Human vascular endothelial cell adhesion receptors for Plasmodium falciparum-infected erythrocytes: roles for endothelial leukocyte adhesion molecule 1 and vascular cell adhesion molecule 1.

The clinical complications associated with severe and cerebral malaria occur as a result of the intravascular mechanical obstruction of erythrocytes infected with the asexual stages of the parasite, Plasmodium falciparum. We now report that a primary P. falciparum-infected erythrocyte (parasitized red blood cell [PRBC]) isolate from a patient with severe complicated malaria binds to cytokine-induced human vascular endothelial cells, and that this adhesion is in part mediated by endothelial leukocyte adhesion molecule 1 (ELAM-1) and vascular cell adhesion molecule 1 (VCAM-1). PRBC binding to tumor necrosis factor alpha (TNF-alpha)-activated human vascular endothelial cells is partially inhibited by antibodies to ELAM-1 and ICAM-1 and the inhibitory effects of these antibodies is additive. PRBCs selected in vitro by sequential panning on purified adhesion molecules bind concurrently to recombinant soluble ELAM-1 and VCAM-1, and to two previously identified endothelial cell receptors for PRBCs, ICAM-1, and CD36. Post-mortem brain tissue from patients who died from cerebral malaria expressed multiple cell adhesion molecules including ELAM-1 and VCAM-1 on cerebral microvascular endothelium not expressed in brains of individuals who died from other causes. These results ascribe novel pathological functions for both ELAM-1 and VCAM-1 and may help delineate alternative adhesion pathways PRBCs use to modify malaria pathology.

Direct effects of IL-4/IL-13 and the nematode Nippostrongylus brasiliensis on intestinal epithelial cells in vitro.

Nematode infections induce the upregulation of mucin- and glycosylation-related genes in intestinal epithelial cells in vivo. However, the factor(s) that induce these changes in epithelial cells have not been fully elucidated. In the present study, we analysed the effects of the Th2 cytokines IL-4 and IL-13 and the excretory-secretory (ES) product of the nematode Nippostrongylus brasiliensis on the gene expression of the major mucin core peptide MUC2, the sialyltransferase ST3GalIV (Siat4c) and the sulphotransferase HS3ST1 in intestinal epithelium-derived IEC-6 cells by quantitative reverse transcription (RT)-PCR. The administration of IL-4 and IL-13 resulted in a significant upregulation of ST3GalIV and HS3ST1 gene transcription, but had no effect on MUC2, in IEC-6 cells. RT-PCR studies also demonstrated the constitutive expression of IL-13Ralpha1 and IL-4R in IEC-6 cells. On the other hand, the ES product induced upregulation of ST3GalIV, but not HS3ST1 or MUC2, while coadministration of IL-13 and the ES product induced a slight but significant upregulation of MUC2. Co-incubation of live N. brasiliensis adult worms with IEC-6 cells resulted in the upregulation of ST3GalIV and MUC2. These results suggested that HS3ST1 gene expression is strictly regulated by IL-4/IL-13, while ST3GalIV and MUC2 gene expressions are regulated by redundant mechanisms.

Immunity-mediated regulation of fecundity in the nematode Heligmosomoides polygyrus--the potential role of mast cells.

Previous studies have shown that host immunity regulates the fecundity of nematodes. The present study was aimed at clarifying the reversible nature of fecundity in response to changes of immunological status and to determine which effector cells are responsible for compromising fecundity in Heligmosomoides polygyrus. Enhanced fecundity was observed in immunocompromised SCID and nu/nu mice compared to those in the corresponding wild-type mice, with significantly fewer numbers of intrauterine eggs produced in the wild-type than in the immunodeficient mice. When 14-day-old adult worms from BALB/c mice were transplanted into naïve BALB/c mice, their fecundity increased significantly as early as 24 h post-transplantation, but not when they were transferred into immune mice, suggesting the plastic and reversible nature of fecundity in response to changes in host immunological status. In mast cell-deficient W/W(v) mice, nematode fecundity was significantly higher than in mast cell-reconstituted W/W(v) or +/+ mice. The serum levels of the mast-cell protease mMCP1 were markedly increased in the wild-type as well as the mast cell-reconstituted W/W(v), but not in the W/W(v), SCID, or nu/nu mice during infection. These findings raise the interesting possibility that certain activities of mast cells, either directly or indirectly, regulate parasite fecundity during infection.

Depleted intestinal goblet cells and severe pathological changes in SCID mice infected with Heligmosomoides polygyrus.

