Sporogony in Pneumocystis carinii: Synaptonemal Complexes and Meiotic Nuclear Divisions Observed in Precysts1

Evidence for meiosis was demonstrated electron microscopically for the first time in Pneumocystis carinii in rat alveoli by the observation of synaptonemal complexes followed by nuclear divisions. Synaptonemal complexes indicating meiotic nuclear divisions were observed in uninuclear precysts. Additionally, owing to the use of tannic acid as a fixative, spindle microtubules were also observed for the first time in the precyst. Based on these facts, a new life cycle of the organism is proposed. The precyst has generally been considered an intermediate form between the trophozoite and the cyst. The present paper proposes that the precyst is additionally defined as the cell in which eight intracystic bodies are produced through meiotic reduction. The most characteristic feature of the precyst is a clump of mitochondria in the cytoplasm. We divide the precyst phase into three forms, which are named early, intermediate, and late. Synaptonemal complexes were only observed in the early precyst, which is a uninuclear cell with a thin pellicle. In the intermediate precyst, nuclear divisions are observed as follows: meiosis I produces two haploid nuclei and each of these divides at meiosis II producing four nuclei. After that, another postmeiotic mitosis takes place, resulting in eight haploid nuclei. In the late precyst, a delimiting membrane originates from the mother plasmalemma and surrounds the daughter nuclei and a small portion of the adjacent cytoplasm. Finally, when the eight intracystic bodies are complete, the precyst changes to a cyst. Thus, we deduce that intracystic bodies resulting from meiotic nuclear division are haploid and, after excystation, they are haploid trophozoites. We consider that this process can be called sporogony. Although we could not distinguish between the haploid and the diploid trophozoite, it is quite plausible that copulation occurs, probably in host alveoli.     

Quantitation of Absolute Pneumocystis carinii Nuclear DNA Content. Trophic and Cystic Forms Isolated from Infected Rat Lungs are Haploid Organisms

The Pneumocystis carinii carinii DNA content in nuclei of trophic forms and cysts (spore cases) containing 2, 4, or 8 intracystic bodies, were compared using quantitative fluorescence image analysis. The nuclear DNA content was found to be lower than the theoretical limits of Feulgen cytophotometry. Several fluorescent DNA dyes provide brighter staining, but these techniques suffer from nonspecific binding to other cellular components, such as RNA. It was demonstrated that the thick glycocalyx surfaces of trophic forms and the cyst walls of P. carinii organisms, as well as the cell wall of S. cerevisiae, bound all fluorescent dyes tested to varying degrees. Hence in this study, measurements were performed on cells in which the outer surfaces of organisms were first removed with lyticase. Two stains that appeared most specific for DNA, DB181 and 4′,6-diamidino-2-phenylindole (DAPI), were used for quantitations; lower deviations of fluorescence intensities were observed with DB181. Haploid wild type Saccharomyces cerevisiae and cdc-28 temperature-sensitive mutant cells, accumulated at the restrictive temperature (37° C), were used as quantitative internal standards for estimating the absolute nuclear DNA content of P. carinii. Haploid wild type and mutant nuclei stained with DAPI had the same relative fluorescence intensities. The P. carinii nuclear DNA content of trophic forms and individual intracystic bodies (spores), regardless of life cycle stage, were not different. The mean values obtained were 6.9 and 6.7 fg DNA/nucleus with DB181 and DAPI, respectively (approximately 9.26 and 8.99 Mbp nucleotides, respectively). Since these would include 2C (G-2 phase) and S-phase nuclei, a 1C population of nuclei was selected by histogram distributions of DB181-stained nuclei. Almost all nuclei analyzed in all life cycle stages fell within this population. The 1C mean of 6.55 fg DNA/nucleus (median, 6.62 fg DNA/nucleus) was estimated as representing 8.79 Mbp nucleotides, assuming only A-T binding of the dye and taking into account the G+C content of S. cerevisiae and P. carinii. A 4C (G-2-phase diploid nuclei) population was not detected in histograms of DB181- or DAPI-stained nuclei. The P. carinii nuclear DNA content values obtained in this study were similar to those independently obtained by calculating the total DNA in the organism's chromosomes resolved by electrophoretic techniques. Together, the data on total chromosome numbers and the estimated DNA content of those chromosomes, with our quantitation of nuclear DNA content of different life-cycle stages demonstrate that P. carinii carinii isolated from infected rat lungs are haploid organisms.     

