Abstract

Forty-seven bronchoalveolar lavage fluid samples from 16 human immunodeficiency virus (HIV)–infected patients were used to test the latency model of Pneumocystis infection in the human host. Identification of DNA sequence polymorphisms at 4 independent loci were used to genotype Pneumocystis jiroveci from the 35 samples that contained detectable P. jiroveci DNA. Eighteen of those 35 samples came from patients who did not have Pneumocystis pneumonia (PCP) and had confirmed alternative diagnoses. Seven patients had asymptomatic carriage of P. jiroveci over periods of ⩽9.5 months after an episode of PCP, and in all 7 cases, a change in genotype from that in the original episode of PCP was observed. The absence of P. jiroveci DNA in one-fourth of the 47 samples and the observed changes in genotype during asymptomatic carriage do not support the latency model of infection. Asymptomatic carriage in HIV-infected patients may play a role in transmission of P. jiroveci and may even supply a reservoir for future infections

The route of transmission of human Pneumocystis infection is still not understood. It was long thought that Pneumocystis jiroveci organisms were acquired during childhood and persisted throughout life in the lungs in a dormant phase [1, 2]. Subsequent immunosuppression of the host, resulting from a variety of causes, permitted the fungus to propagate and to cause disease. According to this theory of transmission, there was little reason to protect immunosuppressed adults from possible exposure to Pneumocystis; acquisition of the infection was thought to take place early in life and not at the time of immunosuppression. The latency hypothesis, however, has recently been challenged. Several studies have suggested that, in the immunocompetent human host, Pneumocystis organisms are frequently acquired and cleared by the immune system, rather than being acquired during childhood and persisting for a lifetime. This reinfection hypothesis has important public health implications, because it suggests that susceptible individuals could be protected from Pneumocystis infection

Data refuting the latency hypothesis come from animal studies. It has been shown in a rat model that Pneumocystis organisms were eliminated from the lungs after Pneumocystis pneumonia (PCP) and that persistence of latent organisms was limited [3]. Similar results have been obtained using the severe combined immunodeficiency (SCID) mouse model [4]. In addition, some studies of human infection have indicated the lack of Pneumocystis in the lungs of immunocompetent adults [5]. However, although it is important to test the latency hypothesis, the hypothesis is difficult to disprove, because it is impossible to demonstrate the complete absence of a particular organism in the human lung. Support for the reinfection model comes from studies involving patients with recurrent episodes of PCP in which a genetically different type of P. jiroveci was associated with each episode [6–9 ]. Further evidence comes from a study in which allelic variation patterns in isolates of P. jiroveci correlated with patients’ place of diagnosis, rather than their place of birth, which strongly supports recent transmission, rather than latency [10]

As another approach, to question the latency hypothesis in the human host, we examined multiple bronchoalveolar lavage (BAL) fluid samples from 16 human immunodeficiency virus (HIV)–infected patients that were collected over a period of months to years. If latent P. jiroveci organisms were present throughout life in these patients, then the immunosuppression associated with HIV infection would provide a permissive environment, and the endogenous P. jiroveci organisms would proliferate

Subjects, Materials, and Methods

SamplesForty-seven BAL fluid samples from 16 HIV-infected patients (2–6 samples/patient) were collected between 1990 and 2001. The interval between samples obtained from a single patient ranged from 2 weeks to 33 months. All patients were known to be HIV-1 antibody positive and had been admitted to the specialist HIV/AIDS inpatient care facility at University College London hospitals for investigation of respiratory symptoms. For each patient, the CD4 lymphocyte count and use of prophylaxis against P. jiroveci pneumonia were recorded. All bronchoscopies were performed by R.F.M. PCP was defined by typical clinical and radiographic presentation [11], with demonstration of P. jiroveci cysts in BAL fluid by Grocott methenamine silver staining and response to anti-Pneumocystis therapy. Bacterial pneumonia was defined by the acute onset of respiratory symptoms, focal or diffuse radiographic abnormalities, identification of a specific bacterial pathogen in BAL fluid and/or blood, and response to specific antibiotics in conventional doses [12]. Disseminated Mycobacterium avium complex infection was defined by staining and culture of the organism from BAL fluid and blood and/or bone marrow. Cytomegalovirus (CMV) pneumonitis was defined by identification in BAL fluid of typical intranuclear and intracytoplasmic inclusions and by a positive result of testing for CMV (detection of early antigen by fluorescent foci) [13]. Pulmonary Kaposi sarcoma was defined by visualization during bronchoscopy of multiple violaceous lesions [14] and identification of multiple skin lesions. Patients received anti-Pneumocystis prophylaxis according to published guidelines [15]. All samples were analyzed by individuals who were blinded to clinical details

