Molecular and Cellular Probes
(1992) 6, 299-304
Detection of Pseudomonas aeruginosa in sputum from cystic fibrosis patients by the polymerase chain reaction 2 lain McIntosh', John R . W . Govan and David J . H . Brock'* 'Human Genetics Unit, Western General Hospital, Edinburgh EH4 2XU and 2 Department of Medical Microbiology, University Medical School, Edinburgh, EH8 9AG, UK (Received 20 January 1992, Accepted 6 March 1992)
A DNA amplification procedure using heat stable Taq polymerase and the polymerase chain reaction is described for the detection of Pseudomonas aeruginosa in specimens from cystic fibrosis patients . A set of primers was selected on the basis of the nucleotide sequence of the algD gene encoding GDP mannose dehydrogenase, a major enzyme in the biosynthesis of alginate by P . aeruginosa. Using this set of primers in conjunction with the polymerase chain reaction, P . aeruginosa could be specifically detected, with a sensitivity approximating 10 bacteria, in sputum harbouring large numbers of other respiratory pathogens, including Staphylococcus aureus and Haemophilus influenzae . These results suggest that amplification of specific sequences within the a/gD gene by the polymerase chain reaction may provide a highly sensitive and specific tool for the detection of P . aeruginosa in the early stages of pulmonary colonization . KEYWORDS : Pseudomonas aeruginosa, polymerase chain reaction, cystic fibrosis
INTRODUCTION Pulmonary colonization with mucoid Pseudomonas aeruginosa is a major cause of morbidity and mortality in patients with cystic fibrosis (CF). Initial asymptomatic and often intermittent colonization of the upper respiratory tract with a non-mucoid strain of P . aeruginosa usually precedes chronic colonization with mucoid variants of the original strain .', ' In CF patients, detection of P . aeruginosa in the early stages of pulmonary colonization would increase the opportunity for early therapeutic intervention and could assist in the identification of primary colonization sites . At present, detection of pulmonary colonization with P . aeruginosa is based on quantitative
serve as an alternative to other methods for the detection of non-mucoid and mucoid forms of P . aeruginosa, particularly in the early stages of pulmonary colonization . In prokaryotes, alginate biosynthesis is restricted to Azotobacter vinelandii, an organism not associated with human colonization, and to P . aeruginosa and a few other pseudomonas species .` We chose the
algD gene as the conserved region of the P . aeruginosa genome to be amplified, since it codes for GDP mannose dehydrogenase, an enzyme essential for alginate biosynthesis :' algD is present in non-mucoid and mucoid P . aeruginosa and is not found in higher organisms .
culture of organisms from homogenized sputa, or indirectly by detection of pseudomonas-specific antibodies using enzyme-linked immunosorbent assay .', ' Since the polymerase chain reaction (PCR) has been used successfully to identify other bacteria, we attempted to develop a rapid and sensitive method for detecting P . aeruginosa . This approach might
MATERIALS AND METHODS Bacteria The bacterial strains used are listed in Table 1 .¢ 14
*Author to whom correspondence should be addressed . 0890-8508/92/040299 + 06 $08 .00/0
299
© 1992 Academic Press Limited
3 00
1. McIntosh et al .
Strains of P. aeruginosa comprised the well-characterised genetic strain PAO1 and its isogenic mucoid variants PA0568, and PA0579 and the mucoid CF isolate 492a . In addition to the bacterial species listed in Table 1, 12 non-mucoid isolates of P . aeruginosa (six CF isolates and six non-CF isolates) and 10 strains of P. cepacia were investigated .
Sputum culture Expectorated sputum was obtained after chest physiotherapy, and after liquefaction with Sputolysin (Behring Diagnostics, La Jolla, California, USA) and suitable dilution (10 2, 104) in sterile physiological saline, quantitative culture for respiratory pathogens was performed by plating out 0 . 1 ml volumes onto blood agar, lysed blood agar incorporating 0 . 5% bacitracin and Pseudomonas isolation agar (Difco Laboratories, Detroit, Michigan, USA) and incubating aerobically at 37 ° C for 48 h . Isolates were identified by standard procedures .'
Table 1 .
