INFECTION AND IMMUNITY, Apr. 1992, p. 1642-1647
Copyright C 1992, American Society for Microbiology
Anti-Infective Effect of Poly-13-6-Glucotriosyl-
13-3-GlucopyranoseGlucan In Vivo
ANDREW B. ONDERDONK,1* RONALD L. CISNEROS,1 PAUL HINKSON,' AND GARY OSTROFF2
Department ofPathology, Channing Laboratory, Brigham and Women's Hospital, Harvard Medical School, 180
Longwood Avenue, Boston, Massachusetts02115,1andAlphaBetaTechnology, Worcester, Massachusetts016052
Received 12 November 1991/Accepted 22 January 1992
Mice challenged with Escherichia coli or Staphylococcus aureus were protected against lethal peritonitis by
the intravenous administration of 10 ,g ofpoly-131-6-glucotriosyl-j31-3-glucopyranose (PGG) glucan per
animal 4 to 6 h prior to bacterial challenge. Subsequent studies with the rat model for intra-abdominal sepsis
indicated that intramuscular doses of 10 to 100 pg per animal 24 and 4 h prior to surgical implantation of the
bacterial inoculum reduced the early mortality associated with the peritonitis phase of this experimental disease
process. Quantitative cultures ofblood obtained from challenged rats showed that significantly fewer organisms
were present in the blood of PGG glucan-treated animals than in that of untreated animals. Quantitative
studies of leukocytes of rats and mice following a single injection of PGG glucan showed a modest transient
increase in the total leukocyte count. The possible mechanisms by which protection occurs in the animal model
system are discussed.
The ability of yeast cell wall constituents to nonspecifi-
cally activate certain components of the host immune re-
sponse has been known since the early 1960s, when a crude
insoluble yeast cell wall extract, zymosan, was shown to be
a stimulant of the reticuloendothelial system (23). Subse-
quent studies identified the active moiety of the yeast cell
wall as ,-glucan (1, 21). These insoluble glucose polymers
alone or as vaccine adjuvants for viral and bacterial antigens
markedly increase resistance to a variety of bacterial, my-
cotic, and viral infections (14, 22, 28). Studies have shown
that0-glucantherapy increases the plasma and splenic levels
of interleukins 1 and 2 in rats, an observation that, in part,
may explain the broad-based anti-infective activity of these
A recent finding by Czop and Austen (6) was the specific
receptor binding site for
3-glucan on the membrane of
human neutrophils and macrophages. It has been shown that
host defense responses related to theP-glucansinclude the
activation of the alternative complement pathway, the re-
lease of lysosomal enzymes by monocytes, and the genera-
tion of leukotrienes by monocytes (7, 9, 13).
Despite the exciting array of biologic activities that appear
to be activated by the ,-glucans, the fact that the material
used in earlier studies was an insoluble extract of yeast cells
rendered it less than desirable for clinical use, except as a
vaccine adjuvant. Recent advances in the understanding of
the composition and structure of the ,3-glucans have dramat-
ically altered the possibilities for use of these materials as
prophylactic and/or therapeutic immunomodulators. One of
these 1-glucans, poly-,B1-6-glucotriosyl-i1-3-glucopyranose
(PGG) glucan, obtained from genetically engineered yeast
cells has been chemically well characterized and is produced
as an unmodified polysaccharide (Betafectin; Alpha Beta
Technology, Worcester, Mass.) (12, 19). The basic structure
of this material is a ,B-D-(1-3)-linked glucopyranosyl back-
bone with periodic ,B-D-(1-6)-linked side chains (3, 24). PGG
glucan is highly branched and has been shown to have a
much higher affinity for the3-glucan receptor of the human
monocyte than do other less-branched materials (2, 15, 16,
We present the results of in vivo mouse and rat studies
designed to determine whether PGG glucan can abrogate the
lethal process that occurs during acute intra-abdominal
MATERIALS AND METHODS
Animals. Male CD-1 virus antibody-free (VAF) mice
(Charles River Laboratories, Wilmington, Mass.) weighing
20 to 24 g each were used for basic characterization studies.
Male VAF Wistar rats (Charles River Laboratories) weigh-
ing 180 to 200 g were used for studies employing the animal
model for intra-abdominal sepsis. All animals were housed,
five per cage, within a VAF facility and were given food and
water ad libitum. Animal experiments were carried out
according to the guidelines of the Harvard Medical Area
Animal Use Committee.
