- Email: [email protected]
Molecular typing of Aspergillus fumigatus strains by sequence-specific DNA primer (SSDP) analysis
FEMS Immunology and Medical Microbiology 17 (1997) 95^102
Molecular typing of Aspergillus fumigatus strains by sequence-speci¢c DNA primer (SSDP) analysis P. Mondon*, M.P. Brenier, E. Coursange, B. Lebeau, P. Ambroise-Thomas, R. Grillot 1 è dicale et Mole è culaire, Laboratoire Relation Ho ê te-Agents Pathoge énes, UPRES-A CNRS 5082, Faculte è de Me è decine, Parasitologie-Mycologie Me è Joseph Fourier Grenoble I, Domaine de la Merci, 38706 La Tronche, France Universite
Received 10 September 1996; revised 5 November 1996; accepted 19 November 1996
Abstract
A PCR typing method has been developed and tested to investigate the polymorphism of clinical strains of Aspergillus . Firstly, the DNA fragments from random amplified polymorphic DNA (RAPD) patterns of nine epidemiologically and geographically non-related monosporal strains of A. fumigatus were cloned and sequenced. The pairs of five sequencespecific DNA primers (SSDP), characteristic of the 5P and 3P extremities of the RAPD products, were then used in high stringency PCR to type 43 clinical strains of A. fumigatus from 13 patients, according to the presence or absence of a single amplified band. This original approach, which uses the advantages of PCR, has made it possible to overcome the difficulties resulting from the low stringency amplification. The SSDP analysis of 51 A. fumigatus strains (9 unrelated monosporal strains and 43 clinical strains from 13 patients) can be classed into 22 different types with a high reproducibility and a high level of discrimination (D = 0.96). The results suggest that seven lung transplant patients with necrotizing aspergillosis, bronchitis aspergillosis and bronchial colonization were infected by multiple strain genotypes, whereas three patients with invasive aspergillosis seem to have been infected by a single strain. fumigatus
Keywords : Aspergillus fumigatus
; Polymerase chain reaction; Molecular marker; Lung transplant; Invasive aspergillosis
1. Introduction
Aspergillus
fumigatus
is an opportunistic fungal
* Corresponding author. Tel.: +33 (4) 76 63 71 01; fax: +33 (4) 76 51 86 67. 1
European Research Group on Biotype and Genotype of (EBGA Network). Members of the EBGA Network who contributed to this study: P. Mondon, F. Symoens, E. Rodriguez, F. Chaib, B. Lebeau, M.A. Piens, A.-M. Tortorano, M. Mallieè, F. Chapuis, A. Carlotti, J. Villard, M.A. Viviani, N. Nolard, J.-M. Bastide, P. Ambroise-Thomas and R. Grillot. Aspergillus
pathogen, responsible for invasive aspergillosis (IA), a fatal disease most often acquired within the hospital environment. Several questions concerning the mode of transmission and the potential origins of patient contamination are to date unsolved. In recent years, several di¡erent molecular methods have been developed to type isolates of A. fumigatus including restriction endonuclease analysis (REA) [1], isoenzyme analysis (IEA) [2], moderately repetitive sequences (MRSs) [3,4] and random ampli¢ed polymorphic DNA (RAPD) [5^7]. Each method has its own advantages and disadvantages and no
0928-8244 / 97 / $17.00 Copyright ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V. PII S 0 9 2 8 - 8 2 4 4 ( 9 6 ) 0 0 1 1 0 - 1
FEMSIM 720 16-4-97
96
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
molecular typing method appears to be clearly better than the rest. The combination of typing methods appears to provide the best discriminatory power [8^11]. With this object in view, we have created a European research group to study the biotype and the genotype of A. fumigatus (EBGA network) using a combination of several molecular typing methods [12^14], the ultimate objective being a better understanding of the epidemiology of invasive aspergillosis. Our team has investigated the RAPD analysis [15,16], which employs a single short primer (10mer) with an arbitrary nucleotide sequence in a low stringency polymerase chain reaction (PCR), to randomly amplify genomic DNA [17]. This method is quick to perform, large numbers of isolates can be analyzed in a few days and it generates a relatively high degree of discrimination among isolates. However, the intra-laboratory reproducibility is not ideal. Occasionally the bands generated have di¡erent staining intensities and this has sometimes made the interpretation of the banding patterns di¤cult and complex in the same laboratory [7^9]. In this context, we propose an original PCR typing strategy using the advantages of RAPD, without the drawback linked to the reproducibility. The RAPD analysis was used to produce banding patterns which made it possible to di¡erentiate the strains. The discriminating fragments were sequenced to design pairs of sequence-speci¢c DNA primers (SSDP) (20-mers) characteristic of the two extremities of the fragments considered. The SSDP was used in higher stringency PCR to study the reality of the presence or the absence of the band suggested as discriminating. The purpose of this study was initially to select SSDP markers which would make it possible to differentiate DNA of nine epidemiologically and geographically non-related monosporal strains. Secondly, these SSDP markers were applied to demonstrate the importance of the SSDP analysis in the investigation of its discriminatory power and the genetic polymorphism of 43 clinical strains from 13 patients, isolated in three EBGA centers (University hospitals of Lyon (France), Milan (Italy) and Grenoble (France)), which were responsible for different types of aspergillosis in invasive aspergillosis and lung transplant patients (bronchial colonization, necrotizing aspergillosis, bronchitis aspergillosis).
2. Materials and methods
2.1. Fungal strains
The nine monosporal strains were from the culture collection of the Institute of Hygiene and Epidemiology Mycology (IHEM Brussels, Belgium) and selected to be epidemiologically and geographically non-related (Table 1). The 43 clinical strains were collected by the three University hospitals involved in the EBGA network (Lyon, Grenoble and Milan), from six patients with proven or highly probable invasive aspergillosis (Table 1, patients A^F) and from seven lung transplant patients with necrotizing aspergillosis, bronchitis aspergillosis and bronchial colonization (Table 1, patients G^N). Each A. fumigatus infection was well documented according to the criteria of aspergillosis classi¢cation [18]. The strains were centralized, preserved by lyophilization at the IHEM and redistributed to the EBGA centers for typing. 2.2. Ampli¢cation conditions
Ampli¢cation was performed using puri¢ed DNA as previously described [19]. The reaction mixtures (30 Wl) contained 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2 , 0.005% Tween 20, 0.005% NP-40, 75 WM each dNTP (Boehringer), 0.5 U of Replitherm0 DNA polymerase (Epicentre), 50 ng of DNA template and 0.2 WM of random primer (10-mers) or 0.1 WM of each speci¢cally designed primer (20-mers). Each reaction mixture was overlaid with mineral oil (30 Wl) to prevent evaporation. In a DNA thermal cycler (Perkin Elmer 480) ampli¢cations were carried out after an initial heat denaturation (95³C for 6 min). For arbitrary short primers 40 cycles consisting of 20 s at 95³C, 30 s at 36³C and 60 s at 63³C were used, while for speci¢c long primers, 30 cycles consisting of 20 s at 95³C, 30 s at 55³C and 60 s at 72³C were performed. Ampli¢cation products were resolved by electrophoresis on 1.5% agarose gels and stained with ethidium bromide (0.5 Wg ml31 ). 2.3. RAPD primers
One hundred and twenty RAPD decamers de-
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
97
signed according to the following theoretical parameters: G-C content greater than 60%, no dryad symmetry and non-self-complementary, were purchased from Operon Technologies Inc. (Alameda, CA), kits (OPAH, OPAI, OPAJ, OPP, OPR, OPQ) and the primer R108 [5]. 2.4. DNA cloning and sequencing
The DNA template was ampli¢ed with a short primer (10-mer) and 2 WCi [K-35S]dATP (Amersham). 2 Wl of each reaction mixture was separated by electrophoresis on an acrylamide (6%)-urea (50%) sequencing gel and subjected to autoradiography. Bands of interest were collected, the gel was homogenized with a conical grinder and 10 Wl of the mixture was ampli¢ed directly as previously described, without further DNA puri¢cation. The ampli¢cation products were cloned using a TA cloning kit (Invitrogen). The insert DNA was sequenced using the Sequenase 2.0 kit (United States Biochemical Corp.) and [K-35S]dATP (Amersham). On the basis of the sequence, the oligonucleotide primer sequences were selected including the sequence of the RAPD primer (10-mer), plus 10 bases characteristic of the fragment excised and synthesized by Eurogentec (Belgium). 2.5. Discriminatory power
The discriminatory power of a typing method, D, is the average probability that the typing system will assign a di¡erent type to two unrelated strains randomly sampled [20]. The discriminatory power was calculated in the manner described by Hunter [21]. 3. Results and discussion
3.1. Development of SSDP sets
One hundred and twenty decamers from Operon technology kits (OPAH, OPAI, OPAJ, OPP, OPR, OPQ) and the primer R108 [5] were tested to generate RAPD markers from genomic DNAs of nine unrelated monosporal strains (8065, 8070, 8075, 8080, 8086, 8095, 9005 8090, and 9000) from the IHEM culture collection (Table 1). Three random
Fig. 1. DNA template of nine unrelated monosporal strains from the IHEM collection (IHEM numbers: 8065, 8070, 8075, 8080, 8086, 8095, 9005, 8090, 9000) ampli¢ed with the RAPD primer R108, 2 WCi [K-35 S]dATP (Amersham) and separated by electrophoresis on an acrylamide (6%)-urea (50%) sequencing gel (lanes 1^9). Lane 0: negative control without DNA.