Polymerase chain reaction Genomic DNA from either bacterial suspensions or Sputolysin-treated sputum samples was obtained by phenol-chloroform extraction and ethanol precipitation . Briefly, 10 pl of sample was diluted to 100µl with 0. 1 M Tris-CI pH 7. 5 and added to an equal volume of phenol/chloroform, heated at 37 ° C for 5 min and centrifuged for 3 min at 10,000 rev/min (Eppendorf microfuge) . The aqueous layer was extracted twice with an equal volume of chloroform and DNA was precipitated by addition of 1/10th volume of 3 M NaCl and 3 volumes of ethanol . The pellet was resuspended in 10 pl of sterile water and amplified by the polymerase chain reaction ." The primers shown in Table 2 were designed using the algD sequence data of Deretic et al.' and obtained from Oswel DNA Service (Department of Chemistry, University of Edinburgh) . The nesting procedure, which allows for greater sensitivity, was performed using the internal primers Pa2 and Pa4 to amplify 2 pl of the PCR product from an initial round of PCR with
Bacterial species and strains used in this study
Species
Strain and reference
P . aeruginosa P. aeruginosa
P . aeruginosa P . aeruginosa P . fluorescens P . putida
P . testosteroni P . acidovorans P . alkaligenes P . mendocina P. diminuta P. stutzeri
Xanthomonas maltophilia Staphylococcus aureus Streptococcus pyogenes Haemophilus influenzae
PAO1 PA0568* PA0579* CF492a* NCIMB 10525 strain 12 NCIMB 10007 strain Cl-B NCIMB 8893 strain 79 NCIMB 9681 type strain NCIMB 9946 type strain NCIMB 10541 type strain CH-50 NCIMB 9393 type strain NCIMB 9040 NCIMB 9203 type strain Clinical isolate Clinical isolate Clinical isolate
(9) (9) (9) (9) (10) (10) (10) (10) (10) (11) (12) (13) (14)
* Mucoid . NCIMB, National Collections of Industrial and Marine Bacteria, Aberdeen U.K.
Table 2.
Sequences used for the oligonucleotide primers
Primer sequence Pal Pa2 Pa3 Pa4
5' 5' 5' 5'
GACAGGTTGAGCTTGTGG CGAACTGGACAAGCAGAC GAATTCCTCCGCGAGAGC GCAGATCACGTCCATCAC
* Sequence from Deretic et al .'
Homologous to algD position* 1297-1280 1067-1084 1005-1022 1274-1257
PCR Detection of P . aeruginosa RESULTS
primers Pal and Pa3 . The appropriate primer pair (50 pmol each) and DNA were mixed in 100 pl of 10 mm Tris-Cl pH 8 . 3, 50 mm KCI, 2-5 mm MgCl, 10% DMSO and amplified through 30 cycles of 91 ° C (30 s), 40° C or 50 ° C (30 s) and 72 ° C (1 min) with 2 units of AmpliTaq DNA polymerase (Perkin-Elmer/Cetus, Hemel Hempstead, England) using an automated Thermal-Cycler (Perkin-Elmer/Cetus). After the 30th cycle, incubation continued at 72 ° C for 10 min to complete elongation of the PCR products . Ten per cent of the reaction product was run on a 2% agarose gel and stained with ethidium bromide to visualize the amplified DNA . To show that DNA prepared directly from small amounts of bacteria can be amplified, the results obtained from serial 10-fold dilutions of a culture P . aeruginosa PAO1 are shown (Fig . 2b) . On the original gel it was possible to see faint bands in lanes containing PCR product from reactions with 30 and three organisms (lanes 10 and 1, respectively) . Again no signal was visible in the negative controls . These results show that nested PCR is sufficiently sensitive and specific to detect very small numbers of P . aeruginosa . To demonstrate that the procedures could detect P. aeruginosa in clinical material, DNA was extracted from 10[d of sputolysin-treated sputum from patients with CF known to be colonized by respira-
For maximal sensitivity and specificity it was found necessary to adopt the 'nested primer' technique (Materials and Methods and Fig . 1) . An initial pair of primers (Pal and Pa3) were used in 30 cycles of PCR to generate a product of 292 by from a relatively large number of genomic templates (3 X 10 5 , . Fig . 1) . Increasing the number of cycles allowed the detection of a smaller number of templates (3 X 10 3 ), but only by using 2% of the original reaction product in a second round of PCR with the internal (`nested') primers Pa2 and Pa4 (Table 2) could as few as 30 genomes be detected (Fig . 1) . The identity of the PCR product was confirmed initially by digestion with restriction enzymes predicted to cut the amplified DNA, and subsequently by DNA sequencing (data not shown) . No signal was visible when no template DNA was included in the reactions : the extreme sensitivity of PCR requires that such negative controls be included in every experiment. The sensitivity and specificity of the technique are illustrated in Fig . 2 . First, genomic DNA of known concentration, from the P. aeruginosa strain PAO1, was serially diluted to give the number of templates shown and amplified as described above . No signal was visible using less than three copies of P . aeruginosa DNA as template or with 10 5 copies of Staphylococcus aureus, Streptococcus pyogenes, Haernophi-
tory pathogens, and amplified by the nested PCR as previously described . The results shown in Fig . 3 indicated that the PCR procedure can specifically detect small numbers of mucoid and non-mucoid P . aeruginosa in sputum against a background of other bacterial species .
