Heat inactivation of any of the serums led to the partial decrease of the expression of both of the tested defensins by cells exposed either to A. fumigatus conidia, HF, or to Il-1β (Figure 2C, D). Kinetics of defensin expression by cells exposed Z-VAD-FMK concentration to A. fumigatus organisms To analyse the kinetics of defensin expression, cells were exposed to A. fumigatus for 4, 8 and 18 hours, and the expression of hBD2 and hBD9 was examined. As a positive control, Il-1β-treated cells were examined. As a negative control, untreated cells or cells exposed to 5 × 106 latex beads were analysed as well. According to the results presented on Figure 3, the expression of both defensins, hBD2 and hBD9, were

induced in the 16HBE cells treated with Il-1β either

for 4, 8 or 18 hours. No hBD2 expression was detected after a 4-h exposure by 16HBE to SC, RC or HF of A. fumigatus, in contrast to hBD9 expression by cells exposed to all morphotypes of A. fumigatus for the same period. Incubation of the cells with both types of conidia or HF for 8 h selleck resulted in a low level of hBD2 expression and a high level of hBD9 expression, comparable to expression by the cells treated with the positive control, Il-1β. Exposure of the cells to conidia or HF for 18 h led to the high expression of both defensins, hBD2 and hBD9. Exposure of the cells to the latex beads did not induce the defensin expression in any of the experiments. The constitutive expression of hBD1 by the cells exposed either to the different morphotypes of A. fumigatus or to the latex beads for the various periods was observed in the current experiment. Figure 3 Kinetics of defensin mRNA expression by 16HBE human epithelial

bronchial cells exposed to A. fumigatus organisms. 16HBE human epithelial tracheal cells (5 × 106) were grown in six well plates for 24 hours. The cells were then exposed to the different morphotypes of A. fumigatus or latex beads for the different periods: 4 h, 8 h and 18 h. After incubation, the cells were washed VAV2 with PBS, mRNA was isolated by TRIzol Reagent, and RT-PCR was performed as described above in Materials and Methods. Specific primer pairs and the conditions of RT-PCR are described in Table 1. The sizes of amplified products are indicated and were as predicted: hBD2, 199-bp product; hBD9, 174 bp product and human GAPDH, 473-bp product. The hBD2 and hBD9 products were sequenced and confirmed to be identical to the predicted KPT-8602 purchase sequence. Cells were cultivated in a control well in the absence of A. fumigatus. GAPDH was uniformly expressed. One of the four results is shown. Similar kinetics of hBD2 and hBD9 expression was observed with A549 cells (data not shown). Real time PCR The relative level of hBD2 and hBD9 expression in 16HBE and A549 cells exposed to different A. fumigatus morphotypes for 18 hours was quantified by real time PCR.

We next attempted to map the transcriptional start sites of these three operons by primer extension using a fluorescent primer protocol. Using this approach, the start of transcription for the preAB operon was identified at -423/424 bp from the start codon, implying that the preAB promoter is internal to ygiW and contains a large, untranslated leader region (Fig. 2). The start site of the ygiW-STM3175 operon was at -161 bp, which is 10 bp internal to the preA open reading Mizoribine frame. Multiple attempts were made to map the mdaB-ygiN

start, however we were unsuccessful at identifying a clear site for transcriptional initiation. Figure 2 Fluorescent primer extension analysis of transcriptional start sites for the preAB and ygiW -STM3175 operons. Electropherograms of the labeled cDNA are shown for preA (A) and ygiW (C). Dashed lines mark the relative fluorescence learn more unit (RFU) cut-off, below which does not give a confident signal strength. Asterisks (*) denote which cDNA peak was analyzed. Labeled cDNA electropherograms (filled peaks) were aligned with sequence chromatograms (open peaks) to identify the base at which transcription starts for both preAB (B) and ygiW-STM3175 (D). Results of transcriptional organization are diagramed as shown with start sites mapped relative to the translational start (E). PreA appears to activate transcription

of each of the three operons defined in the preA region (dashed lines denote positive regulation). Phenotypes of preAB TCS mutants We previously reported that PreA/PreB is orthologous to the E. coli QseBC system, which responds to AI-3 and epinephrine/norepinephrine signals. In response to these signals, the QseC sensor kinase has been reported to affect SIS3 price motility in both E. coli and S. Typhimurium [6, 14]. However, our microarray data did not suggest any major and/or consistent effect of PreA/PreB on transcription of the flagellar operon. Therefore, we assessed the effects of mutations in preA and preB on the motility of S. Typhimurium