Detection of P. jiroveci using DNA amplificationDNA extraction of the samples was carried out using the QIAamp DNA mini kit (Qiagen), according to the manufacturer’s protocol, with 2 exceptions: the proteinase K step was extended to 1 h, and DNA was eluted from the column in 50 μL, rather than 200 μL, of AE buffer. Detection of P. jiroveci DNA was carried out using a nested polymerase chain reaction (PCR) at the genes encoding the mitochondrial large subunit (mt LSU) rRNA and the mitochondrial small subunit (mt SSU) rRNA. For PCR at the mt LSU rRNA locus, the primer pair pAZ102-H and pAZ102-E was used in the first round of amplification, followed by pAZ102-X and pAZ102-Y in the second round of amplification, as described elsewhere [16–18 ]. For PCR at the mt SSU rRNA locus, the primer pair pAZ112-10F and pAZ112-10R was used, followed by pAZ112-13 and pAZ112-14, as described elsewhere [16, 19]. To monitor for false-negative PCR caused by inhibition of DNA amplification, a control PCR assay, using the conserved human β-globin gene sequence, was used. A positive result of human β-globin PCR, using primer pair BGLO1 (5′-ACACAACTGTGTTCACTAGCA-3′) and BGLO2 (5′-CAACTTCATCCACGTTCACC-3′) with an annealing temperature of 55°C and 40 cycles, demonstrated that DNA amplification was possible for a given sample [20]

Taq DNA polymerase (Promega) was used throughout the study. To avoid contamination, negative controls were included in each experiment in both DNA extraction and amplification, DNA extraction and PCR were performed in a laminar flow cabinet, and disposable tips, tubes, and reagent aliquots were used

P. jiroveci multilocus genotypingMultilocus genotyping was carried out at 4 loci, those encoding mt LSU rRNA, mt SSU rRNA, dihydropteroate synthase (DHPS), and superoxide dismutase (SOD). PCR at the mt LSU rRNA and mt SSU rRNA loci was carried out as described above. For the DHPS locus, the primer pair DHPS1 and DHPS2 was used in the first round of PCR, followed by DHPS3 and DHPS4 (corresponding to AHUM and BN) in the second round [21, 22]. For the SOD locus, the primer pair SOD7 and SOD8 was used in the first round, followed by SOD9 and SOD10 in the second round [23]. The PCR protocol for both the DHPS and the SOD locus was the same as that for the mt LSU rRNA locus

Amplification products were purified using the SpinPrep PCR Clean-up Kit (Novagen) and sequenced directly using the ABI Prism Big Dye Terminator Cycling Sequencing Ready Reaction Kit, version 2.0 (Applied Biosystems). The program was 96°C for 30 s, 50°C for 15 s, and 60°C for 4 min for 25 cycles. The reactions were precipitated using Pellet Paint NF Co-Precipitant (Novagen) and run on an ABI 377 automated sequencer (Applied Biosystems). DNA sequence data were analyzed using Gap4 of the Staden package (Genetics Computer Group)

At the mt LSU rRNA locus, genotypes were distinguished on the basis of polymorphisms at nucleotide positions 85 and 248, as described elsewhere [6, 16]. At the mt SSU rRNA locus, genotypes were distinguished by polymorphisms at nucleotide positions 216 and 299 (numbering according to Hunter and Wakefield [19]). At the DHPS locus, genotypes were distinguished on the basis of polymorphisms at nucleotide positions 165 and 171 (corresponding to amino acid substitutions at positions 55 and 57), as described elsewhere [21]. At the SOD locus, genotypes were distinguished by polymorphisms at nucleotide positions 110 and 215 [23]