M
105
lus influenzae or human DNA (Fig. 2a) . The 13 nonmucoid and three mucoid P . aeruginosa investigated each gave a positive reaction (data not shown) . In contrast, no signal was visible using as template 10 5 copies of 10 strains of P. cepacia and the various
I and 3
land 3
30
60
10 3
10
0
10 5
301
10 3
I and 3, 2 and 4
Primers
30 each 10
0
10 5
10 3
Cycles 10
0
292
207
Fig. 1 . The sensitivity of nested-primer PCR . DNA extracted from P . aeruginosa strain PAO1 was diluted in water to give the number of genomes shown and amplified with the primer pairs indicated, as described in Materials and Methods . During the PCR, reactions with primers Pal and Pa3 were annealed at 40°C to maximize sensitivity, those with primers Pa2 and Pa4 were annealed at 50 ° C to maximize specificity . The 292 by and 207 by products are arrowed . Lane M contains DNA size markers (PBR322 DNA digested with
Taq 1) .
30 2
1 . McIntosh et al .
Copies PA01 DNA c
3x 10 3
M
10 4
10 3
10 2
0D 10
0 . 1
1
0
d
E
I
UI
207
( b) Pseudomonas number of bacteria M
10 s
104
10 3
10 2
10
I
0
t CL a U)
E a,
c a E I
207
The sensitivity and specificity of PCR . (a) DNA extracted from PAO1 was serially diluted with water and the number of genomes shown was amplified using the nested-primer PCR procedures . (b) A culture of P . aeruginosa PAO1 was serially diluted, plated on Pseudomonas isolation agar, and viable counts obtained . DNA was extracted from the numbers of bacteria shown and amplified as described in Materials and Methods. Copies of genomic DNA, 3 x 105, from S . aureus, S . pyogenes, H . influenzae and man were similarly amplified in each experiment as controls . Fig. 2 .
other species of Pseudomonas listed in Table 1 (data not shown) .
DISCUSSION
The data presented in this study show that PCR provides a technique as specific and probably more sensitive than the assays explored so far . Quantitative bacterial culture from homogenized sputum by the method described can detect 2 x 10 2 colony-forming units (cfu) ml - ' . Amplification of small amounts of genomic DNA from bacterial cultures indicated that
as little as three copies of PAO DNA could be detected . Three copies of pseudomonas genomic DNA is equivalent to approximately 0 . 01 pg . Investigation of samples of sputum from CF patients containing P. aeruginosa and other respiratory bacteria demonstrated good sensitivity and specificity since as little as 10 3 cfu of P . aeruginosa, identified by bacterial culture, provided a clear signal despite the presence of large numbers of other bacteria . As a diagnostic tool, serology does not necessarily reflect active colonization . In the present study sputum samples culture-negative for P . aeruginosa but providing a clear signal following PCR may indicate a
PCR Detection of P . aeruginosa
A
B
C
D
E
F
G
H
I
'I
303
d Z O o °c
a
a 0 N
CL N
N
E 0 x
c E x
207
Fig . 3 . Specific detection of P . aeruginosa in the presence of other bacterial species . DNA was extracted from a series of sputum samples and amplified as described in Materials and Methods . By bacterial culture, mucoid P. aeruginosa was detected in samples A, B, E, F, G and J (at 3 X 10' ml - ' to 1 x 108 ml - '); non-mucoid P . aeruginosa in sample D (at 1 x 10 3 ml - ') ; S . aureus in samples C, D, H, and I (at 4 X 106 ml - ' to 3 X 10' ml"') : and H. influenzae in samples D, H and I (at 2 x 10' ml - ' to 4 x 10' ml - ') . DNA amplification was not observed in sputum sample C which produced 10 cfu of S . aureus but no P . aeruginosa . Controls are as described in the legend to Fig . 2 .
high sensitivity rather than false-positive reactions . This explanation was supported by the negative PCR results obtained with sputa obtained from non-CF patients and previously shown to be culture-free for P . aeruginosa (data not shown) . It was interesting, and unexpected, to observe that the strains of P. putida, P . fluorescens and P . mendocina examined in the study did not produce a signal with the algD primers employed ; previous studies had shown these strains to be capable of alginate biosynthesis .', " The negative PCR result for these pseudomonas species illustrates the specificity of the chosen algD primers for identification of P . aeruginosa and suggests heterogeneity in the algD sequence within the genus Pseudomonas . In conclusion, we have presented data which suggests that PCR can be used to provide rapid and specific detection of small numbers of P . aeruginosa in sputa against a background of other bacterial species . The use of a/gD primers resulted in an assay which is apparently specific for P . aeruginosa and sufficiently sensitive, first: to indicate the absence of P. aeruginosa colonization which would be useful in the application of vaccination therapy, and second : to detect the initial stages of colonization and the early and beneficial institution of antipseudomonas chemotherapy ." It also seems reasonable to speculate on the potential use of PCR to detect other bacteria in the sputa of CF patients . Pseudomonas aeruginosa may overgrow H . influenza and other potential pathogens in culture . As a consequence, non-capsulated, nongroup B H . influenzae are probably underdiagnosed
and their pathogenic significance underestimated in CF lung disease? Another bacterial candidate for detection by PCR is P . cepacia, a species which has emerged as an important pathogen among CF patients and whose pathogenicity and epidemiology merit further study . 2'' 8
ACKNOWLEDGEMENTS This work was supported by project grants 313 and 327 from the Cystic Fibrosis Research Trust . We thank C . Doherty and A . J . Gilfillan for technical assistance .