on agar plates with DMEM as the culture medium. The results showed a reduction in motility for the preB sensor mutant (Fig. 3) but not for the preA or preAB mutants. As seen with QseC in E. coli, the addition of synthetic AI-2 did not complement the preB mutant motility defect 5-Fluoracil nmr and also did not affect the motility of the wild type strain (Fig. 3A). Additionally, though epinephrine/norepinephrine has been reported to activate motility in both E. coli and S. Typhimurium [6, 15], a slight but non-significant increase in wild type strain motility was observed in our assays using identical conditions and epinephrine concentrations used previously in E. coli. Supplementation of the media with epinephrine did increase the motility of preA, preB and preAB mutants (all statistically significant except preB, Fig. 3B), but as this effect of epinephrine on S. Typhimurium motility was observed only in preA or preB mutant strains, this effect is not mediated by PreA/PreB.

We found a mutant, 18D06, in our mutant library in which XAC3673 was knocked out; the mutation site ALK inhibitor is located inside the response regulator domain [see Additional file 1]. This mutant was observed at a high concentration in planta (Fig. 2) but with no symptom development [see Additional file 1]. Despite the ability of a hybrid histidine kinase to be involved in phosphorylation of any pathogeniCity related gene, we believe that this protein plays a more sophisticated role in the

virulence process in Xcc. Considering the data presented above, namely a protein localized on the inner membrane with high similarity with RpfC, a Xanthomonas exclusive amino terminus, and high mutant cells concentration in planta, led us to propose this role for XAC3673 in Xcc: participation

in the perception and transduction of signals in the quorum sensing system in this Xanthomonas citri subsp. citri. Besides these features, the fact that the response regulator domain (PF00072) from XAC3673 interacts with the domains CheB_methylest (PF01339), Response_reg (00072), Trans_reg_C (PF00486), GGDEF (00990), Hpt (PF01627), P2 (07194), Sigma54_activat (00158), and ANTAR (PF03861) GW572016 [38] gave us more data on which to base this hypothesis. XAC3673 protein can be on the inner membrane and the amino terminus could act as a sensor to perceive host or environmental signals. After signal reception, XAC3673 may be autophosphorylated. The HisKA domain serves as the phosphodonor for the C-terminal AR-13324 mouse receiver domain (response regulator). A histidine phosphotransferase then shuttles the phosphoryl group from the hybrid kinase to a cytoplasmatic response regulator, which could be RpfG or another downstream protein in the signaling chain carrying at least

one of the eight domains with which it could interact [38]. Thus, we are supposing that XAC3673 is an important required member of the signaling transduction process in Xcc (Fig. 4), acting together with RpfC/RpfG and required for complete virulence. When 3-oxoacyl-(acyl-carrier-protein) reductase RpfC, RpfG or XAC3673 is not functional, virulence is abolished, but the mutant is viable. Another observation that we think is important is the site of the mutation on XAC3673: the response regulator domain. The response regulator domain in RpfC and XAC3673 are very similar, indicating that they could share the same protein-protein interactions with RpfG or with other proteins in the downstream signaling pathway. Figure 4 summarizes our hypothesis about the proposed role of XAC3673 in quorum sensing in Xcc. Figure 4 Schematic representation of a suggested DSF signaling model including XAC3673. Schematic representation of a suggested DSF signaling model including XAC3673. At a low cell density, the DSF sensor RpfC forms a complex with the DSF synthase RpfF, which prevents the effective synthesis of the DSF signal.

On the day of the experimental trial, participants were selleck inhibitor asked to ingest 568 ml of water to maintain euhydration, and arrive in a fasted condition. On the morning of each trial, participants presented at an indoor sprint track to perform a standardized warm up (10-min), which consisted of jogging, cruising, sprinting, dynamic

stretching and the RSA protocol. This RSA was used as part of the warm-up and not as a measurement test. Temperature and relative humidity were recorded (Testo, Hampshire, UK) at the start and at the end of each experimental trial to check for changes in environmental conditions. Following the warm-up period, participants initiated the testing phase of the trial by performing the RSA test, followed by a 2-min recovery. Participants then completed the

