A field study on the physiology of digestion in the Antarctic krill, Euphausia superba, with special regard to chitinolytic enzymes

Abstract

Endo- and exochitinase activities were determined in the stomach and midgut gland of the Antarctic krill, Euphausia superba. along a transect west of the Antarctic Peninsula. Activities were compared with the digestive enzymes protease, cellulase
(1,4-β-D-glucanase) and laminarinase (1,3-β-D-glucanase) The chlorophyll and protein contents in the surface water of the
corresponding stations were determined. Enzyme activities were characterized by high individual and spatial variations. Chitinolytic
activity in the stomach correlated well with all digestive enzymes investigated. In the midgut gland, a correlation with cellulase
and laminannase was evident. The amount of chlorophyll a and phytoplankton protein in the surface water was not correlated
with enzyme activity. Specific enzyme activity was higher in the stomach than in the midgut gland. showing individual ratios
for each enzyme. Elevated endochitinase activity in the stomach suggests that chitinous food is digested to oligomers in the
stomach, while the subsequent degradation to amino sugars occurs predominantly in the midgut gland.

Full text

Journal of Plankton Research Vol.18 no.6 pp.895-9O6.19%
A field study on the physiology of digestion in the Antarctic krill,
Euphausia superba, with special regard to chitinolytic enzymes
Friedrich Buchholz and Reinhard Saborowski
Institutfur
Meereskunde,
Diisternbrooker
Weg
20,
D-23105
Kiel,
FRG
Present
address:
Biologische
Anstalt
Helgoland,
Marine
Station,
D-27483
Helgoland,
FRG
Abstract. Endo- and exochitinase activities were determined in the stomach and midgut gland of the
Antarctic krill,
Euphausia
superba.
along a transect west of the Antarctic Peninsula. Activities were
compared with the digestive enzymes protease, cellulase (1.4-p-D-glucanase) and laminarinase
(1,3-fl-D-glucanase). The chlorophyll and protein contents in the surface water of the corresponding
stations were determined. Enzyme activities were characterized by high individual and spatial vari-
ations. Chitinolytic activity in the stomach correlated well with all digestive enzymes investigated. In
the midgut gland,
a
correlation with cellulase and laminannase
was
evident. The amount of chlorophyll
a and phytoplankton protein in the surface water was not correlated with enzyme activity. Specific
enzyme activity
was
higher in the stomach than in the midgut gland, showing individual ratios for each
enzyme. Elevated endochitinase activity in the stomach suggests that chitinous food is digested to oli-
gomers in the stomach, while the subsequent degradation to amino sugars occurs predominantly in the
midgut gland.
Introduction
Krill,
Euphausia
superba,
is a
key organism in the Antarctic
ecosystem.
The annual
production of krill was estimated by Everson (1977) at 500 million tons. While
feeding
on
phytoplankton, the primary producers, krill act as a food source for top
predators. Unlike copepod-dominated food chains, this short krill-dominated
food chain is characterized by a strong concentration of biomass at each of the
trophic levels (Hempel, 1985). In addition to its significance in Antarctic food
chains, krill has an important function in the nutrient flux from the euphotic zone
to deeper water layers and the benthos, respectively (Clarke
et
al.,
1988).
Krill are very effective organisms in the Antarctic environment. Accordingly,
the
physiological characteristics are
high
growth rates when sufficient food
is
avail-
able,
short intermoult periods (Buchholz, 1991) and a generally high energy
demand (Kils,
1982).
Correspondingly, krill must be able to utilize food efficiently.
This also becomes obvious in anatomical and morphological adaptations, e.g. the
filtering efficiency of the highly specialized filter basket (Hamner
et
al.,
1983;
Kils,
1983).
However, effective food utilization
also
demands physiological adaptations.
In this respect, efficient digestive enzymes are required to make nutrients available
for gastrointestinal resorption, since digestive enzymes represent the functional
link between food uptake and food utilization (Mayzaud
et
al.,
1985).
Corresponding to the preferred food, enzymes are present in the digestive tract
of krill that cleave algae-specific substances like the
1,3-fi-D-glucan
laminarin and
the l,4-f$-D-glucan cellulose (Mayzaud et al., 1985; McConville et al., 1986). In
addition, chitinolytic enzymes with high activities are present in the stomach and
the midgut gland. Their significance as digestive enzymes was discussed by
© Oxford University Press 895
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