REFERENCES 1 . Fegan, M ., Francis, P ., Hayward, A . C ., Davis, C . H . G . & Fuerst, J . A. (1990). Phenotypic conversion of Pseudomonas aeruginosa in cystic fibrosis . Journal of Clinical Microbiology 28,1143-46 . 2 . Govan, J . R . W. & Nelson, J . W . (1992) . Microbiology of lung infection in cystic fibrosis . British Medical Bulletin 48 : in press . 3 . Brett, M . M., Ghoneim, A . T . M . & Littlewood, J . M . (1988) . Prediction and diagnosis of early Pseudomonas aeruginosa infection in cystic fibrosis : a follow up study. Journal of Clinical Microbiology 26, 1565-70 . 4. Nelson, J . W ., Barclay, G . R . & Govan, J . R . W . (1990). Diagnosis of chronic Pseudomonas aeruginosa infection in cystic fibrosis by ELISA for anti-pseudomonas LPS IgG antibodies . Serodiagnosis and Immunotherapy in Infectious Diseases 4, 9-16 . 5 . Govan, J . W. R . (1988) . Alginate biosynthesis and other unusual characteristics associated with the pathogenesis of Pseudomonas aeruginosa . In Bacterial infections
304
I . McIntosh et al.
of respiratory and gastrointestinal mucosae . (Griffiths, E ., Donachie, W . & Stephen, J ., eds), pp . 67-96. Oxford: IRL Press . 6 . Govan, J . R . W ., Fyfe, J . A. M. & Jarman, T. R. (1981). Isolation of alginate-producing mutants of Pseudomonas fluorescens, Pseudomonas putida and Pseudomonas mendocina . Journal of General Microbiology 125, 217-20 . 7 . Russell, N . J. & Gacesa, P. (1988) . Chemistry and biology of the alginate of mucoid strains of Pseudomonas aeruginosa in cystic fibrosis . Molecular Aspects of Medicine 10, 1-91 . 8 . Deretic, V ., Gill, J . F . & Chakrabarty, A . M . (1987). Pseudomonas aeruginosa infection in cystic fibrosis : nucleotide sequence and transcriptional regulation of the algD gene . Nucleic Acids Research 15, 4567-81 . 9 . Fyfe, J. A. M. & Govan, J . R. W . (1983) . Synthesis, regulation and function of bacterial alginates . In Progress in Industrial Microbiology ( Bushell, M . E ., ed .), pp. 45-83 . Amsterdam : Elsevier. 10. Stanier, R . Y., Palleroni, N . J. & Doudoroff, M . (1966) . The aerobic pseudomonas: a taxonomic study . Journal of General Microbiology 43, 159-271 . 11 . Palleroni, N . J ., Doudoroff, M ., Stanier, R . Y., Solanes, R. E . & Mandel, M . (1970). Taxonomy of the aerobic
12.
13.
14 .
15 .
16 .
17 .
18 .
pseudomonads : the properties of the pseudomonas stutzeri group . Journal of General Microbiology 60, 215-31 . Leifson, E . & Hugh, R . (1954) . A new type of polar monotrichous flagellation . Journal of General Microbiology 10, 68-70. Wilkinson, S . G . (1970) . Cell walls of Pseudomonas species sensitive to ethylenediaminetetracetic acid . Journal of Bacteriology 104, 1035-44 . Hugh, R. & Ryschenkow, E . (1961). Pseudomonas maltophilia, an alcaligenes-like species . Journal of General Microbiology 26, 123-32 . Saiki, R ., Gelfand, D . H., Stoffel, S. et al. (1988) . Primerdirected enzymatic amplification of DNA with a thermostable DNA polymerase . Science 239, 487-91 . Hacking, A . J ., Taylor, I . W. F ., Jarman, T. R . & Govan, J . R . W . (1983). Alginate biosynthesis by Pseudomonas mendocina . Journal of General Microbiology 129, 3473-80 . Valerius, N . H ., Koch, C. & Holby, N . (1991). Prevention of chronic Pseudomonas aeruginosa colonisation in cystic fibrosis by early treatment . Lancet 338, 725-6. Smith, D . L ., Smith, E . G ., Gumery, L . R . & Stableforth, D . E . (1992) . Pseudomonas cepacia infection in cystic fibrosis . Lancet 339, 252 .
