WO2008138341A2

WO2008138341A2 - Methods for detection of 'pontiac' subgroups of legionella pneumophila serogroup 1

Info

Publication number: WO2008138341A2
Application number: PCT/DK2008/000177
Authority: WO
Inventors: Soren Uldum
Original assignee: Statens Serum Institut
Priority date: 2007-05-11
Filing date: 2008-05-09
Publication date: 2008-11-20
Also published as: WO2008138341A3

Abstract

Methods based on serogroup specific DNA sequences and subgroup- specific gene of Lpn serogroup (sg) 1 strains for detection and discrimination of all monoclonal Lpn sg 1 Pontiac and non-Pontiac subgroups in clinical and environmental samples are described. Primers were designed for the Lipopolysaccharide (LPS) associated (lag-1) gene, which codes for a common LPS epitope specific for the MAb 3/1 of the Dresden monoclonal panel (sg 1 'Pontiac' subgroups), and primers for open reading frame 2 (ORF 2) DNA sequence of the LPS biosynthesis gene cluster of Lpn sg 1. PCR with primers to the lag-l gene and ORF 2 can be used for diagnosis of LD caused by Lpn sg 1 without need for isolation by culture. The PCR method can be used as a rapid method for detection and discrimination between the Pontiac and non-Pontiac subgroups of Lpn sg 1 in clinical and environmental samples before culture and serogroup results can be obtained. The PCR and DNA methods based on lag-1 gene and ORF 2 DNA sequences could be a valuable tool in outbreak investigations and in risk assessment.

Description

Methods for detection of "Pontiac" subgroups of Legionella pneumophila serogroup 1.

Field of invention

A PCR method or other DNA based methods for detection of Legionella pneumophila serogroup 1 strains based on the ORF 2 or the other ORF DNA sequences of the LPS biosynthesis gene cluster and the Iαg-1 gene of the Legionella pneumophila genome, including primers, probes and kits.

General background

Legionella pneumophila (Lpn) is an aquatic bacterium often found in warm/hot water installations and a causative agent of severe pneumonia (Legionnaires' disease, LD). Lpn can be divided in 15 serogroups (sg), of which serogroup 1 (sg 1) is the most common and responsible for approximately 60% of all Danish cases of LD, other countries have reported frequencies of more than 80% (1) for sg 1. Lpn sg 1 can be subdivided in sev- eral subgroups according to reaction with monoclonal antibodies (MAbs) (2). A group of these sg 1 subgroups belongs to a group called "Pontiac" which is defined as having reactivity with MAb 2 of the international panel (2) and MAb 3/1 of the Dresden panel (8) . The "Pontiac" group is considered as the most virulent Legionella group as this group is responsible for almost all cases of LD among healthy individuals (eg: travel associated LD) and the cause of all major community outbreaks in the world.

Diagnosis of legionellosis can be difficult, as clinical features are often indistinguishable from other causes of pneumonia or respiratory diseases. How- ever, early diagnosis of LD is important for guidance in antibiotic therapy. Several laboratory methods are available for diagnosis of Legionella infection including culture, serology (detection of antibodies), polymerase chain reaction (PCR), and urinary antigen test (3).

Risk assessment and source investigation has traditionally been done by a quantitative culture technique followed by serotyping (and/or DNA typing) of the isolates. In recent years, several quantitative real time PCR methods for detection and quantification of Legionella in water samples have been published, and at least one method is commercialized (4).

Traditional methods like culture and serotyping are used for diagnosis of LD (3). PCR enables in vitro amplification of minute amounts of specific DNA into millions of copies. Several PCR assays based on Legionella genus specific genes (16S, 5S, and 23S-5S) and Lpn species specific genes (mip) are available for detection and discrimination between Lpn and other Legionella species(5). However, no PCR methods have been published that can discriminate between serogroups and subgroups of Lpn.

Conventional methods like culture and serotyping are still in use for diagnosing Legionella infection. Culture for Legionella is expensive, needs tech- nical expertise and sufficient processing of samples. Although culture is 100% specific, it is considered as a slow and insensitive diagnostic method (culture - 3 to 7 (10) days with a sensitivity of 50% + 1 to 2 days for serotyping). Culture technique is the standard method for environmental surveillance and outbreak investigations. Culture is the diagnostic tool, which can be used for diagnosis of all Legionella spp. infections, but due to disadvantages like the slow growth (fastidious nature), overgrowth by other microorganisms, and Legionella cells that are viable but nonculturable (VBNC) means that other methods are needed for fast and reliable diagnosis. PCR methods (including quantitative techniques) have recently been introduced as a rapid diagnostic tool, both for diagnosis of LD as well as for detection of Legionella spp. in environmental samples. PCR is a fast, sensitive, and well- established technique for diagnostic purpose in several laboratories.