primers (10-mers) (R108, OPQ6, OPAj12) were selected to generate banding patterns which seemed to discriminate the di¡erent strains. From the RAPD patterns, the discriminating fragment was excised from the acrylamide gel, re-ampli¢ed, and sequenced to design a pair of longer primers (20-mers) characteristic of the two extremities of the considering fragment. The sequence of SSDP included the sequence of the RAPD primer (10-mers), plus 10 bases characteristic of the fragment excised (Table 2). The SSDP marker was then used in higher stringency PCR (annealing temperature: 55³C) with the DNA of the nine unrelated strains to check the presence or the absence of the band suggested as discriminating.
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
98
Table 1 Genotypes of the nine unrelated monosporal strains and the 43 clinical strains of Patient
Strain
Origin
Infection
b
A. fumigatus
c Sample
(IHEM no.)
8090 9000 8080 8075 8095 9005 8065 8070 8086
a a a a a a a a a
obtained by SSDP analysis SSDP analysis SSDP
SSDP
1
2
3
4
5
3
+
+
+
3
11
Iraq
Desert soil
+
+
+
30
Belgium
Lung of buzzard
+
+
32
Belgium
Human
+
+
Belgium
Dust from mattress
Belgium
Seeds of cereal
Milan
Human bronchial secretion
Milan
Intensive care unit env.
Milan
Human skin surgical wound
+
3 3
+
3 3 3
9010^9014
Milan
Abdominal drain (27/01/94)
+
9015
Milan
Hepatic artery (28/01/94)
+
A
9016^9019
Milan
Hepatic ilium (2/2/94)
+
A
9020
Milan
IA
Bile (2/2/94)
+
A
9021
Milan
Liver transplant
Lung ^ autopsy (7/2/94)
+
A
9022
Hepatic artery ^ autopsy
+
Milan
(7/2/94)
30
+
+
30
3
+
+
29
Grenoble
IA
Tracheal aspirate (2/12/92)
+
3
3
+
+
30
3 3
3 3
3 3
3 3
+
24
+
24
+
3
+
+
+
26
3 3 3
3 3 3
Grenoble
G
9615
Grenoble
H
9730
H
9731
I
9733
I
9734
a
+
+
9614
9752
30
+
+
9613
9753
+
BA (31/12/92))
G
M
30
+
IA
G
M
+
Grenoble
7969
a
30
+
26
F
9745
30
+
+
9417
9744
+
+
+
9418^9420
K
+
+
E
K
5
3
E
a
7
+
7965
9739
27
3 3
Bronchial secretions (2/3/95)
D
9740
3 3 3 3 3 3
24
+
IA
7964
J
3 3 3 3 3 3
3 3
+
30
C
J
+
3 3
14
+
Milan
a
+
3
+
31
3
+
a 9637 ^9641
a
+
+
3
B
a
+
+
3
Milan
a
3 3
3 3 3 3 3
+
9023
a
3 3
Portal vein ^ autopsy (7/2/94)
A
a
SSDP
Aircraft fuel
A
a
SSDP
England
A
a
Type
SSDP
Grenoble
IA
BAL (13/1/95)
Grenoble
BMT
BA (13/1/95)
Grenoble
IA
Sinus drainage (19/5/87)
Grenoble
Bronchial colonization
BA (21/4/95) BA (21/4/95)
LT
BA (21/4/95)
+
3 3
+
+
28
+
+
32
+
24
3
Lyon
Necrotizing A
BA (11/6/93)
+
3
3
3
+
22
Lyon
LT
BA (23/6/93)
+
+
+
+
+
25
Lyon
Bronchial colonization
Sputum (24/8/93) Sputum (26/8/93)
3 3
3 3
4
LT
3 3
3
Lyon
+
20
Lyon
Necrotising A
BA (16/11/93)
Lyon
LT
Sputum (24/12/93)
3
3 3
Lyon
Bronchial colonization
Sputum (26/02/93)
3
Lyon
H-L T
Sputum (30/3/93)
+
3 3
Lyon
Bronchitis A
BAL (30/6/93)
Lyon
LT
BAL (21/6/93)
FEMSIM 720 16-4-97
+
3 3
3 3
+ +
+
+
+
26
+
+
+
28
+
20
+
18
+ +
3 3
3 3 3
3
8
+
+
32
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
99
Table 1 (continued) Genotypes of the nine unrelated monosporal strains and the 43 clinical strains of A. fumigatus obtained by SSDP analysis Samplec SSDP analysis Type Patient Strain Origin Infectionb (IHEM no.) SSDP SSDP SSDP SSDP SSDP 1 2 3 4 5 a Lyon Bronchial colonization BAL (15/12/94) + + + 3 + 17 N 9393 N 9394 Lyon BAL (15/12/94) + 3 3 3 + 22 N 9395^9396 Lyon LT BAL (15/12/94) + 3 3 3 3 6 N 9397 Lyon BAL (15/12/94) 3 + 3 3 + 23 a Unrelated strains randomly sampled. b IA: invasive aspergillosis; A: aspergillosis; BMT: bone marrow transplant; LT: lung transplant; HT: heart transplant; H-L T: heart-lung transplant. c BAL: bronchoalveolar lavage; BA: bronchial aspirate.