LIST [16]. The LIST was comprised of 15-min sections of Adriamycin ic50 intermittent shuttle running over a 20-m distance. Each section of the LIST consisted of 11 cycles of a set running protocol. One cycle was comprised of three 20-m AZD3965 molecular weight walks (mean speed: 1.54 m · s-1), one 20 metre sprint, ~ 3 sec of rest, three 20 metre cruises (mean speed: 3.33 m · s-1) and three 20 metre jogs (mean speed: 2.86 m · s-1). Following each section, there was a 3-min recovery period. Appropriate speeds for the walk, cruise and jog shuttles of the LIST were dictated by audible signals from a pre-recorded disc. On completion of the 3-min recovery of the second and fourth section of the LIST, participants completed the RSA test, followed by 2-min recovery period (Figure 1). Throughout the experimental protocol, every attempt was made to ensure that the participants were not distracted. No interaction or encouragement occurred between the investigator and the participants, except for mouth rinse administration. Carbohydrate solutions The CHO solution was a 6.4% maltodextrin solution, containing 64 g of maltodextrin Guanylate cyclase 2C (HighFive, Bardon, England) per 1000 ml

of water. Maltodextrin was used because it is a non-sweet and colourless [5]. The PLA solution was water. To make solutions indistinguishable both treatments contained a non-calorific artificial sweetener consisting of sucralose (FlavDrops, MyProtein, Norwich, England). Each rinse solution was provided as a 25-ml bolus in a pre-weighed plastic cup. Participants were instructed to swirl all of the solution in their mouth for ~ 5 sec, before expectorating the solution back into the cup. Participants rinsed a solution 30 sec prior to each section of the LIST and each RSA test. Participants were also instructed to rinse a solution during the first 20 metre shuttle of the second, fourth, sixth, eighth and tenth cycles of each LIST section. In total, this equated to 27 rinses and 675 ml of solution being rinsed and expectorated during each trial (Figure 1). On completion of the study, participants were asked whether they could distinguish which solution contained CHO.

Each trial contained 3 matches with a 1-hr rest between match 1 and 2 and a 2-hr rest between match 2 and 3. A match contained 3 exercise periods lasting 2 minutes each with a work to rest ratio of 10 seconds: 20 seconds. After each exercise period, a 2 minute rest period was provided before the next exercise period. The load was 0.1 kp/kg body weight. The subjects were asked to pedal as fast as possible with vocal encouragement by research personnel. In the rest periods the load was removed and the subjects were asked to pedal at 60 rpm. The peak and average power of each sprint was recorded. Blood sample collection Blood samples were collected via an indwelled

cannula (20G). The cannula was frequent flushed by sterilized saline to keep it patent throughout the experiment. Ten milliliters of blood sample were collected into an EDTA tube at each sampling time. Hematological analysis was performed Duvelisib immediately after the samples were taken. Thereafter, the rest samples were centrifuged at 1500 × g (Eppendorf 5810, Hamburg, Germany) to extract plasma. The aliquoted plasma samples were stored at -70°C

before analysis. Biochemical and hormone measurements The research personnel who conducted the analysis were blind to the group of the samples. Hemoglobin concentration and hematocrit in whole blood was measured Selleck CH5183284 by a hematology analyzer (KX-21N, Sysmex Corporation, Kobe, Japan) to correct for the change in plasma volume [27]. Plasma NOx concentration was measured with modified Griess reaction using a commercial kit (Sigma, St. Louis, MO, USA). The absorbance at 540 nm was Teicoplanin measured with a microplate spectrophotometer (Benchmark Plus, Bio-Rad, Hercules, CA, USA). Plasma concentrations of insulin were measured by electrochemiluminescence (Elecsys 2010, Roche Diagnostics, Basel, Switzerland) with the kit provided by the manufacturer. Plasma glucose, glycerol and non-esterfied fatty acid (NEFA) were measured with an automatic analyzer (Hitachi 7020, Tokyo, Japan) using commercial kits (Randox, Antrim, UK). Statistical analysis All values were expressed as means ± SEMs. The area under

the curve (AUC) was calculated for plasma concentrations of glucose and insulin, as well as total carbohydrate and fat oxidation, during the 2-hr recovery period after the second match. The changes in exercise performance, plasma concentrations of metabolites, and substrate oxidation rates were analyzed by a two-way analysis of variance with repeated measures. If the treatment or interaction effect was significant, the check details differences among the 3 trials at the same time point were identified by post hoc Bonferroni test. The AUC and total carbohydrate and fat oxidation were analyzed by a one-way analysis of variance with repeated measures. If the main effect was significant, the differences among the 3 trials were identified by post hoc Bonferroni test. The analysis was performed with SPSS for Windows 15.0 (SPSS, Chicago, IL, USA).