Bacterial strains. Staphylococcus aureus (ATCC 29213)
and Eschenchia coli (TVDL 41, originally derived from rat
cecal contents) were used as the challenge strains for studies
with mice. Strains were grown in brain heart infusion broth
supplemented with hemin and menidione (BHIS; Adams
Scientific, Warwick, R.I.) for 24 h at 37°C and washed three
times in phosphate-buffered saline (PBS, pH 7.0); aliquots
were frozen at -80°C until used. Replicates of the frozen
stock cultures were thawed at room temperature, serial
decimal dilutions were made, and 0.1-ml samples of each
dilution were plated onto Trypticase soy blood agar to
determine the viable-cell density for the inoculum. All
counts were reported as the log1o CFU per milliliter.
Cecal content inoculum. The inoculum used for rats as part
of the model for intra-abdominal sepsis was obtained by
removing the cecal contents from rats maintained on a diet of
lean ground beef for a period of 2 weeks and diluting it with
an equal volume of peptone-yeast-glucose broth and 10%
(wt/vol) barium sulfate, as described in previous publications
(18, 27). The bacteriologic characterization of this inoculum
is that it is polymicrobic and includes an array of facultative
Vol. 60, No. 4
ANTI-INFECTIVE EFFECT OF PGG GLUCAN IN VIVO
TABLE 1. Composition of inoculum for intra-abdominal sepsis
model in rats
Escherichia coli ............................
Streptococcus sp. group D .......................................
Streptococcus sp. non-group-A, -B, or -D...................
Streptococcus avium ............................................
Bacteroides fragilis ............................................
Clostridium ramosum ............................................
Peptostreptococcus magnus ............................
Peptostreptococcus productus .....................
Unidentified gram-negative facultative anaerobe ..........
species, such as E. coli, as well as obligate anaerobes (Table
Quantitative cultures. Quantitative cultures of 0.1-ml sam-
ples ofblood and peritoneal fluid were performed. Blood was
obtained by percutaneous transthoracic cardiac puncture;
peritoneal fluid was drawn through an anterior midline
laparotomy incision from animals anesthetized with 0.25 ml
of pentobarbital sodium (Nembutal; 50 mg/ml). Samples of
blood were added directly to 20 ml of molten BHIS agar,
mixed, and poured into sterile plastic petri plates. Once the
agar had solidified, the plates were incubated at 37°C under
anaerobic conditions for 24 to 48 h, and colonies were
enumerated. Peritoneal fluid samples were diluted in sterile
PBS to yield decimal dilutions from 10-2 to 10-7, and
aliquots were plated onto brucella-base blood plates. Fol-
lowing incubation, colonies were enumerated.
Bacterial challenge of mice. Mice were challenged with an
approximately 70% lethal dose (LD70) of either S. aureus or
E. coli as determined by preliminary studies (107-7 CFU ofE.
coli per animal and 107.65 CFU of S. aureus per animal).
Challenge was performed by injection of 0.1 ml of bacterial
suspension into the peritoneal cavity after dilution of the
stock suspension to the appropriate concentration with
sterile PBS. At the time of challenge, an aliquot of the
challenge inoculum was used to verify the viable-cell density
of the stock bacterial suspension. Animals were observed
four times per day for the first 48 h and twice per day
thereafter. Obviously moribund animals were humanely
sacrificed with CO2.
Animal model for intra-abdominal sepsis. The rat model for
intra-abdominal sepsis has been well described in the scien-
tific literature (18, 27). Briefly, a 0.5-ml aliquot of the thawed
cecal content inoculum was placed into a gelatin capsule,
and the capsule was surgically implanted into the peritoneal
cavity of an anesthetized rat through an anterior midline
incision. The incision was closed with 3-0 silk suture, and
then the animals were returned to their cages. Animals were
observed four times per day for the first 48 h and twice per
day thereafter. Moribund animals were humanely sacrificed
Total and differential cell counts. For some experiments,
peripheral blood samples were obtained by percutaneous,
transthoracic cardiac puncture for assessment of total and
differential counts of leukocytes (WBC). A 0.1-ml aliquot of
blood was obtained after animals were lightly anesthetized
with ether. The heparinized sample was placed into standard
saline diluent, and the erythrocytes were lysed for total
WBC determinations, using a Coulter model FN Counter
(Coulter Electronics, Inc., Hialeah, Fla.). Smears of each
sample were stained with Wright Giemsa stain, and the
percentages ofgranulocytes, lymphocytes, and mononuclear
cells were determined. The percentage of each cell type was
then used to calculate the absolute cell counts by multiplying
the percentage by the total cell count for each sample.