The RAPD analysis generates a relatively complex ¢ngerprinting pattern, which is commonly separated on agarose gel and stained with ethidium bromide. However, it has been shown that gel electrophoresis is a major cause of experimental variability [22]. The problem of band intensities could be minimized when the DNA is ampli¢ed in the presence of radioactivity ([K-35 S]dATP) and separated on a 6% denaturing acrylamide gel (Fig. 1). This extra work of running a denaturing acrylamide gel is rewarded by much more and better resolved data than can be obtained using agarose gel. However, it is time consuming and expensive for wide epidemiological application. This powerful detection system has therefore been used to select the discriminating band from the RAPD analysis which will be a potential SSDP marker. The decamer R108 generated nine RAPD fragments which seemed discriminating (Fig. 1). The sequencing of these nine bands has provided nine primer pairs, but only three of these pairs (Table 2: SSDP 3, SSDP 4, SSDP 5) were really discriminating, in each case according to the presence (+) or absence (3) of a single ampli¢ed band (SSDP 4: 0.75 kb, SSDP 3: 0.55 kb, SSDP 5: 0.35 kb) (Table 1). Each of the other six primer pairs ampli¢ed a constant fragment with all of the nine monosporal strains. Six discriminating RAPD fragments were obtained with the primer OPQ6 (GAGCGCCTTG) but only one primer pair (Afp1/Afp2: SSDP 1) characteristic of a 0.95 kb fragment separated the nine unrelated strains into two di¡erent types: 3+ and 63 (Table 1). The decamer OPAj12 (CAGTTCCCGT)
generated three discriminating RAPD fragments, but only one primer pair (AfMp1/AfMp2: SSDP 2) out of three was discriminating and ampli¢ed a 1.1 kb fragment with ¢ve strains out of nine (Table 1). The ¢ve primer pairs (SSDP markers) described in Table 2 and their di¡erent +/3 combinations which de¢ne the genotype (Table 3) made it possible to distinguish all of the nine unrelated monosporal strains (Table 1). Each SSDP analysis was repeated at least twice, the result remaining the same and a primer pair Afc1/Afc2, deduced from the sequencing of a conserved fragment (1.4 kb) from the RAPD pattern of the primer OPQ6 was used as a positive control of each ampli¢cation [16,19]. 3.2. RAPD/SSDP comparison
The cloning and the sequencing of 13 out of 18 fragments, which appeared to di¡erentiate the isolates by RAPD analysis, have shown that in fact they were not discriminating when their characteristic primer pairs were used in higher stringency PCR. This highlights the problem of the band intensity of the RAPD method which uses low stringency PCR (annealing temperature: 36³C) in combination with a single short primer (10-mer). The RAPD technique has the advantage of being relatively quick and straightforward, but has major disadvantages including optimization of protocols, inter-run and interlaboratory reproducibility and interpretation of ampli¢ed patterns [7^10,20,23]. The SSDP analysis uses a high stringency PCR and makes it possible to limit these factors.
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
100
The decamer R108 has been widely described as giving the highest level of variation [5,7,8,11,23]. In our study, the results con¢rm its high discriminatory power (Fig. 1), but not the reproducibility, despite the fact that a higher detection system has been used to visualize the RAPD patterns. Recently, Verweij et
Fig. 2. DNA template of eight clinical strains from four patients
al. [23] have highlighted the necessity to envisage a
(IHEM
new
9745) ampli¢ed with the primer pair Afd1/Afd2 : SSPP 3 (lanes
typing
method
to
achieve
inter-laboratory
standardization. Our aim in view was not to describe a typing system based on the decamer R108, but to develop a new approach of PCR typing using RAPD
numbers :
9739,
9740,
9730,
9731,
9752,
9753,
9744,
1^8) and the positive control primer pair Afc1/Afc2 (lanes 1P^8P). Lanes 0 and 0P : negative control without DNA with Afd1/Afd2 and
Afc1/Afc2,
respectively.
Lane
MW :
1-kb
ladder
marker
(Gibco-BRL).
analysis to select the molecular markers for routine typing. The SSDP analysis has the advantage of combining
the
discriminatory
power
of
di¡erent
3.4. SSDP analysis of clinical strains
RAPD primers (R108, OPQ6 and OPAj12). Furthermore, the typing by each SSDP marker, according to
The feasibility of the SSDP analysis for the typing
the presence or absence of a single ampli¢ed band
of a large number of strains has been con¢rmed with
(Fig. 2), allows quick and easy coding (Table 3) of a
43 clinical strains isolated from seven lung transplant
large number of samples.
patients (bronchial colonization, necrotizing aspergillosis, bronchitis aspergillosis) (Table 1, patients G^
3.3. Discriminatory power of the SSDP analysis
N) and six invasive aspergillosis patients (Table 1, patients A^F). A wide genetic polymorphism has
The ¢ve SSDP markers have been applied to in-
been observed in the lung transplant patients com-
vestigate 43 clinical strains isolated in the three uni-
pared to the invasive aspergillosis. Indeed, in necrot-
versity hospitals of the EBGA network (Lyon, Gre-
izing aspergillosis (patients H and J), bronchitis as-
noble and Milan). The SSDP analysis classi¢ed the
pergillosis (patient M) and bronchial colonization
52 strains of
A. fumigatus
(nine monosporal strains
(patients G, I, K and N), the genotype obtained in
and 43 clinical strains) into 22 di¡erent types (Table
each transplant patient made it possible to separate
1). Among these 52 strains, 22 unrelated strains were
each strain, whereas in the cases of patients G and
randomly sampled (Table 1) and di¡erentiated into
N, three and ¢ve strains, respectively, were isolated
18 types. The SSDP analysis showed a high level of
from the same sample (Table 1). By contrast, the
discrimination :
D=0.96.
Additionally, for a typing
system to be used as a single typing method, should ideally be
s
0.95 [20].
D
patients
A
(14
isolates),
B
(¢ve
isolates)
Sequences of the ¢ve SSDP markers
a
RAPD primer
SSDP marker
Primer
Sequence
OpQ 6
SSDP 1
Afp1
TTGGGGAGATTACCGAACTGG
Afp2
CCTTGACAACCGTCCCATTTC
AfMp1
CAGTTCCCGTCTTGACCTC
AfMp2
CAGTTCCCGTTTTCCTAGTA
Afd1
GTATTGCCCTATAACTTCTT
Afd2
GTATTGCCCTATTCCCAAAG
Afs4
GTATTGCCCTAGCTTACTAA
Afr4
GTATTGCCCTATTACTAAAG
Afs5
GTATTGCCCTAAGGATTCTA
Afr5
GTATTGCCCTAGCTTGCTAA
OPAj12
SSDP 2
OPAj12 R108
SSDP 3
R108 R108
SSDP 4
R108 R108 R108
a
SSDP 5
E
each case to be infected by only one type of strain
Table 2
OpQ 6
and
(four isolates) with invasive aspergillosis seem in
The underlines indicate the sequences of the RAPD primers.
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
101
(Table 1). Moreover, 14 strains of patient A have
In this investigation, we suggest an original meth-
presented the same genotype (type 30) despite the
od with excellent reproducibility and a high discrim-
fact that they were isolated at di¡erent dates and
inatory power, without special optimization of reac-
from
tion conditions for each primer. The extra work to
di¡erent
body
locations
such
as
the
hepatic
artery, the portal vein and the lungs. These observa-
sequence
tions have to be con¢rmed by more IA cases ; how-
possible ¢rstly to check whether a fragment is really
ever, other epidemiological investigations by MRSs
discriminating, secondly to increase the reproducibil-
and a combination of RFLP and RAPD have shown
ity and ¢nally to identify the DNA sequence. The
that patients with IA, three and two cases respec-
SSDP analysis is useful for routine typing of a large
tively,
number of isolates required in our multicenter study
were
although
a
infected
wide
by
variety
only
one
of strain
strain
type,
types existed
in
fragments
generated
by
RAPD
makes
it
of the EBGA network. This molecular method, in
their environment [4,7]. These ¢ndings also support
association
our previous hypothesis that the pathogenicity not
our EBGA partners (Montpellier (France), Brussels
only depended on host immunity, but also on fun-
(Belgium), Leeds (UK), Lyon (France)), will contrib-
A. fumigatus
ute towards a better understanding of the epidemiol-
gus-related
factors
and
that
certain
strains are more virulent than others [16,19].
optimal
jective
Type
SSDP 1
SSDP 2
SSDP 3
laxis.