Results

Detection of P. jiroveci DNAP. jiroveci DNA was detected using PCR at 2 different loci, the locus encoding mt LSU rRNA and the locus encoding mt SSU rRNA. P. jiroveci DNA was detected using a single-round PCR at the mt LSU rRNA and mt SSU rRNA loci in samples from 14 patients with clinical and laboratory-confirmed PCP (table 1). This was consistent with the findings of another study, in which a positive result of a single-round PCR at the mt LSU rRNA locus in a BAL fluid sample indicated clinical disease [17]

Table 1

Detection of Pneumocystis jiroveci DNA in 47 bronchoalveolar lavage fluid samples obtained during respiratory episodes in 16 human immunodeficiency virus–infected patients

Table 1

Detection of Pneumocystis jiroveci DNA in 47 bronchoalveolar lavage fluid samples obtained during respiratory episodes in 16 human immunodeficiency virus–infected patients

Another patient with confirmed PCP had P. jiroveci DNA that was detectable only by nested PCR at the mt LSU and SSU rRNA loci, which indicates lower levels of DNA [21] (table 1). Two patients had an atypical presentation of PCP (table 1); both had respiratory symptoms and negative findings of silver staining of BAL fluid. In one, a thoracic computed tomographic scan showed multiple nodules and patchy focal ground glass shadowing; the patient responded clinically and radiographically to intravenous pentamidine. In the other patient, a chest radiograph showed multiple pneumatocoels with associated patchy ground glass shadowing. A clinical and radiographic response occurred with a combination of clindamycin and primaquine. In both patients, P. jiroveci DNA was detected in BAL fluid by nested PCR

Thirty samples were from patients who had clinical and laboratory diagnoses other than PCP, including bacterial pneumonia, disseminated Mycobacterium avium complex infection, CMV pneumonitis, and pulmonary Kaposi sarcoma. P. jiroveci DNA was found in 16 samples when nested PCR, a more sensitive test capable of detecting much lower levels of Pneumocystis DNA, was used (table 1). The presence of P. jiroveci DNA in these samples, in the absence of clinical symptoms of PCP, suggests that these patients had asymptomatic carriage of Pneumocystis

No P. jiroveci DNA was detected in a total of 12 samples from 7 patients by nested PCR at the mt LSU rRNA and mt SSU rRNA loci (table 1). All of the samples from 2 patients (2 samples from patient 4, obtained 4 months apart, and 3 samples from patient 10, obtained over a period of 33 months) were negative for P. jiroveci DNA, indicating that the finding was consistent (table 2). Two of these patients (patients 4 and 5) died within a month after a sample was obtained in which no P. jiroveci DNA was found; necropsy failed to find P. jiroveci in the lungs of these patients by histochemical staining (table 2). The lack of detectable P. jiroveci in some of the samples from these 7 patients was not due to weaker immunosuppression; the CD4 lymphocyte counts of these patients ranged from 10 to 380 cells/μL (median, 50 cells/μL) and were similar to those of the other patients in this study (range, 10–370 cells/μL; median, 30 cells/μL). Six of the 7 patients were receiving anti-Pneumocystis prophylaxis; patient 15 was not (table 2)

Table 2

Laboratory data for patients for whom at least 1 bronchoalveolar lavage fluid sample had no detectable Pneumocystis jiroveci DNA

Table 2

Laboratory data for patients for whom at least 1 bronchoalveolar lavage fluid sample had no detectable Pneumocystis jiroveci DNA

Change in P. jiroveci genotype in multiple samples from a single patientMultilocus genotyping was carried out on 14 samples from 7 patients in whom asymptomatic carriage of P. jiroveci was detected after an episode of PCP; the interval between samples ranged from 0.75 to 9.5 months (table 3). In all 7 patients, the P. jiroveci genotype found in samples obtained during asymptomatic carriage was different from that found in samples from the episode of PCP. All patients were receiving prophylaxis