PGG glucan administration. PGG glucan (Betafectin) was
used for all experiments. Stock PGG glucan, supplied as a
soluble material at a concentration of 1.5 to 10.0 mg/ml, was
prepared according to "Good Manufacturing Practices"
(GMP) standards. Each lot used in animals was assayed by
the limulus amoebocyte lysate (Associates of Cape Cod,
Woods Hole, Mass.) method for the presence of endotoxin.
Endotoxin levels were below the detectable limits of the
assay (<0.03 endotoxin units per ml) in all materials used in
these studies. The stock material was diluted in sterile,
pyrogen-free saline solution for injection into rats and mice.
Preliminary studies of anti-infective effects in mice indicated
that the intravenous (i.v.) administration of PGG glucan by
the percutaneous, transthoracic cardiac puncture method 4
to 6 h prior to challenge yielded protective results. Subse-
quent studies with rats indicated that two doses of 10 to 100
,ug of PGG glucan per animal given intramuscularly (i.m.) at
24 and 4 h prior to challenge resulted in protection against
lethal bacterial challenge. Except when stated otherwise,
these routes and times of administration were used for all
Experimental design. Mice were used to evaluate the
possible protective effect of PGG glucan in a peritoneal
infection test system using S. aureus and E. coli at lethal
concentrations. Neither animal test system was designed to
simulate human disease but rather provided an assay system
in which dose, response, time of administration, and hema-
tologic changes could be evaluated before the more defini-
tive intra-abdominal sepsis system was employed.
The rat model for intra-abdominal sepsis is a well-docu-
mented animal model that simulates an infectious process
that occurs in humans following accidental peritoneal soilage
with intestinal contents (18, 27). This infection is a two-stage
process in which an early peritonitis associated with a high
mortality and the presence of facultative gram-negative
organisms within the blood and peritoneal cavity is followed
by a more chronic process involving the formation of intra-
abdominal abscesses, during which anaerobes have been
shown to be important constituents (27). On the basis of
preliminary dose and timing studies in mice, this model was
TABLE 2. Effects of various doses of PGG on E. coli-induced
mortality in mice
P value vs
aAll doses were given by percutaneous transthoracic cardiac puncture4 h
prior to challenge.
bSaline-treated control animals; chi-square analysis using Yate's correc-
VOL. 60, 1992
ONDERDONK ET AL.
TABLE 3. Effects of various doses of PGG on S. aureus-induced
mortality in mice
P value vs
aAll doses were given by percutaneous transthoracic cardiac puncture 4 h
prior to challenge.
b Saline-treated control animals; chi-square analysis, Fisher exact test.
used to further characterize dose-response information,
route and time of administration, and peripheral cell re-
sponse following PGG glucan therapy.
Statistical evaluation. Comparison of groups with regard to
mortality was made by chi-square analysis; comparisons of
quantitative data were made with Student's t test, as sup-
plied on commercially available statistical software (Statis-
tix, NH Analytical Software, Ann Arbor, Mich.).
Prevention ofE. coli-induced mortality in mice. Preliminary
experiments with an insoluble PGG glucan (Adjuvax; Alpha
Beta Technology) at doses as low as 100 ,ug per mouse
suggested that mortality in mice challenged intraperitoneally
with E. coli could be abrogated by pretreatment with PGG
glucan. In order to determine whether a soluble PGG glucan
preparation (Betafectin) was also capable of this protective
effect, groups of mice were challenged with E. coli following
percutaneous, transthoracic administration of soluble PGG
glucan at doses ranging from 0.5 to 1,000 ,ug per animal
(Table 2). As can be seen, there is a dose-response relation-
ship between the amount of soluble PGG glucan adminis-
tered to these animals and the degree of protection afforded.
There appears to be an active range for this protective effect
(17), with doses either above or below this range not
providing any protection in this animal test system. Depend-
ing on the route and time of administration, the active range
is between 100 and 0.5 ,ug per animal.