To determine the role of T cells and mast cells in intestinal pathology and immune expulsion of intestinal nematodes, worm burdens, goblet cell responses and villus structures were analysed in T- and B-cell-deficient severe combined immunodeficiency (SCID) mice, athymic nu/nu mice and mast cell deficient W/W(v) mice after infection with the nematode Heligmosomoides polygyrus. SCID and nu/nu mice showed significantly higher worm burdens at week 9 post-infection compared with the wild-type controls. SCID and nu/nu mice showed compromised goblet cell hyperplasia and/or Muc 2 expression, indicating that both events are T-cell dependant. On the other hand, the SCID mice showed increased pathology (villus atrophy and crypt hyperplasia) and increased numbers of proliferating cell nuclear antigen positive cells compared to the wild-type controls. W/W(v) mice, conversely, were able to expel the worms normally, had normal goblet cell hyperplasia, and did not demonstrate the changes in mucosal architecture seen in SCID mice, confirming that a normal mast cell response is not necessarily required for these changes. These results suggest that a functional T-cell response, but not a mast cell response, is necessary for anti-parasite responses, goblet cell function, and maintaining normal mucosal architecture.

COS cell expression cloning of Pfg377, a Plasmodium falciparum gametocyte antigen associated with osmiophilic bodies.

We report the deduced protein sequence and preliminary characterization of Pfg377, a novel sexual stage antigen of Plasmodium falciparum. An initial cDNA clone (Pfg377-1) encoding the N-terminal 755 amino acids of Pfg377 was isolated by transfecting a 3D7 gametocyte cDNA library into COS7 cells and selecting using a pool of anti-Pfs230 monoclonal antibodies. The protein encoded by Pfg377-1 included an N-terminal hydrophobic signal sequence, but no apparent transmembrane anchor. Instead, the particular cDNA clone selected was fused in-frame at its 3' end with the coding sequence for the human decay acceleration factor membrane anchor, which had been deliberately placed downstream of the vector polylinker in order to attach potential fusion proteins onto the COS cell surface. Northern blots probed with the Pfg377-1 cDNA demonstrated cross-hybridization to a single approximately 9.5-kb transcript, which was present only in sexual stages, and not in a sexual stages. DNA hybridization was used to obtain a series of overlapping genomic clones which collectively yielded the complete DNA sequence for Pfg377. There are no introns within the gene, which contains a 9360-bp open reading frame and encodes a 377-kDa protein. The Pfg377 protein is highly hydrophilic, and has an essentially non-repetitive structure, with only four very limited regions of tandem repeats. The Pfg377 gene resides on chromosome 12, and immunoelectron microscopy with two different anti-Pfg377 polyclonal antisera raised against two separate recombinant sub-fragments of the protein both indicated that the antigen is located in electron-dense organelles of the gametocytes--the osmiophilic bodies--which are proposed to play a role in parasite emergence from the erythrocyte during gametocyte maturation in the Anopheles mosquito midgut. Although it was selected with anti-Pfs230 antibodies, comparison of the sub-cellular locations and protein sequences of Pfg377 and Pfs2 show them to be completely distinct antigens. We hypothesize that Pfg377-1 was initially isolated because it expresses an epitope which is recognized by (i.e., cross-reacts with) one of the anti-Pfs230 monoclonal antibodies used to select the original transfected COS cells.

Epitope-selected monospecific antibodies to recombinant antigens from Toxoplasma gondii reacted with dense granules of tachyzoites.

Epitope-selected monospecific antibodies were applied to investigate the localization of antigenic molecules in Toxoplasma gondii by immunoelectron microscopy. Eighty cDNA clones encoding antigenic polypeptides were immunoscreened from lambda gt11 expression library with T. gondii infected mouse sera. Twenty different clones with no crossreactivity were selected from eighty clones. Monospecific antibodies to antigens derived from respective cDNA clones extracted from infected mouse sera by the epitope selection method were used in Western blot analysis and immunoelectron microscopy. Eleven antigens were detected with epitope-selected antibodies in lysates of T. gondii tachyzoites. Five of the antigens with molecular weights of 60, 40, 35, 28, and 27 KD were localized in the dense granules. Monospecific antibodies purified by the epitope selection method were useful for investigating the localization of antigens without preparation of a monoclonal antibody from a hybridoma.

Human cerebral malaria in Thailand: a clinico-pathological correlation.