F.Buchhofcz and R^aborowski
Buchholz (1989). However, our knowledge about the physiology of digestion in
Antarctic krill is still limited, particularly concerning the biochemical character-
istics
of digestive enzymes and mechanisms of enzyme
synthesis.
A further point of
interest for understanding the complex situation in the digestive tract of Antarctic
krill is the physiological interaction between the two principal sites of digestion:
the stomach and the midgut gland (hepatopancreas).
Since most studies on the physiology of digestion in crustaceans have been
carried out on total extracts or
in
only one digestive organ, we paid special atten-
tion to the determination of digestive enzyme activities in the stomach and the
midgut gland, separately. In order to establish whether chitinolytic enzymes are
relevant to digestive processes, we measured the activities of endochitinase
(poly-3-1,4-(2-acetamido-2-deoxy)-D-glucosid-glucanohydrolase) and exochitin-
ase (N-acetyl-(3-D-glucosaminidase, NAGase). The activity patterns were com-
pared with those of protease, cellulase and laminarinase by correlation analysis.
Furthermore, the amount of chlorophyll (chl) a as well as the protein content of
plankton at the sampling sites were related to the enzyme activities in the krill's
digestive organs.
Method
Krill and
phytoplankton
samples
Antarctic krill,
E.superba,
were
caught with a Rectangular Midwater
Trawl,
RMT
1+8
(Roe and Shale, 1979) during the cruise Met
11/4 (21
December
1989-18
Janu-
ary 1990) of the research vessel FS 'Meteor' west of the Antarctic Peninsula (Fig-
ure 1). Sampling was carried out in three depth horizons (Siegel, 1992). Samples
for our investigations were obtained from the upper horizon (60 m to surface).
Immediately after the catch, krill were frozen and stored at -75°C until analysis.
Phytoplankton were obtained from the same stations as the krill catches (Figure
1).
Surface water (140-750
1)
was filtered through 3 u,m membrane filters (Sar-
torius,
11302-293-G) with a pressure filtration system (Sartorius, SM 16277). The
filters were frozen and stored at -75°C until analysis.
Morphometric
data
and
colour index
of the
digestive organs
The lengths and fresh weights of thawed animals were measured, and the sex was
determined according to Makarov and Denys (1980). Animals with no distinct
sexual differentiation were classified as juveniles.
The fullness of the stomach, the midgut gland and the gut was determined visu-
ally using a colour index according to Morris
et
al.
(1983).
Dissection
and
homogenization
The stomach and hepatopancreas were dissected from frozen
animals.
Homogeni-
zation of the organs was carried out in a total volume of
1
ml 0.2 M citrate/phos-
phate buffer (CPB) (pH
5.5)
on ice
with an ultrasonicator (Branson, Sonofier
B-12,
microtip 101-148-063) at 30% of maximal energy for 3 x 15 s, interrupted by a
8%
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

Chhinolytk enzyme activity of E.superba
W 72°68°60°
56"
60°
Fig. 1. Station grid of cruise 11/4 of FS 'Meteor'. Only the sampling stations relevant for the present
investigation are marked by numbers.
break of 20
s.
The homogenates were centrifuged at
15 000 g
and the supernatant
was used for biochemical analysis.
Biochemical investigations
The amount of soluble protein in krill digestive organs
was
determined after Brad-
ford (1976) with the BioRad dye reagent (Cat. 500-0006). Samples were run in
duplicate with parallel bovine serum albumin (BSA) standards.
Protein of phytoplankton samples
was
determined on homogenates of plankton
collected on membrane filters. Pieces of these filters (7.5 cm2 in area) were trans-
ferred into reaction tubes and
3
ml of distilled water were added. Homogenization
was
carried out by ultrasonication (Branson, Sonifier B12) for
3
x
15
s,
interrupted
by a break of
20 s
on ice. After continuous shaking (Vortex),
1
ml of the solution
was transferred into 1.5 ml reaction tubes (Eppendorf
3810)
and centrifuged for
5
min at 15 000 g. The amount of protein in the supernatant was also determined
according to Bradford
(1976):
200
p.1
of dye reagent (BioRad) were added
to 800 ^1
of sample. BSA standards were run in
parallel.
Three different pieces of each filter
were analysed. Each analysis was performed in duplicate.
Chlorophyll a samples were determined according to Jeffrey and Humphrey
(1975).
Five millilitres of acetone were added to each piece of filter (20 cm2) in a
reaction tube. Chlorophyll
a was
dissolved
by
continuous
shaking.
After centrifug-
ing (15 min at 5200 g), the supernatant was measured photometrically. Blanks
were run in parallel with clean filters.
897
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

