Developmentaland ComparativeImmunology,Vol. 17, pp. 319-330, 1993
00
Printed in the USA. All rights reserved.
RECOGNITION OF YEAST CELL WALL GLUCAN BY ATLANTIC
SALMON (Salmo salar L.) MACROPHAGES
Rolf E. Engstad and Borre Robertsen
Department of Marine Biochemistry, Norwegian College of Fishery Science, Universityof Tromso,
N-9037 Troms~, Norway
(Submitted September 1992; Accepted December 1992)
7qAbstract--Phagocytosis of yeast (Saccharo-
myces cerevisiae) glucan particles by Atlantic
salmon (Salmo salar L.) pronephric macro-
phages was studied. The particles contained
>95% glucose linked through 11-1,3-and 11-1,6-
glycosidic linkages. The maerophages rapidly
phagocytized both native and opsonized gluean
particles although the latter were taken up at a
higher rate. Within 30 rain, 40-60% of the
macrophages had taken up >1 native glucan
particle. The uptake of native glucan particles
could be inhibited by preincubating the macro-
phages with laminarin, a soluble 11-1,3-1inked
glucan, and a soluble yeast gluean made by
partial formolysis of gluean particles. Soluble
yeast gluean, on the other hand, did not inhibit
uptake of serum opsonized glucan particles or
sheep red blood cells, which showed that it did
not interfere with phagocytosis in general or
inhibit phagocytosis through complement re-
ceptors. Polyglucoses with glycosidic linkages
other than 11-1,3, like dextran, glycogen, and
pustulan or the polymarmose marman, showed
little or no inhibition of phagocytosis of native
glucan particles. Altogether these observations
indicate that Atlantic salmon macrophages
may have a specific receptor for yeast glucan.
Studies with chelator- and heat-treated salmon
serum showed that glucan particles were op-
sonized primarily by activation of the alterna-
tive complement pathway. However, the data
indicate that serum components other than
complement may also be involved in the op-
sonization of gluean particles.
7qKeywords--Atlantic salmon; Yeast glucan;
Macrophages; Phagocytosis; Gluean receptor;
Complement; Opsonins.
Address correspondence to BOrre Robertsen.
319
0145-305X/93 $6.00 + .
Copyright © 1993 Pergamon Press Ltd.
Introduction
Animals have developed various recog-
nition mechanisms for fungal cell wall
glucans (13-glucans) as a defense against
infection by yeasts and mycelial fungi. In
vertebrates, 13-glucans are known to be
powerful activators of the alternative
complement pathway (1) and to be po-
tent elicitors o f r e s p i r a t o r y b u r s t in
p o l y m o r p h o n u c l e a r phagocytes (2). In
mammals 13-glucans have been shown to
stimulate anti-tumour mechanisms and
to increase the host resistance not only
to fungal pathogens, but also to bacterial
and viral pathogens (3). Recognition of
13-glucans thus appear to activate the
nonspecific defense of animals. Fungal
glucans also appear to have stimulatory
effects on nonspecific defense mecha-
nisms in lower vertebrates like fish. Our
group has recently shown that a 13-1,3-
and 13-1,6-1inked glucan from cell walls of
S a c c h a r o m y c e s cerevisiae enhance the
resistance o f Atlantic salmon against
several gram-negative pathogenic bacte-
ria (4) and i n c r e a s e the a c t i v i t y o f
lysozyme and c o m p l e m e n t in the fish
blood (5). Other fungal 13-glucans have
been shown to enhance the resistance of
carp against bacterial infection (6). Yeast
13-glucan has also been shown to function
as an adjuvant in vaccine against Aero-
m o n a s salmonicida in Atlantic salmon
(7).
Although the mechanisms underlying
the immunostimulatory action of 13-glu-
can in vertebrates are not fully under-
320
stood, several reports have emphasized
the involvement of mononuclear phago-
cytes (8-10). In vitro studies of human
m o n o c y t e s and mouse macrophages
have shown that [3-glucans elicit produc-
tion of leukotrienes (1 l), cytokines, and
prostaglandin E2 (12), messenger mole-
cules that modulate in vivo the function
of leukocytes. Recent experiments have
shown that injection of rainbow trout
with yeast glucan results in enhancement
of the bactericidal activity of head kid-
ney macrophages (13) that indicates that
macrophages are important in the glu-
can-induced resistance of fish.
Because of the in vitro and in vivo
effects of [3-glucans on mononuclear
phagocytes, we have focused our inter-
est on how these compounds are recog-
nized by these cells. Glucan particles are
rapidly phagocytized by mammalian
mononuclear phagocytes and the uptake
has been shown to be inhibitable by sol-
uble polysaccharides with structures ho-
mologous to the particles used (14-16).
A yeast glucan receptor has recently
been isolated from human monocytes
(17). The true specificity of glucan recep-
tor(s) o f m a m m a l i a n m o n o n u c l e a r
phagocytes has, however, not yet been
determined.
The aim of the present work was to
investigate recognition mechanisms for
yeast [3-1,3- and [3-1,6-1inked glucan by
Atlantic salmon macrophages. This was
done by studying phagocytosis of native
and opsonized yeast glucan particles by
salmon macrophages. Ligand inhibition
experiments were performed to examine
the specificity of the uptake of glucan