1
+
+
+
3
+
2
+
+
3 3
3
+
+
SSDP 4
SSDP 5
3 3 3 3 3 3 3 3
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
3 3
3
3 3 3 3
9
+
+
+
+
10
+
+
+
5
+
6
+
7 8
+
3
+
3
3 3
3
13
+
+
14
+
11 12
15 16
3 3
17
+
18
+
19 20
other
methods
developed
by
strategy
for
the
prevention
of
this
fungal
infection in `at risk' patients, with the ultimate ob-
Types of pro¢le with the combination of the ¢ve SSDP markers
4
the
ogy of invasive aspergillosis and development of an
Table 3
3
with
+
3
+
3 +
3
3 3
3
+
+
+
+
+
+
+
3 3 3 3 + + + +
+ + + +
3 3 3 3 3 3 3 3
+
of
designing
recommendations
for
prophy-
Acknowledgments We would like to thank A. Altmann for English proofreading. This work is part of the research program of the EBGA network and was supported in part by funds from the `Recherche Clinique-CHU Grenoble', France. P. Mondon was supported by a grant from the `Agence de l'Environnement et de la è nergie' (ADEME), Paris, France. êtrise de l'E Ma|
References
+ + +
[1] Burnie, J.P., Coke, A. and Matthews, R.C. (1992) Restriction
Aspergillus fumigatus
3 3
3
3 3 3 3
25
+
+
+
+
+
26
+
+
+
+
+
+
+
+
+
+
è, F., Dupouy, C., Derouin, F. [4] Girardin, H., Sarfati, J., Traore
+
+
è, J.P. (1994) Molecular epidemiology of nosocomial and Latge
+
+
+
+
+
+
21 22 23 24
+
3
+
3
3 3
3
29
+
+
30
+
27 28
31 32
3 3
+
3 +
3
3 3 3 3
+ + + +
endonuclease analysis of
DNA. J. Clin.
Pathol. 45, 324^327. [2] Rinyu, E., Varga, J. and Ferenzy, L. (1995) Phenotypic and genotypic
analysis
of
variability
in
Aspergillus fumigatus.
J. Clin. Microbiol. 33, 2567^2575. è, J.P., Srikantha, T., Morrow, B. and Soll, [3] Girardin, H., Latge D.R. (1993) Development of DNA probes for ¢ngerprinting
Aspergillus fumigatus.
J. Clin. Microbiol. 31, 1547^1554.
invasive aspergillosis. J. Clin. Microbiol. 32, 684^690. [5] Aufauvre-Brown, A., Cohen, J. and Holden, D.W. (1992) Use of randomly ampli¢ed polymorphic DNA markers to distin-
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
102
Aspergillus fumigatus.
J. Clin. Microbiol. 30,
by multilocus enzyme electrophoresis. J. Clin. Microbiol. 34,
[6] Loudon, K.W., Burnie, J.P., Coke, A.P. and Matthews, R.C.
[15] Brenier, M.P., Lebeau, B., Mondon, P., Coursange, E., No-
(1993) Application of polymerase chain reaction to ¢nger-
lard, N., Symoens, F. and EBGA Network (1995) Epidemiol-
guish isolates of
2559^2568.
2991^2993.
printing
Aspergillus fumigatus
by random ampli¢cation poly-
ogy of aspergillosis : Origin of contamination by molecular genome typing. Trend Inv. Fung. Infect. III, Brussels, 35,
morphic DNA. J. Clin. Microbiol. 31, 1117^1121. [7] Tang, C.M., Cohen, J., Rees, A.J. and Holden, D.W. (1994) Molecular epidemiological study of invasive pulmonary asper-
54. Abstr. è lu, J., Lebeau, B., Ambroise-Thomas, P. and [16] Mondon, P., The
Aspergillus fumigatus
gillosis in renal transplantation unit. Eur. J. Microbiol. Inf.
Grillot, R. (1995) Virulence of
Dis. 13, 318^321.
investigated by random ampli¢ed polymorphic DNA analysis.
[8] Lin, D., Lehmann, P.F., Hamory, B.H. et al. (1995) Comparison of three typing methods for clinical and environmental isolates of
Aspergillus fumigatus.
J. Clin. Microbiol. 33, 1596^
J. Med. Microbiol. 42, 299^303. [17] Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. and Tingey, S.V. (1990) DNA polymorphisms ampli¢ed by arbitrary primers are useful as genetic markers. Nucleic Acids
1601. [9] Loudon, K.W., Coke, A.P., Burnie, J.P., Lucas, G.S. and Yin Liu, J.A. (1994) Invasive aspergillosis : clusters and sources ?
Res. 18, 6531^6535. [18] Bodey, G.P. and Vartivarian, S. (1989) Aspergillosis. Eur. J. Microbiol. Infect. Dis. 8, 413^437.
J. Med. Vet. Mycol. 32, 217^224. [10] Loudon, K.W. and Burnie, J.P. (1995) Comparison of three typing methods for clinical and environmental isolates of
pergillus fumigatus.
strains
As-
[19] Mondon, P., De Champs, C., Donadille, A., Ambroise-Thomas, P. and Grillot, R. (1996) Variation in virulence of
pergillus fumigatus
J. Clin. Microbiol. 33, 3362^3363.
[11] Anderson, M.J., Gull, K. and Denning, D.W. (1996) Molec-
As-
strains in a murine model of invasive pul-
monary aspergillosis. J. Med. Microbiol. 45, 186^191.
ular typing by random ampli¢ed of polymorphic DNA and
[20] Struelens, M.J. and the Members of ESGEM (1996) Consen-
As-
sus guidelines for appropriate use and evaluation of microbial
[12] Mondon, P., Brenier, M.P., Coursange, E., Lebeau, B., Sy-
[21] Hunter, P. (1990) Reproducibility and indices of discrimina-
moens, F., Nolard, N. and EBGA Network (1996) Multicen-
tory power of microbial typing methods. J. Clin. Microbiol.
M13 southern hybridization of related paired isolates of
pergillus fumigatus.
epidemiologic typing systems. Clin. Microbiol. Inf. 2, 2^11.
J. Clin. Microbiol. 34, 87^93.
tric epidemiological study of
Aspergillus fumigatus
strains. Fo-
28, 1903^1905. [22] He, G., Prakash, C.S., Jarret, R.L., Tuzum, S. and Qiu, J.
cus Fung. Inf. 6, New Orleans (LA), 06 Abstr. [13] Symoens, F., Van Vaerenbergh, B., Bastide, J.-M., Chapuis,
(1994) Comparison of gel matrices for resolving PCR ampli-
è , M., Mondon, P. and F., Grillot, R., Lebeau, B., Mallie
¢ed DNA ¢ngerprinting pro¢les. PCR Methods Appl. 4, 50^
EBGA network. (1995) Typing of
Aspergillus fumigatus
by
random ampli¢ed polymorphic DNA. 2nd European Confer-
è , M., Renaud, F., Sy[14] Rodriguez, E., De Meeus, T., Mallie moens, F., Mondon, P. and EBGA Network (1996) Multi-
Aspergillus fumigatus
P.E.,
Meis,
J.F.G.M.,
Sarfati
J.,
Hoogkamp-
è , J.P. and Melchers, W.J.G. (1996) Korstanje, J.A.A., Latge
ence on Medical Mycology, Brussels, PO 1.12 Abstr.
centric epidemiological study of
51. [23] Verweij,
isolates
Genotypic characterisation of sequential
Aspergillus fumigatus
isolates from patients with cystic ¢brosis. J. Clin. Microbiol. 34, 2595^2597.