F.Buchholz and R.Saborowski
The activity of the endochitinase was determined with CM-Chitin-RBV (No.
04106,
Loewe Biochemica, Otterfing, FRG) as substrate (Wolf and Wirth, 1990)
and that of
the
exochitinase N AGase
with
p-nitrophenyl-N AG (Sigma,
N
9376) as
described by Saborowski et
al.
(1993). All measurements were run in duplicate
with a parallel blank.
The activity of total protease was determined as described by Donachie et al.
(1995) using Azocasein-Na-salt as substrate (Serva 14391).
The cellulase (l,4-(3-D-glucanase) activity
was
determined with the dye-labelled
substrate CM-Cellulose-RBB (No. 04100, Loewe Biochemica, Otterfing, FRG).
The assay was adapted to 1.5 ml Eppendorf reaction tubes. One hundred micro-
litres of substrate (4 mg/ml) were added to
250
u,l
of 0.2
M
CPB (pH 6.0) and
50
u,l
of sample to start the reaction. After 30 min incubation at 35°C, the reaction was
stopped by adding 100
u.1
of
1 M
HC1.
After centrifuging at
15 000 g
for
5
min, the
absorption of the supernatant was read at 600 nm against air.
The determination of laminarinase (1,3-P-D-glucanase) activity was carried out
as described for cellulase, except that the substrate was CM-Curdlan-RBB (No.
04107,
Loewe Biochemica, Otterfing, FRG) and the reaction was terminated with
2 N
HC1.
Statistics
Correlation analyses were performed on the data of enzyme activity within the
stomach and the midgut gland, as well as amounts of chl
a
and protein in the water,
and enzyme activities in the digestive organs. The strength of the association was
expressed by the Pearson correlation coefficients and the Bonferroni-adjusted
probabilities. The statistical analysis was carried out with the computer program
'Systat' (Wilkinson, 1989). The significance level was set at P
=
0.01.
Results
Phytoplankton samples
The average concentration of chl a at all stations was 0.61 mg m 3. Elevated
amounts occurred at stations 28-34 (except at station 29). Low values were
observed at stations
42-86
(Figure
2).
Stations
28-34
were located in the Bransfield
Strait, while the other stations were located in the Drake Passage (see Figure 1).
The protein values were more variable than the
chl a
values.
However, the distri-
bution pattern was similar, except at station
28.
The concentrations ranged from
1.2 mg irr3 at station 63 to 6.9 mg nr3 at station
34.
The protein values correlated
well with chl a concentrations (r
=
0.897).
Morphometric data
ofkrill
and
colour
index
of
the organs
The average length of sampled animals
was 42.0
±
4.5
mm and the average weight
was561.7 ± 216.5
mg.
The largest animals were found at station
51
(51.2 ±
2.9
mm,
1039.0 ±
140.1
mg),
while the smallest individuals
were
caught at station
32
(35.2 ±
1.5
mm,
270.9
±
31.0
mg).
The sex distribution
was
40%
female and
60%
male.
No
juveniles were encountered.
898
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

Chitinolytlc enzyme activity of E-superba
2,0
1.0-
\ 0,5
jf
0,0
" 8 •
6
4 •
2 •
0
Chlorophyll a
M
n^nr-nfi^n
Protein
28 29 30 32 33 34 42 51 62 63 67 76 86
Station
Fig. 2. Chlorophyll a and protein of phytoplankton at the sampling stations (means and SD of three
parallel experiments).
Animals at stations 28-^42 had nearly similar colour indices of their stomachs,
while greater differences occurred in the colour indices of the midgut and the gut
(Figure
3).
Animals at station 34 had the highest indices for stomach and hepato-
pancreas, while the lowest indices in all organs were found at station
51.
Stations
51-86 were characterized by lower indices and greater variability.
Enzyme
activity
of krill
field
samples
Enzyme activities in the krill stomach and midgut gland showed high variability
between regions and between individuals. In general, the highest activities of the
stomach enzymes investigated were found at station 32 and station 63 (Figures 4
and
5).
Furthermore, all enzyme activities increased from station 28 to station 32,
indicating a similar relationship of activities of the different enzymes within each
station. However, this relationship could not be observed at all stations.
6
4
2
0
I 4-
o 2
2
6 •
4 •
2
0
Stomach
Midgut gland
iflM
Gut
h.
ft ft
n
fl fl
ft
1*1
28 29 30 32 33 34 42 51 62 63 67 76 86
Station
Fig. 3. Colour indices of the digestive organs of E.superba [means and 95% confidence interval (CI)
n
=
5\.
899
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

