ELABORATION DES STRATEGIES D’ECHANTILLONNAGE

Cevolet avait pour but de développer une méthodologie d’échantillonnage permettant d’étudier la diversité microbiologique de l’air de différents espaces clos et notamment sa variabilité dans l’espace. Les principaux verrous concernaient l’accessibilité des zones à investiguer, l’acceptabilité par les occupants des métrologies employées et la nécessité de disposer d’une matrice suffisante d’ADN pour permettre les analyses moléculaires. Dans ce contexte, il a été décidé de procéder à deux types de prélèvement : – une collecte ponctuelle à haut débit permettant d’étudier la variabilité de la diversité microbienne aéroportée dans le temps, – une collecte intégrée sur près d’un mois permettant d’accéder à une quantité d’ADN suffisante pour l’étude de la diversité spatiale dans différents espaces clos sans géner les activités qui s’y déroulent.

ARTICLE 1 : PROCEDURE EXPERIMENTALE OPTIMISEE POUR L’ANALYSE MOLECULAIRE DE L’AIR DE TROIS ENVIRONNEMENTS INTERIEURS.

Une étude a été menée afin de développer une procédure expérimentale optimisée pour l’analyse moléculaire de l’air, appliquée aux environnements intérieurs. Le système choisi devait répondre à plusieurs caractéristiques : – Le temps de prélèvement devait être suffisamment court pour permettre un suivi dans le temps et pour gêner au minimum les occupants (facilité de mise en œuvre), – La réalisation des analyses moléculaires dans des conditions optimales (aspect qualitatifs et quantitatifs). Notre approche s’est divisée en deux parties. Il fut tout d’abord nécessaire de valider les protocoles de collecte pouvant être appliqués dans de tels environnements et pour la technique analytique utilisée. Pour y parvenir, des mesures ont été effectuées dans un environnement où les conditions de ventilation et de circulation des personnes lors de l’analyse, étaient contrôlées et maintenues constantes. La faisabilité de la méthodologie a ensuite été éprouvée en environnement réel dans un bureau paysager et deux musées. Les résultats de cette étude sont présentés sous la forme d’un article accepté dans Clean-Soil, Air, Water.

Sample preparation

Before concentration on a filter, all polyester filters and polyurethane foams were rinsed five times with sterile water up to a volume of 100mL. The samples were concentrated by filtering the solution through a 47 mm, 0.2 µm polyestersulfone sterile filter (Supor 200, Pall Corporation, USA). The filter was then cut using a sterile scalpel and each half-filter was placed in a 1.5 mL sterile tube and frozen at -80°C.

Extraction and purification of total genomic DNA 

DNA extraction was performed directly on the filter using the protocol described by Moletta [19] but slightly modified. The filter was ground to a powder using a sterile plastic stick while being kept on dry ice. The main steps of the DNA extraction and purification are detailed below. Volumes of 380 µL 4 M guanidine thiocyanate in 0.1 M Tris–HCl (pH 7.5) and 120 µL 10% Nlauroylsarcosine were added to the ground filter which was frozen at -20°C. After, a volume of 500 µL 5% N-lauroylsarcosine in 0.1 M phosphate buffer (pH 8) was added to the frozen filter and the tube was immediately incubated at 95°C for 10 min. After the thermal treatment, the samples were placed in a vibroshaker with 0.1 mm zirconium beads. Polyvinylpolypyrrolidone was then added to remove any inhibiting substances. DNA was precipitated using isopropanol. Total DNA was then purified using a QiaAmp DNA microkit (Qiagen, Hilden, Germany) and collected in 50 µL DNA-free water. After extraction, the DNA was stored at -80°C. A field control was carried out using molecular biology-grade water for each of the sampling devices and these checked by passing through all extraction and amplification stages. 