Characterization of Pneumocystis carinii PHR1, a pH-Regulated Gene Important for Cell Wall Integrity

Pneumocystis carinii remains an important opportunistic fungal pathogen causing life-threatening pneumonia in patients with AIDS and malignancy. Currently, little is known about how the organism adapts to environmental stresses and maintains its cellular integrity. We recently discovered an open reading frame approximately 600 bp downstream of the region coding GSC-1, a gene mediating β-glucan cell wall synthesis in P. carinii. The predicted amino acid sequence of this new gene, termed P. carinii PHR1, exhibited 38% homology to Saccharomyces cerevisiae GAS1, a glycosylphosphatidylinositol-anchored protein essential to maintaining cell wall integrity, and 37% homology to Candida albicans PHR1/PHR2, pH-responsive genes encoding proteins recently implicated in cross-linking β-1,3- and β-1,6-glucans. In view of its homology to these related fungal genes, the pH-dependent expression of P. carinii PHR1 was examined. As in C. albicans, P. carinii PHR1 expression was repressed under acidic conditions but induced at neutral and more alkaline pH. PHR1-related proteins have been implicated in glucan cell wall stability under various environmental conditions. Although difficulties with P. carinii culture and transformation have traditionally limited assessment of gene function in the organism itself, we have successfully used heterologous expression of P. carinii genes in related fungi to address functional correlates of P. carinii-encoded proteins. Therefore, the potential role of P. carinii PHR1 in cell wall integrity was examined by assessing its ability to rescue an S. cerevisiae gas1 mutant with absent endogenous Phr1p-like activity. Interestingly, P. carinii PHR1 DNA successfully restored proliferation of S. cerevisiae gas1 mutants under lethal conditions of cell wall stress. These results indicate that P. carinii PHR1 encodes a protein responsive to environmental pH and capable of mediating fungal cell wall integrity.     

Freeze-Fracture Localization of Filipin-Sterol Complexes in Plasma- and Cyto-Membranes of Pneumocystis carinii1

The polyene antibiotic, filipin, was used as the probe for demonstrating sterols in the freeze-fractured plasma- and cytomembranes of Pneumocystis carinii. The distribution of filipin-sterol complexes was homogeneous on the plasma membrane throughout all developmental stages from trophozoite to cyst; however, the density of the complexes gradually decreased with the progress of development. In the trophozoite, the density of the complexes was 485 ± 42/μm² on the P face and 341 ± 27/μm² on the E face. It was 249 ± 50 on the P face and 132 ± 48 on the E face in the precyst and 138 ± 24 and 59 ± 20, respectively, in the cyst. The membranes of nucleus, mitochondria, and small round bodies showed more or fewer complexes while no complexes were found in the membranes of one endoplasmic reticulum. In nuclear and mitochondrial membranes, some small scattered clusters of complexes were observed. Two types of vacuoles were distinguished: one having many complexes in its membrane and the other having none at all.     

Freeze-fracture studies on Pneumocystis carinii. I. Structural alteration of the pellicle during the development from trophozoite to cyst

Pneumocystis carinii has generally been distinguished in three developmental stages, namely, trophozoite, precyst and cyst. The fine structure of the pellicle--the plasma membrane and the outer layer existing outside this plasma membrane--of each stage was studied by freeze-fracture technique. By this technique, P. carinii was cleaved through the cytoplasm or through the hydrophobic region of the plasma membrane, and the cross-fractured face of the outer layer was revealed on the replicas. The outer layer, which is electron-dense in the thin section, consisted of numerous fine granules about 15 nm in diameter in freeze-fracture images, whereas the electron-lucent middle layer which appeared in the precyst and cyst was less granular. Measurement of the intramembranous particles (IMP) also was carried out. The number of IMP per square micrometer of the plasma membrane of the trophozoite was 1,512 +/- 125 on the P face and 417 +/- 44 on the E face. In the precyst, the IMP density decreased, and 1,037 +/- 56 on the P face and 262 +/- 22 on the E face. In the cyst, it further decreased, nd 875 +/- 59 and 150 +/- 20 respectively. It is generally assumed that the density of IMP is related to the physiological activity of the cell membrane, so that the present results obtained in P. carinii suggest that the trophozoite is the most active stage, and that metabolic activity of the pellicle gradually decreases with the progress of development to the precyst then to the cyst.     