Based on the cerebral malaria coma scale, 39 falciparum malaria autopsy cases from the Hospital for Tropical Diseases, Mahidol University, Bangkok, Thailand were divided into two groups of patients that had either cerebral malaria or non-cerebral malaria. We then studied significant pathological differences, such as parasitized erythrocyte (PRBC) sequestration, ring hemorrhages and cerebral edema, between these two groups in order to investigate the correlation between the clinical coma scale and pathological findings. Patients with a coma grade of 2 and higher were designated as having cerebral malaria, and had erythrocyte PRBC sequestration in cerebral microvessels. Ninety four percent (94%) of cerebral microvessels showed PRBC sequestration when quantitatively analyzed. On the other hand, only 13% of cerebral microvessels showed sequestration in non-cerebral malaria patients with a coma grade of 1 and lower, although some degree of PRBC sequestration was found in 50% of these patients. Our study, therefore, clearly demonstrated that the degree of the PRBC sequestration in cerebral microvessels appeared to correlate closely with the clinical coma scale.

Bromodeoxyuridine labeling studies on the proliferation of intestinal mucosal mast cells in normal and athymic rats.

The proliferation of mucosal-type mast cells (MMC) in rat small intestine was studied using a bromodeoxyuridine (BrdU)-labeling method. After 24-h cumulative injections of BrdU into adult SD rats, 3.5% of MMC were labeled, while only 0.3% and 0.1% of mast cells were labeled in back skin and ear respectively. From the results, it was concluded that MMC division occurred more than 10 times as frequently as the division of skin mast cells. Similar results were obtained in athymic adult rats (F344/N Jcl-rnu) in which the number of MMC was similar to that in heterologous animals. Thus, thymic factor(s) or T cells may not have an important role in MMC division in normal states. When SD rats were infected with Nippostrongylus brasiliensis, vigorous proliferation of MMC was brought about 13 to 15 days after infection. At that period, 40% of MMC were labeled by a single injection of BrdU, and 85% of MMC were labeled by 9-h cumulative injections of BrdU, with the result that most MMC rapidly proliferated in the intestinal mucosa during this period. Mitotic figures of MMC were sometimes observed. On the contrary, hyperplasia of MMC was not observed in athymic rats infected with nematodes. Therefore, MMC hyperplasia after nematode infection is dependent on thymic factor or T cells, and its mechanism is different from that of MMC division in normal states, in which thymic factor(s) or T cells are not essential.

Mucosal mast cell proliferation following normal and heterotopic infections of the nematode Nippostrongylus brasiliensis in rats.

Infections of intestinal nematodes induce the T cell-dependent proliferation of intestinal mucosal mast cells (MMC). To examine whether nematode-induced MMC proliferation is affected by the site of infestation, adult-stage nematode Nippostrongylus brasiliensis (NB) was transplanted into the normal infection site, the duodenum, or into heterotopic sites, the peritoneal cavity (i.p.) or subcutaneous tissue (s.c.), of rats. Two weeks after duodenal inoculation, MMC numbers in the small intestine had increased 6.5-fold. In contrast, i.p. and s.c. inoculation induced only slight increases of intestinal MMC. After i.p. inoculation, worm granulomas developed in the connective tissues adhering to stomach and duodenum, and large numbers of mast cells appeared around the granulomas. The majority of the latter mast cells showed histochemical features similar to MMC: they were formalin sensitive, berberine sulfate-, alcian blue+/safranine-, and rat mast cell protease (RMCP) II+. After s.c. inoculation, worm granulomas developed at the inoculation site, but the number of mast cells around the granulomas was not significantly increased. These results suggest that intense proliferation of MMC or MMC-like cells is induced only by the infections on mucosa or in mucosa-associated tissues.

Lack of active lung anaphylaxis in congenitally mast cell-deficient Ws/Ws rats sensitized with the nematode Nippostrongylus brasiliensis.

Ws/Ws rats are deficient in both mucosal- and connective tissue-type mast cells. To study the role of mast cells in active anaphylaxis, changes in vascular permeability in the trachea upon intravenous antigen challenge with Evans blue dye were examined in Ws/Ws, heterogenic Ws/+, and normal +/ + rats sensitized with the nematode Nippostrongylus brasiliensis. Antigen challenge resulted in fatal anaphylactic shock in some +/+ and Ws/+ rats, but not in Ws/Ws rats. Marked dye leakage developed within 30 min in the trachea of +/+ and Ws/+ rats, while Ws/Ws rats showed no substantial increases in the levels of vascular permeability. Ex vivo stimulation of sensitized lung fragments from +/+ animals with specific antigen induced significant releases of histamine and leukotriene (LT) C4, while sensitized Ws/Ws rat-lung fragments did not. In Ws/Ws rats, levels of nematode-specific IgE, IgG1 and IgG2a antibodies as well as levels of lung eosinophilia were not significantly different from those in +/+ rats. These results show that mast cell-deficient Ws/Ws rats fail to develop active anaphylaxis, and this is mediated probably by the lack of mast cell-derived mediators required for initiation of the reaction.