(1992) 6, 299-304
Detection of Pseudomonas aeruginosa in sputum from cystic fibrosis patients by the polymerase chain reaction 2 lain McIntosh', John R . W . Govan and David J . H . Brock'* 'Human Genetics Unit, Western General Hospital, Edinburgh EH4 2XU and 2 Department of Medical Microbiology, University Medical School, Edinburgh, EH8 9AG, UK (Received 20 January 1992, Accepted 6 March 1992)
A DNA amplification procedure using heat stable Taq polymerase and the polymerase chain reaction is described for the detection of Pseudomonas aeruginosa in specimens from cystic fibrosis patients . A set of primers was selected on the basis of the nucleotide sequence of the algD gene encoding GDP mannose dehydrogenase, a major enzyme in the biosynthesis of alginate by P . aeruginosa. Using this set of primers in conjunction with the polymerase chain reaction, P . aeruginosa could be specifically detected, with a sensitivity approximating 10 bacteria, in sputum harbouring large numbers of other respiratory pathogens, including Staphylococcus aureus and Haemophilus influenzae . These results suggest that amplification of specific sequences within the a/gD gene by the polymerase chain reaction may provide a highly sensitive and specific tool for the detection of P . aeruginosa in the early stages of pulmonary colonization . KEYWORDS : Pseudomonas aeruginosa, polymerase chain reaction, cystic fibrosis
INTRODUCTION Pulmonary colonization with mucoid Pseudomonas aeruginosa is a major cause of morbidity and mortality in patients with cystic fibrosis (CF). Initial asymptomatic and often intermittent colonization of the upper respiratory tract with a non-mucoid strain of P . aeruginosa usually precedes chronic colonization with mucoid variants of the original strain .', ' In CF patients, detection of P . aeruginosa in the early stages of pulmonary colonization would increase the opportunity for early therapeutic intervention and could assist in the identification of primary colonization sites . At present, detection of pulmonary colonization with P . aeruginosa is based on quantitative
serve as an alternative to other methods for the detection of non-mucoid and mucoid forms of P . aeruginosa, particularly in the early stages of pulmonary colonization . In prokaryotes, alginate biosynthesis is restricted to Azotobacter vinelandii, an organism not associated with human colonization, and to P . aeruginosa and a few other pseudomonas species .` We chose the
algD gene as the conserved region of the P . aeruginosa genome to be amplified, since it codes for GDP mannose dehydrogenase, an enzyme essential for alginate biosynthesis :' algD is present in non-mucoid and mucoid P . aeruginosa and is not found in higher organisms .
culture of organisms from homogenized sputa, or indirectly by detection of pseudomonas-specific antibodies using enzyme-linked immunosorbent assay .', ' Since the polymerase chain reaction (PCR) has been used successfully to identify other bacteria, we attempted to develop a rapid and sensitive method for detecting P . aeruginosa . This approach might
MATERIALS AND METHODS Bacteria The bacterial strains used are listed in Table 1 .¢ 14
*Author to whom correspondence should be addressed . 0890-8508/92/040299 + 06 $08 .00/0
299
© 1992 Academic Press Limited
3 00
1. McIntosh et al .
Strains of P. aeruginosa comprised the well-characterised genetic strain PAO1 and its isogenic mucoid variants PA0568, and PA0579 and the mucoid CF isolate 492a . In addition to the bacterial species listed in Table 1, 12 non-mucoid isolates of P . aeruginosa (six CF isolates and six non-CF isolates) and 10 strains of P. cepacia were investigated .
Sputum culture Expectorated sputum was obtained after chest physiotherapy, and after liquefaction with Sputolysin (Behring Diagnostics, La Jolla, California, USA) and suitable dilution (10 2, 104) in sterile physiological saline, quantitative culture for respiratory pathogens was performed by plating out 0 . 1 ml volumes onto blood agar, lysed blood agar incorporating 0 . 5% bacitracin and Pseudomonas isolation agar (Difco Laboratories, Detroit, Michigan, USA) and incubating aerobically at 37 ° C for 48 h . Isolates were identified by standard procedures .'
Table 1 .