Prevention ofS. aureus-induced mortality in mice. In order
to determine whether the protection against E. coli-induced
mortality was specific only for facultative gram-negative
rods, groups of mice were challenged with an =LD70 of S.
aureus following treatment with soluble PGG glucan at doses
ranging from 1 to 100 ,ug per animal (Table 3). Protection
against the lethal challenge was provided by a dose of 1 ,ug
per animal in this test system but not by higher doses of
soluble PGG glucan.
TABLE 5. Effects of PGG on mortality in a rat model for intra-
P value vs
aTwo doses of 100 ,ug of PGG per animal given i.m. at 24 and 4 h prior to
Effect of soluble PGG glucan on peripheral WBC counts in
mice. Further characterization of the in vivo response of
mice following administration of PGG glucan was achieved
by obtaining peripheral blood samples from animals follow-
ing a single dose of 10 ,ug of PGG glucan per animal (Table
4). Total WBC counts appear to peak between 24 and 48 h
after a single i.v. dose of PGG glucan. The percentage
increase in the numbers of monocytes and granulocytes is
greater than anticipated from the increase in total counts
relative to those for saline-treated animals. Total and differ-
ential counts returned to approximately pretreatment levels
within 72 h of a single dose of PGG glucan (17).
Effect of soluble PGG glucan on mortality in the rat model
for intra-abdominal sepsis. The preliminary mouse studies
provided important dose and route of administration infor-
mation for subsequent experiments in an animal model
system simulating intra-abdominal sepsis. Because the use
of percutaneous, transthoracic cardiac puncture for i.v.
administration is known to increase mortality during this
experimental infection, total and differential WBC counts
following i.m. and i.v. injection were compared to determine
whether i.m. injection was an alternative method for admin-
istration of soluble PGG glucan. No differences between
these two methods of administration were noted for total and
differential cell counts (17); thus, administration of soluble
PGG glucan for all experiments involving rats was by the
The results of representative experiments with similar
PGG glucan preparations and the same bacterial inoculum
are shown in Table 5. As can be seen, the administration of
two doses of PGG glucan at 24 and 4 h prior to challenge
significantly reduces the mortality in this model system
compared with that for untreated controls. Because the
mortality during this experimental infection is known to be
due to facultative gram-negative rods and is associated with
relatively large numbers of organisms in the blood and
peritoneal cavity, quantitative bacteriologic studies were
performed to determine whether there were any differences
between soluble PGG glucan-treated animals and untreated
TABLE 4. Effects of PGG treatment on peripheral WBC counts in mice
24 h after treatment
48 h after treatment
% Increase vs saline
8,129 + 1,294b
13,220 + 60d
1,430 ± 233
2,919 + 284d
604 ± 117
2,381 + 577d
6,310 + 1,926
10,990 + 1,013"
1,522 + 83
2,979 ± 414d
389 + 120
2,010 + 266
aAbsolute cell counts.
bMeans + standard errors; three or more animals sampled per group.
c Ten micrograms.
dp < 0.05; two-sample t test versus saline controls.
ANTI-INFECTIVE EFFECT OF PGG GLUCAN IN VIVO1645
TABLE 6. Comparison of quantitative peritoneal fluid cultures
in a rat model for intra-abdominal sepsis
Time (h) post-
4.83 ± 0.75
5.66 ± 0.09
4.85 ± 2.08
2.83 ± 0.42
531 ± 0.16
4.79 ± 0.68
2.46 ± 0.76c
1.16 ± 0.16c
aTwo to five samples per group.
bTwo doses of 100 p,g per animal given i.m. at 24 and 4 h prior to challenge.
c P< 0.05, two-sample t-test.
FIG. 1. Comparison of quantitative blood cultures 4
rats following development of intra-abdominal sepsis.
100 ,ug per animal were given i.m. at 24 and 4 h prior
Counts are expressed as the meanlog1oCFU per millili
of two to nine samples per time point. A large in
animals was used for blood culture data because of de;
expected to occur over the course of the experime
animals were sacrificed humanely and were not used
blood culture data. Comparison of counts at 24 h
yielded a P value of <0.05 for PGG-treated animals
with untreated control animals (U), using a two-samp
controls. The results of quantitative blood culti
indicate that there is an approximate 20-fold diffc
numbers of organisms per milliliter, with the s
glucan-treated animals having significantly fewe
present at 24 h, a time at which most deaths c
model system. A continued decrease in the
organisms present occurs in the surviving recif
cecal challenge over the next several days, witl
PGG glucan-treated animals appearing to clea
from the blood more quickly than untreated an]
tional studies have yielded similar quantitati
treatment and control groups.