Table 3

Laboratory data for patients with alternative diagnoses and asymptomatic Pneumocystis jiroveci carriage after an episode of Pneumocystis pneumonia (PCP)

Table 3

Laboratory data for patients with alternative diagnoses and asymptomatic Pneumocystis jiroveci carriage after an episode of Pneumocystis pneumonia (PCP)

Asymptomatic carriage of P. jiroveci was found in 2 patients (patients 1 and 3) who had bacterial pneumonia 4 and 9.5 months before an episode of PCP. Both patients were receiving prophylaxis. Multilocus genotyping showed that, in both of these patients, the type of P. jiroveci observed during the subclinical infection was the same as that causing the clinical disease

Duration of asymptomatic carriage of P. jiroveciIn 5 patients, the duration of asymptomatic carriage of P. jiroveci ranged from 0.5 to 16.5 months (table 4). In 2 of these patients (patients 6 and 9), a change of genotype was detected in samples obtained up to 14.5 months apart; in the other 3 patients, the P. jiroveci genotype was the same in all samples. Four of the 5 patients were receiving prophylaxis; patient 11 was not

Table 4

Laboratory data for patients with alternative diagnoses and asymptomatic Pneumocystis jiroveci carriage for whom P. jiroveci DNA was detected in ⩾2 bronchoalveolar lavage fluid samples

Table 4

Laboratory data for patients with alternative diagnoses and asymptomatic Pneumocystis jiroveci carriage for whom P. jiroveci DNA was detected in ⩾2 bronchoalveolar lavage fluid samples

Discussion

This study sought to question the hypothesis that P. jiroveci infection in the human host is caused by latent carriage of fungus acquired during childhood by examining multiple BAL fluid samples obtained from immunosuppressed HIV-infected patients with and without PCP. We found that not all samples contained detectable P. jiroveci DNA. This finding indicated that, although patients were immunologically challenged by HIV infection, no proliferation of Pneumocystis had occurred, which suggests that these individuals were not carrying latent P. jiroveci infection in their lungs. Samples were obtained over a period of 2.5 years from 1 patient with a CD4 lymphocyte count of <200 cells/μL, and there was no evidence of P. jiroveciat any point. Since the 1980s, increasingly sensitive techniques have been developed for detecting P. jiroveci DNA in human respiratory samples [2, 5, 6, 8, 17, 18, 22, 24–26 ]. This increase in sensitivity has enabled detection of P. jiroveci in individuals in whom it was not expected, for example, in adults with mild immunosuppression induced by HIV infection [26, 27] and immunocompetent individuals with primary pulmonary disorders [28–31 ]. By contrast, although the use of nested PCR at the mt LSU rRNA locus to distinguish genotypes is very sensitive and can identify 1 or 2 P. jiroveci organisms in a respiratory sample [32], it is possible that the BAL fluid samples in this study in which no P. jiroveci DNA was found by nested PCR at the mt LSU rRNA locus did, in fact, contain P. jiroveci at very low levels, levels below the threshold for detection by this technique

In the present study, identification of P. jiroveci DNA at the mt LSU rRNA locus after a single round of PCR on a BAL fluid sample was only made in samples from patients with clinical and laboratory-confirmed PCP, samples from patients who had recently recovered from PCP, and samples obtained immediately before presentation with PCP. This finding confirms and extends observations published elsewhere [17, 24]. Patients with radiologically atypical PCP, despite negative findings of histochemical staining, also had detectable P. jiroveci DNA in BAL fluid, but this could only be identified by nested PCR, a more sensitive technique, at the mt LSU and SSU rRNA loci. This finding has been described elsewhere for patients with atypical (granulomatous) PCP [33]