Serotyping with polyclonal or MAbs is the most common technique used for typing of Lpn isolates (clinical and environmental). Several commercial reagents are available that can discriminate Lpn sg 1 form other Lpn serogroups and to a certain degree Lpn from other Legionella species. Subtyping of Lpn sg 1 with MAbs (Dresden MAb Panel) (8), raised against Lipopolysac- charide epitopes, by Enzyme linked- immunosorbent assay (ELISA) and /or Immunofluorescence test (IFT) technique can discriminate up to 10 monoclonal subgroups (MAb subgroups) of Lpn sgl(6) . The subgroups can be di- vided in those that react with the MAb 3/1 of the Dresden panel (MAb 2 of the International Panel) (8, 2) as "Pontiac" group and those that do not as "non- Pontiac" group. For surveillance and risk assessment, it is important to be able to detect the "Pontiac" subgroups, as these subgroups are the cause of all major community outbreaks in the world. However, serotyping requires cultured isolates, which is a time consuming (3-10 days) method and the subtyping technique is restricted to few reference laboratories.

Serotyping or subtyping of Lpn sg 1 isolates with MAbs (eg. the Dresden panel), is the only tool to discriminate between Lpn serogroups and all monoclonal subgroups of Lpn sg 1, including monoclonal subgroups of "Pontiac" and "non-Pontiac" group.

The patent application WO05049642 describes a PCR kit and method to de- tect Lpn and discloses a range of primers. However, this paper does not disclose the identification or diagnosis of the "Pontiac" subgroup among the Lpn sg 1, and is therefore not able to distinguish and detect the high risk of virulence associated with the " "Pontiac" subgroup.

PCR methods are rapid diagnostic tools for early detection of Legionella and can be used as adjunct to culture and serotyping for diagnosis of LD and typing of Lpn. Currently available PCR methods are only able to detect Le- gionella spp and Lpn in clinical and environmental samples, but cannot discriminate between Lpn serogroups and its subgroups.

Lpn sg 1 is cause of 80% of LD cases and 66.8% cases of LD are associated with the "Pontiac" group and 11.7% cases with the less virulent subgroup "non-Pontiac" (6). Due to predominance cause of LD due to Lpn sg 1, makes it necessary to detect and discriminate the group /and subgroups from other serogroups, a rapid method can be used to detect and prevent outbreaks of LD.

A PCR method based on serogroup and subgroup specific genes of Lpn sg 1, can be a method to detect and discriminate between Lpn serogroups and monoclonal subgroups of Lpn sg 1 into "Pontiac" and "non-Pontiac" groups in clinical and environmental samples, with in short time for early diagnosis and prevention of outbreaks without the need for culture and serotyping.

Summary of the invention

Lpn is the most common causative agent of LD. Lpn strains were serotyped into 1-15 serogroups based on their antigenic differences of LPS and MAb binding patterns, of which Lpn sg 1 is responsible for 80% of cases of LD. Lpn sg 1 is further subtyped based on presence or absence of LPS associated epitope, one epitope is characterised by reactivity with the MAb 3/1 of the Dresden Panel and can divide Lpn sgl into "Pontiac" group (MAb 3/1- positive) and "non-Pontiac" group (MAb 3/1-negative) strains, respectively. The LPS associated gene lag-1 is present in all monoclonal subgroups of "Pontiac" group and deleted in "non-Pontiac" group strains. "Pontiac" group is proved as the most virulent subgroup of Lpn sg 1 and responsible for majority cases of LD and its outbreaks. In all "Pontiac" group (MAb 3/1- positive) strains an ORF present in opposite direction to the downstream re- gion of lag-1 gene is with identical size and high degree of nucleotide sequence similarity to ORF 2 present on Lpn LPS biosynthesis gene cluster (30kb locus) of Lpn sg 1 strain OLDA of "non-Pontiac" group. Thus, ORF 2 of LPS biosynthesis gene cluster is detected as a common gene for Lpn sg 1 strains and its monoclonal subgroups of "Pontiac" and "non-Pontiac" group strains.

