Lipids are fundamental components of many biological structures, and their composition is diet dependent. Differences in lipid composition can impact the functioning of cellular membranes and proteins, subsequently altering the organism’s ability to respond to environmental conditions. The American lobster (Homarus americanus) is an economically important shellfish in New England and is frequently kept in lobster impoundments (pounds) for prolonged periods where they are typically fed herring, which differs from the natural diet of wild-caught lobsters. In this study, we compared the lipid composition of lobster muscle, heart, and hepatopancreas from wild caught and from pound kept lobsters that were fed different diets. We performed lipidomic analysis, mitochondrial function tests, tested gene expression of cellular stress markers, and evaluated thermal stress tolerance assessing heart and ventilation rate, as well as hemolymph oxygenation during a fast progressive temperature challenge. We find a significant shift in lipid composition in pound kept lobster hearts together with reduced mitochondrial function, and increased gene transcription of cellular stress markers HSP70 and AMPK, indicating a worse nutritional state, compared to wild fed lobsters. The shift in the various parameters did not lead to a shift in thermal thresholds, indicating a substantial plasticity and tolerance to compensate for adverse diet-induced conditions. This study links mechanistically diet, lipid composition, mitochondrial function and thermal tolerance and highlights the need of a more detailed understanding of cellular processes to understand climate change-induced impacts on marine invertebrates.

Frozen tissue samples were used in mass spectrometry-based multiple precursor ion scan (MPIS) lipidomic analysis (Simons et al., 2012) at the MaineHealth Institute for Research (MHIR) Proteomics and Lipidomics core facility in Scarborough, Maine. Lipids were isolated using a modified Bligh and Dyer protocol (Bligh and Dyer, 1959; Liaw et al, 2016; Okeoma et al., 2024) using dichloromethane and methanol. Briefly, around 20 mg of tissue was transferred to an Eppendorf cup with 200 µL of 66% methanol (MeOH) (Alfa Aesar-Themo Fisher, Waltham, MA, USA) in LC-MS grade water (Burdick and Jackson-Honeywell, Muskegon, MI, USA) and kept on ice all the time. Tissues were homogenized with a pellet pestle motor (Kimble, Millville, NJ, USA) and then by sonication using a Branson Sonifier 250 (VWR Scientific, Radnor, PA, USA) for ten 30-second cycles at 30% until the tissue was dissolved. Samples sat on ice for 10 min to allow for debris to settle. Supernatants were then transferred to a glass screw-cap tube. 66% MeOH and 0.9 mL dichloromethane (DCM) (Sigma Aldrich, St. Louis, MO, USA) was added to allow all samples to reach a volume of 3 mL. Samples were then vortexed and 50-100 µL MeOH was added to allow for a DCM-MeOH-sample monophase. Samples were incubated at room temperature for 30 minutes before the aqueous and organic phases were separated by the subsequent addition of 1 mL water and 0.9 mL DCM. All samples were centrifuged at 1200 rpm for 10 minutes and the lower DCM organic phase was removed to a new glass tube and dried under nitrogen and a partial vacuum using a Visiprep manifold (Supelco, Bellefonte, PA, USA). Lipid extracts were dissolved in lipid load solution: MeOH/DCM (50:50, v/v) containing 10 mM ammonium acetate (NH₄Ac) LC/MS grade (Fisher, Waltham, MA, USA) for MS analysis. Approximately 100 µL of a 4-fold diluted lipid extract in lipid load solution, was delivered by PAL3 System (LEAP Technologies, Naperville, IL, USA) into the source of a Sciex QTRAP 4000 mass spectrometer. A PAL3-RSI (Trajan Inc.) robotic sample handling instrument was used to deliver lipid load solution at a rate of 0.25 mL min^-1 during all the runs. Source parameters included gases GS1 at 15 and GS2 at 20, curtain gas at 10, IonSpray voltage 5300, temperature at 150, and collision energy was 50 eV. Each sample was injected twice and, positive and negative polarity modes were acquired simultaneously in serial experiments from a single sample infusion. The instrument was controlled, and quality control for instrument performance included bovine heart extract (Avanti Research/Croda Inc., Alabaster, AL, USA).