. Diseases of Aquatic Organisms 19, 121–128.
41 Relf, J. M., Chisholm, J. R. S., Kemp, G. D. Smith, V. J. (1999). Purification and
characterization of a cysteine-rich 11.5-kDa antibacterial protein from the granular
haemocytes of the shore crab, Carcinus maenas. European Journal of Biochemistry
264, 350–357.
42 Noga, E. J., Arroll, T. A. Fan, Z. (1996b). Specificity and some physiochemical
characteristics of the antibacterial activity from blue crab Callinectes sapidus. Fish
Shellfish Immunology 6, 403–412.
43 Xylander, W. R., Ullrich, G. Kaiser, H. E. (1997). Antibacterial immune response
in Astacus leptodactylus (Crustacea, Decapoda). In vivo 11, 195–200.
44 Fenouil, E. Roch, P. (1991). Evidence and characterization of lysozyme in
6 species of freshwater crayfishes from astacidae and cambaridae families.
Comparative Biochemistry and Physiology B 99, 43–49.
45 Cantacuzène, J. (1912). Sur certains anticorps naturels observés chez Eupagurus
prideauxii. Comptes Rendus des Séances de la Société de Biologie 73, 663–664.
46 Cantacuzène, J. (1920). Formation d’hemolysins dans le serum de Maia Squinado
inoculées avec des hematies de Mammiferes. Existance dans ce serum d’une
substance antagoniste qui empêche ou retarde hemolyse. Comptes Rendus des
Séances de la Société de Biologie 83, 1512–1514.
47 Weinheimer, P. F., Evans, E. E., Stroud, R. M., Acton, R. T. Painter, B. (1969).
Comparative immunology: Natural hemolytic system of the spiny lobster, Panulirus
argus. Proceedings of the Society for Experimental Biology and Medicine 130,
322–326.
48 Guzman, M. A., Ochoa, J. L. Vargas-Albores, F. (1993). Hemolytic activity in
the brown shrimp (Penaeus californiensis Holmes) hemolymph. Comparative
Biochemistry and Physiology A 106, 271–275.
49 Indap, M. M., Thakur, N. L., Zarapkar, S. S. Shivalkar, S. A. (1996). Isolation and
characterization of bioactive steroid from a prawn. Indian Journal of Experimental
Biology 34, 588–589.