The resulting cultures were subsequently selleck inhibitor used for further BIBF 1120 supplier Bacterial selection. Panel B shows the changes in the richness of bacterial populations during the selection process for

DON-transforming bacteria. The number of DGGE DNA bands decreased during the process of selection until a single colony isolate was obtained, which demonstrated a single major DNA band in the DGGE gel (Lane 3). Figure 4 PCR-DGGE bacterial profiles showing the richness of bacterial populations . A) Bacterial profiles before and after antibiotic treatments. Lane 1: large intestinal digesta sample (LIC); Lane 2: start culture that was the first subculture from the digesta (LIC) before lincomycin treatment; Lanes 3 and 4: same start culture after the treatment with lincomycin at 60 and 30 μg ml-1, respectively; Lanes 5 and 6: same start culture after the treatment with tylosin at 80 and 40 μg ml-1, respectively. B) Changes of PCR-DGGE bacterial profiles through the selection by antibiotics and AIM+CecExt medium. Lane 1: start culture (1st subculture from the digesta) before antibiotic and AIM+CecExt treatments; Lane 2: the same culture (in Lane 1) after antibiotic and AIM+CecExt treatments; Lane 3: a pure culture of a single colony isolate with DON-transforming activity (Isolate LS-61). Note: Lane 1, lanes 2 – 4, and lanes

5 – 6 of Panel A were from three separate DGGE gels. The migration tetracosactide of their DNA bands was not identical among the different gels. Identification of DON-transforming bacterial Rabusertib nmr isolates The sequence similarity analysis of partial 16S rRNA genes (~700 bp) of the 10 isolates with DON-transforming activity indicated that they belonged to four different bacterial groups, Clostridiales, Anaerofilum, Collinsella, and Bacillus (Table 2). Isolates within the same group had sequence similarities greater than 99%. However, isolates located in different groups showed sequence similarities less than 85%. One isolate, named LS-100, had 99% similarity in the partial sequence of 16S rRNA gene compared with that of Bacillus arbutinivorans. Table 2 Putative identity

of the selected DON-transforming bacterial isolates     Blast search     RDP Classifier Groups Isolates Closest relatives Accession # Homology (%) Closest identification 1 SS-3 Uncultured bacterium clone p-662 AF371567.1 98 Clostidiales order   LS-61 Uncultured bacterium clone B778 AY984815.1 96 Clostidiales order   LS-107 Uncultured bacterium clone B778 AY984815.1 96 Clostidiales order 2 LS-72 Unidentified bacterium clone CCCM8 AY654968.1 99 Anaerofilum genus   LS-83 Unidentified bacterium clone CCCM8 AY654968.1 99 Anaerofilum genus 3 LS-94 Coriobacterium sp. EKSO3 AJ245921.1 97 Collinsella genus   LS-117 Coriobacterium sp. EKSO3 AJ245921.1 97 Collinsella genus   LS-121 Coriobacterium sp. EKSO3 AJ245921.

2-q22 regions. This CGH profile is represented in Figure 7. Figure 7 CGH profile of FU-MFH-2 cell line showing high-level amplification of 9q31-q34, gains of 1p12-p34.3, 2p21, 2q11.2-q21, 3p, 4p, 6q22-qter, 8p11.2, 8q11.2-q21.1, 9q21-qter, 11q13, 12q24, 15q21-qter, 16p13, 17, 20, and X, and losses of 1q43-qter, 4q32-qter, 5q14-q23, 7q32-qter, 8p21-pter, 8q23, 9p21-pter, 10p11.2-p13, and 10q11.2-q22. The line in the middle (gray) is the baseline ratio (1.0); the left (red) and right (green) lines indicate ratio values of 0.8 and 1.2, respectively. Bars to the left (red) and right (green) of each frame indicate losses and gains, respectively. selleck kinase inhibitor The terminology 1(10) represents 10 aberrations detected

on chromosome 1. The same applies to other chromosomes shown in the profile. Discussion We established the FU-MFH-2 cell line derived from human pleomorphic MFH and used various analytical