Cell Wall Antigens of Pneumocystis carinii Trophozoites and Cysts: Purification and Carbohydrate Analysis of these Glycoproteins

We have purified and biochemically analyzed individual cell wall glycoproteins of Pneumocystis carinii. Our results show that corresponding core glycoproteins constitute the cell wall antigens in both trophozoites and cysts, and glycosylation of these glycoproteins does not appear to be significantly altered during development. Cysts and trophozoites in rat-derived organism preparations were separated from each other by counterflow centrifugal elutriation, then treated with Zymolyase to obtain the cell wall fractions. Gel electrophoresis patterns of these fractions from both life-cycle stages were qualitatively similar. Ten major antigenic glycoproteins in these fractions were purified by preparative continuous elution gel electrophoresis. All ten glycoproteins from cysts and trophozoites contained mannose, glucose, galactose. and N-acetylglucosamine, and some contained traces of fucose. The glycoproteins of cysts had more mannose than their trophozoite counterparts. The trophozoite glycoproteins differed from those of the cyst by the presence of xylose. To examine the species-specificity of glycoprotein glycosylation, preparations of human-derived P. carinii (comprised of mixed life-cycle stages) were also examined and found to contain the same sugars as those found in rat-derived organisms. Most of the purified rat-derived glycoproteins bound Concanavalin A, which was abolished by treatment with N-glycanase. This suggested that the majority of the oligosaccharides were N-linked to the proteins, but attempts to identify carbohydrate linkage sites by amino acid sequencing were hampered by apparent modifications of residues. The peptides derived by cyanogen bromide cleavage revealed distinct size patterns for each glycoprotein, suggesting that they were distinct proteins. Most of the glycoproteins reacted with monoclonal antibodies which recognize a highly conserved epitope on rat P. carinii. Four of the individually purified glycoprotein preparations elicited in vitro cellular immune responses, implicating their involvement in the recognition of P. carinii by host T cells. The identification and characterization of P. carinii cell wall proteins will be helpful in analyzing the relationship of the organism to its mammalian host. Supplementary key words. Biochemical analysis, developmental stages, opportunistic pathogen, structure.     

Population Structure of Rat-Derived Pneumocystis carinii in Danish Wild Rats

The rat model of Pneumocystis carinii pneumonia is frequently used to study human P. carinii infection, but there are many differences between the rat and human infections. We studied naturally acquired P. carinii in wild rats to examine the relevance of the rat model for human infection. P. carinii DNA was detected in 47 of 51 wild rats and in 10 of 12 nonimmunosuppressed laboratory rats. Evidence for three novel formae speciales of rat-derived P. carinii was found, and these were provisionally named Pneumocystis carinii f. sp. rattus-secundi, Pneumocystis carinii f. sp. rattus-tertii, and Pneumocystis carinii f. sp. rattus-quarti. Our data suggest that low-level carriage of P. carinii in wild rats and nonimmunosuppressed laboratory rats is common and that wild rats are frequently coinfected with more than one forma specialis of P. carinii. We also examined the diversity in the internally transcribed spacer (ITS) regions of the nuclear rRNA operon of Pneumocystis carinii f. sp. carinii by using samples from wild rats and laboratory rats and spore trap samples. We report a lack of variation in the ITS1 and ITS2 regions that is consistent with an evolutionary bottleneck in the P. carinii f. sp. carinii population. This study shows that human- and rat-derived P. carinii organisms are very different, not only in genetic composition but also in population structure and natural history.     

Resting cysts of Parentocirrus hortualis Voß, 1997 (Ciliophora,Hypotrichia), with preliminary notes on encystation and various types of excystation

Resting cysts of Parentocirrus hortualis were investigated, using live observation, SEM and TEM. Processes during encystation and excystation were observed in vivo under the light microscope. During encystation, the trophic body becomes globular, the ciliature is resorbed in an anterior direction, the macronuclear nodules fuse into an elongated mass, and finally a cyst wall develops. As typical for oxytrichids, the resting cysts of P. hortualis are of the kinetosome-resorbing type and their wall is made of four layers: ectocyst, mesocyst, endocyst, and metacyst. The beginning of excystation is indicated by the formation of an excystation vacuole that helps the regenerating specimen to break the cyst wall. The excysting specimen leaves the resting cyst in a thin membrane that is gradually resorbed in the outer environment. Also two other excystation modes were observed. During the rare mode, the excystation vacuole breaks the thin membrane instead of the cyst wall that ruptures under the pressure of the body of the regenerating specimen. During the reproduction mode, the regenerating specimen divides within the resting cyst, producing two to four tomites. This is the first report of division in resting cysts of oxytrichids, but reproduction in division cysts was already described in keronopsids.     