F.Bochhob
and R5aborowski
STOMACHMIDGUT GLAND
28 29 30 32 33 34 42
51
62
63 67 76 86 28 29 30 32 33 34 42
51
62 63 67 76 86
StationStation
Fig. 4. Soluble protein and chitinolytic enzyme activity of the stomach and the midgut gland of
E.superba
(means and
95%
CI. n
=
5).
In the midgut
gland,
the enzyme activities showed
a
similar level of variability
as
in the stomach. However, the activity pattern was not similar to that of the
stomach. The apparently divergent values of chitinase and NAGase over all sta-
tions are remarkable.
Correlation analysis
Enzyme
activities in digestive
organs.
In the stomach, significant correlations were
found between all enzymes investigated (Table I), although, in some cases, the
correlation coefficients were weak. In the midgut gland, there was a good corre-
lation between chitinase and
cellulase,
as
well
as between chitinase and laminarin-
ase,
and also between cellulase and laminarinase. No correlation was found
between NAGase and other digestive enzymes.
Enzyme
activities
of
the stomach versus activities
of the
midgut
gland.
There were
only two significant correlations between enzyme activities
in
the stomach and the
midgut gland. Stomach laminarinase was significantly correlated with midgut
gland protease (r
=
0.508)
and laminarinase (r
=
0.636).
Plankton
protein,
chl
a
and
colour
index of the
organs versus
enzyme
activities
of
digestive
organs.
In addition to chl a and protein in the surface water, the colour
index of the digestive organs was included in the correlation analysis. This index
represents the amount of phytoplankton already ingested by the animals. No sig-
nificant correlations existed between the stomach enzyme activities and either the
amount of
chl
a or the protein content. Only the colour index of the stomach was
weakly, but significantly correlated with the laminarinase activity (r
=
0.471).
900
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

Chltinolytlc enzyme activity of Ksuperba
STOMACH
Ml
DGUT GLAND0.08
28 29 30 32 33 34 42
51
62 63 67 76 86
Station28 29 30 32 33 34 42 51 62 63 67 76 86
Station
Fig.
S.
Digestive enzyme activity in the stomach and the midgut gland of E.superba (means and 95% CI,
/i
=
5).
In the midgut gland, the colour index was inversely correlated with the protein
content of the organ (r = -0.487). Furthermore, there was a significant negative
correlation between the chl a content and NAGase activity (r
=
-0.477).
Chl a and plankton protein
versus
colour
indices.
There was a weak correlation
between the colour indices of the stomach and the midgut gland and the chl
a
and
plankton protein (Table II), but the colour index for the gut was not correlated
with these parameters of the plankton.
Ratio
of enzyme
activities
in
the stomach
and the midgut
gland
The ratios of enzyme activities in these organs were calculated as the quotient of
enzyme activity in the stomach versus the enzyme activity in the midgut gland. The
highest ratio occurred for protease (9.84 ± 1.45) which indicated that the protease
activity in the stomach
is
nearly
10
times higher than that in the midgut gland (Fig-
ure
6).
High ratios also occurred for cellulase and laminarinase. The chitinase and
NAGase had the lowest
ratios.
The activity of NAGase was only slightly higher in
stomach than in the midgut gland. The amount of protein was higher
in
the midgut
gland than in the stomach (ratio 0.86 ± 0.07).
Discussion
Chitinolytic enzymes are widely distributed in nature (Jeuniaux, 1966). In arthro-
pods,
they are known as moulting enzymes and they also act as digestive enzymes,
for example in fish which feed on crustaceans (Danulat, 1986,1987; Rehbein
etal.,
1986;
Lindsay, 1987).
901
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

