Determination of available soil phosphorus
Ortho-phosphate (PO₄^-2) content of the soil was measured by extracting 10 g of soil in 50 ml of 0.5M NaHCO₃ at pH 8.5. Extracts were shaken for 90 min at 170 rpm, then decanted into centrifuge tubes and spun at 27,216 x g for 10 min. Supernatant was poured into Nalgene bottles and acidified to approximately pH 2. The acidified filtrate was stored at -20 ºC, shipped to Dartmouth College, and analyzed on a Lachat QuikChem 8500 (Lachat Instruments)¹.

Soil sampling and nematode isolation
Soils were sub-sampled within 48 hours of collection under a laminar flow hood in at McMurdo Station, Antarctica to provide material for invertebrate and chemical analyses. Nematodes were extracted from sieved soil samples using a sugar-centrifugation technique modified for Antarctic soils².

Determination of somatic phosphorus
Nematodes were identified based on morphology under light microscopy and picked into RNAlater® (Ambion Inc.) solution. Nematodes preserved in RNAlater® were shipped to Brigham Young University, Provo for RNA extraction and further analyses. Nematode samples preserved in RNAlater were washed twice with a 5% PBS (Phosphate-Buffered Saline) solution and shipped to ALS Laboratory Group at Salt Lake City for total phosphorus analysis. To test if preservation in PBS had any effect on nematode P-content, samples containing 10, 50, and 100 Heterorhabditis bacteriophora nematodes stored in distilled water and PBS for two weeks were analyzed and compared for body P-content. For Scottnema lindsayae and Plectus murrayi, six bulk samples containing 25 nematodes each were used for total phosphate measurement and the P-content of individual nematodes was calculated. Measurement of total P in whole nematode body was done by using inductively coupled plasma mass spectrometry (ICP-MS). The total nematode body phosphorus content was measured on pooled samples (25 individual nematodes per sample) using Inductively Coupled Plasma Mass Spectrometry (ICP-MS, ALS laboratory group in Salt Lake City) with external calibration set according to the manufacturer’s standard protocol.

Nematode body size and biomass
In 2002 the climate regime of MCM soil ecosystems dramatically shifted from a decadal cooling trend³ to a warmer, wetter, climate with greater interannual variability⁴. To account for potential effects of this climate shift on habitat suitability and nematode phenotypes, we measured adult S. lindsayae of each sex from each till type (N = 118 (1999), 94 (2004) for Ross Sea till and N = 46 (1999), 54 (2004) for Taylor II till) sampled from 6 different soil monitoring plots per till type during the 1999-2000 and 2004-2005 austral summers. All nematode samples were preserved in 10% hot formalin (~ 60ºC). Animals were photographed using SPOT 3.0 imaging software for microscope digital cameras (SPOT Imaging, Diagnostics Instruments). The images were measured (lengths from tail to mouth, widths from just behind the basal pharyngeal bulb) using Carnoy 2.0 digital image analysis program (Biovolution). Biomass calculations were made using the formula of Andrassy (1956)⁵ as directed by Freckman (1982)⁶.

Experimental evolution: Culturing conditions
Cultures of P. murrayi were established according to Adhikari et al (2010)⁷. Caenorhabditis elegans strain N2 (CGC, University of Minnesota) and P. murrayi were reared on media of two different P concentrations for 46 months which is 23-31 generations for P. murrayi and around 345 generations for C. elegans. Nematodes were grown on phosphorous sand agar media, which includes 15g Agar, 965 mL H₂O, 20 ml BMB for both P-poor and P-rich plates, and 1.033 mg K₂HPO₄ for P-poor and 10.33 mg K₂HPO₄ for P-rich. The pH was adjusted 7.0, H₂O was added to 1.0 L, and the mixture was autoclaved for 20 minutes at 120 °C. Sterile builder’s sand was poured onto cooled 60-mm plates and stored at 4 °C. The plates were then inoculated with 30µL of stock OP-50 and incubated at 37 °C for 2 days. Nematodes were transferred onto the plates and incubated at 27 °C for 1 week, followed by incubation at 15 °C for 3 weeks. The process was then repeated for each successive generation⁷. For all subsequent comparative tests, populations were reared on common plates of P-rich and P-poor nematode growing media (NGM)⁸ for at least two generations.

RNA extraction and real-time PCR
Nematodes stored in RNAlater solution were washed twice with a 5% solution of phosphate buffer saline (PBS) before RNA extraction. Total RNA for Quantitative real-time PCR (qRT-PCR) was extracted using Trizol reagent (Molecular Research Center Inc.). Three replications of each sample were used for RNA extraction, yielding three independent RNA extracts for each bulk sample. Nematodes were directly homogenized in liquid nitrogen, mixed with Trizol Reagent, and the suspension was exposed to three freeze-thaw cycles using liquid nitrogen and 37ºC water bath. The suspension was ground using mortar and pestle, vortexed, and phase separated using chloroform. After centrifugation (15 min, 12000 g, 4 °C), the aqueous phase containing RNA was separated from the other phases, which were stored for DNA preparation (see below). The colorless upper aqueous phase was transferred into fresh vials to precipitate the RNA by addition of 100 ml isopropyl alcohol. The samples were incubated for 10 min and centrifuged (20 min, 12000 g, 4 °C). The RNA precipitates were then washed twice with 75% ethanol, air-dried, eluted in nuclease-free water, and quantified and quality-checked via spectrophotometer (A260/A280>1.9; NanoDrop ND-1000, NanoDrop Technologies, Thermo Fisher Scientific Inc.) and agarose gel electrophoresis.