FEMSIM 720 16-4-97
Molecular typing of Aspergillus fumigatus strains by sequence-speci¢c DNA primer (SSDP) analysis P. Mondon*, M.P. Brenier, E. Coursange, B. Lebeau, P. Ambroise-Thomas, R. Grillot 1 è dicale et Mole è culaire, Laboratoire Relation Ho ê te-Agents Pathoge énes, UPRES-A CNRS 5082, Faculte è de Me è decine, Parasitologie-Mycologie Me è Joseph Fourier Grenoble I, Domaine de la Merci, 38706 La Tronche, France Universite
Received 10 September 1996; revised 5 November 1996; accepted 19 November 1996
Abstract
A PCR typing method has been developed and tested to investigate the polymorphism of clinical strains of Aspergillus . Firstly, the DNA fragments from random amplified polymorphic DNA (RAPD) patterns of nine epidemiologically and geographically non-related monosporal strains of A. fumigatus were cloned and sequenced. The pairs of five sequencespecific DNA primers (SSDP), characteristic of the 5P and 3P extremities of the RAPD products, were then used in high stringency PCR to type 43 clinical strains of A. fumigatus from 13 patients, according to the presence or absence of a single amplified band. This original approach, which uses the advantages of PCR, has made it possible to overcome the difficulties resulting from the low stringency amplification. The SSDP analysis of 51 A. fumigatus strains (9 unrelated monosporal strains and 43 clinical strains from 13 patients) can be classed into 22 different types with a high reproducibility and a high level of discrimination (D = 0.96). The results suggest that seven lung transplant patients with necrotizing aspergillosis, bronchitis aspergillosis and bronchial colonization were infected by multiple strain genotypes, whereas three patients with invasive aspergillosis seem to have been infected by a single strain. fumigatus
Keywords : Aspergillus fumigatus
; Polymerase chain reaction; Molecular marker; Lung transplant; Invasive aspergillosis
1. Introduction
Aspergillus
fumigatus
is an opportunistic fungal
* Corresponding author. Tel.: +33 (4) 76 63 71 01; fax: +33 (4) 76 51 86 67. 1
European Research Group on Biotype and Genotype of (EBGA Network). Members of the EBGA Network who contributed to this study: P. Mondon, F. Symoens, E. Rodriguez, F. Chaib, B. Lebeau, M.A. Piens, A.-M. Tortorano, M. Mallieè, F. Chapuis, A. Carlotti, J. Villard, M.A. Viviani, N. Nolard, J.-M. Bastide, P. Ambroise-Thomas and R. Grillot. Aspergillus
pathogen, responsible for invasive aspergillosis (IA), a fatal disease most often acquired within the hospital environment. Several questions concerning the mode of transmission and the potential origins of patient contamination are to date unsolved. In recent years, several di¡erent molecular methods have been developed to type isolates of A. fumigatus including restriction endonuclease analysis (REA) [1], isoenzyme analysis (IEA) [2], moderately repetitive sequences (MRSs) [3,4] and random ampli¢ed polymorphic DNA (RAPD) [5^7]. Each method has its own advantages and disadvantages and no
0928-8244 / 97 / $17.00 Copyright ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V. PII S 0 9 2 8 - 8 2 4 4 ( 9 6 ) 0 0 1 1 0 - 1
FEMSIM 720 16-4-97
96
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
molecular typing method appears to be clearly better than the rest. The combination of typing methods appears to provide the best discriminatory power [8^11]. With this object in view, we have created a European research group to study the biotype and the genotype of A. fumigatus (EBGA network) using a combination of several molecular typing methods [12^14], the ultimate objective being a better understanding of the epidemiology of invasive aspergillosis. Our team has investigated the RAPD analysis [15,16], which employs a single short primer (10mer) with an arbitrary nucleotide sequence in a low stringency polymerase chain reaction (PCR), to randomly amplify genomic DNA [17]. This method is quick to perform, large numbers of isolates can be analyzed in a few days and it generates a relatively high degree of discrimination among isolates. However, the intra-laboratory reproducibility is not ideal. Occasionally the bands generated have di¡erent staining intensities and this has sometimes made the interpretation of the banding patterns di¤cult and complex in the same laboratory [7^9]. In this context, we propose an original PCR typing strategy using the advantages of RAPD, without the drawback linked to the reproducibility. The RAPD analysis was used to produce banding patterns which made it possible to di¡erentiate the strains. The discriminating fragments were sequenced to design pairs of sequence-speci¢c DNA primers (SSDP) (20-mers) characteristic of the two extremities of the fragments considered. The SSDP was used in higher stringency PCR to study the reality of the presence or the absence of the band suggested as discriminating. The purpose of this study was initially to select SSDP markers which would make it possible to differentiate DNA of nine epidemiologically and geographically non-related monosporal strains. Secondly, these SSDP markers were applied to demonstrate the importance of the SSDP analysis in the investigation of its discriminatory power and the genetic polymorphism of 43 clinical strains from 13 patients, isolated in three EBGA centers (University hospitals of Lyon (France), Milan (Italy) and Grenoble (France)), which were responsible for different types of aspergillosis in invasive aspergillosis and lung transplant patients (bronchial colonization, necrotizing aspergillosis, bronchitis aspergillosis).
2. Materials and methods
2.1. Fungal strains
The nine monosporal strains were from the culture collection of the Institute of Hygiene and Epidemiology Mycology (IHEM Brussels, Belgium) and selected to be epidemiologically and geographically non-related (Table 1). The 43 clinical strains were collected by the three University hospitals involved in the EBGA network (Lyon, Grenoble and Milan), from six patients with proven or highly probable invasive aspergillosis (Table 1, patients A^F) and from seven lung transplant patients with necrotizing aspergillosis, bronchitis aspergillosis and bronchial colonization (Table 1, patients G^N). Each A. fumigatus infection was well documented according to the criteria of aspergillosis classi¢cation [18]. The strains were centralized, preserved by lyophilization at the IHEM and redistributed to the EBGA centers for typing. 2.2. Ampli¢cation conditions
Ampli¢cation was performed using puri¢ed DNA as previously described [19]. The reaction mixtures (30 Wl) contained 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2 , 0.005% Tween 20, 0.005% NP-40, 75 WM each dNTP (Boehringer), 0.5 U of Replitherm0 DNA polymerase (Epicentre), 50 ng of DNA template and 0.2 WM of random primer (10-mers) or 0.1 WM of each speci¢cally designed primer (20-mers). Each reaction mixture was overlaid with mineral oil (30 Wl) to prevent evaporation. In a DNA thermal cycler (Perkin Elmer 480) ampli¢cations were carried out after an initial heat denaturation (95³C for 6 min). For arbitrary short primers 40 cycles consisting of 20 s at 95³C, 30 s at 36³C and 60 s at 63³C were used, while for speci¢c long primers, 30 cycles consisting of 20 s at 95³C, 30 s at 55³C and 60 s at 72³C were performed. Ampli¢cation products were resolved by electrophoresis on 1.5% agarose gels and stained with ethidium bromide (0.5 Wg ml31 ). 2.3. RAPD primers
One hundred and twenty RAPD decamers de-
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
97
signed according to the following theoretical parameters: G-C content greater than 60%, no dryad symmetry and non-self-complementary, were purchased from Operon Technologies Inc. (Alameda, CA), kits (OPAH, OPAI, OPAJ, OPP, OPR, OPQ) and the primer R108 [5]. 2.4. DNA cloning and sequencing
The DNA template was ampli¢ed with a short primer (10-mer) and 2 WCi [K-35S]dATP (Amersham). 2 Wl of each reaction mixture was separated by electrophoresis on an acrylamide (6%)-urea (50%) sequencing gel and subjected to autoradiography. Bands of interest were collected, the gel was homogenized with a conical grinder and 10 Wl of the mixture was ampli¢ed directly as previously described, without further DNA puri¢cation. The ampli¢cation products were cloned using a TA cloning kit (Invitrogen). The insert DNA was sequenced using the Sequenase 2.0 kit (United States Biochemical Corp.) and [K-35S]dATP (Amersham). On the basis of the sequence, the oligonucleotide primer sequences were selected including the sequence of the RAPD primer (10-mer), plus 10 bases characteristic of the fragment excised and synthesized by Eurogentec (Belgium). 2.5. Discriminatory power
The discriminatory power of a typing method, D, is the average probability that the typing system will assign a di¡erent type to two unrelated strains randomly sampled [20]. The discriminatory power was calculated in the manner described by Hunter [21]. 3. Results and discussion
3.1. Development of SSDP sets
One hundred and twenty decamers from Operon technology kits (OPAH, OPAI, OPAJ, OPP, OPR, OPQ) and the primer R108 [5] were tested to generate RAPD markers from genomic DNAs of nine unrelated monosporal strains (8065, 8070, 8075, 8080, 8086, 8095, 9005 8090, and 9000) from the IHEM culture collection (Table 1). Three random
Fig. 1. DNA template of nine unrelated monosporal strains from the IHEM collection (IHEM numbers: 8065, 8070, 8075, 8080, 8086, 8095, 9005, 8090, 9000) ampli¢ed with the RAPD primer R108, 2 WCi [K-35 S]dATP (Amersham) and separated by electrophoresis on an acrylamide (6%)-urea (50%) sequencing gel (lanes 1^9). Lane 0: negative control without DNA.