F.Bochbolz
and R-Saborowski
Table I. Matrix of correlation coefficients between the activities of different enzymes in the stomach
and the midgut gland of
E.superba
Stomach
Protein
Chitinase
NAGase
Protease
Cellulase
Midgut gland
Protein
Chitinase
NAGase
Protease
Cellulase
*P<O.0l.n =
Chitinase
0.648*
-
-
-
-
0.002
-
-
_
-
65.
NAGase
0.761*
0.756*
-
-
-
0.015
0.148
-
_
-
Protease
0.820*
0.712*
0.719*
-
-
0.206
0.427*
-0.107
_
-
Cellulase
0.691*
0.636*
0.549*
0.856*
-
0.114
0.743*
-0.168
0.638*
-
Laminarinase
0.476*
0.532*
0.521*
0.671*
0.548*
-0.002
0.701*
-0.005
0.451*
0.734*
Buchholz (1989) described the chitinolytic activity in the stomach and midgut
gland of the Antarctic krill
E.superba
and the Northern krill
Meganyctiphanes
nor-
vegica,
and discussed their significance as digestive enzymes. In the present study,
we
investigated the occurrence of chitinases
in
the stomach and in the midgut gland
of field samples of E.superba and compared the activities with the digestive
enzymes protease, cellulase and laminarinase.
Since Antarctic krill are predominantly herbivorous, phytoplankton field
samples were used to estimate the food available to the
krill.
The amounts of
chl a
and protein were used as a measure of the phytoplankton concentration. For tech-
nical reasons, sampling could only be performed in the surface water.
The chl
a
concentrations (average 0.6
mg
nr3) were similar to those reported by
other authors: Smith and Nelson (1985) found
chl a
values of
0.25-1.6
mg
nr3
in
the
surface waters of the Ross Sea. Deeper water layers (0-25 m) had only slightly
higher amounts of chl
a.
High concentrations occurred in areas of phytoplankton
blooms (up to
8
mg nr3). El-Sayed and Turner (1977) reported a mean chl a con-
centration of 0.26
mg
nr3 in
the
surface waters of the Pacific sector of
the
Antarctic
Ocean. El-Sayed and Taguchi (1981) found 1.65 mgchl
a
nr3 in the southern Wed-
dell Sea. References to phytoplankton protein concentrations in Antarctic waters
are given by Mayzaud
et
al.
(1985). The reported amounts are about one order of
magnitude higher than those found
in
our
study.
This discrepancy can be explained
by the differences between the extraction procedures. With our method, we deter-
mined the water-soluble protein, which is certainly lower than the NaOH heat
extraction performed by Mayzaud
et
al.
(1985). However, the protein data of our
investigations correlated well with the amount of
chl
a(r
=
0.897), indicating that
the protein
was
of phytoplankton origin and thus
was
available as
a
food source for
krill. In general, chl
a
and protein values were higher at the stations located in the
Bransfield Strait than at those located in the Drake Passage. This could be due to
the hydrographic conditions prevailing during the cruise (Stein, 1992).
The activities of the enzymes investigated showed different patterns in the
stomach and the midgut gland. In the stomach, significant correlations between
902
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

Chitinolytic enzyme activity of
ELsuperba
Table II. Correlation coefficients between chl a and plankton protein and the colour indices of the
digestive organs of
E.superba
Chl
a
Plankton protein
•/><0.01.n=65.
Colour index
Stomach
0.579*
0.539*
Midgut gland
0.595*
0.589*
Gut
0.247
0.120
chitinase and protease, as well as between protease and cellulase, occurred, but
there was no correlation between cellulase and laminarinase or between chitinase
and laminarinase. Similar differences in correlation analysis of digestive enzymes
in E.superba were reported by Mayzaud et
al.
(1985) for the enzymes amylase,
laminarinase, cellulase,
3-galactosidase
and trypsin. The results of our field inves-
tigations indicate a selective availability of digestive enzyme activity. If enzymes
were always present in similar proportion to one another, significant relationships
should exist. Statistical relationships between the nutrients carbohydrate and pro-
tein, chlorophyll and the digestive enzymes amylase, maltase, cellubiase, lactase,
trehalase as well as acid and alkaline phosphatase were reported by Mayzaud and
Conover (1975). The investigations were performed on zooplankton total
samples, dominated by the copepods
Calanus
minutus,
Acartia clausi
and Temora
longicornis.
Corresponding investigations by Mayzaud
et
al.
(1985) on
E.superba
did not give similar results, and our results also did not show correlations between
chl a or protein values and enzyme activities. The chemical components of phy-
toplankton at only
one
depth are probably not fully representative of
the
resources
available to the
krill.
Mayzaud et
al.
(1985) investigated the phytoplankton at
15 m
depth and we did so in the surface water.
Furthermore, misinterpretations are likely because of the high mobility and fil-
tering efficiency of krill: in contrast to relatively small copepods, E.superba is
characterized by a much higher mobility and filtering efficiency. Phytoplankton
can thus be depleted rapidly and effectively by krill swarms. Consequently, deter-
mining the potential food in the water may lead to underestimations at high graz-
ChiurusoNAGasc
Fig.
6.
Ratios of enzyme activities between
the
stomach and midgut gland of
E.superba
(means and 95 %
CI,n
=
65).
903
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