particles. The nature of glucan opsonins
in salmon serum was also studied.
Materials and Methods
Chemicals
Laminarin from Laminaria digitata,
mannan from Saccharomyces cerevi-
R.E. Engstad and B Robertsen
siae, oyster glycogen, and dextran (M w
70,000) were obtained from Sigma Co.
(St. Louis, MO). Pustulan was obtained
from Calbiochem Co. (La Jolla, CA).
Glutaraldehyde fixed sheep red blood
cells (SRBC) were obtained from Sigma.
Microparticulate (3-4 ~m) yeast glucan
from cell walls of S. cerevisiae (M-
Glucan), was obtained from KS Biotec-
Mackzymal, (Troms¢, Norway). Soluble
yeast glucan was prepared by partial for-
molysis of particulate yeast glucan in
90% formic acid at 100°C for 30 min. Af-
ter evaporating the formic acid the resi-
due was boiled in distilled water for 3 h,
dialysed (cut off 5000 Mw) against tap
water overnight, and lyophilized.
Determination of Linkage
Composition of the Yeast Glucan
(Methylation Analysis)
The yeast glucan was methylated by
the method of Hakamori (18), whereafter
the permethylated glucan was hydro-
lysed and acetylated by the method of
Albersheim et al. (19). The partially
methylated alditol acetates were anal-
ysed by G C - M S on a Hewlett Packard
5890 A gas chromatograph and a 5970 B
series mass selective detector. The cap-
illary column (30 m x 0.25 mm i.d., film
thickness 0.2 ~m), was a SP-2380 with
partially crosslinked cyanosilicone as
stationary phase (Supelco).
Isolation of Head
Kidney Macrophages
Atlantic salmon, 1-2 kg weight, were
obtained from the Centre of Aquacul-
ture, Troms¢. After anaesthetizing the
fish with p-aminobenzoate (0.04 g L 1)
the fish were bled from the caudal vein
using evacuated tubes (Venoject, Te-
rumo-Europe, Belgium) before the head
kidney was removed. Macrophages were
isolated by a procedure adapted from
Recognition of yeast p-glucan
Braun-Nesje et al. (20). The head kid-
ney was divided into four equal parts and
each part pushed through a steel mesh in
a petri dish with L-15 medium (380
mOsm) containing 2% fetal calf serum
(FCS), 100 units m L - 1 penicillin, 0.1 mg
mL-1 streptomycin and 20 units m L - l
heparin (Gibco). The cell suspension was
layered on top of a 37%/51% discontinu-
ous Percoll (Pharmacia) density gradient
and centrifuged at 400 x g for 40 min.
The macrophage enriched cell band at
the gradient interface was suspended in
L-15 with 0.1% FCS to a concentration
of 1 x 10 6 cells mL -1, and a 1-mL cell
suspension was seeded in 24-well culture
plates (Nunc) supplied with 12-mm cov-
erslips. The macrophages were allowed
to adhere for 2 h at 14°C before nonad-
herent cells were washed off. After
washing with PBS, the cells were supple-
mented with L-15 containing 0.1% Med-
icult ® Synthetic Serum Replacement 2
(SSR2) obtained from Medi-Cult A/S
(Copenhagen, Denmark) instead of FCS.
The cells were cultured overnight and
washed three times before phagocytosis
assays were performed.
Phagocytosis
Phagocytosis of glucan particles was
studied by adding 5 x 10 6 yeast glucan
particles in L-15 containing 0.1% SSR2
to each well, after which the culture
plates were centrifuged at 140 x g for 2
min to bring the particles in contact with
the macrophages. The culture plates
were then incubated at 18°C for 5-30
min. The macrophage layers were next
washed three times with PBS to remove
free glucan particles and fixed with cold
2.5% glutaraldehyde in PBS for a mini-
mum of 2 h. After fixation the coverslips
were washed three times with distilled
H20, stained with Giemsa, and mounted
on micro slides with histokit. For each
well, 200-400 macrophages were exam-
ined by light microscopy. The number of
321
macrophages that had phagocytized ~<1
particle, > I particle, and the total num-
ber of particles p h a g o c y t i z e d were
counted. Results are presented as per-
cent macrophages with > 1 particle and
phagocytic index, defined as the number
of glucan particles phagocytized divided
by the total number of macrophages
counted on each coverslip.
Inhibition of phagocytosis was stud-
ied by incubating the macrophages with
various soluble polysaccharides in cul-
ture medium for 10 min prior to the ad-
dition of glucan particles and then incu-
bating with particles for 30 min unless
otherwise stated.
Glucan particles were opsonized by
incubating yeast glucan particles (6 x 10 6
m L - I ) for 1 h at 14°C in 10% salmon
serum in either GVB 2÷ buffer (21), GVB
with 5 mM EDTA, or GVB with 5 mM
MgCI2 and 4 mM EGTA, or in 10% heat
inactivated (45°C, 30 min) salmon serum.
After incubation, the glucan particles
were centrifuged, washed, and resus-
pended in L-15 containing SSR2 by ul-
trasonication at 60 W for 30 s using a
Bandelin Sonopuls HD 60 sonifier.
Statistics
The results are presented as mean av-
erages (-SD). Student's t-test was used
to analyse the data and calculate the p
value.
Results
Linkage Composition of the
Yeast Glucan
Previous studies have shown that
yeast glucan particles contain more than
95% g l u c o s e (4). The m e t h y l a t i o n
analysis resulted in the appearance of
1,5-di-O-acetyl-2,3,4 ,6-tetra-O-methyl-
D-glucitol; 1,3,5-tri-O-acetyl-2,4 ,6-tri-O-
methyl-D-glucitol; 1,5,6-tri-O-acetyl-
322 R.E. Engstad and B. Robertsen