Migration of neutrophils is dependent on mast cells in nonspecific pleurisy in rats.

To determine the role of mast cells in the recruitment of neutrophils and eosinophils, acute nonspecific pleurisy was induced by injecting isologous serum into normal +/+ and mast cell-deficient Ws/Ws rats. In +/+ rats, neutrophil infiltration peaked 4 h after serum administration, followed by influx of eosinophils after 24-48 h. The levels of neutrophil influx after 4 h as well as the activity of myeloperoxidase (MPO) in pleural lavage-cell extract were significantly lower in Ws/Ws rats than in +/+ rats. In contrast, numbers of eosinophils as well as activity of eosinophil peroxidase (EPO) did not differ significantly between Ws/Ws and +/+ rats. For local reconstitution of mast cells, +/+ rat peritoneal mast cells (PMC) or mesenteric lymph node cells (MLNC) as a control were transferred into the Ws/ Ws pleural cavity. Serum injection into animals with PMC transfer 7 days previously triggered augmented neutrophil influx by approximately 4.7-fold as compared to that in MLNC-transferred animals. Mast cells recovered from the pleural cavity of PMC-transferred rats showed histamine contents equivalent to 20% of that of freshly isolated PMC and retained the reactivity to compound 48/80. These results indicated that dependency of neutrophil recruitment on resident mast cells is greater than that of eosinophils in isologous serum-induced pleurisy.

Renal pathology in owl monkeys vaccinated with Plasmodium falciparum asexual blood-stage synthetic peptide antigens.

Renal specimens from Aotus monkeys were studied by light microscopy and immunohistochemistry to examine pathologic changes following vaccination with synthetic peptides corresponding to the 35-kD, 55-kD, and 83-kD asexual blood stage antigens of Plasmodium falciparum. The monkeys were vaccinated and later challenged with P. falciparum. In the monkeys vaccinated with Centers for Disease Control peptides (group I), specimens from four of six postvaccinated animals had mild to severe mesangial proliferation and two had diffuse interstitial nephritis. Specimens from three monkeys vaccinated with Colombia peptides (group II) had mild to severe mesangial proliferation and one had interstitial nephritis. In the hybrid polymer-vaccinated monkeys (group III), specimens from three animals had mild to moderate mesangial proliferation and one had severe interstitial nephritis. On the other hand, the control group immunized with bovine serum albumin (group IV) showed that specimens from three animals had mild to severe mesangial proliferation and two had severe interstitial nephritis. In the nonimmunized group (group V), specimens from three animals had moderate to severe mesangial proliferation and two had severe and mild interstitial nephritis. Immunohistochemical analysis using the peroxidase-antiperoxidase method revealed mesangial deposits of P. falciparum antigens in 11 of 14 vaccinated monkeys and in five of 10 unvaccinated controls. These results show that treatment of monkeys with prospective malaria vaccines does not increase the frequency of occurrence or of the severity of renal lesions. These data thus provide a baseline for assessing the safety of synthetic malarial vaccines in the future.

Placental pathology in Plasmodium berghei-infected rats.

The pathologic changes in placentae of pregnant rats infected with Plasmodium berghei at different stages of gestation were studied using light and electron microscopy and immunohistochemistry. The major changes observed were thickening and duplication of the trophoblastic basement membrane, and accumulation of parasitized erythrocytes and occasional mononuclear cells in the maternal blood space. Immunohistochemical examination of nine placentae revealed that six stained positively for IgG, two for IgM, and four for P. berghei antigen. No C3 deposition was detected. The findings in this study indicate that the variable parasitologic-clinical course from benign to fatal of P. berghei infection in pregnant rats makes it a potentially valuable model of human gestational malaria infection.

Pathology of falciparum malaria in Vietnam.

