Pneumocystis carinii f. sp. hominis Is Not Infectious for SCID mice

Abstract

The infectious power of Pneumocystis carinii f. sp. hominis was explored by inoculating SCID mice intranasally with either P. carinii f. sp. hominis or P. carinii f. sp. muris isolates. Only mice inoculated with mouse parasites developed Pneumocystis pneumonia, as assessed by microscopy and PCR. These results suggest that humans do not contract pneumocystosis from animals.

The expression “Pneumocystis carinii” describes a group of formae speciales (6) which may be distinguished by a cluster of host species-related genomic and phenotypic differences, including morphology (22) and selective infectivity to hosts of a given species (10, 15, 21, 26). Cross-infection experiments confirmed that Pneumocystis from a given mammal could not infect other host species (2, 3, 19, 20). The tested Pneumocystis sp. strains, mostly from laboratory mammals, were strongly host species specific (Table 1).

TABLE 1.

Host species specificity of Pneumocystis formae speciales as revealed by cross-infection experiments

Origin of inoculated P. carini isolates	Inoculation route	Recipient hosts	P. carinii detection	Detection methodsa	Reference(s)
Mouse	Intranasal	SCID mice	Yes	ME, PCRb	2, 3
Mouse	Intratracheal	SCID mice	Yes	ME, PCR, MAbs	20
Mouse	Intratracheal	Nude rats	No	ME, PCRb	3
Rat	Intratracheal	Nude rats	Yes	ME, PCRb	3
Rat	Intranasal	SCID mice	No	ME, PCRb	2, 3
Rabbit	Intranasal	SCID mice	No	ME, PCRb	2, 3
Rabbit	Intratracheal	Nude rats	No	ME, PCRb	3
Ferret	Intratracheal	SCID mice	No	ME, PCR, MAbs	20
Rhesus monkey	Intranasal	SCID mice	No	ME	19
Rhesus monkey	Intranasal	Nude rats	No	ME	19
Human	Intratracheal	SD ratsc	No	ME, PCR	8
Human	Intranasal	SCID mouse	No	ME, PCR	Present report

^aME, microscopical examination; MAbs, Pneumocystis-specific monoclonal antibodies. PCR was used both for confirming the absence of the parasite in uninfected hosts and for identifying strains developing in positive hosts by using strain-specific primers or hybridization with strain-specific DNA probes.\

^bAliouat et al., unpublished data.\

^cIn this experiment, recipient animals were already parasitized with Pneumocystis before being inoculated with human-derived Pneumocystis organisms.

Previous attempts to infect animals with parasites from humans were often performed by using recipient hosts latently infected with Pneumocystis (8, 9). In addition, molecular tools to identify accurately the parasite isolates were not available in the oldest studies (18, 23, 27, 28). Different results in terms of the infection route were reported (27, 28). Defined cryopreservation methods applied to Pneumocystis (16) and significant improvements in available in vitro systems (4, 5) make it possible to isolate viable human-derived Pneumocystis organisms. In the present work, the infectious power of cryopreserved human Pneumocystis samples to the SCID mouse, a highly susceptible Pneumocystis animal model, was evaluated.

Pneumocystis isolates.

Pneumocystis isolates were obtained from bronchoalveolar lavage fluid samples from five human immunodeficiency virus-negative patients with Pneumocystis carinii pneumonia (PCP). Mouse-derived parasites were used in order to verify the susceptibility of SCID mice to Pneumocystis nasal inoculation. P. carinii organisms were obtained from corticosteroid-treated BALB/c/BU mice (Lille Pasteur Institute, Lille, France), and parasite extraction was performed as previously described (2). Parasite isolates were identified as Pneumocystis carinii f. sp. hominis or P. carinii f. sp. muris by PCR amplification of the mitochondrial gene encoding the large subunit rRNA with primers pAZ102-H and pAZ102-E (25) and direct sequencing. When no amplification product was visualized on agarose gels, hybridization of PCR products was performed with two oligonucleotide probes, one specific to P. carinii f. sp. hominis (IPL-H) (17) and the other specific to P. carinii f. sp. muris (IPL-M). Probes for human- or mouse-derived P. carinii were, respectively, as follows: 5′-AACTATTTCTTAAAATAAATAATC-3′ (IPL-H) and 5′-CGTAATTTGAATTACAAGAAGGGA-3′ (IPL-M). They were synthesized and 5′-end labeled with digoxigenin obtained from Eurogentec (Seraing, Belgium). The membranes were incubated with the probes for at least 4 h, at either 42°C (IPL-H) or 55°C (IPL-M). The IPL-H probe is highly specific to P. carinii f. sp. hominis and has been used for a considerable time in our laboratory (17). The absence of cross-reactivity of the IPL-M probe was tested in the present work. Parasite samples from human or mouse origin were cryopreserved in liquid nitrogen for less than 1 year or 6 years, respectively (16). The selected cryopreservation method has proved to be efficient in obtaining viable, infectious, and ultrastructurally well-preserved Pneumocystis parasites (16).