2,3,4-tri-O-methyl-D-glucitol; and 100

A
1,3,5,6-tetra-O-acetyl-2,4-di-O-methyl-
D-glucitol in a molar ratio of about 1.0: 80
13.8:1.0:0.8, respectively. These data
show that the yeast glucan is composed 6o
of about 6% nonreducing terminals, 83%
[3-1,3-1inked glucosyl residues, 6% [3-1,6- 4o
linked glucosyl residues, and 5% [3-1,3,6- .-~
"~ 20
linked glucosyl residues. The glucan is K
thus a branched polysaccharide with
[3-1,3,6-1inked glucose as the branch ~ 0 0 5 10 15 20 25 30
points. 7 Time (min)

100 "

Phagocytosis of Nonopsonized and 80" B

Opsonized Yeast Glucan Particles

by Salmon Macrophages E
60"

The rate of phagocytosis of nonop-

sonized glucan particles (NG) and op-
sonized glucan particles (OG) by salmon
macrophages were studied in two sepa-
rate experiments with cells from two dif-
ferent fish (Fig. 1). Both OG and NG
were rapidly phagocytized by salmon
m a c r o p h a g e s . OG w e r e , h o w e v e r ,
phagocytized at a higher rate than NG in
both experiments. In experiment 1 [Fig.
l(a)] 65% of the m a c r o p h a g e s had
phagocytized >10G and 37% had
phagocytized >1 NG after 30 min; in ex-
periment 2 [Fig. 1(b)] 90% of the macro-
phages had phagocytized > 1 0 G and
49% >1 NG after 30 min. Phagocytosis
of OG reached a plateau after 10 min in
experiment 1 and 5 min in experiment 2,
while the phagocytosis of NG seemed to
reach a plateau after 20 min in both
experiments.
Although there were differences in the
total phagocytic activity of macrophages
isolated from different fish, the ligand in-
hibition and opsonization experiments
described gave similar results for all in-
dividuals tested. As a consequence of
the differences seen in phagocytic activ-
ity of macrophages isolated from differ-
ent fish, all studies that are referred to as
one experiment were performed simulta-
neously and with macrophages isolated
20
0 •
0 5 10 15 20 25 30
Time (min)
Figure 1. Rate of phagocytosis of opsonized
(-C)-), and nonopsonized ( - B - ) glucan parti-
cles by salmon pronephric macrophages iso-
lated from two different fish (A and B). Values
are mean percent m a c r o p h a g e s w i t h > 1 g l u c a n
particle in two (A) and four wells (B), vertical
bars are standard deviation.
from the same fish. As a control in each
experiment, macrophages given NG and
OG were always included. Each experi-
ment described is representative of at
least three experiments.
Inhibition of Phagocytosis of Yeast
Glucan Particles by Preincubation
With Soluble Polysaccharides
To determine if phagocytosis of glucan
particles by salmon macrophages is me-
diated through a ligand inhibitable recep-
tor, the macrophages were pretreated
with different soluble polysaccharides
prior to addition of particles. Pretreat-
ment of macrophages with increasing
Recognition of yeast I~-glucan
doses of the soluble 13-1,3-glucan, lami-
narin, resulted in a dose-dependent re-
duction in uptake (Fig. 2). When the
macrophages were pretreated with solu-
ble polyglucoses with linkages other than
13-1,3, such as glycogen, dextran, and
pustulan or with the polymannose, man-
nan, no, or at best a minor, inhibition of
uptake was observed (Table 1). A five-
fold increase in the concentration of