Polymerase chain reaction Genomic DNA from either bacterial suspensions or Sputolysin-treated sputum samples was obtained by phenol-chloroform extraction and ethanol precipitation . Briefly, 10 pl of sample was diluted to 100µl with 0. 1 M Tris-CI pH 7. 5 and added to an equal volume of phenol/chloroform, heated at 37 ° C for 5 min and centrifuged for 3 min at 10,000 rev/min (Eppendorf microfuge) . The aqueous layer was extracted twice with an equal volume of chloroform and DNA was precipitated by addition of 1/10th volume of 3 M NaCl and 3 volumes of ethanol . The pellet was resuspended in 10 pl of sterile water and amplified by the polymerase chain reaction ." The primers shown in Table 2 were designed using the algD sequence data of Deretic et al.' and obtained from Oswel DNA Service (Department of Chemistry, University of Edinburgh) . The nesting procedure, which allows for greater sensitivity, was performed using the internal primers Pa2 and Pa4 to amplify 2 pl of the PCR product from an initial round of PCR with
Bacterial species and strains used in this study
Species
Strain and reference
P . aeruginosa P. aeruginosa
P . aeruginosa P . aeruginosa P . fluorescens P . putida
P . testosteroni P . acidovorans P . alkaligenes P . mendocina P. diminuta P. stutzeri
Xanthomonas maltophilia Staphylococcus aureus Streptococcus pyogenes Haemophilus influenzae
PAO1 PA0568* PA0579* CF492a* NCIMB 10525 strain 12 NCIMB 10007 strain Cl-B NCIMB 8893 strain 79 NCIMB 9681 type strain NCIMB 9946 type strain NCIMB 10541 type strain CH-50 NCIMB 9393 type strain NCIMB 9040 NCIMB 9203 type strain Clinical isolate Clinical isolate Clinical isolate
(9) (9) (9) (9) (10) (10) (10) (10) (10) (11) (12) (13) (14)
* Mucoid . NCIMB, National Collections of Industrial and Marine Bacteria, Aberdeen U.K.
Table 2.
Sequences used for the oligonucleotide primers
Primer sequence Pal Pa2 Pa3 Pa4
5' 5' 5' 5'
GACAGGTTGAGCTTGTGG CGAACTGGACAAGCAGAC GAATTCCTCCGCGAGAGC GCAGATCACGTCCATCAC
* Sequence from Deretic et al .'
Homologous to algD position* 1297-1280 1067-1084 1005-1022 1274-1257
PCR Detection of P . aeruginosa RESULTS
primers Pal and Pa3 . The appropriate primer pair (50 pmol each) and DNA were mixed in 100 pl of 10 mm Tris-Cl pH 8 . 3, 50 mm KCI, 2-5 mm MgCl, 10% DMSO and amplified through 30 cycles of 91 ° C (30 s), 40° C or 50 ° C (30 s) and 72 ° C (1 min) with 2 units of AmpliTaq DNA polymerase (Perkin-Elmer/Cetus, Hemel Hempstead, England) using an automated Thermal-Cycler (Perkin-Elmer/Cetus). After the 30th cycle, incubation continued at 72 ° C for 10 min to complete elongation of the PCR products . Ten per cent of the reaction product was run on a 2% agarose gel and stained with ethidium bromide to visualize the amplified DNA . To show that DNA prepared directly from small amounts of bacteria can be amplified, the results obtained from serial 10-fold dilutions of a culture P . aeruginosa PAO1 are shown (Fig . 2b) . On the original gel it was possible to see faint bands in lanes containing PCR product from reactions with 30 and three organisms (lanes 10 and 1, respectively) . Again no signal was visible in the negative controls . These results show that nested PCR is sufficiently sensitive and specific to detect very small numbers of P . aeruginosa . To demonstrate that the procedures could detect P. aeruginosa in clinical material, DNA was extracted from 10[d of sputolysin-treated sputum from patients with CF known to be colonized by respira-
For maximal sensitivity and specificity it was found necessary to adopt the 'nested primer' technique (Materials and Methods and Fig . 1) . An initial pair of primers (Pal and Pa3) were used in 30 cycles of PCR to generate a product of 292 by from a relatively large number of genomic templates (3 X 10 5 , . Fig . 1) . Increasing the number of cycles allowed the detection of a smaller number of templates (3 X 10 3 ), but only by using 2% of the original reaction product in a second round of PCR with the internal (`nested') primers Pa2 and Pa4 (Table 2) could as few as 30 genomes be detected (Fig . 1) . The identity of the PCR product was confirmed initially by digestion with restriction enzymes predicted to cut the amplified DNA, and subsequently by DNA sequencing (data not shown) . No signal was visible when no template DNA was included in the reactions : the extreme sensitivity of PCR requires that such negative controls be included in every experiment. The sensitivity and specificity of the technique are illustrated in Fig . 2 . First, genomic DNA of known concentration, from the P. aeruginosa strain PAO1, was serially diluted to give the number of templates shown and amplified as described above . No signal was visible using less than three copies of P . aeruginosa DNA as template or with 10 5 copies of Staphylococcus aureus, Streptococcus pyogenes, Haernophi-
tory pathogens, and amplified by the nested PCR as previously described . The results shown in Fig . 3 indicated that the PCR procedure can specifically detect small numbers of mucoid and non-mucoid P . aeruginosa in sputum against a background of other bacterial species .