Detection of Lpn sg 1 and its subgroups of" Pontiac" group is important in diagnosis of LD and prevention of outbreaks, as these strains are known to be most virulent among all serogroups and responsible for most cases of LD and its outbreaks. Early and rapid detection of these serogroup and subgroups strains have become essential to prevent and control legionellosis. Until now, no PCR method is available to detect and discriminate Lpn sgl and its subgroups from Lpn sg 2-15 strains except methods like culture and MAb serotyping, which are time consuming. We have developed a PCR method to detect Lpn sg 1 strains based on ORF 2 as Lpn sg 1 specific DNA sequence for all "Pontiac "and "non-Pontiac" groups and lag-1 gene as subgroup specific gene for only "Pontiac" subgroup strains. The PCR method with primers to lag-1 gene detected all monoclonal subgroups of "Pontiac" group and primers to the ORF- 2 DNA sequence detected all monoclonal subgroups of "Pontiac" group and "non-Pontiac" group of Lpn sg 1 strains in clinical and environmental isolates and samples. Detailed disclosure of the invention

The present invention discloses methods for detection of Legionella pneumophila serogroup 1 strains based on the ORF 2 DNA sequences and the lag-1 gene. The method is for detection of the "Pontiac" group, by detecting both the lag-1 gene and the ORF 2, or the "non-Pontiac" group, by detecting only the ORF 2, of Legionella pneumophila serogroup 1.

A preferred method of the invention is PCR.

The invention also discloses specific primers for amplification of the lag-1 gene and ORF 2 and other ORF regions of LPS biosynthesis gene cluster.

The method can be performed on an environmental sample or on sample from human. The application is both for clinical samples and/or isolates

(human) and environmental samples and/or isolates (e.g. water) and both for diagnostics, epidemiology, surveillance and risk assessment.

A kit for detection of Legionella pneumophila serogroup 1 strains based on the ORF 2 (or the other ORFs) and the lag-1 gene is also claimed for detection of the "Pontiac" subgroup or "non-Pontiac" subgroup of Legionella pneumophila serogroup 1

Definitions

Legionellosis: Legionella is responsible for two known diseases in humans - legionnaires' disease and "Pontiac" fever, the generic term for these diseases is Legionellosis. The reservoirs of legionellae are natural or man-made water systems and their natural hosts are various amoebae species. The route of infection is most often by inhalation of aerosolized droplets containing the organism or probably in some cases by aspiration. Aerosolized droplets can be found in eg: cooling towers and hot water installations (5).

Legionnaires' disease: A severe pneumonia caused by Lpn and other Le- gionella species. Lpn accounts for the majority LD cases. Lpn sgl and its subgroups are the most frequent causes of LD cases. The mortality rate of LD can reach 15% or higher and it has an attack rate less than 5% (5). Pontiac fever: It is an acute non-pneumonic flu-like illness caused by Lpn and other Legionella species with high attack- rate over 90% and no mortality. It has no clinical evidence of pneumonia (5).

International MAb Panel: A panel of seven MAbs used for subtyping of Lpn sgl in to 10 major subgroups (2).

Dresden panel: A Panel of MAbs produced against LPS epitopes of Le- gionella and designated the Dresden Panel. MAb panel of Dresden along with MAb 3 of the international panel allows the subtyping of Lpn 1 into 15 serogroups and Lpn sgl into 10 monoclonal subgroups (8).

Pontiac group: The serological monoclonal subgroups Knoxville, Philadelphia, enidorm, ranee and Allentown belong to so-called "Pontiac" group or MAb 3/1- positive group, which expresses the virulence-associated epitope recognized by MAb 3/1 of Dresden panel (MAb 2 of International panel). This epitope is only found in strains that posses the lag-1 gene and the epitope is not found in any other sero- and subgroups. Monoclonal subgroups of this group are with highest virulence among all subgroups of Lpn sg 1 (6)

Non-Pontiac group:

The serological monoclonal subgroups OLDA, Oxford, Bellingham, Hey- sham andCamperdown belong to so called "non-Pontiac" group or MAb

3/1 -negative group, which do not expresses the virulence associated epitope recognized by MAb 3/1 of Dresden panel (MAb 2 of International panel).

The lag-1 gene is deleted (not present) or non-functional in this subgroup and has least virulence (6).

Lag-1 gene

The gene which codes for a polypeptide involved in O-acetylation of the 8-

O-acetylgroup of Legionaminic acid in the LPS core of Lpn sgl is identified as iαg-1 (Lipopolysaccharide associated gene). The iσg-1 gene (of 1074 bp) is determinant for virulence-associated epitope, which reacts with MAb 3/1 of

Dresden Panel. Based on presence or absence of iαg-1 gene and its reactivity with MAb 3/1 of Dresden Panel, Lpn sg 1 strains were subtyped into "Pontiac" (MAb 3/1-positive) and strains that lost iαg-1 gene and do not react with MAb 3/1 are "non-Pontiac" (MAb 3/1-negative) strains. All monoclonal subgroups of "Pontiac" group strains harbours lag-1 gene (9).