primers (10-mers) (R108, OPQ6, OPAj12) were selected to generate banding patterns which seemed to discriminate the di¡erent strains. From the RAPD patterns, the discriminating fragment was excised from the acrylamide gel, re-ampli¢ed, and sequenced to design a pair of longer primers (20-mers) characteristic of the two extremities of the considering fragment. The sequence of SSDP included the sequence of the RAPD primer (10-mers), plus 10 bases characteristic of the fragment excised (Table 2). The SSDP marker was then used in higher stringency PCR (annealing temperature: 55³C) with the DNA of the nine unrelated strains to check the presence or the absence of the band suggested as discriminating.
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
98
Table 1 Genotypes of the nine unrelated monosporal strains and the 43 clinical strains of Patient
Strain
Origin
Infection
b
A. fumigatus
c Sample
(IHEM no.)
8090 9000 8080 8075 8095 9005 8065 8070 8086
a a a a a a a a a
obtained by SSDP analysis SSDP analysis SSDP
SSDP
1
2
3
4
5
3
+
+
+
3
11
Iraq
Desert soil
+
+
+
30
Belgium
Lung of buzzard
+
+
32
Belgium
Human
+
+
Belgium
Dust from mattress
Belgium
Seeds of cereal
Milan
Human bronchial secretion
Milan
Intensive care unit env.
Milan
Human skin surgical wound
+
3 3
+
3 3 3
9010^9014
Milan
Abdominal drain (27/01/94)
+
9015
Milan
Hepatic artery (28/01/94)
+
A
9016^9019
Milan
Hepatic ilium (2/2/94)
+
A
9020
Milan
IA
Bile (2/2/94)
+
A
9021
Milan
Liver transplant
Lung ^ autopsy (7/2/94)
+
A
9022
Hepatic artery ^ autopsy
+
Milan
(7/2/94)
30
+
+
30
3
+
+
29
Grenoble
IA
Tracheal aspirate (2/12/92)
+
3
3
+
+
30
3 3
3 3
3 3
3 3
+
24
+
24
+
3
+
+
+
26
3 3 3
3 3 3
Grenoble
G
9615
Grenoble
H
9730
H
9731
I
9733
I
9734
a
+
+
9614
9752
30
+
+
9613
9753
+
BA (31/12/92))
G
M
30
+
IA
G
M
+
Grenoble
7969
a
30
+
26
F
9745
30
+
+
9417
9744
+
+
+
9418^9420
K
+
+
E
K
5
3
E
a
7
+
7965
9739
27
3 3
Bronchial secretions (2/3/95)
D
9740
3 3 3 3 3 3
24
+
IA
7964
J
3 3 3 3 3 3
3 3
+
30
C
J
+
3 3
14
+
Milan
a
+
3
+
31
3
+
a 9637 ^9641
a
+
+
3
B
a
+
+
3
Milan
a
3 3
3 3 3 3 3
+
9023
a
3 3
Portal vein ^ autopsy (7/2/94)
A
a
SSDP
Aircraft fuel
A
a
SSDP
England
A
a
Type
SSDP
Grenoble
IA
BAL (13/1/95)
Grenoble
BMT
BA (13/1/95)
Grenoble
IA
Sinus drainage (19/5/87)
Grenoble
Bronchial colonization
BA (21/4/95) BA (21/4/95)
LT
BA (21/4/95)
+
3 3
+
+
28
+
+
32
+
24
3
Lyon
Necrotizing A
BA (11/6/93)
+
3
3
3
+
22
Lyon
LT
BA (23/6/93)
+
+
+
+
+
25
Lyon
Bronchial colonization
Sputum (24/8/93) Sputum (26/8/93)
3 3
3 3
4
LT
3 3
3
Lyon
+
20
Lyon
Necrotising A
BA (16/11/93)
Lyon
LT
Sputum (24/12/93)
3
3 3
Lyon
Bronchial colonization
Sputum (26/02/93)
3
Lyon
H-L T
Sputum (30/3/93)
+
3 3
Lyon
Bronchitis A
BAL (30/6/93)
Lyon
LT
BAL (21/6/93)
FEMSIM 720 16-4-97
+
3 3
3 3
+ +
+
+
+
26
+
+
+
28
+
20
+
18
+ +
3 3
3 3 3
3
8
+
+
32
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
99
Table 1 (continued) Genotypes of the nine unrelated monosporal strains and the 43 clinical strains of A. fumigatus obtained by SSDP analysis Samplec SSDP analysis Type Patient Strain Origin Infectionb (IHEM no.) SSDP SSDP SSDP SSDP SSDP 1 2 3 4 5 a Lyon Bronchial colonization BAL (15/12/94) + + + 3 + 17 N 9393 N 9394 Lyon BAL (15/12/94) + 3 3 3 + 22 N 9395^9396 Lyon LT BAL (15/12/94) + 3 3 3 3 6 N 9397 Lyon BAL (15/12/94) 3 + 3 3 + 23 a Unrelated strains randomly sampled. b IA: invasive aspergillosis; A: aspergillosis; BMT: bone marrow transplant; LT: lung transplant; HT: heart transplant; H-L T: heart-lung transplant. c BAL: bronchoalveolar lavage; BA: bronchial aspirate.