M
105
lus influenzae or human DNA (Fig. 2a) . The 13 nonmucoid and three mucoid P . aeruginosa investigated each gave a positive reaction (data not shown) . In contrast, no signal was visible using as template 10 5 copies of 10 strains of P. cepacia and the various
I and 3
land 3
30
60
10 3
10
0
10 5
301
10 3
I and 3, 2 and 4
Primers
30 each 10
0
10 5
10 3
Cycles 10
0
292
207
Fig. 1 . The sensitivity of nested-primer PCR . DNA extracted from P . aeruginosa strain PAO1 was diluted in water to give the number of genomes shown and amplified with the primer pairs indicated, as described in Materials and Methods . During the PCR, reactions with primers Pal and Pa3 were annealed at 40°C to maximize sensitivity, those with primers Pa2 and Pa4 were annealed at 50 ° C to maximize specificity . The 292 by and 207 by products are arrowed . Lane M contains DNA size markers (PBR322 DNA digested with
Taq 1) .
30 2
1 . McIntosh et al .
Copies PA01 DNA c
3x 10 3
M
10 4
10 3
10 2
0D 10
0 . 1
1
0
d
E
I
UI
207
( b) Pseudomonas number of bacteria M
10 s
104
10 3
10 2
10
I
0
t CL a U)
E a,
c a E I
207
The sensitivity and specificity of PCR . (a) DNA extracted from PAO1 was serially diluted with water and the number of genomes shown was amplified using the nested-primer PCR procedures . (b) A culture of P . aeruginosa PAO1 was serially diluted, plated on Pseudomonas isolation agar, and viable counts obtained . DNA was extracted from the numbers of bacteria shown and amplified as described in Materials and Methods. Copies of genomic DNA, 3 x 105, from S . aureus, S . pyogenes, H . influenzae and man were similarly amplified in each experiment as controls . Fig. 2 .
other species of Pseudomonas listed in Table 1 (data not shown) .
DISCUSSION
The data presented in this study show that PCR provides a technique as specific and probably more sensitive than the assays explored so far . Quantitative bacterial culture from homogenized sputum by the method described can detect 2 x 10 2 colony-forming units (cfu) ml - ' . Amplification of small amounts of genomic DNA from bacterial cultures indicated that
as little as three copies of PAO DNA could be detected . Three copies of pseudomonas genomic DNA is equivalent to approximately 0 . 01 pg . Investigation of samples of sputum from CF patients containing P. aeruginosa and other respiratory bacteria demonstrated good sensitivity and specificity since as little as 10 3 cfu of P . aeruginosa, identified by bacterial culture, provided a clear signal despite the presence of large numbers of other bacteria . As a diagnostic tool, serology does not necessarily reflect active colonization . In the present study sputum samples culture-negative for P . aeruginosa but providing a clear signal following PCR may indicate a
PCR Detection of P . aeruginosa
A
B
C
D
E
F
G
H
I
'I
303
d Z O o °c
a
a 0 N
CL N
N
E 0 x
c E x
207
Fig . 3 . Specific detection of P . aeruginosa in the presence of other bacterial species . DNA was extracted from a series of sputum samples and amplified as described in Materials and Methods . By bacterial culture, mucoid P. aeruginosa was detected in samples A, B, E, F, G and J (at 3 X 10' ml - ' to 1 x 108 ml - '); non-mucoid P . aeruginosa in sample D (at 1 x 10 3 ml - ') ; S . aureus in samples C, D, H, and I (at 4 X 106 ml - ' to 3 X 10' ml"') : and H. influenzae in samples D, H and I (at 2 x 10' ml - ' to 4 x 10' ml - ') . DNA amplification was not observed in sputum sample C which produced 10 cfu of S . aureus but no P . aeruginosa . Controls are as described in the legend to Fig . 2 .
high sensitivity rather than false-positive reactions . This explanation was supported by the negative PCR results obtained with sputa obtained from non-CF patients and previously shown to be culture-free for P . aeruginosa (data not shown) . It was interesting, and unexpected, to observe that the strains of P. putida, P . fluorescens and P . mendocina examined in the study did not produce a signal with the algD primers employed ; previous studies had shown these strains to be capable of alginate biosynthesis .', " The negative PCR result for these pseudomonas species illustrates the specificity of the chosen algD primers for identification of P . aeruginosa and suggests heterogeneity in the algD sequence within the genus Pseudomonas . In conclusion, we have presented data which suggests that PCR can be used to provide rapid and specific detection of small numbers of P . aeruginosa in sputa against a background of other bacterial species . The use of a/gD primers resulted in an assay which is apparently specific for P . aeruginosa and sufficiently sensitive, first: to indicate the absence of P. aeruginosa colonization which would be useful in the application of vaccination therapy, and second : to detect the initial stages of colonization and the early and beneficial institution of antipseudomonas chemotherapy ." It also seems reasonable to speculate on the potential use of PCR to detect other bacteria in the sputa of CF patients . Pseudomonas aeruginosa may overgrow H . influenza and other potential pathogens in culture . As a consequence, non-capsulated, nongroup B H . influenzae are probably underdiagnosed
and their pathogenic significance underestimated in CF lung disease? Another bacterial candidate for detection by PCR is P . cepacia, a species which has emerged as an important pathogen among CF patients and whose pathogenicity and epidemiology merit further study . 2'' 8
ACKNOWLEDGEMENTS This work was supported by project grants 313 and 327 from the Cystic Fibrosis Research Trust . We thank C . Doherty and A . J . Gilfillan for technical assistance .