these polysaccharides gave no difference
in inhibition of uptake of glucan parti-
cles. Pretreatment of macrophages with
glucose or mannose (1.0 mg m L - l ) had
no effect on the uptake of glucan parti-
cles (not shown). On the other hand, a
soluble 13-1,3 and 13-1,6-1inked glucan,
made by partial formolysis of glucan par-
ticles, proved to be a very effective in-
hibitor of phagocytosis (Table 1). On a
weight basis this polyglucose prepara-
tion was an even more effective inhibitor
than laminarin. When the macrophages
were pretreated with 1.0 mg mL-~ solu-
ble glucan, the inhibition was more than
90% whereas 1.0 mg mL -1 laminar-
in gave 54% inhibition and at 0.2 mg
mL-~; soluble yeast glucan and lami-
narin showed 60% and 34% inhibition,
respectively.
60
~ 50
~ 4o
~ .
A
~ 3o-
20-
,11
~°t~ 10"
N 0
• 6 "16o
Laminarin concentration (lag mE1)
Figure 2. Inhibition of phagocytosis of glucan particles by salmon macrophages with increasing
concentrations of laminarin. Laminarin was added to the macrophages 10 min prior to the addition
of glucan particles. Values are mean percent macrophages with >1 glucan particle -+ SD, n = 4.
323
Comparison of the inhibitory effect of
the different polysaccharides on the ba-
sis of phagocytic index (number of glu-
can particles per macrophage) gave es-
sentially the same picture as comparison
on the basis of percent macrophages
with > 1 particle (Table 1).
The effect of lower concentrations
(10-200 Ixg mL-1) of soluble glucan on
the inhibition of phagocytosis of glucan
particles by macrophages is shown in
Figure 3. Pretreatment with the soluble
glucan was significantly inhibitory even
at 10 Ixg mL-1, and pretreatment with 50
~g mL-1 soluble glucan resulted in a re-
duction in phagocytosis of more than
50%.
Soluble Yeast Glucan Does Not
Inhibit Phagocytosis of Particles by
Salmon Macrophages in General
To examine if soluble yeast glucan af-
fected phagocytosis in general or inter-
fered with complement receptors, two
sets of experiments were performed.
First, we asked whether pretreatment of
macrophages with soluble yeast glucan
• | • ! - • i • | •
'~o' "5bd ' ' ~ooo
324 R.E. Engstad and B, Robertsen

Table 1. Effect of Different Soluble Polysaccharldes on Phagocytosis of Glucan Particles by

Atlantic Salmon Macrophages.

Macrophages With Phagocytic Index

Composition >1 Glucan (Glucan Particles
Particle (%) per Mqb)
Poly- Mono-
saccharide saccharide Linkages (0.2 mg mL 1) (1.0 mg mL 1) (0.2 mg mL 1) (1.0 mg m L ~)

None -- -- 49.8 2.72

Glycogen Glucose (x-1,4 46.8 46.9 2.70 2.58
Dextran Glucose (x-1,6 41.0" 44.4" 2.50 2.38
Mannan Mannose a-1,6 40.4* 39.4* 2.23 2.07*
Pustulan Glucose 13-1,6 42.1 * 45.0* 2.14* 2.38
Laminarin Glucose 13-1,3 32.5t 22.51 1.671 1.091
Soluble Glucose 13-1,3;
glucan 13-1,6 19.61" 4.2t 0.80t 0.191"

The macrophages were pretreated for 10 min with two concentrations (0.2 and 1.0 mg m L 1) of soluble polysac-
charides before the addition of glucan particles. Each value represents mean uptake in triplicate wells with
macrophages isolated from one fish; 200-400 macrophages were examined in each welt.
*? For significant difference from untreated macrophages, *p < 0.05, IP < 0.01.