Autopsy samples from the brains of 20 patients who died of falciparum malaria were examined by light microscopy and by an immunohistologic method. Particular attention was paid to a comparison of the pathologic features of the white matter and the cortex. In the high-sequestration (greater than 50%) group (n = 8), the mean +/- SD percentage of cerebral microvessels that showed parasitized red blood cell (PRBC) sequestration was 71.2 +/- 8.1% in the cortex and 84.0 +/- 6.7% in the white matter. The difference in the PRBC sequestration rate between cortex and white matter was statistically significant (P less than 0.01). Perivascular and ring hemorrhages were seen more frequently in the white matter than in the cortex. Deposition of IgG and Plasmodium falciparum antigen in the cerebral microvessels was more highly significant in the white matter than in the cortex (P less than 0.01). Our study demonstrated that the localized concentration of PRBC sequestration in the brain correlated with the marked immunohistologic differences in the microvessels of cortex and white matter.

Renal pathology in Saimiri monkeys during a vaccine trial using the recombinant circumsporozoite protein of Plasmodium vivax.

Renal specimens of squirrel monkeys (Saimiri sciureus boliviensis) were studied by light microscopy and immunohistochemistry to examine the pathologic changes during vaccine trials with four recombinant circumsporozoite (CS) proteins (rPvCS-1, rPvCS-2, rPvCS-3, NS1(81) V20) of Plasmodium vivax. The monkeys were vaccinated and later challenged with P. vivax sporozoites. Among the 33 posttrial biopsies, 17 had mild to moderate mesangial proliferation and nine had interstitial nephritis. Immunohistochemistry by the peroxidase-antiperoxidase (PAP) method revealed IgG deposits in only three of 24 specimens and failed to demonstrate C3 deposits and P. vivax antigens in their glomeruli. There was no relationship between the severity of nephropathy and intensity of parasitemia. The intensity of parasitemia was the same in the vaccinated and control groups. Vaccinated monkeys from the groups (rPvCS-1, rPvCs-2, rPvCS-3) had no differences in renal pathology from the unvaccinated controls, but one group vaccinated with NS1(81) V20 did not develop renal changes.

Immunoglobulin complex deposits in Plasmodium falciparum-infected placentas from Malawi and Papua New Guinea.

Term placentas from 35 patients infected with Plasmodium falciparum were obtained in Malawi in southeast Africa and six term placentas from patients infected with P. falciparum were obtained in Wewak, Papua New Guinea, Melanesia. The placental tissues were examined by light microscopy and by an immunohistologic method to compare the pathologic changes of placentas in the two malaria-endemic countries. Using the number of parasitized red blood cells (PRBC) in intervillous spaces, pregnant women from Malawi with placental parasitemia were categorized into three groups. In the high PRBC group (> 20%, group I), there was no deposition of IgE in fetal blood vessels. In contrast, IgE was observed in fetal blood vessels of the intermediate PRBC group (1-10%, group II) and low PRBC group (< 1%, group III). In all six placentas from Papua New Guinean women, deposition of immune complexes, including IgE, was observed in the fetal blood vessels. All placentas with deposition of IgE in fetal blood vessels showed no sequestration of malaria parasites in intervillous spaces. Our data indicate that the amount of deposition of IgE in the placenta from women infected with P. falciparum is inversely correlated with the degree of parasitemia at that site.

A primate model for human cerebral malaria: Plasmodium coatneyi-infected rhesus monkeys.

A major factor in the pathogenesis of human cerebral malaria is blockage of cerebral microvessels by the sequestration of parasitized human red blood cells (PRBC). In vitro studies indicate that sequestration of PRBC in the microvessels is mediated by the attachment of knobs on PRBC to receptors on the endothelial cell surface such as CD36, thrombospondin (TSP), and intercellular adhesion molecule-1 (ICAM-1). However, it is difficult to test this theory in vivo because fresh human brain tissues from cerebral malarial autopsy cases are not easy to obtain. Although several animal models for human cerebral malaria have been proposed, none have shown pathologic findings that are similar to those seen in humans. In order to develop an animal model for human cerebral malaria, we studied brains of rhesus monkeys infected with the primate malaria parasite, Plasmodium coatneyi. Our study demonstrated PRBC sequestration and cytoadherence of knobs on PRBC to endothelial cells in the cerebral microvessels of these monkeys. Cerebral microvessels with sequestered PRBC were shown by immunohistochemical analysis to possess CD36, TSP, and ICAM-1. These proteins were not evident in the cerebral microvessels of uninfected control monkeys. Thus, our study indicates, for the first time, that rhesus monkeys infected with P. coatneyi can be used as a primate model to study human cerebral malaria. By using this animal model, we may be able to evaluate strategies for the development of vaccines to prevent human cerebral malaria.