F.Buchhoh and R^aborowski
ing
activity of
krill.
This could explain the lack of correlation between amounts of
food and digestive enzyme activity reported above. Krill are also fast swimmers
and can leave grazed areas rapidly. Therefore, we cannot be sure that krill caught
in a certain area were also feeding there. A more appropriate estimate of the
amount of food ingested by
E.superba
may be the colour index of the digestive
organs. However, the correlation analysis of the colour index with the digestive
enzyme activity showed a significant relationship only in the case of laminarinase,
while no significant correlation was found between the colour index and the other
enzymes investigated. Based
on
these
results,
it can be suggested that laminarin (or
generally
1,3-fi-D-glucan)
was
the main component of the phytoplankton ingested
by the krill and that the corresponding enzyme activity was induced as postulated
by Mayzaud and Mayzaud (1981).
The activities of
all
the enzymes investigated were higher in the stomach than in
the midgut gland, especially for protease which showed activities nearly 10 times
higher in the stomach. This effect could be explained by a higher concentration of
enzymes in this organ. Enzymes are synthesized in the midgut gland and released
into the stomach, where they occur in a relatively high quantity compared to the
midgut gland. Furthermore, expressing enzyme activity per unit fresh weight leads
inevitably to a high result for the stomach, since the stomach contains only the
material to be digested and the digestive enzymes, and has a relatively low weight.
In contrast, the midgut gland
is
a more complex organ with a high amount of con-
nective tissue that contributes
to
the weight and this leads to relatively lower activi-
ties than in the stomach.
However, the relationships between the activities of
the
enzymes investigated in
the stomach and the midgut gland should be similar for all enzymes. Since this is
not the case, it can be suggested that digestive enzymes are selectively induced or
activated in the stomach of
E.superba.
A stimulation of digestive enzymes was
reported by De-La-Ruelle
et
al.
(1992). These authors determined an increase in
aminopeptidase activity in the hepatopancreas of the freshwater crayfish
Procam-
barus clarkii following the addition of millimolar quantities of different sub-
stances, among them cysteine. Similar mechanisms could operate with the
proteolytic enzymes of
E.superba
and explain the high activities in the stomach
compared to the midgut gland.
In the case of the chitinolytic enzymes, the activity of endochitinase is distinctly
higher in the stomach than in the midgut gland (stomach:midgut ratio
3:8).
In con-
trast, the activity of exochitinase (NAGase) was nearly equal in both organs (ratio
1:2). These results can be explained by the different biochemical functions of both
enzymes. While the endochitinase
cleaves
chitin to
oligomers,
NAGase hydrolyses
the
oligomers to amino
sugars.
Consequently, the chitinase reaction
in
the stomach
constitutes the first step in chitin digestion. The further degradation to amino sug-
ars takes place in the midgut gland. Similar, but technically more difficult investi-
gations for endo- and exoenzymes of cellulase and laminarinase would provide
more information on this point.
The possible contribution of bacteria to digestive enzyme activity cannot be
neglected (Especje et
al.,
1987; Rakusa-Suszczewski and Zdanowski, 1989) par-
ticularly for proteases, as shown by Donachie et
al.
(1995) in the Northern krill,
904
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

