IMIS

Cruise track, sampling sites and an overview of geochemical and biological properties have been detailed elsewhere (Yager et al., 2012). Additional information can be found in the BCO-DMO database (http://osprey.bco-dmo.org/project.cfm?id=146&flag=view). For the 2010–2011 cruise, water samples (3–10 L) for microbial community sequence analyses were passed over a 20 μm mesh and collected onto 0.2 μm Sterivex membrane filter cartridges by pressure filtration (Whatman Masterflex L/S series). Since high biomass caused rapid clogging of the filters, the sampling volumes varied between stations. Two distinct plankton size classes (0.2–3 μm and 3–200 μm) were fractionated for samples collected during the 2007–2008 cruise. This sampling effort (10–20 L) was done along a depth profile that spanned the full water column and the microbial community analysis was part of the International Consensus for Marine Microbes project. Filters were quickly frozen in the headspace of a LN2 Dewar and stored at −80°C prior to DNA extraction. We note that the 2007–2008 data were determined on samples from a single depth profile inside the ASP. It was decided to incorporate these data in order to derive first hints regarding the reproducibility of bacterial community compositions that accompany Phaeocystis blooms in the ASP and to gain early insights into the bacterial taxa that may associate with Phaeocystis colonies and other particles.

Study Extent: Water samples were collected at various sites across the Amundsen Sea polynya (ASP) during the austral summers of 2007–2008 (aboard the icebreaker R/V “Oden”; depth profiles) and 2010–2011 (ASPIRE cruise aboard the R/V “Nathaniel B Palmer,” horizontal grid of surface samples).

Method step description:

1. DNA extraction was performed using the Puregene kit (Gentra®) after disruption of the cells with lytic enzyme coupled to proteinase K (Sinigalliano et al., 2007). DNA concentrations were quantified using a Nanodrop 2000 instrument (Thermo Fisher Scientific, Wilmington, DE).
2. The V6 hypervariable region of the 16S rRNA gene (typically 60–65 bp in length) was amplified (25 cycles using HiFi buffer 1X, MgSO4 2 mM, dNTPs 0.2 mM, combined primers 0.2 mM and four units of platinum HiFi) in triplicate PCR reaction from 10 ng of environmental DNA templates with reverse primer (1046R) “CGACRRCCATGCANCACCT” and the forward primer mix (967F) “CTAACCGANGAACCTYACC,” “ATACGCGARGAACCTTACC,” “CNACGCGAAGAACCTTANC,” and “CAACGCGMARAACCTTACC.” PCR cycle conditions were defined as follow: 30 s at 94°C followed by 45 s at 60°C and 1 min at 72°C. The PCR started with 3 min at 94°C and ended with 2 min at 72°C followed by a rapid stepdown to 4°C. Negative controls (no template DNA) were run for each of the index primer combinations in the PCR reactions. V6 amplicon sequences from samples collected during the 2007–2008 R/V “Oden” cruise (n = 12) were obtained on a GS-FLX pyrosequencing platform. Sequence reads were subsequently trimmed for low-quality sequences (Huse et al., 2007). For samples collected on the ASPIRE cruise during the 2010–2011 austral summer (n = 23), a paired-end sequencing strategy for Illumina Hiseq platform was employed with custom fusion primers described previously (Eren et al., 2013b) targeting the V6 hypervariable region of the 16S rRNA gene.