Appendix A. Molecular methods in detail.
Roots not used for assessing AMF colonization microscopically were frozen in liquid nitrogen and lyophilized. A random sample of lyophilized root was selected (approximately 20-25 mg) and the weight of each sample recorded. Extraction of DNA was done using the Qiagen DNeasy plant mini kit following the manufacturer recommendations for the purification of total DNA from plant tissue (Qiagen Sciences, Germantown, Maryland, USA). The qPCR reactions were done using a LightCycler 2.0 (Roche Applied Science, Rotkreuz, Switzerland). Reagent amounts per reaction were: 4.55 ml H2O, 0.18 ml forward primer (25 µM), 0.18 ml reverse primer (25 µM), 0.04 ml hydrolysis probe (25 µM), 1.8 ml Roche LightCycler® TaqMan® Master mix (Roche Applied Science, Rotkreuz, Switzerland) and 2.25 ml of template DNA. The sequences of primers and hydrolysis probes for the nuclear large ribosomal subunit (LSU) specific to G. mosseae, G. intraradices and G. claroideum, are outlined elsewhere (Thonar 2009). The oligonucleotides were synthesized at Microsynth (Balgach, Switzerland), hydrolysis probes were labelled with fluorescein and BHQ-1 quencher at the 5`and 3`s ends, respectively. The primers were purified by polyacrylamide gel electrophoresis and hydrolysis probes by preparative HPLC. For the qPCR quantification of LSU copies of D. celata, novel primers and a hydrolysis probe were designed. Forward primer sequence was 5`-TCGGAGGTTGTAAAATACTTGG-3`, the reverse primer was 5`-CAAAGGCATTTGCTGCAATC-3`, hydrolysis probe sequence was 5`-AAGGTCTATAACACTCTCCCGAAG-3`. This combination targeted a 106-bp fragment of LSU of D. celata. Specificity of amplification was confirmed using genomic DNA preparation from spores of 19 different AMF strains as well as root colonized by the different AMF species included in this study (data not shown). The PCR program for G. intraradices and G. claroideum consisted of an initial denaturation and Taq-polymerase activation step at 95 °C for 15 min, followed by 50 cycles of: denaturation at 95 °C for 15 s, annealing at 52 °C for 30 s, and elongation at 72 °C for 1 s. Following the 50 cycles, the program concluded by cooling at 40 °C for 30 s. This same program structure was used for G. mosseae and D. celata with an annealing conditions of 54 °C for 10 s and 62 °C for 15 s, respectively.
The cycle threshold (Cp) values as determined by the automatic function using the LighCycler software were recorded and used to estimate the number of LSU copies per mg of lyophilized root by the equation:
Where MR is the mass of lyophilized root in mg used for DNA extraction and a and b are constants derived from the calibration equations of each AMF primer and hydrolysis probe set. The calibration of each primer set used was done using serial dilution in series of 10 ´ dilutions of a plasmid carrying target DNA sequence, with a known concentration (see Jansa et al. 2008 for determining DNA concentration) and plotting the regression line between the log of the concentration of template DNA and the Cp of the reaction curve. The slope of the regression line is the constant a and the intercept is the constant b used in equation 1. The factor 200 stands for DNA preparation volume (in microliters), the calibration was derived as number of LSU copies per microliter template. In samples where inoculated AMF were not detected the number of LSU copies per mg of lyophilized root was assumed to be zero and was analyzed as such within all statistical calculations.