Chitinolytk enzyme activity ot
E.superba
M.norvegica.
However,
we
concluded that the digestive enzymes of
E.superba
are
mainly endogenous. It seems unlikely that a population of bacteria big enough to
make a significant contribution to enzymes can continuously survive in krill stom-
achs.
Because food is continuously ingested (Antezana
el
al.,
1982) and there is a
high turnover rate of food in the stomach, bacteria would be egested through the
gut. Furthermore, krill moult frequently every 2-3 weeks during the summer
(Buchholz, 1991) and the chitinous cuticule of the stomach is replaced. Therefore,
the stomach contents would also be replaced. After each moult, it would take some
time for a new bacterial population to become established. If such a population
contributed significantly to digestive enzyme activity, a decrease in enzyme
activity should be evident after each moult. However, this was not observed in
laboratory studies (Buchholz, 1989).
Since Antarctic phytoplankton can be dominated by diatoms that are charac-
terized
by
the presence of chitinous spines (Johansen and Fryxell,
1985),
the ability
to digest chitin
is
obviously significant. In conclusion, our study has confirmed that
chitinolytic enzymes
in the
stomach and midgut gland of
field
samples
of E.superba
are digestive enzymes due to their similar properties compared to other digestive
enzymes. In order to satisfy their energy demand, krill are able to utilize a wide
range of nutrients.
Acknowledgements
We thank the staff of the research vessel FS 'Meteor' for excellent support on
board. Technical assistance was contributed by Mrs Sonja Bohm and Mr Gerrit
Sahling. Constructive comments and discussion was contributed by Dipl. Biol.
Gerrit Peters, Dr Ralf-Achim Vetter and Dr Matthew Dring, who also corrected
the English. This work was supported by a grant of the German Research Council
(DFG Bu 548/2).
References
Antezana.T., Ray.K. and Melo.C. (1982) Trophic behavior of
Euphausia superba
Dana in laboratory
conditions.
Polar
Biol.,
1, 77-82.
Bradford.M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of
protein utilizing the principle of protein-dye binding. Anal.
Biochem.,
72,
248-254.
Buchholz,F. (1989) Moult cycle and seasonal activities of chitinolytic enzymes in the integument and
the digestive tract of the Antarctic krill,
Euphausia
superba.
Polar
Biol.,9,311-317.
Buchholz,F. (1991) Moult cycle and growth of Antarctic krill,
Euphausia
superba,
in the laboratory.
Mar.
Ecol.
Prog.
Ser.,
69,217-229.
Clarke,A.,
Quetin,L.B.
and Ross.R.M. (1988) Laboratory and field estimates of the rate of faecal pellet
production by Antarctic krill,
Euphausia
superba.
Mar.
Biol.,
98,
557-563.
Danulat,E. (1986) The effects of various diets on chitinase and beta-glucosidase activities and the con-
dition of cod,
Gadus morhua
(L).
J.
Fish.
Biol.,
28,191-197.
Danulat,E. (1987) Digestibility of chitin in cod, Cadus morhua, in vivo.
Helgol.
Meeresunters.,
41,
425-436.
De-La-Ruelle,M., Hajjou,M., Van-Herp,F. and Le-Gal,Y. (1992) Aminopeptidase activity from the
hepatopancreas of
Procambarus
clarkii.
Biochem.
Sysl.
Ecol.,
20,331-337.
Donachie.S., Saborowski.R., Peters.G. and Buchholz,F. (1995) Bacterial digestive enzyme activity in
the stomach and hepatopancreas of
Meganyctiphanes norvegica
(M.Sars,
1857).
J.
Exp.
Mar.
Biol.
Ecol.,
188,151-165.
EI-Sayed,S.Z. and
Taguchi,S.
(1981) Primary production and standing
crop
of phytoplankton along the
ice-edge in the Weddell Sea.
Deep-Sea
Res.,
28A, 1017-1032.
905
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from