A nonhuman primate model for human cerebral malaria: effects of artesunate (qinghaosu derivative) on rhesus monkeys experimentally infected with Plasmodium coatneyi.

We studied the effects of artesunate on rhesus monkeys infected with Plasmodium coatneyi. Sixteen rhesus monkeys were divided in four groups. Group I consisted of three monkeys that were splenectomized and were treated with three doses (loading dose: 3.3 mg/kg, maintenance doses: 1.7 mg/kg) of artesunate, group II consisted of three monkeys that were treated with three doses of artesunate (same as group I), group III consisted of two monkeys that were treated with one dose (3.3 mg/kg) of artesunate, and group IV consisted of five untreated monkeys. Parasitemias of these groups ranged from 13.3% to 19.5% before treatment. Twenty-four hours after administration, the parasitemia was reduced to 2.2% in group I and to < 0.1% in group II; parasitemia was lowered to 10.6% in group III only 3 hr after drug administration. The rate of sequestration in the cerebral microvessels, which was 29.4% in untreated animals, was < 0.1% in groups I and II (24 hr after treatment), and 2.0% in group III (3 hr after treatment). These data clearly indicate that artesunate not only reduced parasitemia, but also reduced the rate of parasitized red blood cell (PRBC) sequestration in cerebral microvessels. In an immunohistologic study, endothelial-leukocyte adhesion molecule-1 (ELAM-1) was not detected in group I after treatment with artesunate, although the presence of CD36, thrombospondin, intercellular adhesion molecule-1, IgG, and C3 in the cerebral microvessels was not altered. This is the first in vivo study to show that artesunate interferes with continued PRBC sequestration in the cerebral microvessels in cerebral malaria.(ABSTRACT TRUNCATED AT 250 WORDS)

Detection of surface carbohydrates on Pneumocystis carinii by fluorescein-conjugated lectins.

Lectins react with a wide range of different carbohydrates (Table 1). Even so-called monospecific anti-H(O) lectins from Lotus tetragonolobus, Ulex europaeus, and Anguilla anguilla react not only with the anti-H determinant but also with several fucosylated carbohydrates. Consequently, the type of lectin receptor existing on the surface of Pneumocystis carinii should be determined, because only a carbohydrate analysis can demonstrate the kind of carbohydrates which exist on the cell surface of this parasite. For the purpose of this study we used fluorescent isothiocyanata (FITC)-conjugated lectins. Concanavalin A (Con A) and Maclura pomifera (MPA) agglutinin reacted to P. carinii at low concentrations, and the fluorescence intensity was gradually increased with the concentration of the lectins. With lectins from Bauhinia purpurea (BPA), Dolichos biflorus (DBA), Glycine max (SBA), Griffonia simplicifolia (GS-I, GS-II), and Triticum vulgaris (WGA), fluorescence was emitted at high concentrations, while Arachis hypogaea (PNA) and Ulex europaeus (UEA-I) agglutinins did not show fluorescence. The results suggest that P. carinii has abundant Con A- and MPA-specific carbohydrates on the surface.

Yeast glucan in the cyst wall of Pneumocystis carinii.

Ultrastructurally, the cyst wall of Pneumocystis carinii consists of an electron-dense outer layer, an electron-lucent middle layer, and an innermost plasmalemma. This is similar in appearance to the cell wall of some yeasts, e.g. Saccharomyces cerevisiae, which consists of an outer dense layer of mannan, a middle lucent layer of beta-1,3-glucan (yeast glucan) and an innermost plasmalemma. The cyst wall of P. carinii, as well as the cell wall of S. cerevisiae, can be labeled by a variety of methods which stain polysaccharides, such as Gomori's methenamine silver (GMS) and by Aniline blue, a dye which selectively stains beta-1,3-glucan. The treatment of P. carinii cysts with Zymolyase, which the key enzyme is beta-1,3-glucan laminaripentaohydrolase, results in lysis of the outer 2 layers of the cyst wall and the loss of positive staining by both GMS and Aniline blue. The lysis of elements of the cyst wall of P. carinii is achieved under the same conditions and concentration at which Zymolyase lyses the outer 2 layers of the cell wall of viable cells of S. cerevisiae. These observations indicate that a major component of the cyst wall of P. carinii is beta-1,3-glucan.