had an effect on the phagocytosis of op-
sonized glucan particles. As seen in Fig-
ure 3, soluble glucan had no effect on this
process.
In the second experiment we deter-
mined whether soluble glucan affected
the uptake of opsonized and nonop-
sonized SRBC. As seen in Table 2, both
opsonized and untreated glutaraldehyde
fixed SRBC were p h a g o c y t i z e d by
salmon macrophages. The uptake of op-
sonized SRBC was higher than of un-
treated SRBC, similar to what was ob-
served with glucan particles. Soluble
yeast glucan, however, did not inhibit
the uptake of either untreated or op-
sonized SRBC. Uptake of both native
and opsonized glucan particles and the
i n h i b i t o r y effect of soluble glucan
showed a similar pattern in this experi-
ment, as in the previous experiments.
Nature of Glucan Opsonins in
Atlantic Salmon Serum
In order to study the nature of serum
proteins involved in the opsonization of
glucan particles, salmon serum was heat
inactivated or treated with chelators that
bind Mg 2÷ and/or Ca 2+ before incuba-
tion with the glucan particles. Since the
macrophages isolated for this experi-
ment were much more active in phago-
cytosis than macrophages isolated ear-
lier, the preincubation time with soluble
glucan was reduced from 10 to 5 min and
the time of phagocytosis was reduced
from 30 to 12 min.
As seen in Figure 4, chelation of both
Ca 2+ and Mg 2+ in serum (treatment with
EDTA) during opsonization resulted in a
marked reduction in uptake of opsonized
glucan particles by the macrophages.
This experiment also demonstrated that
serum that was heat treated to abolish
complement activity had a diminished
opsonizing effect on the glucan particles.
Glucan particles incubated with EDTA-
treated serum or heat inactivated serum
were, however, still phagocytized at a
higher rate than native particles. This in-
dicates that serum proteins other than
complement might be able to opsonize
and thereby facilitate phagocytosis of
glucan particles by salmon macrophages.
The lack of Ca 2+ ions (treatment with
Mg/EGTA) during o p s o n i z a t i o n ap-
peared to have no effect on the phagocy-
tosis of glucan particles compared with
the uptake of particles opsonized with
nontreated serum, indicating that the
glucan particles activate the alternative
complement pathway (Fig. 5).
Recognition of yeast I~-glucan 325

*~ 60
A
50-
; •
~ 40-
A

30-
?,
~o 20-
,=;

o 10-
E
0 i i i i i

0 50 100 150 200

A 2,5 "

e~ 2,0"

~ 1,5"
&
x

~ 1,0"
~J

g 0,5
r~

0,0 i i i i i

0 50 100 150 200

Soluble glucan concentration (I-tgm 11)
Figure 3. Inhibition of phagocytosis of opsonized (-O-), and nonopsonized (-m-) glucan particles
by macrophages preincubated with different concentrations of soluble yeast glucan. (A) Mean
percent macrophages with >1 glucan particle +- SD; (B) mean phagocytic index -+ SD, n = 3.

To study whether glucan structures Discussion

contributed to the uptake of the different
opsonized glucan particles, macrophages
were pretreated with soluble yeast glu-
can prior to addition of glucan particles
(Figs. 4 and 5, hatched bars). As in the
experiments described above, phagocy-
tosis of particles opsonized with normal
or Mg/EGTA treated serum was not in-
hibited by soluble glucan. On the other
hand, soluble yeast glucan caused a
small but significant (p < 0.05) reduction
in phagocytosis of glucan particles op-
sonized with EDTA-treated serum and a
marked inhibition of phagocytosis of glu-
can particles opsonized with heat inacti-
vated serum.
Fungal cell wall 13-glucans appear to
be important for the recognition of
yeasts and fungi by vertebrate macro-
phages. 13-glucans also have the ability to
stimulate the nonspecific defense of ver-
tebrates and to potentiate the effect of
vaccines (3-7,22). Mononuclear phago-
cytes are believed to be the main target
cell for the immunomodulatory effects of
13-glucans (8-10). In mammals 13-1,3 glu-
cans are known to be potent activators of
the alternative complement pathway,
and fungi, like other microorganisms,
may be phagocytized through comple-
ment receptors after being opsonized by
326 R. E. Engstad and B. Robertsen