F.Budiholz
and Ritaborowski
El-Sayed.S Z. and TurnerJ.T. (1977) Productivity of the Antarctic and the tropical-subtropical
regions: A comparative study. In Dunbar M. (ed.).
Polar
Oceans.
Proceedings
of
the
SCOR/SCAR
Polar Oceans
Conference.
Montreal. 1974. Antarctic Institute of North America. Calgary, pp.
463-503.
Especje.M.E.. Molina.M.G. and Fraile.E.R. (1987) Enzymatic activities of psychrotrophic bacteria
from Antarctic krill. BIO MASS
Sci.
Ser
.
No. 7. 133-139.
Everson.I. (1977) The living resources of the Southern Ocean. Rome FAO United Nations Develop-
ment Programme, Southern Ocean Fisheries Survey Programme
GLO/SO/77/1.
156 pp.
Hamner.W.M.. Hamner.P.P., Strand.S.W. and GilmerR.W. (1983) Bahavior of Antarctic krill.
Euphausia
superba:
chemoreception. feeding, schooling, and moulting.
Science.
220.433-435.
Hempel.G. (1985) Antarctic marine food weds. In Siegfried.W.R., Condy.P.R. and Laws.R.M. (eds).
Antarctic Nutrient Cycles
and
Food
Webs.
Springer-Verlag, Berlin, pp. 266-270.
Jeffrey.S.W. and Humphrey.G.F. (1975) New spectrophotometnc equations for determining chloro-
phyll a. b, cl and c2 in higher plants, algae and natural phytoplankton.
Biochem.
Physiol.
Pflanzen.
167.191-194.
Jeuniaux.C. (1966) Chitinases.
Methods
Enzymot..
8, 644-650.
JohansenJ. and Fryxell.G A. (1985) The genus
Thalassiosira
(Bacillariophyceae): Studies on species
occurring south of the Antarctic Convergence Zone.
Phycologica,
24,
155-179.
Kils.U. (1982) The swimming behavior, swimming performance and energy balance of Antarctic krill.
Euphausia
superba.
BIOMASS
Set.
Ser..
No.
3.
122 pp.
Kils.U. (1983) Swimming and feeding of Antarctic krill,
Euphausia superba—some
outstanding ener-
getics and dynamics, some unique morphological details. Ber
Polarforsch.
Sonderh.,
4, 130-155.
Lindsay.GJ.H. (1987) Seasonal activities of chitinase and chitobiase in the digestive tract and serum of
cod. Gadus
morhua
(L.).
J.
Fish.
Biol..
30,495-500.
Makarov.R.R. and Denis.C.J. (1980) Stages of sexual maturity of
Euphausia
superba.
BIOMASS
Handbook
Scr,
11,
1-11.
Mayzaud.P. and Conover.R.J. (1975) Influence of potential food supply on the activity of digestive
enzymes of nentic zooplankton. In Persoone.G. and Jaspers.E. (eds),
Proceedings
of the 10th
European
Symposium of
Marine
Biology,
Osiend,
1975.
Universa Press. Wettern. Belgium, Vol. 2,
pp.
415-427.
Mayzaud.P. and Mayzaud,O. (1981) Kinetic properties of digestive carbohydrases and proteases of
zooplankton.
Can.
J.
Fish.
Aquat.
Sci.,
38,
535-543.
Mayzaud.P., Farber-Lorda J. and Corre.M.C. (1985) Aspects of the nutritional metabolism of
two
Ant-
arctic
euphausiids:
Euphausia
superba
and Thysanoessa
macrura.
In Siegfried.W.R., Condy.P.R. and
Laws.R.M. (eds),
Antarctic Nutrient Cycles
and
Food
Webs.
Springer-Verlag. Berlin, pp. 330-338.
McConville,M.J., Ikeda.T., Bacic.A. and Clarke,A.E. (1986) Digested carbohydrases from the hepato-
pancreas of
two
antarctic euphausiid species (£up/u7U5iasii/w6a and
E.crystallorophias).
Mar.
Biol.,
90.371-378.
Morris.DJ.. Ward.P. and Clarke.A. (1983) Some aspects of feeding in the Antarctic krill,
Euphausia
superba.
Polar
Biol.,
2. 21-26.
Rakusa-Suszczewski.S. and Zdanowski.M.K. (1989) Bacteria in krill (Euphausia superba Dana)
stomach. Acta Protozooi, 28.87-90.
Rehbein.H., Danulat,E. and Leinemannjvi. (1986) Activities of chitinase and protease and concen-
tration of fluoride
in
the digestive tract of Antarctic
fishes
feeding
on
krill (Euphausia
superba
Dana).
Comp.
Biochem.
Physiol.,
85A.
545-551.
Roe.H.S.L. and Shale.D.M. (1979) A new multiple rectangular midwater trawl (RMT 1+8M) and some
modifications to the Institute of Oceanographic Sciences" RMT
1+8.
Mar.
Biol.,
SO,
283-288.
Saborowski.R., Buchholz.F., Vetter.R.-A.H., Wirth.S J. and Wolf.G.A. (1993) A soluble, dye-labelled
chitin derivative adapted for the assay of krill chitinase.
Comp.
Biochem.
Physiol..
105B, 673-678.
Siegel.V. (1992) Assessment of the krill
(Euphausia superba)
spawning stock off the Antartic Penin-
sula. Arch.
Fischereiwiss.,
41,
101-130.
Smith.W.O.Jr and Nelson.D.M. (1985) Phytoplankton biomass near a receding ice-edge in the Ross
Sea. In Siegfried.W.R.. Condy.P.R. and Laws.R.M. (eds),
Antarctic Nutrient Cycles and Food
Webs.
Springer-Verlag. Berlin, pp. 70-77.
Stein.M. (1992) Variability of local upwelling off the Antarctic Peninsula, 1986-1990.
Arch.
Fischcrei-
wiss..
41,
131-158.
Wilkinson.L. (1989) SYSTA
T:
The
System
for
Statistics.
SYSTAT
Inc..
Evanston. IL, 638 pp.
Wolf.G.A. and Wirth.S J. (1990) Application of soluble chromogenic substrates for assays of poly-
saccharide endo-hydrolase activity. In KlemenuZ.. RudoJpJiJC. and SandsJ5.C. (eds). Methods in
Phytobacteriology.
Akadfimia Kiad6. Budapest, pp. 409-413.
Received
on
January
24,
1995;
accepted
on
January
15,
1996
906
by guest on June 1, 2012http://plankt.oxfordjournals.org/Downloaded from