The RAPD analysis generates a relatively complex ¢ngerprinting pattern, which is commonly separated on agarose gel and stained with ethidium bromide. However, it has been shown that gel electrophoresis is a major cause of experimental variability [22]. The problem of band intensities could be minimized when the DNA is ampli¢ed in the presence of radioactivity ([K-35 S]dATP) and separated on a 6% denaturing acrylamide gel (Fig. 1). This extra work of running a denaturing acrylamide gel is rewarded by much more and better resolved data than can be obtained using agarose gel. However, it is time consuming and expensive for wide epidemiological application. This powerful detection system has therefore been used to select the discriminating band from the RAPD analysis which will be a potential SSDP marker. The decamer R108 generated nine RAPD fragments which seemed discriminating (Fig. 1). The sequencing of these nine bands has provided nine primer pairs, but only three of these pairs (Table 2: SSDP 3, SSDP 4, SSDP 5) were really discriminating, in each case according to the presence (+) or absence (3) of a single ampli¢ed band (SSDP 4: 0.75 kb, SSDP 3: 0.55 kb, SSDP 5: 0.35 kb) (Table 1). Each of the other six primer pairs ampli¢ed a constant fragment with all of the nine monosporal strains. Six discriminating RAPD fragments were obtained with the primer OPQ6 (GAGCGCCTTG) but only one primer pair (Afp1/Afp2: SSDP 1) characteristic of a 0.95 kb fragment separated the nine unrelated strains into two di¡erent types: 3+ and 63 (Table 1). The decamer OPAj12 (CAGTTCCCGT)
generated three discriminating RAPD fragments, but only one primer pair (AfMp1/AfMp2: SSDP 2) out of three was discriminating and ampli¢ed a 1.1 kb fragment with ¢ve strains out of nine (Table 1). The ¢ve primer pairs (SSDP markers) described in Table 2 and their di¡erent +/3 combinations which de¢ne the genotype (Table 3) made it possible to distinguish all of the nine unrelated monosporal strains (Table 1). Each SSDP analysis was repeated at least twice, the result remaining the same and a primer pair Afc1/Afc2, deduced from the sequencing of a conserved fragment (1.4 kb) from the RAPD pattern of the primer OPQ6 was used as a positive control of each ampli¢cation [16,19]. 3.2. RAPD/SSDP comparison
The cloning and the sequencing of 13 out of 18 fragments, which appeared to di¡erentiate the isolates by RAPD analysis, have shown that in fact they were not discriminating when their characteristic primer pairs were used in higher stringency PCR. This highlights the problem of the band intensity of the RAPD method which uses low stringency PCR (annealing temperature: 36³C) in combination with a single short primer (10-mer). The RAPD technique has the advantage of being relatively quick and straightforward, but has major disadvantages including optimization of protocols, inter-run and interlaboratory reproducibility and interpretation of ampli¢ed patterns [7^10,20,23]. The SSDP analysis uses a high stringency PCR and makes it possible to limit these factors.
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
100
The decamer R108 has been widely described as giving the highest level of variation [5,7,8,11,23]. In our study, the results con¢rm its high discriminatory power (Fig. 1), but not the reproducibility, despite the fact that a higher detection system has been used to visualize the RAPD patterns. Recently, Verweij et
Fig. 2. DNA template of eight clinical strains from four patients
al. [23] have highlighted the necessity to envisage a
(IHEM
new
9745) ampli¢ed with the primer pair Afd1/Afd2 : SSPP 3 (lanes
typing
method
to
achieve
inter-laboratory
standardization. Our aim in view was not to describe a typing system based on the decamer R108, but to develop a new approach of PCR typing using RAPD
numbers :
9739,
9740,
9730,
9731,
9752,
9753,
9744,
1^8) and the positive control primer pair Afc1/Afc2 (lanes 1P^8P). Lanes 0 and 0P : negative control without DNA with Afd1/Afd2 and
Afc1/Afc2,
respectively.
Lane
MW :
1-kb
ladder
marker
(Gibco-BRL).
analysis to select the molecular markers for routine typing. The SSDP analysis has the advantage of combining
the
discriminatory
power
of
di¡erent
3.4. SSDP analysis of clinical strains
RAPD primers (R108, OPQ6 and OPAj12). Furthermore, the typing by each SSDP marker, according to
The feasibility of the SSDP analysis for the typing
the presence or absence of a single ampli¢ed band
of a large number of strains has been con¢rmed with
(Fig. 2), allows quick and easy coding (Table 3) of a
43 clinical strains isolated from seven lung transplant
large number of samples.
patients (bronchial colonization, necrotizing aspergillosis, bronchitis aspergillosis) (Table 1, patients G^
3.3. Discriminatory power of the SSDP analysis
N) and six invasive aspergillosis patients (Table 1, patients A^F). A wide genetic polymorphism has
The ¢ve SSDP markers have been applied to in-
been observed in the lung transplant patients com-
vestigate 43 clinical strains isolated in the three uni-
pared to the invasive aspergillosis. Indeed, in necrot-
versity hospitals of the EBGA network (Lyon, Gre-
izing aspergillosis (patients H and J), bronchitis as-
noble and Milan). The SSDP analysis classi¢ed the
pergillosis (patient M) and bronchial colonization
52 strains of
A. fumigatus
(nine monosporal strains
(patients G, I, K and N), the genotype obtained in
and 43 clinical strains) into 22 di¡erent types (Table
each transplant patient made it possible to separate
1). Among these 52 strains, 22 unrelated strains were
each strain, whereas in the cases of patients G and
randomly sampled (Table 1) and di¡erentiated into
N, three and ¢ve strains, respectively, were isolated
18 types. The SSDP analysis showed a high level of
from the same sample (Table 1). By contrast, the
discrimination :
D=0.96.
Additionally, for a typing
system to be used as a single typing method, should ideally be
s
0.95 [20].
D
patients
A
(14
isolates),
B
(¢ve
isolates)
Sequences of the ¢ve SSDP markers
a
RAPD primer
SSDP marker
Primer
Sequence
OpQ 6
SSDP 1
Afp1
TTGGGGAGATTACCGAACTGG
Afp2
CCTTGACAACCGTCCCATTTC
AfMp1
CAGTTCCCGTCTTGACCTC
AfMp2
CAGTTCCCGTTTTCCTAGTA
Afd1
GTATTGCCCTATAACTTCTT
Afd2
GTATTGCCCTATTCCCAAAG
Afs4
GTATTGCCCTAGCTTACTAA
Afr4
GTATTGCCCTATTACTAAAG
Afs5
GTATTGCCCTAAGGATTCTA
Afr5
GTATTGCCCTAGCTTGCTAA
OPAj12
SSDP 2
OPAj12 R108
SSDP 3
R108 R108
SSDP 4
R108 R108 R108
a
SSDP 5
E
each case to be infected by only one type of strain
Table 2
OpQ 6
and
(four isolates) with invasive aspergillosis seem in
The underlines indicate the sequences of the RAPD primers.
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
101
(Table 1). Moreover, 14 strains of patient A have
In this investigation, we suggest an original meth-
presented the same genotype (type 30) despite the
od with excellent reproducibility and a high discrim-
fact that they were isolated at di¡erent dates and
inatory power, without special optimization of reac-
from
tion conditions for each primer. The extra work to
di¡erent
body
locations
such
as
the
hepatic
artery, the portal vein and the lungs. These observa-
sequence
tions have to be con¢rmed by more IA cases ; how-
possible ¢rstly to check whether a fragment is really
ever, other epidemiological investigations by MRSs
discriminating, secondly to increase the reproducibil-
and a combination of RFLP and RAPD have shown
ity and ¢nally to identify the DNA sequence. The
that patients with IA, three and two cases respec-
SSDP analysis is useful for routine typing of a large
tively,
number of isolates required in our multicenter study
were
although
a
infected
wide
by
variety
only
one
of strain
strain
type,
types existed
in
fragments
generated
by
RAPD
makes
it
of the EBGA network. This molecular method, in
their environment [4,7]. These ¢ndings also support
association
our previous hypothesis that the pathogenicity not
our EBGA partners (Montpellier (France), Brussels
only depended on host immunity, but also on fun-
(Belgium), Leeds (UK), Lyon (France)), will contrib-
A. fumigatus
ute towards a better understanding of the epidemiol-
gus-related
factors
and
that
certain
strains are more virulent than others [16,19].
optimal
jective
Type
SSDP 1
SSDP 2
SSDP 3
laxis.