Asymptomatic carriage of P. jiroveci was shown by detection of P. jiroveci DNA in a BAL fluid sample only by nested PCR at the mt LSU rRNA locus. Not all of these patients had detectable P. jiroveci DNA at the mt SSU rRNA locus, which indicates that PCR was more sensitive at the mt LSU rRNA locus than at the mt SSU rRNA locus. In 7 patients, P. jiroveci DNA was detected in BAL fluid obtained 0.75–9.5 months after treatment of an episode of PCP. In all 7 patients, the genotype found at follow-up was different from that identified during the acute episode of PCP. It is postulated that high-dose treatment of PCP resulted in clearance of the infection and that, because patients were vulnerable as a result of their level of immunosuppression, reinfection from an external reservoir took place despite prophylaxis. These data support the reinfection hypothesis

On the other hand, use of Pneumocystis prophylaxis did not always result in clearance of infection. In some patients, asymptomatic carriage was demonstrated in consecutive samples obtained over the course of several months, which suggests that, although prophylaxis had prevented development of PCP in these individuals, it merely contained the infection. In 2 of these patients, asymptomatic carriage subsequently led to development of PCP

In this study, multilocus typing was carried out at 4 independent loci. Five different types of P. jiroveci have been identified on the basis of differences at the mt LSU rRNA locus [6], 2 have been identified on the basis of differences at the mt SSU rRNA locus [19], and 4 types have been identified on the basis of differences at the DHPS locus [21], which is under selective pressure from sulfa drug exposure [34–36 ]. Two genotypes have been identified at the SOD locus [23]. Using these 4 loci to genotype P. jiroveci enabled discrimination between 9 different genotypes (5.6%) of a potential total of 160 genotypes (4 types of P. jiroveci were identified at the mt SSU rRNA locus in this study). By contrast, genotyping at the internal transcribed spacer (ITS) regions shows 27 types at ITS region 1 and 30 types at ITS region 2, yielding a potential of 810 different types, of which 87 (10.7%) have been described [26, 37]

The chief advantage of multilocus typing is that there is a greater sensitivity for detecting differences when several loci are examined concurrently than when a single locus is examined [38, 39]. In addition, by looking at polymorphisms at multiple loci, further information is available for strain characterization, and “strain” typing is achieved, rather than simply “gene” typing. Hauser et al. [40] used multilocus typing at 4 independent loci—ITS region 1, the intron of the nuclear 26S rRNA gene, the mt LSU rRNA locus, and the intron 6 region of the β-tubulin gene—to describe differences in isolates of P. jiroveci. Amplification by PCR was followed by identification using single-strand confirmation polymorphism analysis. A total of 35 genotypes (16.2%) of a potential total of 216 types were identified

Multilocus typing was not possible for all samples from patients with asymptomatic carriage of P. jiroveci in the present study. In some of these, we were able to genotype only at multicopy genes (i.e., at the mt LSU and SSU rRNA loci), and genotyping at the single-copy DHPS and SOD genes was not possible, an observation consistent with the presence of low levels of P. jiroveci DNA in these samples [18]. The change in genotype of P. jiroveci identified by multilocus typing in these sequential samples suggests that these patients were exposed to different types of P. jiroveci over time [18] and provides further evidence to support the model of reinfection, rather than latency

When we identified changes at the DHPS locus, they were always accompanied by changes at ⩾1 other independent locus. If changes at the DHPS locus were due to selective pressure resulting from exposure to sulfa drugs [32–34 ], then it would be predicted that no differences would be observed at other independent loci. This suggests that mutations at the DHPS locus were not evolving within individual patients and further supports the idea that exogenous reinfection by different genotypes of P. jiroveci is occurring [41]

A large number of samples from patients in this study without clinically apparent disease had detectable P. jiroveci DNA at low levels in BAL fluid (35 of 47 samples; 74.5%). This suggests that asymptomatic individuals who had demonstrable carriage of P. jiroveci DNA over periods of up to several months might act a reservoir of P. jiroveci infection. Furthermore, this patient group might represent an important and as yet unrecognized element in the transmission of disease [27, 29]