Table 2. Effect of Soluble Yeast Glucan on the glucans, that is, laminarin and solubi-
Phagocytosls of Opsonlzed and
Nonopsonlzed Glutaraldehyde Fixed SRBC
lized yeast glucan particles. Polyglu-
and Glucan Particles by Atlantic cases with different glycosidic linkages
Salmon Macrophages. such as glycogen (a- 1,4-linked), dextran
Macrophages With
(a-l ,6-linked), and pustulan (p-1,6-
>2 Particles (%) linked), had no or very little inhibitory
effect on uptake of glucan particles by
MQ, Given
Untreated Soluble
salmon macrophages. Neither was the
Particle Treatment M@ Glucan uptake markedly inhibited by pretreat-
ment with mannan, a polymannose from
Glucan None 32 ? 1.1 17 k 2.0
Glucan Opsonized 50 2 3.0 49 k 5.5
S. cerevisiae cell walls. The fact that sol-
SRBC None 29 2 13.3 27 k 13.9 uble yeast glucan did not inhibit the up-
SRBC Opsonized 39 * 8.1 44 2 3.2 take of glucan particles opsonized with
The soluble glucan (200 kg mL_‘) was added to the
normal salmon serum, or the uptake of
macrophages IO min prior to the addition of SRBC or opsonized and nonopsonized SRBC,
glucan particles. Each value represent mean uptake in
showed that the soluble yeast glucan did
triplicate wells with macrophages isolated from one
fish; 200-400 macrophages were examined in each not bind to complement receptors or in-
well.
terfere with phagocytosis in general.
This experiment thus strongly suggests
the interaction of p-1,3 glucan structures that phagocytosis of glucan particles by
and complement (1). Recent work also salmon macrophages is mediated
indicates that yeast cells may be phago- through a specific receptor similar to the
cytized by macrophages through specific one described for human monocytes.
receptors for p-1,3 glucan. Czop and The fact that pustulan had such a small
Austen (14) have shown that yeast p-glu- inhibitory effect on phagocytosis of
can particles are phagocytized by human yeast glucan particles compared to lam-
monocytes through a ligand inhibitable inarin indicates that p-1,3-linked gluco-
receptor and have isolated a receptor syl chains are more important in the rec-
from human monocytes that appears to ognition of yeast glucan by salmon mac-
be specific for yeast P-glucan (17). Oth- rophages than p-1,6-linked glucosyl
ers have obtained results that indicate chains. Pustulan was in fact not more in-
that mouse macrophages have a specific hibitory than mannan. The observation
receptor for P-glucan (15,16). that soluble yeast glucan inhibited
In the present work we have shown phagocytosis of glucan particles more ef-
that similar recognition mechanisms for fectively than laminarin indicates that
P-glucan exist in Atlantic salmon macro- there are oligosaccharides present in the
phages by studying phagocytosis of na- former mixture that are either more ef-
tive and opsonized glucan particles ob- fective because they are smaller than
tained from S. cerevisiae cell walls. The laminarin or they possess branched
glucan particles are composed of 95% structures that are better ligands than
glucose linked through p-1,3- and p-1,6- laminarin. Glucose or mannose did not
linkages with p-1,3,6-linked branch inhibit phagocytosis, so the putative glu-
points. Salmon macrophages rapidly can receptor must recognize oligosac-
phagocytized both native and opsonized charide chains and not only terminal
glucan particles. Opsonized particles sugar residues. Janusz et al. (23) recently
were, however, taken up at a higher rate isolated a heptaglucoside from yeast glu-
than native particles. can that showed potent inhibitory activ-
The uptake of native glucan particles ity of phagocytosis of yeast glucan par-
was strongly inhibited by pretreatment ticles by human monocytes. Because
of the macrophages with soluble p-1,3- smaller oligosaccharides derived from
Recognition of yeast 13-glucan 327

loo
A [] Unlreated Me,
[] /V~ pretncubated with soluble glucan
80

A
60

~, 40
OO

20

o
None Salmon Serum Heat
seFuIn with inactivated
EDTA serum

5,0"
T
4,0
"d

3,0 -r
X

.~

o
2,0

1,o

o,o
FL
None Salmon Serum Heat
serum with inactivated
EDTA seruin

Treatment of glucan particles

Figure 4. Effect of EDTA and heat treatment on the ability of salmon serum to opsonize glucan
particles. The bars show phagocytosis by salmon macrophages of nonopsonized glucan particles;
particles opsonized with normal salmon serum; particles opsonized with EDTA-treated serum; or
particles opsonized with heat inactivated serum. Macrophages were either untreated (open bars) or
preincubated (5 min) with 500 p,g mL -1 soluble yeast glucan (hatched bars) before the addition of
glucan particles. Conditions in (A) and (B) as in Figure 3.

yeast glucan showed much lower inhibi-
tion of phagocytosis, they suggested that
the monocytes are able to recognize hep-
tasaccharide structures in yeast glucan
(23). The structure of the heptasaccha-
ride and thus the precise specificity of
the r e c e p t o r is, however, presently
unknown.
The small, but significant (p < 0.05),
inhibition of glucan uptake by dextran
and mannan could be due to steric hin-
drance because of binding of these
polysaccharides to the macrophage sur-
face via other receptors or nonspecific
forces.
Glucan particles treated with normal
serum appeared to be opsonized mainly
through the alternative pathway, because
EDTA and heat treatment (45°C, 30 min)
caused a marked reduction in the opso-
nizing ability of salmon serum whereas
treatment of serum with Mg/EGTA did
not affect is ability to opsonize glucan
particles.
328 R.E. Engstad and B. Robertsen