ORF 2-gene

An open reading frame with putative glycosyltransferase function present on the Lpn LPS biosynthesis gene cluster identified in strain RCl (Lpn sg 1, mo- clonal subgroup OLDA). LPS biosynthesis gene cluster is a 30kb gene locus comprises of 30 putative open reading frames (ORFs), which shows homologies to proteins involved and required for Lipopolysaccharide biosynthesis in Lpn. ORF 2 is the located at the second ORF position at 1316-2209 (894 bp) nucleotide position on 30 kb LPS biosynthesis gene cluster of OLDA. The 30 kb gene locus is specifically present in all Lpn sg 1 strains and only parts of DNA of locus may be found in other serogroups of Lpn from sg 2-14.(11).

The genus Legionella comprises more than 48 species, which can be divided into 70 serogroups. Lpn was classified into serogroups on the basis of antigenic differences on Lipopolysaccharide (LPS) .The serogroup specificity of Lpn is related to its LPS characteristics and classified serologically into 1-15 serogroups by Monoclonal antibodies (MAb) and/or rabbit polyclonal antis- era (1,2, 7). Monoclonal antibody reactivity with LPS epitope is used as virulence marker for Lpn sg 1

Up to 80% of LD, cases are caused by Lpn sgl.This serogroup was further subdivided into subgroups based on the presence or absence of a virulence- associated epitope on LPS, which is recognized by the MAb 3/1 of Dresden panel into Lpn sgl "Pontiac" and "non-Pontiac" group (6,8) (see Figure 1). This epitope is not present on strains belonging to any other sub/serogroups, except Lpn sgl Pontiac group strains.

The MAb 3/1 recognized epitope on LPS determining the "Pontiac" group is associated with the 8-O-acetylgroup on the Legionaminic acid of Lpn sgl LPS. Acetylation of Lpn sgl LPS mediate virulence. The gene responsible for O-acetylation was identified as lag-1 (Lipopolysaccharide associated gene), which encodes a polypeptide involved in O-acetylation of the 8-O- acetylgroup of Legionaminic acid in the LPS core of Lpn sgl(9)

The genetic basis for virulence differences among sg 1 strains was related to the presence or absence of a functional iαg-1 gene. The iαg-1 gene is determinant for virulence-associated epitope, which reacts with MAb 3/1. Thus, strains having lag-1 gene and react with MAb 3/1 are MAb 3/1 -positive and strains that lost iαg-1 gene and do not react with MAb 3/1 are MAb 3/1- negative. Based on presence or absence of the epitope recognized by the MAb 3/1 and several other LPS epitopes recognized by other MAbs, sg 1 strains were subdivided into 10 monoclonal subgroups or 10 phenons as per Dresden MAb panel (6). According to the Pan European study theMAb 3/1- positive ("Pontiac" group) cases were most frequent (66.85%) and majority of LD outbreaks are associated with this most virulent subgroups of "Pontiac" group and 11.7% of the isolates belonged to MAb 3/1-negative se- rogroup 1 subgroups of "non-Pontiac" group with least virulence and 21.5% to other serogroups (6).

All monoclonal subgroups of "Pontiac" group (MAb 3/1 -positive) of Lpn sg 1 strains harbour the iαg-1 gene of identical size of 1074 bp and it is deleted (or non-functional) in all monoclonal subgroups of "non-Pontiac" group (MAb 3/1-negative) strains. In most of the monoclonal subgroups of "Pontiac" group of Lpn sg 1, the iαg-1 gene is present on an unstable genetic element bordered by direct repeats of open reading frames (ORFs) with high DNA sequence homology in each of the pairs of strains (Figure 2) (10). The DNA sequences in each of the identical sets of direct repeats found in a strain may differ from other strain. Due to this high DNA sequence homology with in a set of direct repeats present on either side of iαg-1 gene may be responsible in making iαg-1 gene region as unstable element and in- volved in insertion or deletion of iαg-1 gene. Strains not reacting with MAb 3/1 either lost the complete gene or contained a mutation or insertions within the iαg-1 gene. In all investigated [..pneumophila sg 1 Pontiac group strains, downstream region of the iαg-1 gene, in an opposite orientation an ORF (or in some strains a direct repeat) was identified which showed identi- cal size and high degree of nucleotide sequence homology to ORF 2 of the Lpn LPS biosynthesis gene cluster found in strain RCl (Lpn sgl,moclonal subgroup OLDA) (see Figure 3) (11). The LPS biosynthesis gene cluster is a 30 kb gene locus involved in LPS biosynthesis of Lpn sgl found in monoclonal subgroup OLDA. The 30 kb gene locus is specifically present in all Lpn sgl strains and only parts of the DNA locus may be found in other sero- groups of Lpn from sg 2-14.