Quantitative peritoneal cultures. The numbei
isms present in free-flowing peritoneal exudatc
mals treated with soluble PGG glucan and fro
animals were determined at various times afte
Within 96 h, the number of organisms measured
exudates from treated animals was significantly I
CFU/ml) than that noted for untreated animals
ml). Within 5 days of challenge, discreet intr
abscesses had formed in most animals and per
dates had decreased in quantity (Table 6).
An assessment of the total peripheral WBC c
ing administration of soluble PGG glucan yie
similar to those for mice, with total WBC couni
almost 100% within 5 days in treated animals co]
those in untreated animals (Table 7).
Previous studies have documented the nonsp
nomodulatory properties of insoluble polysacch
as zymosan (4, 14, 22, 23, 28), in animal test s
earlier work with yeast glucans in animals involved a variety
of uncharacterized, insoluble materials. Although it was
clearly shown that such materials served as excellent adju-
vants and as immunostimulants in their own right, the
methods for administering such materials limited their use-
fulness to the laboratory.
The observation that yeast glucans could abrogate the
effects of a variety of microbial pathogens
prompted considerable interest indetermining exactlyhow
such compounds interact with the host immune system. The
finding of a specific cell surface receptor forP-1,3glucans on
macrophages and granulocytes suggested that the protective
effects for the glucans are provided by such cells (5, 6, 10,
26), possibly by cytokine production and release, improved
phagocytic killing, or increased production of phagocytic
cells. Because PGG glucans bind to cells with an array of
mechanism(s) of protection provoked by such materials are
Additional research has found soluble glucan polymers
that appear to act in much the same manner as their insoluble
parent compounds (9) and that might be used either thera-
peuticallyorprophylacticallyin humans at risk for microbial
disease. Our research with mice was aimed at two essential
tasks: (i) to find an in vivo test system to determine whether
various preparations of soluble PGG glucans were biologi-
cally active inpreventinglethal bacterial infections and(ii)to
explore the basic host response following administration of
soluble preparations by a variety of routes, using various
doses and times of administration. Initialexperimentsin this
laboratoryrevealed theprotective effects of both soluble and
insoluble parent materials in mice challenged with E. coli,
effects that extend togram-positive organismson the basis of
our experience with S. aureus. It should be noted that the
protective effect was not related to endotoxin-induced toler-
ance, since none of the PGG glucan compounds contained
measurable endotoxin. With either soluble or insoluble com-
pounds there appeared to be an optimal doserange. Doses
that were either higheror lower than thisoptimal rangedid
not protect as well against the lethal challenge. When the
numbers and types of WBC were evaluated after adntinis-
tration of soluble PGG glucan, a transient increase in the
granulocytes and monocytes were observed. Althoughnot
addressed in this study, i.v.-administered soluble PGGglu-
can at doses as high as 5 mg per animal was not toxic to
Because the animal testsystem usingmice did notrepre-
sent a true animal model for an infectious process, we
elected to continue the in vivo characterization of soluble
Two doses of
r to challenge.
iter for groups
itial group of
aths that were
as a source of
le t test.
it is likely that the actual
erence in the
a corresponding increase
VOL. 60, 1992
ONDERDONK ET AL.
TABLE 7. Effects of PGG treatment on peripheral WBC counts in rats
4,035 + 1,035b
5,067 ± 1,528
1,814 ± 518
1,465 + 1,436
94 + 31
3,307 ± 463
12,260 + 2,452d
1,021 + 116
5,883 + 588d
131 ± 18
808 + 220
aAbsolute cell counts.
bMeans + standard errors; minimum of two observations per sample time.
cOne hundred micrograms.
dp < 0.05; two-sample t test versus untreated animals.
PGG glucan in the rat model for intra-abdominal sepsis. This
model has been well characterized microbiologically (18,
27), and implantation of the cecal content inoculum routinely
results in an "50% mortality, with all animals exhibiting a
gram-negative septicemia during the first several days fol-
lowing challenge. It has also been shown that death in these
against facultative gram-negative rods, such as E. coli. Initial
trials with soluble PGG glucan in this model system indi-
cated that a reproducible reduction in mortality could be
observed when soluble PGG glucan was administered i.m. in
two doses at 24 and 4 h prior to challenge.