70
A [] Untxeated M ~
$:h 60 [] M ~ p r e i n c u b a t e d w i t h soluble g l u c a n

50
%
v~
A 40
"-r-
30

I
0/)
20
t:h
O
b 10

0
None Salmon Serum with
serum Mg/EGTA

3,0"

T Y
2,0"
x
Y

I
•~ 1,0"
O

0,0
None Salmon Serum with
serum Mg/EGTA

Treatment of glucan particles

Figure 5. Effect of Mg/EGTA on the ability of salmon serum to opsonize glucan particles. The bars
show phagocytosis by salmon macrophages of nonopsonized glucan particles; particles opsonized
with normal salmon serum; or particles opsonized with Mg/EGTA-treated serum. Macrophages
were either untreated (open bars) or preincubated (5 min) with 200 i~g mL 1 soluble glucan
(hatched bars) before the addition of glucan particles. Conditions in (A) and (B) as in Figure 3.

The fact that both EDTA-treated and
heat inactivated serum still possessed a
significant ability to opsonize glucan par-
ticles indicates that salmon serum also
contains other opsonins beside comple-
ment. Konopski et al. (15), working with
mouse macrophages, reported that se-
rum fibronectin has the ability to opso-
nize glucan particles and that fibronectin
receptors are directly involved in phago-
cytosis of glucan particles. As fibronec-
tin is also an important component in se-
rum of salmonids (24), fibronectin might
have been involved in the opsonization
of glucan particles by heat inactivated
and EDTA-treated salmon serum. One
can, however, not exclude the possibility
that serum levels of spontaneous formed
complement protein C3b, caused by C3
"tickover," might be sufficient to opso-
nize glucan particles and thus increase
the rate of phagocytosis of particles in-
cubated with serum treated with EDTA
or heat inactivated. At present, too little
is known about the nature of the possible
opsonins in serum treated as described,
to draw any conclusions from this exper-
iment. Phagocytosis of glucan particles
opsonized with heat inactivated serum
and EDTA-treated serum was partially
inhibited by soluble yeast glucan that in-
dicates that the [3-1,3 glucan receptor
Recognition of yeast 13-glucan
c o n t r i b u t e s to the uptake of t h e s e
particles.
In conclusion, the present investiga-
tion demonstrates that macrophages of
teleost fish, like mononuclear phago-
cytes from mammals, are able to recog-
nize 13-1,3-glucan structures in yeast cell
walls. The glucan is recognized either di-
rectly through a specific receptor or in-
directly because it binds complement or
other serum opsonins.
Fungal 13-1,3-glucans have also been
shown to enhance the capacity of verte-
brates including fish to resist attacks
from microbial pathogens (3,4,22). Sev-
eral reports indicate that the glucan-
induced disease resistance of vertebrates
is mediated via the monocyte/macro-
References
1. Czop, J. K.; Austen, K. E Properties of gly-
cans that activate the human alternative com-
plement pathway and interact with the human
monocyte 13-glucan receptor. J. Immunol. 135:
3388-3393; 1985.
2. Ross, G. D.; Cain, J. A.; Myones, B. L.; New-
mann, S. L.; Lachmann, P. J. Specificity of
membrane complement receptor type three
(CR3) for 13-glucans. Complement 4:61-74;
1987.
3. Di Luzio, N. R. Update on the immunomodu-
lating activities of glucans. Springer Semin.
Immunopathol. 8:387-400; 1985.
4. Robertsen, B.; RCrstad, G.; Engstad, R. E.;
Raa, J. Enhancement of nonspecific disease re-
sistance in Atlantic salmon, Salmo salar L., by
a glucan from Saccharomyces cerevisiae cell
walls. J. Fish Dis. 13:391-400; 1990.
5. Engstad, R. E.; Robertsen, B.; Frivold, E.
Yeast glucan induces increase in lysozyme and
complement-mediated activity in Atlantic
salmon blood. Fish Shellfish Immunol. 2:287-
297; 1992.
6. Yano, T.; Matsuyama, H.; Mangindaan,
R. E. P. Polysaccharide-induced protection of
carp, Cyprinus carpio L., against bacterial in-
fection. J. Fish Dis. 14:577-582; 1991.
7. RCrstad, G.; Aasjord, E M.; Robertsen, B.
Adjuvant effect of a yeast glucan in vaccines