Early acquisition of Pneumocystis carinii in neonatal rats as evidenced by PCR and oral swabs

The complete life cycle of Pneumocystis carinii has not been defined, but accumulating evidence suggests that the mammalian host may acquire this organism early in life. In the present study, the initial time of P. carinii acquisition was determined in rats by amplification of P. carinii DNA in oral swabs from seven sets of pups and dams and from fetal tissue obtained by cesarean section of three gravid female rats. DNA extracted from all samples was amplified by using PCR primers directed to the P. carinii mitochondrial large subunit rRNA. Amplicons were produced from 80% (28 of 35) of pups within 2 h after birth; from 97% (34 of 35) after 24 h, and in all of the serially sampled pups by 48 h. No P. carinii amplicons were produced from 48 fetuses or their placentae taken by cesarean section. Thus, P. carinii is acquired almost immediately after birth, and placental transmission occurs rarely, if ever, in rats.     

An Ultrastractural Study of the Trophozoite and Cyst Stages of Protostelium pyriformis Olive & Stoianovitch, 1969 (Eumycetozoea,Protosteliia)1

ABSTRACT. Electron microscope observations of the trophic and cyst stages of the protostelid Protostelium pyriformis Olive & Stoianovitch, 1969 indicate a perinuclear golgi-vesicular region in the amoeboid cell. Within this region is a microtubule organizing center (MTOC) and associated microtubules which radiate into the cytoplasm. The cyst wall is revealed as having three distinct regions, certain areas of the wall possibly corresponding to exit pores. Protostelium pyriformis shares many similarities in trophic cell structure with other eumycetozoans and members of the non-mycetozoan family Acanthamoebidae.     

Interstitial pneumonia in immunocompetent laboratory rats caused by natural infection with Pneumocystis carinii

Pneumocystis (P.) carinii is known to cause fatal pneumonia in immunocompromised rats. Cases of P. carinii interstitial pneumonia in immunocompetent rats have been shown histologically to present with perivascular lymphoid cuffs, which have previously been attributed to rat respiratory virus. This study aims to determine the prevalence and pathological characteristics of P. carinii in immunocompetent laboratory rats in experimental facilities in Japan. An epidemiological survey for this agent was performed using PCR to assess 1,981 immunocompetent rats from 594 facilities in Japan. We observed that 6 of the 1,981 rats (0.30%) from 4 out of 594 facilities (0.67%) were positive for P. carinii without infection of other known pathogens. Gross pulmonary lesions were found in 4 of the 6 affected rats. The lungs of these rats contained scattered dark red/gray foci. Histopathologically, the lungs exhibited interstitial pneumonia with lymphoid perivascular cuffs: Pneumocystis cysts were observed using Grocott’s methenamine silver stain. To our knowledge, this report is the first to reveal the prevalence of natural P. carinii infection in immunocompetent laboratory rats in Japan.     

Spontaneous cystogenesis in vitro of a Brazilian strain of Toxoplasma gondii

Conversion of Toxoplasma gondii tachyzoites to the bradyzoite stage and tissue cyst formation in the life cycle of the parasite have crucial roles in the establishment of chronic toxoplasmosis. In this work we investigated the in vitro cystogenesis and behavior of the EGS strain, isolated from human amniotic fluid. We observed that tachyzoites of the EGS strain converted to intracellular cysts spontaneously in LLC-MK₂ epithelial cells, HSFS fibroblasts and C6 glial cell lineage. The peak of conversion occurred in the LLC-MK₂ cells after 4 days of infection, when 72.3 ± 15.9 of the infected cells contained DBA positive cysts. Using specific markers against bradyzoite, tachyzoite and cyst wall components, we confirmed stage conversion and distinguished immature from mature cysts. It was also observed that the deposition of cyst wall components occurred before the total conversion of parasites. Transmission electron microscopy confirmed the fully conversion of parasites presenting the typical characteristics of bradyzoites as the posterior position of the nucleus and the presence of amylopectin granules. A thick cyst wall was also detected. Besides, the scanning microscopy revealed that the intracyst matrix tubules were shorter than those from the parasitophorous vacuole intravacuolar network and were immersed in a granular electron dense material. The EGS strain spontaneously forms high burden of cysts in cell culture without artificial stress conditions, and constitutes a useful tool to study this stage of the T. gondii life cycle.     