It is important, from a public health perspective, to determine whether the reinfection or the latency model of human P. jiroveci infection occurs; if reinfection during adult life can occur, then susceptible patients should be protected from potential sources of infection. The data from this study demonstrate that multilocus typing may be used to define differences in P. jiroveci isolates and that not all immunosuppressed HIV-infected patients have detectable P. jiroveci in their lungs. The absence of P. jiroveci in some patients and the observed change in genotype in patients with asymptomatic carriage strongly support the reinfection model of human infection. Asymptomatic carriage in HIV-infected patients may play a role in transmission of P. jiroveci. This patient group may be a reservoir for future infections in other susceptible patients

References

1
Cushion
MT
Wakefield
AE
New nomenclature for the genus Pneumocystis
J Eukaryot Microbiol
 , 
2001
, vol. 
49
 
Suppl
(pg. 
184S
-
9S
)
2
Beard
CB
Miller
RF
Wakefield
AE
A new name (Pneumocystis jiroveci) for Pneumocystis from humans
Emerg Infect Dis
 , 
2002
, vol. 
8
 (pg. 
891
-
6
)
3
Vargas
SL
Hughes
WT
Wakefield
AE
Oz
HS
Limited persistence in and subsequent elimination of Pneumocystis carinii from the lungs after P. carinii pneumonia
J Infect Dis
 , 
1995
, vol. 
172
 (pg. 
506
-
10
)
4
Chen
W
Gigliotti
F
Harmsen
AG
Latency is not an inevitable outcome of infection with Pneumocystis carinii
Infect Immun
 , 
1993
, vol. 
61
 (pg. 
5406
-
9
)
5
Peters
SE
Wakefield
AE
Sinclair
K
Millard
PR
Hopkin
JM
A search for Pneumocystis carinii in post mortem lungs by DNA amplification
J Pathol
 , 
1992
, vol. 
166
 (pg. 
195
-
8
)
6
Baughman
RP
Linke
MJ
Walzer
PD
Genetic variation among Pneumocystis carinii hominis isolates in recurrent pneumocystosis
J Infect Dis
 , 
1995
, vol. 
172
 (pg. 
595
-
8
)
7
Source of Pneumocystis carinii in recurrent episodes of pneumonia in AIDS patients
AIDS
 , 
1996
, vol. 
10
 (pg. 
881
-
8
)
8
Sequences of Pneumocystis carinii f sp hominis strains associated with recurrent pneumonia vary at multiple loci
J Clin Microbiol
 , 
1997
, vol. 
35
 (pg. 
2745
-
7
)
9
Tsolaki
AG
Miller
RF
Underwood
AP
Banerji
S
Wakefield
AE
Genetic diversity at the internal transcribed spacer regions of the rRNA operon among isolates of Pneumocystis carinii from AIDS patients with recurrent pneumonia
J Infect Dis
 , 
1996
, vol. 
174
 (pg. 
141
-
56
)
10
Beard
CB
Carter
JL
Keely
SP
, et al.  . 
Genetic variation in Pneumocystis carinii isolates from different geographic regions: implications for transmission
Emerg Infect Dis
 , 
2000
, vol. 
6
 (pg. 
265
-
72
)
11
Miller
RF
Millar
AB
Weller
IVD
Semple
SJG
Empirical treatment without bronchoscopy for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome
Thorax
 , 
1989
, vol. 
44
 (pg. 
559
-
64
)
12
Miller
RF
Loveday
C
Holton
J
Sharvell
Y
Patel
G
Community-based respiratory viral infections in HIV positive patients with lower respiratory tract disease: a prospective bronchoscopic study
Genitourin Med
 , 
1996
, vol. 
72
 (pg. 
9
-
11
)
13
Millar
AB
Patou
G
Miller
RF
, et al.  . 
Cytomegalovirus in the lungs of patients with AIDS
Am Rev Respir Dis
 , 
1990
, vol. 
141
 (pg. 
1474
-
7
)
14
Miller
RF
Tomlinson
MC
Cottrill
CP
Donald
JJ
Spittle
MF
Bronchopulmonary Kaposi’s sarcoma in patients with AIDS
Thorax
 , 
1992
, vol. 
47
 (pg. 
721
-
5
)
15
US Public Health Service (USPHS) and Infectious Diseases Society of America (IDSA)
1999 USPHS/IDSA guidelines for the prevention of opportunistic infections in persons infected with human immunodeficiency virus
MMWR Morb Mortal Wkly Rep
 , 
1999
, vol. 
48
 