In all strains of monoclonal "Pontiac" subgroups (MAb 3/1-positive) strains, ORF 2 (894 bp) of LPS biosynthesis cluster of Lpn sgl strain OLDA, showed high similarity to ORF found in downstream region to iαg-1 gene region (10,12) . In all strains that lost iαg-1 gene, ORF 2 of the LPS biosynthesis cluster of Lpn sg 1 was present adjacent to ORF 3 (1149bp) strains of Lpn sgl (12,13).

The ORF 2 (894 bp) of LPS biosynthesis gene cluster found in Lpn sgl strain OLDA (MAb 3/1 -negative strain), has high sequence homology with the ORF found in opposite direction to the downstream region of iαg-1 gene in all iαg-1 harbouring strains (MAb 3/1-positive strains). It showed that ORF 2 of LPS biosynthesis gene cluster from Lpn sg I₁ strain OLDA is present in all MAb 3/1 positive ("Pontiac" group) strains as an ORF towards downstream region of iαg-1 gene and present adjacent to ORF 3 in iαg-1 negative strains (MAb 3/1-negative strains/ "non-Pontiac" group). Thus, ORF 2 DNA sequence is conserved in all Lpn sg 1 strains.

The present invention discloses a PCR method for detection of all "Pontiac" subgroups (iαg-1 gene positive and ORF 2 positive) strains and "non- Pontiac" subgroups (iαg-1 negative but ORF2 positive). This PCR method can differentiate an Lpn PCR positive patient or water samples into Lpn se- rogroup 1, non-serogroup 1 (no reaction with either ORF2 or lag-1 primers), "Pontiac" and "non-Pontiac" subgroups, without the need for culture and testing for reactivity with MAb 3/1. However, it is important to notice that this is a DNA based typing method that not can be expected to be 100 % homologue to a phenotypical typing method (serotyping). Detection of the "Pontiac" subgroup in water samples is an indication of high risk and can be used for assessment of risk. LD is normally acquired by inhalation or aspira- tion of Legionella from contaminated environmental source such as hot water systems and cooling towers or hospital water systems. A high density of Legionella in water is an indicative of risk for legionellosis. Detection of "Pontiac" subgroup strains, in water samples is an indication of high risk as these are most virulent group of Legionella frequently found in most of cases of LD and its outbreaks.

The "non-Pontiac" group does not react with the iαg-1 gene and in general no other Lpn serogroups (or species) react with the primers to ORF 2 (the ORF 2 region is not present in other serogroups) - one exception is however the reference strain for serogroup 7 and to some degree the reference strain of serogroup 11, but we have not detected the sequence in any non- serogroup 1 clinical or environmental strain.

The PCR with lag-1 and ORF2 can be used for epidemiological and outbreak investigations by linking to possible environmental strain e.g. to detect high risk of Lpn in water samples. A kit optimized for PCR of water samples could be used for analysing water distribution systems.

The PCR can be used in routine and laboratory work for diagnosis and prognosis of LD. A PCR kit with iαg-land ORF 2 gene can significantly accelerates screening of clinical samples into sg 1 and non sg 1 with in short time as compared to culture/serotyping. It can be an alternative option to commer- cial serological kits used for screening colonies of Lpn into Lpn sg 1 and non sg 1 strains before running them for subtyping with MAbs by ELISA or IFA. The PCR can also compensate the use of ELISA technique for subtyping of Lpn sgl strains into "Pontiac" and "non-Pontiac" subgroups unless monoclonal subgroups are needed to be identified. Early diagnosis with a PCR kit, instead for testing with culture, Oxoid and ELISA results, can save around 3- 10 days.

PCR is not the only method for detection of DNA sequences but encompasses all other DNA amplification and probe techniques for detection of specific sequences of the lag-1 gene and ORF 2 DNA sequence. Likewise other ORF regions could be used for Lpn sg 1 detection The application is both for clinical samples and/or isolates (human) and environmental samples and/or isolates (e.g. water) and both for diagnostics, epidemiology, surveillance and risk assessment.