Use of Recombinant Cellulose-Binding Domains of Trichoderma reesei Cellulase as a Selective Immunocytochemical Marker for Cellulose in Protozoa

Some unicellular organisms are able to encyst as a protective response to a harmful environment. The cyst wall usually contains chitin as its main structural constituent, but in some cases, as in Acanthamoeba, it consists of cellulose instead. Specific cytochemical differentiation between cellulose and chitin by microscopy has not been possible, due to the similarity of their constituent β-1,4-linked hexose backbones. Thus, various fluorescent brightening agents and lectins bind to both cellulose and chitin. We have used a recombinant cellulose-binding protein consisting of two cellulose-binding domains (CBDs) from Trichoderma reesei cellulases linked together in combination with monoclonal anticellulase antibodies and anti-mouse immunoglobulin fluorescein conjugate to specifically stain cellulose in the cysts of Acanthamoeba strains for fluorescence microscopy imaging. Staining was observed in ruptured cysts and frozen sections of cysts but not in intact mature cysts. No staining reaction was observed with the chitin-containing cyst walls of Giardia intestinalis, Entamoeba dispar, or Pneumocystis carinii. Thus, the recombinant CBD can be used as a marker to distinguish between cellulose and chitin. Thirteen of 25 environmental or clinical isolates of amoebae reacted in the CBD binding assay. All 13 isolates were identified as Acanthamoeba spp. Five isolates of Hartmannella and seven isolates of Naegleria tested negative in the CBD binding assay. Whether cyst wall cellulose really is a unique property of Acanthamoeba spp. among free-living amoebae, as suggested by our findings, remains to be shown in more extensive studies.     

In vitro attachment of Pneumocystis carinii from mouse and rat origin

Summary— The attachment of Pneumocystis carinii to lung cells could play a role in the pathophysiology of P carinii pneumonia. The trophozoite attaches to type I alveolar epithelial cells. Physical, chemical, and extracellular matrix factors, involved in the mouseor rat-derived P carinii attachment to fibroblastic cells in culture, were examined using a new model of in vitro adherence. The development of parasite filopodia penetrating deeply the host cell cytoplasm was observed using transmission electronic microscopy. Killed P carinii organisms were unable to attach to cultured cells. Also, parasites were unable to attach to killed target cells. The P carinii in vitro attachment was partially inhibited by cytochalasin B. In contrast, the parasite attachment was not affected when the target cell cytoskeleton was altered. In our work conditions, sialic acids were not involved in the attachment process. Present results showed that fibronectin (Fn) plays a role in the parasite attachment, and suggest that a specific Fn-binding receptor is present at the surface of mouse-derived P carinii organisms.     

Characterization of Pneumocystis Major Surface Glycoprotein Gene (msg) Promoter Activity in Saccharomyces cerevisiae

Major surface glycoprotein (Msg), the most abundant cell surface protein of Pneumocystis, plays an important role in the interaction of this opportunistic pathogen with host cells, and its potential for antigenic variation may facilitate evasion of host immune responses. In the present study, we have identified and characterized the promoter region of msg in 3 species of Pneumocystis: P. carinii, P. jirovecii, and P. murina. Because Pneumocystis cannot be cultured, promoter activity was measured in Saccharomyces cerevisiae, a related fungus, using a yeast vector modified to utilize the gene coding for Renilla luciferase as a reporter gene. The 5′-flanking sequences of msg from all three Pneumocystis species showed considerable promoter activity, with increases in luciferase activity up to 15- to 44-fold above baseline. Progressive deletions helped define an ∼13-bp sequence in each Pneumocystis species that appears to be critical for promoter activity. Electrophoretic mobility shift analysis using P. carinii-specific msg promoter sequences demonstrated binding of nuclear proteins of S. cerevisiae. The 144-bp 5′-flanking region of P. murina msg showed 72% identity to that of P. carinii. The 5′-flanking region of P. jirovecii msg showed 58 and 61% identity to those of P. murina and P. carinii, respectively. The msg promoter is a good candidate for inclusion in a construct designed for genetic manipulation of Pneumocystis species.     