methods to characterize this cell line. FU-MFH-2 cells exhibited a spindle and polygonal shape, similar to other pleomorphic MFH cell lines established previously [5, 13, 15]. The immunophenotype of FU-MFH-2 GSK872 cells in vitro and in vivo was similar to that of the original tumor cells. In addition, FU-MFH-2 cells could grow in vivo to produce tumors with histopathologic features similar to those of the original tumor in SCID mice. Furthermore, FU-MFH-2 and the original tumor had the same DNA sequence copy number changes by CGH. These findings suggested that this cell line has retained the characteristics of the original tumor. Cytogenetic analyses of pleomorphic MFH have revealed highly complex karyotypes lacking specific https://www.selleckchem.com/products/gsk126.html structural or numerical aberrations [1, 22]. Recurrent breakpoints are seen in chromosome bands 1p36, 1q11, 1q21, 3p12, 11p11, 17p11, and 19p13 [23–25]. As expected, the FU-MFH-2 cells had MEK inhibitor complex karyotypes with a number of numerical and

structural alterations, including marker chromosomes. Using M-FISH analysis, we were able to decipher the origin of marker chromosomes and complex chromosomal rearrangements. These results emphasize the usefulness of M-FISH in the description of complex changes occurring in pleomorphic MFH cell lines. CGH studies have indicated that chromosomal gains seem to be more frequent than losses in pleomorphic MFH. Genomic imbalances frequently include gains of 1p31, 5p, 6q22-q24, 7q32, 9q31-q34, 12q13-q15, and 17q and losses of 9p21-pter and 13q14-q21 [26–30]. The FU-MFH-2 cells also had gains of 1p12-p34.3, 6q22-qter, 9q21-qter, and 17 and loss of 9p21-pter. Moreover, a high-level amplification at 9q31-q34 was detected in FU-MFH-2 cells, suggesting a critical role in pleomorphic MFH progression. Interestingly, Tarkkanen et al. reported that gain of 9q32-qter was one of the most frequent genomic imbalances in MFH of bone [31]. Several candidate genes have been mapped to this chromosomal region, including VAV2, ABL1, Notch1, and Tenascin-C (TNC).

Nano Lett 2011,11(6):2311–2317.CrossRef 6. Alonso-Álvarez D, Taboada AG, Ripalda JM, Alén B, González Y, González L, García JM, Briones F, Martí A, Luque A, Sánchez ATM Kinase Inhibitor concentration AM, Molina SI: Carrier recombination effects in strain compensated quantum dot stacks embedded in solar cells. Appl Phys Lett 2008,93(12):123114.CrossRef

7. Zhou D, Sharma G, Thomassen SF, Reenaas TW, Fimland BO: Optimization towards high density quantum dots for intermediate band solar cells grown by molecular beam epitaxy. Appl Phys Lett 2010,96(6):061913.CrossRef 8. Wu J, Shao D, Li Z, Manasreh MO, Kunets VP, Wang ZM, Salamo GJ: Intermediate-band material based on GaAs quantum rings for solar cells. Appl Phys Lett 2009,95(7):071908.CrossRef 9. Wu J, Wang ZM, Dorogan VG, Li S, Zhou Z, Li H, Lee J, Kim ES, Mazur YI, Salamo GJ: Strain-free ring-shaped nanostructures by droplet

epitaxy for photovoltaic application. Appl Phys Lett 2012, 101:043904.CrossRef 10. Jo M, Mano T, Sakoda K: Lasing in ultra-narrow emission from GaAs quantum dots coupled with a two-dimensional layer. Nanotechnology 2011,22(33):335201.CrossRef 11. Wu J, Shao D, Dorogan VG, Li AZ, Li S, DeCuir EA, Manasreh MO, Wang ZM, Mazur YI, Salamo GJ: Intersublevel infrared photodetector with strain-free GaAs quantum dot pairs grown by high-temperature droplet Selleckchem EPZ 6438 epitaxy. Nano Lett 2010,10(4):1512–1516.CrossRef 12. Jolley G, McKerracher I, Fu L, Tan HH, Jagadish C: The conduction band absorption spectrum of interdiffused InGaAs/GaAs quantum dot infrared photodetectors. J Appl Phys 2012,111(12):123719.CrossRef 13. Pankratov EL: Optimization of pulse laser annealing to increase sharpness of implanted-junction Cobimetinib chemical structure rectifier in semiconductor heterostructure. Nano-Micro Lett 2010, 2:256–267. 14. Martí A, Antolín E, Linares PG, Luque A: Understanding experimental characterization of intermediate band solar