Experimental design.

P. carinii-free 7-week-old CB17 SCID mice from a colony bred at the Pasteur Institute of Lille were used as recipient hosts. Microscopy and molecular methods confirmed the absence of latent Pneumocystis infection. Recipient SCID mice were anesthetized and then intranasally inoculated with parasites suspended in 25 μl of Dulbecco's modified Eagle's culture medium (2, 11). They were divided into eight groups of three SCID mice each that were housed in separated cages. All groups were administered dexamethasone at 2 mg/ml in drinking water from 15 days before inoculation to the end of the experiment (11). Mice in groups 1 to 5 were inoculated with 10⁵ to 10⁶ P. carinii f. sp. hominis organisms per animal; organisms were obtained from PCP cases 1 to 5, respectively. Mice in groups 6 and 7 were inoculated with 10⁵ and 10⁶ P. carinii f. sp. muris organisms per animal, respectively. Group 8 consisted of uninfected SCID mice used as negative controls. Until day 21 postinfection, all mice were administered enrofloxacine (125 mg/ml in drinking water) and fluconazole (65 μg/ml in drinking water) in order to preclude development of other infections potentially present in the inoculum. All experiments were performed in separate HEPA-filtered air isolators. Ten weeks postinfection, all SCID mice were euthanized and lungs were removed for microscopic and Pneumocystis PCR analysis. Both parasite inocula and lung samples were stained with toluidine blue O, which stains cystic forms, and methanol-Giemsa, which stains cytoplasm and nuclear structures of all Pneumocystis stages, as previously described (1). Animal experiments were performed according to the conditions stipulated in European guidelines (7).

Short-term culture of human-derived Pneumocystis isolates.

In order to further evaluate the growth ability of cryopreserved human-derived Pneumocystis samples, isolates 1 to 4 were cultivated on L2 monolayer lung epithelial-like cells of rat origin (ATCC CCL 149) in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum as previously described (4, 16), with some modifications. Briefly, cultures were made in quadruplicate on glass coverslips placed in the wells of 24-well flat-bottom plates. Plates were incubated for 96 h (37°C; 5% CO₂). Every 24 h, one microculture per cultured isolate was stopped in order to assess parasite development. Coverslips were washed to eliminate unattached parasites. The attached Pneumocystis organisms were detected microscopically by using methanol-Giemsa stain. As P. carinii f. sp. hominis organisms cluster closely in cultures, counting individual parasites is usually difficult. For this reason, in order to evaluate the in vitro growth of P. carinii f. sp. hominis isolates, parasite clusters were enumerated at low magnification.

Human-derived isolates 1 to 4 showed detectable in vitro development, and no morphological alteration was observed at the photonic microscopic level (Fig. 1). The number of parasite clusters increased from about 5 to 24 per coverslip within the 24 to 48 h, reaching the stationary phase by day 3. Vegetative and cystic forms, including mature cysts containing well-visible intracystic bodies, were clearly observed (Fig. 1).

FIG. 1.

In vivo and in vitro P. carinii f. sp. hominis organisms. (a and b) Parasites in a fresh bronchoalveolar lavage fluid sample stained with ortho-toluidine blue (a) or methanol-Giemsa (b) stains. (c and d) Clustered vegetative and cystic forms attached to L2 epithelial alveolar cells in short-term culture after 24 h (c) or 48 h (d). Arrows, Pneumocystis cysts containing well-visible intracystic bodies; arrowheads, trophozoites. Bar, 10 μm.

Ten weeks postinfection, no parasite was detected by microscopy in lung samples from SCID mice inoculated with human-derived Pneumocystis (groups 1 to 5). All SCID mice infected with P. carinii f. sp. muris samples developed PCP (groups 6 and 7) (Table 2). No apparent clinical changes were observed in SCID mice inoculated with human or mouse Pneumocystis isolates during the experiment. PCR was only positive in lung samples from mice inoculated with P. carinii f. sp. muris (groups 6 and 7). Direct sequencing and hybridization with the IPL-M probe identified P. carinii f. sp. muris. PCRs with IPL-H- or IPL-M probe hybridizations were negative in lung samples from SCID mice inoculated with P. carinii f. sp. hominis (groups 1 to 5). No parasite was detected in control mice (group 8) with microscopy or a PCR-hybridization assay (Fig. 2).