Real-time PCR (Polymerase Chain Reaction) amplification and quantification 

All PCR assays were performed on a RotorGene 3000 (Corbett Research, Australia). The calibration curve was generated using RotorGene software, version 6.0. During this study, two different real-time PCR systems were used to assess the concentrations of bacteria and fungi in indoor air. They were all based on the 16S ribosomal RNA gene sequences and 18S ribosomal RNA gene sequences. All primers and the TaqMan® probe were synthesized by Sigma Proligo (Paris, France). The cycle threshold (CT) was calculated as the cycle number at which the reaction became exponential. The cycle threshold of each sample was then compared to a standard curve and the result was expressed as a numerical value of the number of target genomes in the sample. The standard curves were generated by amplifying serial 5-fold dilutions of the total genomic DNA of Escherichia coli and Aspergillus fumigatus. Concentrations of 4.13×107 Genome Equivalent (GE) bacteria/µL and 2.08×106 GE fungi/µL of the total genomic DNA solution were determined using the Genequant Pro (Amersham Biosciences). Each standard was tested in triplicate. The current specificity of all real-time amplification systems already published was checked in silico using Probematch software from the ARB database and NCBI Blast [24] against all 16S rRNA gene sequences and 18S rRNA gene sequences available in RDP and/or Genbank.  

Evaluation of total bacterial load

A 441 bp (base pair) fragment of the bacterial 16S rRNA gene sequence was amplified with bacterialdirected primers and a probe set (see details in Table 14) [25]. PCR was performed using the Quantitect Probe PCR kit (Qiagen) with 12.5 µL of Master Mix, 0.25 µL of forward primer, 0.25 µl of reverse primer and 0.25 µL of probe, with water added to a final volume of 20 µl. 5 µL of the sample were added to PCR mix. The amplification was carried out under the following conditions: 95°C for 15 min, then 50 cycles at 94°C for 15 s, 60°C for 60 s and 72°C for 20 s. No-template controls with molecular biology-grade water rather than DNA were included in each run. Serial dilutions of Escherichia coli (DSMZ 5695, ATCC 12435) DNA were used as the standard to determine bacterial concentrations with real-time PCR. All samples were included in triplicate in each run. The reaction efficiency of the standard curve for bacterial load evaluations was 0.83 and the correlation coefficient R2 >0.99. 

Evaluation of total fungal load 

A 379 bp fragment of the fungal 18S rDNA was amplified with fungi-directed primers (see details in Table 14). PCR was performed using the Quantitect Sybr Green PCR kit (Qiagen) with 12.5 µL of Master Mix, 2.25 µL of forward primer and 2.25 µL of reverse primer and water being added to a final volume of 20 µL. 5 µL of the sample were added to the PCR mix. Amplification was carried out under the following conditions: 95°C for 15 min, then 45 cycles at 94°C for 15 s, 52°C for 30 s and 72°C for 45 s. No-template controls with molecular biology-grade water rather than DNA were included in each run. The Q-PCR system used in this study can amplify the 18S rRNA of the four major fungal phyla: Ascomycetes, Basidiomycetes, Zygomycetes and Chytridomycetes [26]. Serial dilutions of the DNA of Aspergillus fumigatus (Institut d’Hygiène et d’Epidémiologie de Bruxelles – Mycology Section) were used as the standard to determine fungal concentrations with real-time PCR. All samples were included in triplicate in each run. The reaction efficiency of the standard curve for fungal load evaluations was 0.78 and the correlation coefficient R2 >0.99. 

Limits of quantification 

The limit of quantification was defined using the lowest concentration of the standard within the linear range at which 95% of PCR results were positive. The limit of quantification of the number of GE bacteria or GE fungi per cubic meter of collected air was then calculated for each sample using these values, taking into account of the equivalent volume of air used for DNA extraction and the dilution factor applied prior to Q-PCR. In the case of bacteria, the limit of detection was 8.8×102 GE/m3 of air. For the fungal system, the limit of detection was 2×102 GE/m3 of air and was given by the higher dilution of the standard in the amplified range. A lower limit of detection values was found for fungi than for bacteria, mainly because of the presence in Q-PCR reagents of contaminating 16S rRNA gene sequences arising from the manufacturing process [27]. The dispersion of values was measured for each system based on an analysis of the Q-PCR standards conducted in triplicate, and were found to be those typical in the quantitative experiment; i.e. between 8% and 30%, depending on the collection system. The dispersion observed for Q-PCR triplicate analyses was 3%.