cells. J Mater Chem 2012, 22:22832–22839.CrossRef 15. Hsu TM, Lan YS, Chang W, Yeh NT, Chyi J: Tuning the energy levels of self-assembled InAs quantum dots by rapid thermal annealing. Appl Phys Lett 2000,76(6):691.CrossRef 16. Fu L, McKerracher I, Tan HH, Jagadish C, Vukmirovic N, Harrison P: Effect of GaP strain compensation layers on rapid thermally annealed InGaAs/GaAs quantum dot infrared photodetectors grown by metal-organic chemical-vapor deposition. Appl Phys Lett 2007,91(7):073515.CrossRef 17. Pierz K, Ma Z, Keyser UF, Haug RJ: Kinetically limited quantum dot formation on AlAs(100) surfaces. J Cryst Growth 2003,249(3–4):477–482.CrossRef 18. Sanguinetti S, Watanabe K, Kuroda T, Minami F, Gotoh Y, Koguchi N: Effects of post-growth annealing on the optical selleck compound properties of self-assembled GaAs/AlGaAs quantum dots. J Cryst Growth 2002,242(3–4):321–331.CrossRef Competing interests The authors declare that they have no competing interests.

Amplified Fragment Length Polymorphism (AFLP) Genomic DNA from individual symbiont strains was used for AFLP as described by [47]. Briefly,

DNA was digested with the two restriction AZD2281 enzymes ApaI (4U) and TaqI (4U), and ApaI and TaqI adapters were added (Additional file 8: Table S5). After pre-amplifying the Adriamycin mw ligation product, selective amplifications were conducted using the two differently labeled primers TaqI-G (IRDye 700) and TaqI-C (IRDye 800) in combination with one out of ten ApaI primers with two selective nucleotides (see Additional file 8: Table S5). Amplified products were separated based on size with a LI-COR DNA Analyzer 4300. A formamide-dye stop solution was added to the AFLP reactions, and samples were heat-denatured before electrophoresis.

For separation, a 6.5% polyacrylamide gel was used, and a labeled size standard was loaded at each end. Gels were run for 2.5 h and subsequently scored using the software AZD3965 cell line AFLP-Quantar™ Pro 1.0 (KeyGene Products, Wageningen, The Netherlands). Scoring results of 202 AFLP markers were converted into ‘pseudo-sequences’ (with presence = ‘A’, absence = ‘T’, and unknown = ‘N’), imported into MEGA5.01 [45], and used to construct a neighbour-joining phylogeny including 100 replicates for bootstrap analysis. Acknowledgements We are grateful to Tobias Engl, Sabrina Köhler (MPI-CE, Germany), Christine Michel (Germany), and Erol Yildirim (Atatürk University, Turkey) for help with collecting beewolf specimens for symbiont isolation. We thank Astrid Groot and Susanne Donnerhacke (MPI-CE, Germany) for help with the AFLP analysis, Benjamin Weiss and Ulrike Helmhold (MPI-CE, Germany) for assistance with bacterial strain identification and Susanne Linde (Centre for Electron Microscopy, Germany) for electron microscopy. Collecting permits were issued by the nature conservation boards

of KwaZulu Natal (Permit No. 4362/2004), Eastern Cape Province (WRO 44/04WR, WRO9/04WR, WRO74/06WR, WRO75/06WR, CRO135/11CR, CRO136/11CR, CRO179/10CR, and CRO180/10CR) and Western Cape Province (001-202-00026, 001-506-00001, AAA004-00053-0035, AAA004-00089-0011, Guanylate cyclase 2C AAA004-00683-0035, and 0046-AAA004-00008) of South Africa, and the Brazilian Ministry of the Environment (MMA/SISBIO/22861-1). We gratefully acknowledge financial support from the Max Planck Society (MK) and the German Science Foundation (DFG-KA2846/2-1 [MK]). Supporting data The data set supporting the results of this article is available at the http://​www.​biomedcentral.​com/​bmcmicrobiol/​. Additional files Additional file 1: Table S1. Composed media recipes. Additional file 2: Table S2. Composition of commercial cell line media used in this work (amounts in mg/L). Additional file 3: Table S3. Number of ‘S. philanthi’ CFUs isolated from different females’ antennal samples. Additional file 4: Table S4. Accession numbers of actinobacterial sequences included in the phylogenetic analyses shown in Figure 3.