When tested at various times after challenge, soluble PGG
glucan-treated animals had significantly fewer bacteria in
their blood than did untreated animals, with a difference of at
least 20-fold noted at 24 h postchallenge, a time when most
deaths occur in infected animals. Peripheral WBC counts in
rats given a single dose of soluble PGG glucan yielded the
same proportional increases in total numbers of WBC as
were noted in mice.
There is some evidence that soluble PGG glucan treatment
primes granulocytes and macrophages for subsequent cytok-
ine release (interleukin-1 and -2) when bacterial challenge
occurs (26) and that soluble PGG glucan treatment acts in a
manner similar to treatment with colony-stimulating factors
(20). However, the timing of administration makes it unlikely
that cytokines are the exclusive mechanism by which pro-
tection occurs. It is also clear that the differences in the
number of bacteria present in blood cannot be explained
simply by the presence of more phagocytic cells in the
On the basis of the observations regarding the total WBC
counts and the number of organisms in the blood of treated
animals, there are at least two additional possible explana-
tions for the activity of soluble PGG glucan: (i) phagocytic
killing is much more efficient following soluble PGG glucan
treatment or (ii) bacteria do not leave the peritoneal cavity
and enter the blood as effectively in soluble PGG glucan-
treated animals because of improved phagocytic killing
within the peritoneal cavity or some other mechanism.
Although there are reports of increased phagocytic killing of
both bacteria and yeast cells by monocytes and neutrophils
after soluble PGG treatment, it is difficult to accept a 20- to
50-fold-greater killing efficiency ofWBC as the entire expla-
nation for the biologic activity of this material (11). We are
currently addressing the possibility that E. coli simply does
not enter the blood as effectively in animals treated with
soluble PGG glucan as it does in untreated control animals.
Preliminary data suggest that trapping of organisms within
the peritoneal cavity is an important part of PGG glucan-
Irrespective of the ultimate explanation for the in vivo
activity of soluble PGG glucan, this apparently nontoxic,
is abrogated by antimicrobial therapy directed
glucose polymer is an interesting immunomodulatory drug.
If the activity noted in both animal test systems and the
animal model system for intra-abdominal sepsis applies to
larger animals, including humans, it is possible that such
materials may either be used adjunctively with existing
therapeutic modalities or replace less-appealing therapeutic
methods of preventing and treating serious bacterial sepsis.
1. Bacon, J. S. D., V. C. Farmer, D. Jones, and I. F. Taylor. 1969.
The glucan component of the cell wall of Baker's yeast (Sac-
charomyces cerevisiae) considered in relation to its ultrastruc-
ture. Biochem. J. 114:556-557.
2. Bluhm, T., Y. Deslandes, and R. Marchessault. 1982. Solid-state
and solution conformation of scleroglucan. Carbohydr. Res.
3. Bluhm, T. L., and A. Sarko. 1977. The triple helical structure of
lentinan, a linearP-1-3-D-glucan.Can. J. Chem. 55:293-299.
4. Cook, J. A., T. W. Holbrook, and W. J. Dougherty. 1982.
Protective effect of glucan against visceral leishmaniasis in
hampsters. Infect. Immun. 37:1261-1269.
5. Czop, J. K. 1986. Phagocytosis of particulate activators of the
alternative complement pathway: effects of fibronectin. Adv.
6. Czop, J. K., and K. F. Austen. 1985. A P-glucan inhibitable
receptor on human monocytes: its identity with the phagocytic
receptor for particulate activators of the alternative complement
pathway. J. Immunol. 134:2588-2593.
7. Czop, J. K., and K. F. Austen. 1985. Generation of leukotrienes
by human monocytes upon stimulation of their 13-glucan recep-
tor during phagocytosis. Proc. Natl. Acad. Sci. USA 82:2751-
8. Czop, J. K., and K. F. Austen. 1985. Properties of glycans that
activate the human alternative complement pathway and inter-
act with the human monocyte I-glucan receptor. J. Immunol.
9. Czop, J. K., D. T. Fearson, and K. F. Austen. 1978. Opsonin-
independent phagocytosis of activators of the alternative com-
plement pathway by human monocytes. J. Immunol. 120:1132-
10. Easson, D. D., Jr., G. R. Ostroff, and S. James. 1990. Macro-
phage-targeted carbohydrate microcapsules for antigen and
drug delivery. Abstr. Int. Congr. Infect. Dis. 1990, abstr. 689, p.