(RR-10)
(pg. 
1
-
59
61–6
16
Tsolaki
AG
Beckers
P
Wakefield
AE
Pre-AIDS era isolates of Pneumocystis carinii f sp hominis: high genotypic similarity with contemporary isolates
J Clin Microbiol
 , 
1998
, vol. 
36
 (pg. 
90
-
3
)
17
Wakefield
AE
Pixley
FJ
Banerji
S
, et al.  . 
Detection of Pneumocystis carinii with DNA amplification
Lancet
 , 
1990
, vol. 
336
 (pg. 
451
-
3
)
18
Miller
RF
Ambrose
HE
Wakefield
AE
Pneumocystis carinii f sp hominis DNA in immunocompetent health care workers in contact with patients with P. carinii pneumonia
J Clin Microbiol
 , 
2001
, vol. 
39
 (pg. 
3877
-
82
)
19
Hunter
JAC
Wakefield
AE
Genetic divergence at the mitochondrial small subunit ribosomal RNA gene among isolates of Pneumocystis carinii from five mammalian host species
J Eukaryot Microbiol
 , 
1996
, vol. 
43
 
Suppl
(pg. 
24S
-
5S
)
20
Saiki
RK
Scharf
S
Faloona
F
, et al.  . 
Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anaemia
Science
 , 
1985
, vol. 
230
 (pg. 
1350
-
4
)
21
Lane
BR
Ast
JC
Hossler
PA
, et al.  . 
Dihydropteroate synthase polymorphisms in Pneumocystis carinii
J Infect Dis
 , 
1997
, vol. 
175
 (pg. 
482
-
5
)
22
Miller
RF
Ambrose
HE
Novelli
V
Wakefield
AE
Probable mother-to-infant transmission of Pneumocystis carinii f sp hominis infection
J Clin Microbiol
 , 
2002
, vol. 
40
 (pg. 
1555
-
7
)
23
Denis
CM
Mazars
E
Guyot
K
, et al.  . 
Genetic divergence at the SODA locus of six different formae specialis of Pneumocystis carinii
Med Mycol
 , 
2000
, vol. 
38
 (pg. 
289
-
300
)
24
Tsolaki
AG
Miller
RF
Wakefield
AE
Oropharyngeal samples for genotyping and monitoring of response to treatment in AIDS patients with Pneumocystis carinii pneumonia
J Med Microbiol
 , 
1999
, vol. 
48
 (pg. 
897
-
905
)
25
Wakefield
AE
DNA sequences identical to Pneumocystis carinii f sp carinii and Pneumocystis carinii f sp hominis in samples of air spora
J Clin Microbiol
 , 
1996
, vol. 
34
 (pg. 
1754
-
9
)
26
Lee
CH
Helweg-Larsen
J
Tang
X
, et al.  . 
Update on Pneumocystis carinii f sp hominis typing based on nucleotide sequence variations in internal transcribed spacer regions of rRNA genes
J Clin Microbiol
 , 
1998
, vol. 
36
 (pg. 
734
-
41
)
27
Leigh
TR
Kangro
HO
Gazzard
BG
Jeffries
DJ
Collins
JV
DNA amplification by the polymerase chain reaction to detect sub-clinical Pneumocystis carinii colonization in HIV-positive and HIV-negative homosexuals with and without respiratory symptoms
Respir Med
 , 
1993
, vol. 
87
 (pg. 
525
-
9
)
28
Nevez
G
Raccurt
C
Jounieaux
V
Dei-Cas
E
Mazars
E
Pneumocystosis versus pulmonary Pneumocystis carinii colonization in HIV-negative and HIV-positive patients
AIDS
 , 
1999
, vol. 
13
 (pg. 
535
-
6
)
29
Sing
A
Roggenkamp
A
Autenrieth
IB
Heesemann
J
Pneumocystis carinii carriage in immunocompetent patients with primary pulmonary disorders as detected by single or nested PCR
J Clin Microbiol
 , 
1999
, vol. 
37
 (pg. 
3409
-
10
)
30
Armbruster
C
Hassl
A
Kriwanek
S
Pneumocystis carinii colonization in the absence of immune suppression
Scand J Infect Dis
 , 
1997
, vol. 
29
 (pg. 
591
-
3
)
31
Calderon
EJ
Regordan
C
Medrano
FJ
Ollero
M
Varela
JM
Pneumocystis carinii infection in patients with chronic bronchial disease
Lancet
 , 
1996
, vol. 
347
 pg. 
977
 
