SEARCH

SEARCH BY CITATION

Fig. S1. Sampling site near the Kuparuk River at the Toolik LTER Field Station, Alaska. Map modified after Walker and Maier (2008) and http://www.arcticatlas.org.

Fig. S2. Core T08S1C3 prior to and after preparation for aseptic subsampling. Selected intervals for geochemistry and radiocarbon dating, exoenzyme activity assays and molecular analysis marked with numbered rectangles. Depths and radiocarbon ages are denoted above or below the selected intervals. See Table S1 for calendar age data. After removal of the outer 1 cm of core, the frozen intervals were split and subsamples were taken from the central part of each section.

Fig. S3. Denaturing gradient gel electrophoresis of PCR-amplified bacterial 16S rDNA (i.e. DNA) and reverse transcribed 16S rRNA (i.e. cDNA). Totally, 123 bands were excised from the gels and all were successfully sequenced. Bands melting at identical positions in the DGGE contained identical sequences and carry the same number. The 123 sequenced DGGE bands represented 30 unique bacterial OTUs based on a 98% sequence similarity cut-off. Sequenced DGGE bands that cluster within the same phylogenetic group (compare Fig. 1 of the main article) carry a unique colour. Control PCR reactions (35 cycles) with DNase-treated RNA were run parallel with cDNA samples from pristine deeper permafrost soils and at day 6 of the thaw experiment for the deepest soil sample. Subsequent sequencing analysis of DGGE bands (numbered 5 in the gel) stemming from the control PCRs revealed a single phylotype (T08S1C3Bac5; Fig. S4) and was discarded as a member of the indigenous soil bacterial population. Other samples required fewer cycles during PCR and parallel control reactions remained negative.

Fig. S4. Neighbour-Joining tree showing the bacterial 16S rRNA gene and transcript sequences recovered from the pristine active layer and permafrost samples and after 6–11 days of incubation at 4°C (soil OTUs marked with white text in black boxes) and the most similar sequences present in the NCBI database. The soil OTUs (e.g. T08S1C3-Bac1) are depicted from corresponding sequenced DGGE bands (e.g. DGGE band #1) of Fig. S3. T08S1C3-Bac5 related to γ-Proteobacteria was recovered from PCR-amplified DNase-treated RNA controls and therefore not considered to be part of the indigenous soil community.

Fig. S5. Denaturing gradient gel electrophoresis of PCR-amplified eukaryal 18S rDNA (i.e. DNA) and reverse transcribed 18S rRNA (i.e. cDNA). Excised bands melting at identical positions in the DGGE contained identical sequences and carry the same number. The 86 out of 90 successfully sequenced DGGE bands represented 35 unique OTUs based on a 98% sequence similarity cut-off (Fig. S6). Sequenced DGGE bands that cluster within the same phylogenetic group (compare Fig. 3 of the main article) carry a unique colour. Control PCR reactions (35 cycles) with DNase-treated RNA were run parallel with cDNA samples from deeper permafrost soils at day 6 of the thaw experiment. Resulting DGGE bands of these ‘controls’ were sequenced and represented fungal OTU EukF6, or plant sequences (Viridiplantae; OTUs V2 or V8). The plant sequences marked V1–V8 in the gel were not of interest for this article and therefore not discussed. Other samples required fewer cycles during PCR and parallel control reactions remained negative.

Fig. S6. Neighbour-Joining tree showing the fungal 18S rRNA gene and transcript sequences recovered from the pristine active layer and permafrost samples and after 6–11 days of incubation at 4°C (soil OTUs marked with white text in black boxes) and the most similar sequences present in the NCBI database. The soil OTUs (e.g. T08S1C3-EukF1) are depicted from corresponding sequenced DGGE bands (e.g. DGGE band F1) of Fig S5. T08S1C3-EukF5 was recovered from PCR-amplified DNase-treated RNA controls and therefore not considered to be part of the indigenous soil community.

Table S1. Analysed soil depths with uncorrected radiocarbon years before the present (14C age BP) and calibrated calendar years before the present (Cal BP). AL, active layer.

FilenameFormatSizeDescription
EMI_2489_sm_REV.docx7569KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.