11. Jamas, S., D. D. Easson, Jr., and G. R. Ostroff. 1990. PGG-a
class of macrophage
Abstr. Int. Congr. Infect. Dis. 1990, abstr. 698, p. 143.
12. Jamas, S., D. D. Easson, Jr., G. R. Ostroff, and A. B. Onder-
donk. Submitted for publication.
13. Janusz, M. J., K. F. Austen, and J. K. Czop. 1987. Lysosomal
enzyme release from human monocytes by particulate activa-
tors is mediated by f-glucan inhibitable receptors. J. Immunol.
14. Kokoshis, P. L., D. L. Williams, J. A. Cook, and N. R. DiLuzio.
1978. Increased resistance to Staphyococcus aureus infection
and enhancement of serum lysozyme activity by glucan. Sci-
15. Norisuye, T., T. Yanaki, and H. Fujita. 1980. Triple helix of a
ANTI-INFECTIVE EFFECT OF PGG GLUCAN IN VIVO
schizophyllum commune polysaccharide in aquaeous solution.
J. Polym. Sci. Phys. Ed. 188:547-558.
16. Ogawa, K., and T. Watanabe. 1973. The dependence of the
conformation of a (1--3)-j3-D-glucan on chain-length in alkaline
solution. Carbohydr. Res. 29:397-403.
17. Onderdonk, A. B., R. L. Cisneros, P. Hinkson, and G. R.
Ostroff. Unpublished data.
18. Onderdonk, A. B., W. M. Weinstein, N. M. Sullivan, J. G.
Bartlett, and S. L. Gorbach. 1974. Experimental intra-abdominal
abscesses in rats: quantitative bacteriology of infected animals.
Infect. Immun. 10:1256-1259.
19. Ostroff, G. R., D. D. Easson, Jr., and S. Jamas. 1989. Manipu-
lation of yeast glucan structure: molecular weight, branch
frequency and branch length. 198th Am. Chem. Soc. Natl.
Meet. 1989, abstr. MBTD 019.
20. Patchen, M. L., and T. J. Macvittie. 1986. Hemopoietic effects
of intravenous soluble glucan administration. J. Immunophar-
21. Phaff, H. J. 1963. Cell wall of yeasts. Annu. Rev. Microbiol.
22. Reynolds, J. A., M. D. Kastello, D. G. Harrington, C. L. Crebs,
C. J. Peters, J. V. Jemski, G. H. Scott, and N. R. DiLuzio. 1980.
Glucan-induced enhancement of host resistance to selected
infectious disease. Infect. Immun. 30:51-57.
23. Riggi, S. J., and N. R. DiLuzio. 1961. Identification of a
reticuloendothelial stimulating agent in zymosan. Am. J. Phys-
24. Saito, H., T. Ohki, and T. Sasaki. 1977. A 13C nuclear magnetic
resonance study of gel-forming (1--3)-P-D-glucans. Evidence of
the presence of single-helical conformation in a resilient gel of a
curdlan-type polysaccharide 13140. Biochemistry 16:908-914.
25. Saito, H., T. Ohki, N. Takasuka, and T. Susaki. 1977. A
13C-NMR-spectral study of a gel-forming, branched(1-*3)-P-D-
glucan (lentinan) from Lentinus edodes, and its acid-degraded
fractions, structure, and dependence of conformation on the
molecular weight. Carbohydr. Res. 58:293-305.
26. Sherwood, E. R., D. L. Williams, R. B. McNamee, E. L. Jones,
I. W. Browdwe, and N. R. DiLuzio. 1987. Enhancement of
interleukin-1 and interleukin-2 production of soluble glucan. Int.
J. Immunopharmacol. 9:261-267.
27. Weinstein, W. M., A. B. Onderdonk, J. G. Bartlett, and S. L.
Gorbach. 1974. Experimental intra-abdominal abscesses in rats:
development of experimental model. Infect. Immun. 10:1250-
28. Williams, D. L., J. A. Cook, F. 0. Hoffman, and N. R. DiLuzio.
1978. Protective effects of glucan in experimentally induced
candidiasis. J. Reticuloendothel. Soc. 23:479-490.
VOL. 60, 1992