32
Peters
SE
Wakefield
AE
Banerji
S
Hopkin
JM
Quantification of the detection of Pneumocystis carinii by DNA amplification
Mol Cell Probes
 , 
1992
, vol. 
6
 (pg. 
115
-
7
)
33
Wakefield
AE
Miller
RF
Guiver
LA
Hopkin
JM
Granulomatous Pneumocystis carinii pneumonia: DNA amplification studies on bronchoscopic alveolar lavage samples
J Clin Pathol
 , 
1994
, vol. 
47
 (pg. 
664
-
6
)
34
Huang
L
Beard
CB
Creasman
J
, et al.  . 
Sulfa or sulfone prophylaxis and geographic region predict mutations in the Pneumocystis carinii dihydropteroate synthase gene
J Infect Dis
 , 
2000
, vol. 
182
 (pg. 
1192
-
8
)
35
Kazanjian
P
Armstrong
W
Hossler
PA
, et al.  . 
Pneumocystis carinii mutations are associated with duration of sulfa or sulfone prophylaxis exposure in AIDS patients
J Infect Dis
 , 
2000
, vol. 
182
 (pg. 
551
-
7
)
36
Ma
L
Kovacs
JA
Genetic analysis of multiple loci suggests that mutations in the Pneumocystis carinii f sp hominis dihydropteroate synthase gene arose independently in multiple strains
Antimicrob Agents Chemother
 , 
2001
, vol. 
45
 (pg. 
3213
-
5
)
37
Nimri
LF
Moura
INS
Huang
L
, et al.  . 
Genetic diversity of Pneumocystis carinii f sp hominis based on variations in nucleotide sequences of internal transcribed spacers of rRNA genes
J Clin Microbiol
 , 
2002
, vol. 
40
 (pg. 
1146
-
51
)
38
Maiden
MCJ
Bygraves
JA
Feil
E
, et al.  . 
Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms
Proc Natl Acad Sci
 , 
1998
, vol. 
95
 (pg. 
3140
-
5
)
39
Feavers
IM
Gray
SJ
Urwin
R
, et al.  . 
Multilocus sequence typing and antigen gene sequencing in the investigation of a meningococcal disease outbreak
J Clin Microbiol
 , 
1999
, vol. 
37
 (pg. 
3883
-
7
)
40
Hauser
PM
Blanc
DS
Sudre
P
, et al.  . 
Genetic diversity of Pneumocystis carinii in HIV-positive and -negative patients as revealed by PCR-SSCP typing
AIDS
 , 
2001
, vol. 
15
 (pg. 
461
-
6
)
41
Huang
L
Friedly
J
Morris
AM
, et al.  . 
Pneumocystis carinii dihydropteroate synthase genotypes in HIV-infected persons residing in San Francisco: possible implications for disease transmission
J Eukaryot Microbiol
 , 
2001
, vol. 
48
 
Suppl
(pg. 
137S
-
8S
)
This study was carried out within the guidelines of the Middlesex Hospital Clinical Investigations Panel. All patients gave informed written consent for bronchoscopy
Financial support: Royal Society (to A.E.W.); Wellcome Trust (to A.E.W.); Medical Research Council (to H.E.A. and C.-M.D.); Fifth Framework Programme of the European Commission (grant QLK2-CT-2000-01369 to A.R.L.); Camden Primary Care Trust (R.F.M.)
Deceased