Examples

Example 1; Design oi primers

The PCR method with lag-1 gene and ORF 2 was developed with initial design of primers, (Table 1), optimization of primers and PCR conditions. The lag-1 gene (IpgO777) DNA sequence (1074bp) of Lpn sg 1, strain Philadelphia 1 (Accession No. AE017354) and ORF 2 DNA sequence (894bp) of Lpn sg 1, OLDA strain (Accession No. AJ007311) were used as reference sequences to search for all available sequences for concerned genes through NCBI BLAST (Basic Local Alignment Search Tool) tool (http : //www. ncbi . nlm . nih . gov/BL AST/) . Primers for lag-1 gene and ORF 2 were designed from conserved regions of multialignment of all the lag-1 gene (of Pontiac group strains) and ORF 2 DNA sequences (of Pontiac and non-Pontiac group strains) respectively, from all available (published) sequences for of Lpn sg 1 strains (see Table 2) found in Gene Bank database in NCBI (National centre for Biotechnology Information, USA) at: http://www.ncbi.nlm. nih.qov/entrez/σuerv.fcqi?CMD=search&DB-genome,

Table 1 List of Primer sequences used in study for lagΛ and ORF2 gene

*primes are from reference (10) and other primers are designed in this study. F=forward; R=reverse

Table 2. List of L.pneumophila SgI strains used in this study from Gene Bank

Strain Description Accession

Pontiac group

1 Philadelphia 1 Complete genome σil52627367lσblAE017354.111526273671

2. Paris Complete genome gil53749768lemblCR628336 ll[53749768]

3. Isolate Sweden 3 ORF2, lag-1 gene and ORF3 αil22208365lemblAJ504790 1l[222083651

4. Philadelphia 1 O-acetyltransf erase(Jαg- 1 ) gil974327lgblU32118.1ILPU32118[974327]

5. Lens Complete genome gil53752796lemblCR628337.1l[53752796]

6. Corby lαg-1 (O-acetyltransferase) oilll493207lemblAJ300467 Hf 114932071

Non-Pontiac group

7. OLDA LPS biosynthesis gene cluster gil6688579lemblAJ007311 ll[6688579] Example 2; PCR optimization and conditions

Optimization was done for primer concentrations, annealing temperatures, PCR components (MgCl₂, Taq polymerase, etc.) and thermal cycler conditions. The primers used for lag-1 gene PCR are Lag-1 6F/478 R and ORF-2 200F/630 R primers for ORF 2 PCR,respectively in this study for further experiments and other primer were used as confirmatory primers for corresponding PCRs (Table 1). Sensitivity and specificity of the lag-1 and ORF 2 PCR was investigated on several Lpn reference strains (Lpn sg 1-15) as well as on Danish clinical (n=100) and environmental isolates (n=75) of Lpn sg 1 ("Pontiac" and "non Pontiac" subgroups) and Lpn non-sg 1 (sg 2-15). Clinical and environmental isolates selected in this study include all available monoclonal subgroups from "Pontiac" and "non-Pontiac" group of Lpn sgl, which were previously sero- typed by MAbs of the Dresden Panel (see Fig.l). Except Sg 1 Non-Pontiac monoclonal subgroup Heysham and among Sg 2-15 isolates Sg 7, 9, 11, 13 were not included as Danish isolates were not available. All Clinical and environmental samples previously found PCR positive for Lpn by the macrophage infectivity potentiator (mip) gene based PCR as previously described (14) were investigated. All these isolates were referred to the National reference centre for Legionella, SSI for Legionella test.

PCR with lag-1 gene and ORF 2 was performed on basis of two selection criteria: i) PCR test on patient and environmental cultured isolates, ii) PCR test on patient and environmental samples.

A total of 100, patient isolates (n=75) and environmental isolates (n=25) were selected and cultured based on previously known MAb serotyping results. The selected isolates included Lpn serogroups from Lpn sg 1-15 and all available Lpn sg 1 monoclonal subgroups of "Pontiac" and "non-Pontiac" group strains. All the selected isolates were tested with PCR for lag-1 and ORF 2.

The PCR for samples was performed directly on patient samples (n=80) and environmental samples (n= 50). PCR for patient samples include: i) 60 of Lpn PCR positive and culture positive and ii) 20 of Lpn PCR positive and culture negative samples. AU the samples were tested based on previously exam- ined PCR and culture results. The selected samples were previously found positive for Lpn (mip gene) by Lpn PCR (14) and also differentiated into culture positive and culture negative samples.

All patient and environmental isolates or samples found PCR positive for iαg-1 gene and ORF 2 are grouped into Lpn sgl "Pontiac" group and Isolates or samples found only positive for ORF 2 were considered as Lpn sg 1 non-pontiac group strains. Patient and environmental samples from Lpn PCR positive samples found negative for both lag-1 gene and ORF 2 were consid- ered under Lpn sg 2-15

Evaluation of PCR assay results was done based on obtained PCR results: PCR positive Patient and environmental isolates were correlated with the known MAb serotyping results and PCR positive patient and environmental samples were correlated with known Lpn PCR and culture results

Example 3:

PCR results of patient (n=75) and environmental (n=25) isolates with lag-1 PCR has shown 100% correlation between monoclonal "Pontiac" strains (Mab 3/1 -positive strains) and lag-1 gene reactivity by showing positive result for 33 of 33 (100%) monoclonal "Pontiac" isolates and negative for 18 "non-Pontiac" Lpn sg 1 and for 49 Lpn sg 2-15 isolates (Table 3). ORF 2 was detected in 51 of 51 (100%) Lpn sg 1 isolates but in none 49 non-sg 1 isolates (Table 3) , except in Lpn reference strains for Lpn sg 7 and 11. PCR with the lag-1 gene and ORF 2 primers on Lpn PCR and culture positive clinical samples showed respectively a 96% (24/25) and a 97% (37/38) correlation with Lpn serotyping results (Table 4) Only one sample of sg 1 Pontiac group showed negative result with both lag-1 and ORF 2 PCR (Table 4). Among 20 [..pneumophila PCR positive but culture negative patient samples, 10 were lag-1 and ORF 2 negative (non-sgl). Among the remaining 10 samples, 6 were lag-1 and ORF 2 positive (Pontiac) and, 4 were only ORF 2 positive (non-Pontiac) (Table 5). PCR on environmental samples was less sensitive; however we were able to detect Lpn sg 1 "Pontiac" groups (5 of 9) and "non-Pontiac" group (6 of 13) strains in samples with high copy number of Lpn (Table 5) . PCR results in this study are represented in Table 3, Table 4 and Table 5, respectively.

Table 3. PCR results with primers to lagΛ gene and ORF 2 on Patient (n=75) and Environmental isolates (n=25)

Lpn serogroup Patient isolates Environmental isolates and subgroup PCR positive/ no. tested PCR positive/ no. tested lag-1 ORF 2 lag-l ORF 2

Sg 1 Pontiac 25/25 25/25 8/8 8/8

Sg 1 non-Pontiac 0/13 13/13 0/5 5/5

Sg 2-15 0/37 0/37 0/12 0/12

Table 4. PCR results for Patient samples with primers to lagΛ gene and ORF 2

Lpn serogroup and No. of PCR positive /no. tested with sub groups culture positive (60) lagΛ ORF 2 mip*

Pontiac 24/25 24/25 25/25 non Pontiac 0/13 13/13 13/13 non sg 1 0/22 0/22 0/22

Culture negative (20)

Unknown serogroup 6/20 10/20 20/20 *Patient samples positive for Lpn with mip gene was previously tested with Lpn PCR (14).

Table 5. PCR results with primers to lagΛ gene and ORF 2 on environmental sam- pies

Lpn serogroup and PCR Results sub groups (n) culture positive lagΛ ORF 2 mip*

Pontiac (9) 5 7 9 non Pontiac (13) 0 6 13 non sg 1 (18) 0 0 18

Total (40) 5 13 40 Culture negative (10)

Unknown serogroup 0 0 10

*Patient samples positive for Lpn with mip gene was previously tested with Lpn PCR (14)

PCR assay with primers to iσg-1 gene and ORF 2 could detect -3.2 genome copies of Lpn sg 1, which is the limit of detection (LOD) or sensitivity of the PCR assay with primers to lag-l and ORF 2.

The PCR assay has 100% sensitivity with the clinical and environmental isolates. PCR with iαg-lgene showed positive result with all monoclonal sub- groups (5 phenoms) of "Pontiac" group but with none non-"Pontiac" sub groups. PCR with ORF-2 detected all monoclonal subgroups of "Pontiac" and "non-Pontiac" groups of Lpn sgl.PCR with iαg-land ORF 2 shown 96% and 97% correlation with Lpn serotyping results, respectively. PCR assay with lag-l and ORF 2 has less sensitivity with the environmental samples, it is may be due to less sensitivity of PCR assay for water samples, more PCR inhibitors in water samples and insufficient sampling methods for this PCR assay. The limited sensitivity of the present method on water samples which does not retract the value of the PCR method as a useful tool for the detection of "Pontiac" and "non-Pontiac" subgroups of Lpn sgl and assessment risk of disease.

Figure legends:

Figure 1: Flowchart for monoclonal sub grouping of Legionella pneumophila serogroup 1 using the Dresden Panel plus MAb 3, obtained from the ATCC (6).

Figure 2: Schematic representation of the gene organization in the 2αg-l region in the MAb 3/1 -positive patient strain of Lpn sgl strain Sweden isolates 3. Open reading frames are represented as open boxes, with arrowheads indicating the orientation. The closed box indicates the fragment that was deleted in the MAb 3/1-negative strain. Two repeats of 797 bp that showed identities of 99.6% are also shown (10). Figure 3 : Diagram of LPS biosynthesis gene locus of Lpn sgl strain RCl (OLDA). It represents 30 kb gene locus involved in LPS biosynthesis of Lpn. Designation of ORFs is indicated above gene blocks and direction of tran- scrip tion is indicated by arrows. Plasmids and Hybridization probes used in this study are indicated (11).