REFERENCES 1 . Fegan, M ., Francis, P ., Hayward, A . C ., Davis, C . H . G . & Fuerst, J . A. (1990). Phenotypic conversion of Pseudomonas aeruginosa in cystic fibrosis . Journal of Clinical Microbiology 28,1143-46 . 2 . Govan, J . R . W. & Nelson, J . W . (1992) . Microbiology of lung infection in cystic fibrosis . British Medical Bulletin 48 : in press . 3 . Brett, M . M., Ghoneim, A . T . M . & Littlewood, J . M . (1988) . Prediction and diagnosis of early Pseudomonas aeruginosa infection in cystic fibrosis : a follow up study. Journal of Clinical Microbiology 26, 1565-70 . 4. Nelson, J . W ., Barclay, G . R . & Govan, J . R . W . (1990). Diagnosis of chronic Pseudomonas aeruginosa infection in cystic fibrosis by ELISA for anti-pseudomonas LPS IgG antibodies . Serodiagnosis and Immunotherapy in Infectious Diseases 4, 9-16 . 5 . Govan, J . W. R . (1988) . Alginate biosynthesis and other unusual characteristics associated with the pathogenesis of Pseudomonas aeruginosa . In Bacterial infections
304
I . McIntosh et al.
of respiratory and gastrointestinal mucosae . (Griffiths, E ., Donachie, W . & Stephen, J ., eds), pp . 67-96. Oxford: IRL Press . 6 . Govan, J . R . W ., Fyfe, J . A. M. & Jarman, T. R. (1981). Isolation of alginate-producing mutants of Pseudomonas fluorescens, Pseudomonas putida and Pseudomonas mendocina . Journal of General Microbiology 125, 217-20 . 7 . Russell, N . J. & Gacesa, P. (1988) . Chemistry and biology of the alginate of mucoid strains of Pseudomonas aeruginosa in cystic fibrosis . Molecular Aspects of Medicine 10, 1-91 . 8 . Deretic, V ., Gill, J . F . & Chakrabarty, A . M . (1987). Pseudomonas aeruginosa infection in cystic fibrosis : nucleotide sequence and transcriptional regulation of the algD gene . Nucleic Acids Research 15, 4567-81 . 9 . Fyfe, J. A. M. & Govan, J . R. W . (1983) . Synthesis, regulation and function of bacterial alginates . In Progress in Industrial Microbiology ( Bushell, M . E ., ed .), pp. 45-83 . Amsterdam : Elsevier. 10. Stanier, R . Y., Palleroni, N . J. & Doudoroff, M . (1966) . The aerobic pseudomonas: a taxonomic study . Journal of General Microbiology 43, 159-271 . 11 . Palleroni, N . J ., Doudoroff, M ., Stanier, R . Y., Solanes, R. E . & Mandel, M . (1970). Taxonomy of the aerobic
12.
13.
14 .
15 .
16 .
17 .
18 .
pseudomonads : the properties of the pseudomonas stutzeri group . Journal of General Microbiology 60, 215-31 . Leifson, E . & Hugh, R . (1954) . A new type of polar monotrichous flagellation . Journal of General Microbiology 10, 68-70. Wilkinson, S . G . (1970) . Cell walls of Pseudomonas species sensitive to ethylenediaminetetracetic acid . Journal of Bacteriology 104, 1035-44 . Hugh, R. & Ryschenkow, E . (1961). Pseudomonas maltophilia, an alcaligenes-like species . Journal of General Microbiology 26, 123-32 . Saiki, R ., Gelfand, D . H., Stoffel, S. et al. (1988) . Primerdirected enzymatic amplification of DNA with a thermostable DNA polymerase . Science 239, 487-91 . Hacking, A . J ., Taylor, I . W. F ., Jarman, T. R . & Govan, J . R . W . (1983). Alginate biosynthesis by Pseudomonas mendocina . Journal of General Microbiology 129, 3473-80 . Valerius, N . H ., Koch, C. & Holby, N . (1991). Prevention of chronic Pseudomonas aeruginosa colonisation in cystic fibrosis by early treatment . Lancet 338, 725-6. Smith, D . L ., Smith, E . G ., Gumery, L . R . & Stableforth, D . E . (1992) . Pseudomonas cepacia infection in cystic fibrosis . Lancet 339, 252 .