Synthesis and SAR of alkanediamide-linked bisbenzamidines with anti-trypanosomal and anti-pneumocystis activity

A series of alkanediamide-linked bisbenzamidines was synthesized and tested in vitro against a drug-sensitive strain of Trypanosoma brucei brucei, a drug-resistant strain of Trypanosoma brucei rhodesiense and Pneumocystis carinii. Bisbenzamidines linked with longer alkanediamide chains were potent inhibitors of both strains of T. brucei. However, bisbenzamidines linked with shorter alkanediamide chains were the most potent compounds against P. carinii. N,N′-Bis[4-(aminoiminomethyl)phenyl] hexanediamide, 4 displayed potent inhibition (IC₅₀ = 2–3 nM) against T. brucei and P. carinii, and was non-cytotoxic in the A549 human lung carcinoma cell line. The inhibitory bioactivity was significantly reduced when the amidine groups in 4 were moved from the para to the meta positions or replaced with amides.     

Molecular Genetic Distinction of Pneumocystis carinii from Rats and Humans

Pneumocystis carinii from rats and from humans were compared with respect to electrophoretic karyotype, presence of DNA sequences known to be repeated in rat-derived P. carinii, overall DNA sequence homology, and the sequences at two genetic loci. The organisms from each host species were different in each respect. Neither of two repeated DNAs from rat-derived P. carinii was found in the genome of human-derived organisms, and total DNA from rat-derived P. carinii failed to hybridize to human-derived P. carinii DNA. The sequences of the α-tubulin genes from the two P. carinii were strikingly different and the base composition of the α-tubulin gene from rat-derived P. carinii was rich in adenine and thymine, while the base composition of this gene from human-derived P. carinii was rich in guanine and cytosine. The sequence from the 18S rRNA gene of human-derived P. carinii was twice as divergent from that of rat-derived P. carinii as the sequence from the corresponding region of Candida albicans was from that of Candida tropicalis. These data show that rats and humans can harbor distinct types of P. carinii that are sufficiently different to suggest that P. carinii from the two hosts could be different species.     

Ultrastructural localization of epitopes recognized by monoclonal antibodies produced to rat Pneumocystis carinii.

The binding sites of five monoclonal antibodies (MAb) developed against rat Pneumocystis carinii were examined at the ultrastructural level by using a post-embedding labeling method. Although all five MAb reacted with the pellicle of P. carinii, they were divided into two groups by localization of binding sites. The MAb 168.2.1, 174.2.1, and 215.2.1 reacted mainly with the electron-dense outer layer, whereas MAb 227.1.1 and 228.1.1 labeled both the outer dense layer and the middle lucent layer. With in the first group of MAb, no significant differences were observed in the reactivity patterns seen with the different stages of P. carinii. In the second group, however, the intensity of labeling of the electron-dense layer was higher in the precyst, cyst, and ruptured cyst stages than in the trophozoite stage. These latter results indicate that there may be an increase in antigen accumulation during development from the trophozoite to the cyst stages, or that antigens may be modified the development.     

Evidence of Airborne Excretion of Pneumocystis carinii during Infection in Immunocompetent Rats. Lung Involvement and Antibody Response

To better understand the role of immunocompetent hosts in the diffusion of Pneumocystis in the environment, airborne shedding of Pneumocystis carinii in the surrounding air of experimentally infected Sprague Dawley rats was quantified by means of a real-time PCR assay, in parallel with the kinetics of P. carinii loads in lungs and specific serum antibody titres. Pneumocystis-free Sprague Dawley rats were intratracheally inoculated at day 0 (d0) and then followed for 60 days. P. carinii DNA was detected in lungs until d29 in two separate experiments and thereafter remained undetectable. A transient air excretion of Pneumocystis DNA was observed between d14 and d22 in the first experiment and between d9 and d19 in the second experiment; it was related to the peak of infection in lungs. IgM and IgG anti-P. carinii antibody increase preceded clearance of P. carinii in the lungs and cessation of airborne excretion. In rats receiving a second challenge 3 months after the first inoculation, Pneumocystis was only detected at a low level in the lungs of 2 of 3 rats at d2 post challenge and was never detected in air samples. Anti-Pneumocystis antibody determinations showed a typical secondary IgG antibody response. This study provides the first direct evidence that immunocompetent hosts can excrete Pneumocystis following a primary acquired infection. Lung infection was apparently controlled by the immune response since fungal burdens decreased to become undetectable as specific antibodies reached high titres in serum. This immune response was apparently protective against reinfection 3 months later.