References

1. Yu et al 2002. Distribution of Legionella species and serogroups isolated by culture in patients with sporadic community-acquired legionellosis: an inter- national collaborative survey Infect Dis. 2002 JuI Ij 186(1): 127-8. Epub 2002 May 21.

2. JoIy et al 1986.Developemnt of standardized subgrouping scheme for lpnsgl using monoclonal antibodies. j.clin.Microbiol.23:768-771.

3. David R.Murdoch2003. Diagnosis of legionella infection, clin infe disease; 36: 64-9

4. JoIy et al., 2006. Quantitative real-time Legionella PCR for environmental water samples: data interpretation, Appl. Environ. Microbiol. 72, pp. 2801-

2808.

5. Fields BS. et al.,2002. Legionella and Legionnaires' disease: 25 years of investigation. Clin Microbiol Rev. JuI; 15 (3) :506-26. Review.

6. Helbig, J. H et al 2002. Pan-European study on culture -proven Legionnaires' disease: distribution of Legionella pneumophila serogroups and monoclonal subgroups. Eur. J. Clin. Microbiol. Infect. Dis. 21:710-716

7. Ciesielki, et al.1986. Serogroup specificity of Legionella pneumophila is related to Lipopolysaccharide characteristics, immum. 51:397-404.

8. Hebig JH et al 1997.Antigenic lipopolysaccharide components of Legionella pneumophila recognized by monoclonal antibodies: possibilities and limitations for division of the species into serogroups. J Clin Microbiol. Nov; 35(ll):2841-5.

9. Zou,et al.1999. Molecular cloning and characterization of a locus responsible for O acetylation of the O polysaccharide of Legionella pneumophila serogroup 1 Lipopolysaccharide. J. Bacteriol. 181:4137-4141. 10. Bernander S, et al.2003. A hospital-associated outbreak of Legionnaires' disease caused by Legionella pneumophila serogroup 1 is characterized by stable genetic fingerprinting but variable monoclonal antibody patterns. J Clin Microbiol Jun_;41(6):2503-8.

11. Luneberg, et al 2000. Cloning and functional characterization of a 30 kb gene locus required for Lipopolysaccharide biosynthesis in Legionella pneumophila. Int. J. Med. Microbiol., 290: 37-49

12. Luck, P.C. et al(2002). Changes in the iσg-1 locus of Legionella pneumophila serogroup 1 strains results in different Lipopolysaccharide recognized by monoclonal antibodies but do not influence virulence. Legionella, ASM press, Washington, DC pp 52-55

13. Luck PC.et al 2001. A point mutation in the active site of Legionella pneumophila O-acetyltransferase results in modified lipopolysaccharide but does not influence virulence. Int J Med Microbiol. Nov; 291(5):345-52.

14. Søren A.Uldum.et al 2002. PCR as a routine method for diagnosis of Le- gionnaires disease. In Legionella,ASM press, edited by Reinhard Marre et al.pp 213-215

Claims

1. A method for detection of Legionella pneumophila serogroup 1 strains based on the ORF 2 or other ORF DNA sequence of the LPS biosynthesis gene cluster and the lag-l gene of the Legionella pneumophila genome.

2. A method according to claim 1 which is a PCR

3. A method according to claim 1-2 for detection of the "Pontiac" group or the "non-Pontiac" group of Legionella pneumophila serogroup 1.

4. A method according to claim 1-3 for detection of the "Pontiac" group of Legionella pneumophila serogroup 1 by detecting both the lag-l gene and the ORF 2 or other ORF DNA sequence of the LPS biosynthesis gene cluster

5. A method according to claim 1-4 for detection of "non-Pontiac" group of Legionella pneumophila serogroup 1 by detecting only the ORF 2 or other ORF DNA sequence of the LPS biosynthesis gene cluster.

6. A method according to claim 1-5 where the lag-l gene and ORF 2 is amplified by using the primers of table 1.

7. A method according to any preceding claim where the sample is an envi- ronmental or clinical samples or isolates.

8. A kit for detection of Legionella pneumophila subgroup 1 strains based on the ORF 2 or other ORF DNA sequence of the LPS biosynthesis gene cluster and the lag-l gene.

9. A kit according to claim 8 for detection of the "Pontiac" group of Le- gionella pneumophila serogroup 1

10. A kit according to claim 8-9 with the primers of table 1.

11. Primers according to table 1.