1
+
+
+
3
+
2
+
+
3 3
3
+
+
SSDP 4
SSDP 5
3 3 3 3 3 3 3 3
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
3 3
3
3 3 3 3
9
+
+
+
+
10
+
+
+
5
+
6
+
7 8
+
3
+
3
3 3
3
13
+
+
14
+
11 12
15 16
3 3
17
+
18
+
19 20
other
methods
developed
by
strategy
for
the
prevention
of
this
fungal
infection in `at risk' patients, with the ultimate ob-
Types of pro¢le with the combination of the ¢ve SSDP markers
4
the
ogy of invasive aspergillosis and development of an
Table 3
3
with
+
3
+
3 +
3
3 3
3
+
+
+
+
+
+
+
3 3 3 3 + + + +
+ + + +
3 3 3 3 3 3 3 3
+
of
designing
recommendations
for
prophy-
Acknowledgments We would like to thank A. Altmann for English proofreading. This work is part of the research program of the EBGA network and was supported in part by funds from the `Recherche Clinique-CHU Grenoble', France. P. Mondon was supported by a grant from the `Agence de l'Environnement et de la è nergie' (ADEME), Paris, France. êtrise de l'E Ma|
References
+ + +
[1] Burnie, J.P., Coke, A. and Matthews, R.C. (1992) Restriction
Aspergillus fumigatus
3 3
3
3 3 3 3
25
+
+
+
+
+
26
+
+
+
+
+
+
+
+
+
+
è, F., Dupouy, C., Derouin, F. [4] Girardin, H., Sarfati, J., Traore
+
+
è, J.P. (1994) Molecular epidemiology of nosocomial and Latge
+
+
+
+
+
+
21 22 23 24
+
3
+
3
3 3
3
29
+
+
30
+
27 28
31 32
3 3
+
3 +
3
3 3 3 3
+ + + +
endonuclease analysis of
DNA. J. Clin.
Pathol. 45, 324^327. [2] Rinyu, E., Varga, J. and Ferenzy, L. (1995) Phenotypic and genotypic
analysis
of
variability
in
Aspergillus fumigatus.
J. Clin. Microbiol. 33, 2567^2575. è, J.P., Srikantha, T., Morrow, B. and Soll, [3] Girardin, H., Latge D.R. (1993) Development of DNA probes for ¢ngerprinting
Aspergillus fumigatus.
J. Clin. Microbiol. 31, 1547^1554.
invasive aspergillosis. J. Clin. Microbiol. 32, 684^690. [5] Aufauvre-Brown, A., Cohen, J. and Holden, D.W. (1992) Use of randomly ampli¢ed polymorphic DNA markers to distin-
FEMSIM 720 16-4-97
P. Mondon et al. / FEMS Immunology and Medical Microbiology 17 (1997) 95^102
102
Aspergillus fumigatus.
J. Clin. Microbiol. 30,
by multilocus enzyme electrophoresis. J. Clin. Microbiol. 34,
[6] Loudon, K.W., Burnie, J.P., Coke, A.P. and Matthews, R.C.
[15] Brenier, M.P., Lebeau, B., Mondon, P., Coursange, E., No-
(1993) Application of polymerase chain reaction to ¢nger-
lard, N., Symoens, F. and EBGA Network (1995) Epidemiol-
guish isolates of
2559^2568.
2991^2993.
printing
Aspergillus fumigatus
by random ampli¢cation poly-
ogy of aspergillosis : Origin of contamination by molecular genome typing. Trend Inv. Fung. Infect. III, Brussels, 35,
morphic DNA. J. Clin. Microbiol. 31, 1117^1121. [7] Tang, C.M., Cohen, J., Rees, A.J. and Holden, D.W. (1994) Molecular epidemiological study of invasive pulmonary asper-
54. Abstr. è lu, J., Lebeau, B., Ambroise-Thomas, P. and [16] Mondon, P., The
Aspergillus fumigatus
gillosis in renal transplantation unit. Eur. J. Microbiol. Inf.
Grillot, R. (1995) Virulence of
Dis. 13, 318^321.
investigated by random ampli¢ed polymorphic DNA analysis.
[8] Lin, D., Lehmann, P.F., Hamory, B.H. et al. (1995) Comparison of three typing methods for clinical and environmental isolates of
Aspergillus fumigatus.
J. Clin. Microbiol. 33, 1596^
J. Med. Microbiol. 42, 299^303. [17] Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. and Tingey, S.V. (1990) DNA polymorphisms ampli¢ed by arbitrary primers are useful as genetic markers. Nucleic Acids
1601. [9] Loudon, K.W., Coke, A.P., Burnie, J.P., Lucas, G.S. and Yin Liu, J.A. (1994) Invasive aspergillosis : clusters and sources ?
Res. 18, 6531^6535. [18] Bodey, G.P. and Vartivarian, S. (1989) Aspergillosis. Eur. J. Microbiol. Infect. Dis. 8, 413^437.
J. Med. Vet. Mycol. 32, 217^224. [10] Loudon, K.W. and Burnie, J.P. (1995) Comparison of three typing methods for clinical and environmental isolates of
pergillus fumigatus.
strains
As-
[19] Mondon, P., De Champs, C., Donadille, A., Ambroise-Thomas, P. and Grillot, R. (1996) Variation in virulence of
pergillus fumigatus
J. Clin. Microbiol. 33, 3362^3363.
[11] Anderson, M.J., Gull, K. and Denning, D.W. (1996) Molec-
As-
strains in a murine model of invasive pul-
monary aspergillosis. J. Med. Microbiol. 45, 186^191.
ular typing by random ampli¢ed of polymorphic DNA and
[20] Struelens, M.J. and the Members of ESGEM (1996) Consen-
As-
sus guidelines for appropriate use and evaluation of microbial
[12] Mondon, P., Brenier, M.P., Coursange, E., Lebeau, B., Sy-
[21] Hunter, P. (1990) Reproducibility and indices of discrimina-
moens, F., Nolard, N. and EBGA Network (1996) Multicen-
tory power of microbial typing methods. J. Clin. Microbiol.
M13 southern hybridization of related paired isolates of
pergillus fumigatus.
epidemiologic typing systems. Clin. Microbiol. Inf. 2, 2^11.
J. Clin. Microbiol. 34, 87^93.
tric epidemiological study of
Aspergillus fumigatus
strains. Fo-
28, 1903^1905. [22] He, G., Prakash, C.S., Jarret, R.L., Tuzum, S. and Qiu, J.
cus Fung. Inf. 6, New Orleans (LA), 06 Abstr. [13] Symoens, F., Van Vaerenbergh, B., Bastide, J.-M., Chapuis,
(1994) Comparison of gel matrices for resolving PCR ampli-
è , M., Mondon, P. and F., Grillot, R., Lebeau, B., Mallie
¢ed DNA ¢ngerprinting pro¢les. PCR Methods Appl. 4, 50^
EBGA network. (1995) Typing of
Aspergillus fumigatus
by
random ampli¢ed polymorphic DNA. 2nd European Confer-
è , M., Renaud, F., Sy[14] Rodriguez, E., De Meeus, T., Mallie moens, F., Mondon, P. and EBGA Network (1996) Multi-
Aspergillus fumigatus
P.E.,
Meis,
J.F.G.M.,
Sarfati
J.,
Hoogkamp-
è , J.P. and Melchers, W.J.G. (1996) Korstanje, J.A.A., Latge
ence on Medical Mycology, Brussels, PO 1.12 Abstr.
centric epidemiological study of
51. [23] Verweij,
isolates
Genotypic characterisation of sequential
Aspergillus fumigatus
isolates from patients with cystic ¢brosis. J. Clin. Microbiol. 34, 2595^2597.
FEMSIM 720 16-4-97