Reverse transcription (RT) was performed with 1µg of total RNA extracted from pooled sample of nematodes. The RT of polyadenylated mRNA to cDNA was done using the ImPromp-II™ reverse transcriptase (Promega Corporation) and a random hexamer primer. Total RNA was incubated with 20 pmol random hexamer primer at 70 °C for 5 min and then quickly chilled on ice. The reverse transcription mixture (20 µl) was mixed with RNA template and incubated at 25 °C for 5 min to encourage annealing and the first strand was extended for 60 min at 42°C. The cDNA was precipitated in 100% ethanol and washed twice with 75% ethanol, air-dried, and dissolved in DEPC-treated water.

Quantitative real-time PCR was performed with LightCycler 480 SYBER Green I Mastermix (three replicate samples for each extraction) and gene specific primers in a Light Cycler 480 RT-PCR system (Roche Applied Science) equipped with LightCycler 480 software with the following program: 3 min at 95 °C; 45 cycles of 30 s at 94 °C, 30 s at 58 °C and 1 min at 72 °C followed by a standard melt curve. The RT-PCR reaction had a final volume of 10 μl including SYBR Green Mastermix (Roche Applied Science), and template DNA. Negative control reactions containing water in place of cDNA were included in each batch of PCR reactions to ensure that contamination was not a problem. We used nematode ß-actin, ß-tubulin and GAPDH genes as internal control for natural population of S. lindsayae and P. murrayi and ß-actin for laboratory population of P. murrayi and C. elegans.

Life history trait observations
P. murrayi and C. elegans were grown in 50 of P-poor and 50 of P-rich plates as described above, 10 cultured plates with healthy populations were randomly selected for further observation (without contamination). For each of the 10 cultured nematode populations, 20 pregnant females were individually picked onto fresh petri dishes that contain regular P and seeded with OP50 Escherichia coli as a food source. All plates were kept at 15 ℃ and checked daily to record eggs laying. After hatching, juveniles were individually transferred to new plates and observed three times a day. The hatching day was recorded as day 0, the development duration was defined by their body volume based on de Tomasel’s study⁹. The duration of each stage and total egg abundance of each culture plate were observed and recorded. Photographs were taken daily, and juvenile body sizes were captured using a CKX 41 Olympus inverted microscope and body sizes, including body lengths and volumes, were measured and calculated by the modified Andrássy formula 9 in MatLab (MATLAB 6.1, 2000).

Analyses notes
Total biomass calculations were calculated by determining the average individual dry weight from individual lengths and widths and corrected for water content using standard calculations^5,7. For qRT-PCR experiments, gene expression level changes (fold change) of the rRNA gene was normalized to single-copy nuclear gene expression (ß-actin, ß-tubulin and GAPDH). The fold change in expression of rRNA in samples from Taylor Ⅱ till is as equal to one. ß-actin, ß-tubulin and GAPDH were also used as internal controls to minimize genomic template contamination bias and correct the variations among samples¹⁰. Environmental analyses were performed by the BYU Environmental Analytics Lab.

Temporal range
Samples from which animals were isolated and measured for body size  and biomass were collected and preserved during the 1999-2000 and 2004-2005 austral summer and accessioned into the BYU Life Science Museum. Soil samples from which animals were originally isolated and cultured, measured for somatic phosphorus, and for which the soil chemistry analyses were conducted, were collected during the 2008-2009 austral summer. Real-time PCR experiments were conducted in 2009 and 2015. Measurements of life history of cultured animals were completed in 2015. Transcriptome data were collected in 2016.

Geographic location
Soil samples used in this study were collected from Taylor Valley, Antarctica, from the southeast end of the Lake Fryxell basin at -77° 36.48 S, 163° 14.94 E, and from the southwest end of the Lake Bonney basin at -77°43.56 S 162°18.84 E.

¹ Ball, B. A., Adams, B. J., Barrett, J. E., Wall, D. H. & Virginia, R. A. Soil biological responses to C, N and P fertilization in a polar desert of Antarctica. Soil Biology and Biochemistry 122, 7-18, doi:https://doi.org/10.1016/j.soilbio.2018.03.025 (2018).
² Freckman, D. & Virginia, R. Low-diversity Antarctic soil nematode communities: Distribution and response to disturbance. Ecology, 363-369 (1997).
³ Doran, P. T. et al. Antarctic climate cooling and terrestrial ecosystem response. Nature 415, 517-520, doi:10.1038/nature710 (2002).
⁴ Gooseff, M. N. et al. Decadal ecosystem response to an anomalous melt season in a polar desert in Antarctica. Nature Ecology & Evolution 1, 1334-1338, doi:10.1038/s41559-017-0253-0 (2017).
⁵ Andrassy, I. Die Rauminhalts und Gewichtsbestimmung der fadenwurmer (Nematoda). Acta Zoologica Hungary 2, 1-15 (1956).
⁶ Freckman, D. & DW, F. Parameters of the nematode contribution to ecosystems.  (1982).
⁷ Adhikari, B. N., Tomasel, C. M., Li, G., Wall, D. H. & Adams, B. J. Culturing the Antarctic nematode Plectus murrayi. Cold Spring Harb Protoc 2010, pdb prot5522, doi:10.1101/pdb.prot5522 (2010).
⁸ Chaudhuri, J., Parihar, M. & Pires-daSilva, A. An introduction to worm lab: from culturing worms to mutagenesis. J Vis Exp, 2293, doi:10.3791/2293 (2011).
⁹ de Tomasel, C. M., Adams, B. J., Tomasel, F. G. & Wall, D. H. The Life Cycle of the Antarctic Nematode Plectus murrayi Under Laboratory Conditions. Journal of Nematology 45, 39-42 (2013).
¹⁰ Vandesompele, J. et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology 3 (2002).