against furunculosis in Atlantic salmon (Salmo
salar L.). Fish Shellfish Immunol. (in press).
8. Di Luzio, N. R. Lysozyme, glucan-activated
macrophages and neoplasia. J. Reticuloendo-
thel. Soc. 26:67-81; 1979.
9. Seljelid, R.; BOgwald, J.; Lundwall, ,~. Glycan
stimulation of macrophages in vitro. Exp. Cell.
Res. 131:121-129; 1981.
10. BOgwald, J.; Johnson, E.; Seljelid, R. The cy-
totoxic effect of mouse macrophages stimu-
329
phage cell lineage (8-10). Macrophages
may be the target cells of glucan action
also in Atlantic salmon (5,13). Fungal
glucans are likely to interact with mac-
rophages either in their native state or
after being opsonized by serum compo-
nents. Whether the immunostimulatory
e f f e c t s of 13-glucans are m e d i a t e d
through interaction with a specific glu-
can receptor or with complement recep-
tors is still a matter of investigation.
Acknowledgements---The authors would like
to thank Ms. Marit H. J o h a n s e n for her tech-
nical assistance. This w o r k was supported by
grants from the Royal N o r w e g i a n Council for
Scientific & Industrial R e s e a r c h and f r o m
K/S B i o t e c - M a c k z y m a l , Troms¢, Norway.
lated in vitro by a 13-1,3-o-glucan from yeast
cell walls. Scand. J. Immunol. 15:297-304;
1982.
11. Czop, J. K.; Austen, K. F.; Generation of leu-
kotrienes by human monocytes upon stimula-
tion of their 13-glucan receptor during phagocy-
tosis. Proc. Natl. Acad. Sci. 82:2751-2755;
1985.
12. Doita, M.; Rasmussen, L.-T.; Seljelid, R.; Lip-
sky, P. Effect of soluble aminated ~-I,3-D-
polyglucose on human monocytes: Stimulation
of cytokine and prostaglandin E2 production
but not antigen-presenting function. J. Leuko.
Biol. 49:342-351; 1991.
13. JCrgensen, J. B.; Sharp, G. J. E.; Secombes,
C. J.; Robertsen, B. Effect of yeast cell wall
glucan on the bactericidal activity of rainbow
trout macrophages. Fish Shellfish Immunol.
(in press).
14. Czop, J. K.; Austen, K. F. A 13-glucan-
inhibitable receptor on human monocytes: Its
identity with the phagocytic receptor for par-
ticulate activators of the alternative comple-
ment pathway. J. Immunol. 134:2588-2593;
1985.
15. Konopski, Z.; Rasmussen, L.-T.; Seljelid, R.;
Eskeland, T. Phagocytosis of 13-1,3-D-glucan-
derivatized microbeads by mouse peritoneal
macrophages involves three different recep-
tors. Scand. J. Immunol. 33:297-306; 1991.
16. Goldman, R. Characteristics of the 13-glucan
receptor of murine macrophages. Exp. Cell
Res. 174:481-490; 1988.
17. Czop, J. K.; Kay, J. Isolation and character-
ization of 13-glucan receptors on human mono-
nuclear phagocytes. J. Exp. Med. 173:1511-
1520; 1991.
18. Hakamori, S. A rapid permethylation of gly-
colipid and polysaccharide catalyzed by meth-
330
ylsulfinyl carbanion in dimethyl sulfoxide. J.
Biochem. 55:205-208; 1964.
19. Albersheim, P.; Nevins, D. J.; English, P. D.;
Karr, A. A method for the analysis of sugars in
plant cell-wall polysaccharides by gas-liquid
chromatography. Carbohydr. Res. 5:340-345;
1967.
20. Braun-Nesje, R.; Bertheussen, K.; Kaplan,
G.; Seljelid, R. Salmonid macrophages: Sepa-
ration, in vitro culture and characterization. J.
Fish. Dis. 4:141-151; 1981.
21. Mayer, M. M. Complement and complement
fixation. In: Kabat, E. A., ed., Experimental
Immunochemistry. 2nd ed. Springfield, IL:
Charles C. Thomas; 1971:134-240.
R.E. Engstad and B. Robertsen
22. Seljelid, R.; Rasmussen, L.-T.; Larm, O.;
Hoffman, J. The protective effect of [3-1,3D-
glucan-derivatized plastic beads against Es-
cherichia coli infection in mice. Scand. J. Im-
munol. 25:55-60; 1987.
23. Janusz, M. J.; Austen, K. E; Czop, J. K. Iso-
lation of a yeast heptaglucoside that inhibits
monocyte phagocytosis of zymosan particles.
J. lmmunol. 142:959-965; 1989.
24. Lee, L. E. J.; Bols, N. C. Identification of fi-
bronectin in rainbow trout and influence of
cortisol on fibronectin accumulation by rain-
bow trout cell cultures. J. Aquat. Anita. Health
3:31-40; 1991.