Rickettsia conorii is a gram-negative bacterium responsible for Mediterranean spotted fever, a rickettsiosis transmitted to humans by tick bite inoculation of rickettsiae into the skin [1]. The isolation of rickettsial isolates from various samples including blood, skin or infected ticks can be achieved on shell vial assays [2]. This approach, initially developed for R. conorii, was later successfully applied for several strains, including R. prowazekii and R. felis, the major drawback being related to the selection of convenient animal cell lines and to the choice of appropriate growth condition parameters, among which are temperature, culture medium, and fetal calf serum proportion. The development of standardised culture conditions allowing efficient growth of various strains would be advantageous to investigate the differences among various species of rickettsiae [3]. Moreover, a feature of the post-genomic era is the requirement of important amounts of bacteria for large-scale investigations such as proteomic and transcriptomic approaches [4]. The aim of this work was thus to quantify the growth of R. conorii within three different cell lines and using real-time quantitative PCR (qPCR). Eukaryotic cells used in this study were the African green monkey kidney cells (Vero cell, American Type Culture Collection ATCC CRL 1587) and the murine fibrobastic L929 cell line (ATCC CCL1), both maintained in Eagle’s minimum essential medium (MEM; Invitrogen, Paisley, UK) supplemented with 4% fetal bovine serum (FBS; Invitrogen) and 1% L-glutamine (Invitrogen). We also used 3T3 cell lines (ATCC CCL163) maintained in Dulbecco minimum essential medium (DMEM; Invitrogen) supplemented with 10% FBS and 1% L-glutamine. Confluent cells grown on coverslips were infected with 10 rickettsiae resuspended in 300 lL of the appropriate medium. After centrifugation (3452 · g; 15 min), the supernatants were removed, and following a gentle washing, 500 lL of fresh buffer was added in each shell vial and further incubated at 37 C and under 5% CO2. This point was considered as the starting point (t0) of infection. For each kinetic point, samples were processed four times, two coverslips being used for Gimenez or immunofluorescence staining while quantification of bacteria was carried out on remaining shell vials. DNA was extracted using the QIAamp DNA mini kit (Qiagen, Hilden, Germany) and amplification was performed with the 16sRNA primers forward 5¢-TGATGAAGGCCTTAGGGTTG-3¢ and reverse 5¢-TAAACCGCCTACGCACTCTT3¢ using the LightCycler system together with the SYBR Green master mix (Roche Diagnostics, Mannheim, Germany). In some experiments, the determination of the gene copy numbers was also performed using the probes targeting rickA, a gene exclusive for bacteria from the Rickettsia genus (F 5¢-AGTTACAAAAGATAGTAAGTAA-3¢; R 5¢-CCRGYTTTTTAA CCGTAGTAG-3¢). The yield of qPCR was assessed by plotting the cycle threshold values vs. the log10 numbers of input DNA copies, the standard curve being made with genomic DNA of R. conorii (from 1–10 copies) as template and using the same primers. The qPCR data obtained indicated that entrance of R. conorii into the three eukaryotic cell lines tested in this study was not significantly different. Thus, under our experimental conditions, almost all bacteria were taken up into the host cell, the percentages being 94 ± 8%, 80 ± 6.5% and 84 ± 3.2% for Vero, L929 and 3T3 cell lines, respectively. Despite a similar invasion efficacy, we thereafter observed variable replication rates (Fig. 1). Thus, 96 h after the cell infection, the number of R. conorii appeared around six times higher in L929 than in Vero cells or 3T3. These results were confirmed with qPCR assays targeting rickA, as well as through the optical Corresponding author and reprint requests: P. Renesto, Unite des Rickettsies, URMITE IRD-CNRS 6236, Faculte de Medecine, 27 Bd Jean Moulin, 13385 Marseille, France E-mail: prenesto@embl.fr