TABLE 2.

Results of cross-infection experiments with P. carinii

Pneumocystis origin	Group no.	Recipient host	Cyst count (106)	Total parasites (106)	PCPa	Positive PCRa	P. carinii f. sp. murisa, b	P. carinii f. sp. hominisa, b
Human	1 to 5	SCID mouse	0	0	0/15	0/15	0/15	0/15
Mouse	6	SCID mouse	0.9 ± 0.4	20.9 ± 5.1	3/3	3/3	3/3	0/3
Mouse	7	SCID mouse	2.3 ± 0.2	85.5 ± 30.3	3/3	3/3	3/3	0/3
Control SCID mice	8		0	0	0/5	0/5	0/5	0/5

^aNumber of positive results per total number of examined SCID mice.

^bP. carinii f. sp. muris and P. carinii f. sp. hominis were identified with oligonucleotidic-specific probes by hybridization.

FIG. 2.

Hybridization of amplified DNA with the digoxigenin-labeled 24-bp specific probe for P. carinii f. sp. muris (IPL-M) (top) and P. carinii f. sp. hominis (IPL-H) (bottom), defined in the mitochondrial gene encoding the large subunit rRNA. Lanes 1 to 3, control SCID mice (group 8); lanes 4 to 9, mice inoculated with P. carinii f. sp. hominis; lane 10, PCR negative control (with no added DNA); lane 11, DNA ladder; lanes 12 and 13, mice inoculated with P. carinii f. sp. muris; lane 14, P. carinii f. sp. muris organisms used as inoculum for mice from groups 6 and 7; lane 15, positive DNA control of P. carinii f. sp. hominis. The arrow indicates the size (350 bp) of the specific amplified product.

The present work provides the first evidence of the inability of viable P. carinii f. sp. hominis to develop in Pneumocystis-free SCID mice. In 1992 it was reported that human-derived Pneumocystis could develop in SCID mice latently infected with Pneumocystis (23). After nasal inoculation with human-derived Pneumocystis, mice developed PCP caused both by mouse-derived and human-derived Pneumocystis organisms. This result appeared to contradict cross-infection experiments performed using non-human-derived parasites (2, 3, 19, 20). However both models and technical approaches were different for the following reasons: (i) recipient SCID mice were latently infected with mouse-derived parasites; and (ii) detection and identification of parasites in mouse lungs were performed by using fluorescent monoclonal antibodies (23). A previous infection might also induce changes in the alveolar environment, rendering it receptive to Pneumocystis organisms from other host species. Two different groups (20; E. M. Aliouat et al., unpublished data), have previously tested this hypothesis and were unable to confirm Sethi's (23) observations. In these experiments, Pneumocystis-infected recipient hosts were inoculated with unspecific parasites, i.e., parasites derived from another host species; PCR and hybridization were used in order to identify strains on completion of the experiments. Only specific Pneumocystis organisms, i.e., the strain specific to recipient hosts, were found in their lungs (20) (Aliouat et al., unpublished data). These data suggest that coinfection with two Pneumocystis strains does not influence the host species specificity. More recently, human-derived Pneumocystis organisms were found to be unable to develop in corticosteroid-treated rats which were latently infected with rat Pneumocystis isolates and intratracheally inoculated with parasites obtained from four AIDS patients (8). In another experiment, human-derived Pneumocystis organisms were inoculated by the endotracheal route into corticosteroid-treated owl monkeys (9). Thirty weeks later, animals were euthanized and parasites were found in their lungs. The dihydropteroate synthase gene sequence found in the lung parasites showed close similarity with that of inoculated human-derived Pneumocystis isolates but, of note, the two sequences were not identical (9).

In this study, we established experimentally that the cryopreservation method (16) kept intact P. carinii f. sp. muris infectivity for at least 6 years. The same method was used for human-derived isolates that were stored in liquid nitrogen for less than 1 year. Viability and infectivity of human Pneumocystis isolates were proved by the fact that isolates showed significant growth when cultured on L2 monolayer alveolar epithelial cells.

Our results are consistent with the following facts. P. carinii f. sp. hominis has not been found in animals until now (17). P. carinii f. sp. hominis is the unique forma specialis that we have identified in humans after more than 5 years of routine molecular laboratory diagnosis of PCP. As expected on the basis of recent phylogenetic studies (13, 15), parasites of human origin were not infectious to SCID mice. These observations suggest that humans do not contract PCP from infected animals and that pneumocystosis should not have a zoonotic pattern. Consistently, more and more evidence supports the hypothesis that humans play a key role in the circulation of parasites in the community (14), representing probably the most important infection source for susceptible human subjects (12, 24).