Optimized lipid extraction and annotation pipeline customization for individual chitinous mesozooplankton using UPLC-HRMS

Abstract

High-resolution lipidomics at the scale of individual mesozooplankton offers a powerful tool for understanding trophic interactions and carbon cycling in marine ecosystems, but chitinous exoskeletons present challenges for efficient lipid extraction. Here, we developed and validated an optimized Bligh and Dyer–based extraction protocol that incorporates in-line glass bead homogenization, yielding a 2.5-fold increase in lipid recovery and, when combined with an increased injection volume, a 4.4-fold gain in signal intensity. This workflow enables robust detection of intact lipid species from single Calanus copepods without additional homogenization equipment or extended extraction steps, making it broadly accessible for analytical applications. Furthermore, to address the limitations of current annotation pipelines, we compared adduct-hierarchy (LOBSTAHS) and fragmentation-based (MS-DIAL) approaches directly, finding systematic biases that reshape lipidomic profiles depending on the computational strategy employed. Additionally, by integrating a wax ester-specific fragmentation library, we demonstrated improved annotation of marine-relevant lipid classes largely absent from conventional databases. Together, this extraction and hybrid annotation pipeline enables high-resolution, compound-specific lipidomics of individual mesozooplankton, capturing biological heterogeneity while remaining scalable to pooled samples. Our approach provides a critical methodological advance for tracing lipid metabolism across trophic levels and for quantifying the role of mesozooplankton lipids in marine biogeochemical cycles.

Introduction

The advent of high-resolution mass spectrometry coupled with ultra-high performance liquid chromatography (UPLC-HRMS) has enabled the detection and quantification of increasingly more compounds in less biomass, with single-cell lipidomics beginning to emerge as a tractable technique (Hunter et al., 2021; Wang et al., 2023). However, the smallest biological unit for which reliable and routine lipidomic measurements have been reported is the individual multicellular eukaryotic organism, like those living at the micro- (20-200µm) and mesoscale (0.2-20mm) (Sieburth et al., 1978). In the marine ecosystem, small invertebrates within these size ranges often occupy lower trophic positions in food webs and serve as key intermediaries in the transfer of essential lipids (i.e., lipids that are not produced de novo by most metazoans and must be obtained from their diet) - such as docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) - to higher trophic levels, underscoring the importance of lipidomic profiling at this scale.

Many of these organisms, including copepods and other small crustaceans, possess chitinous exoskeletons. Chitin, a nitrogen-containing polysaccharide, has a supramolecular structure, stabilized by extensive inter- and intramolecular hydrogen bonding that makes it insoluble in common organic solvents (Rinaudo, 2006). Because chitin is the most abundant amino polysaccharide in nature, its insolubility represents a significant challenge to researchers looking to extract the lipidome of chitin-encased species. Currently, most extraction methods for this taxa rely on heavily extended versions of traditional lipid extraction protocols that employ longer processing steps such as shaking, sonication, or incubation. These methods, while effective, are often time-consuming, labor-intensive, depend on large quantities of biomass, require extra equipment, or are otherwise not optimized for organisms encased in chitin (Supplementary Table 1, Figure 1).

Figure 1

The original Bligh and Dyer method was first developed in the 1950s to extract lipids from fish muscle. In this system, the lower organic phase contains lipids and other relatively non-polar compounds, while the upper aqueous phase retains more polar metabolites. The protocol was modified in the 2000s by substituting dichloromethane (DCM) in for chloroform to reduce toxicity and improve handling (Cequier-Sánchez et al., 2008; Sturt et al., 2004; Van Mooy et al., 2008) and was subsequently optimized for the extraction of marine phytoplankton (Popendorf et al., 2013; Van Mooy and Fredricks, 2010). Here, we present a hybridized lipid extraction method adapted from Van Mooy and Fredericks’ version of the classic Bligh and Dyer protocol, that enhances extraction efficiency from Calanus spp., a genus of cosmopolitan zooplankton belonging to the class Copepoda. Other versions of the Bligh and Dyer method have been applied to analyze copepod metabolomes (Hansen et al., 2013; Mayor et al., 2015), as well as one instance of a Matyash lipid extraction, that uses Methyl tert-butyl ether (MTBE) instead of chloroform or DCM, on commercially sourced Calanus copepod specimens (Matyash et al., 2008; Wood et al., 2023). Given the physiological diversity within copepods and the differences in experimental context, direct comparisons with our results are limited. For these reasons, we focus on the development and optimization of the Bligh and Dyer protocol adapted for marine microalgae, as established and applied within our research group.

Alongside challenges in extraction, accurate compound annotation remains a challenge in zooplankton lipidomics. In lipidomics, identification of compounds typically relies on a multi-pronged strategy: (1) MS/MS fragmentation patterns, and (2) the “adduct hierarchy” of lipids, which infers compound identity and deconvolutes isomeric and isobaric compounds based on preferred ionization behavior during electrospray ionization (ESI), and (3) comparison of retention times to authentic standards. Although all approaches are widely used and experimentally valid, each presents specific limitations. For example, adduct-based identification tools, such as LOBSTAHS (Lipid and Oxylipin Biomarker Screening Through Adduct Hierarchy Sequences), are constrained by empirical validation of adduct hierarchies for additional compound classes, and regioisomers within the database resulting in competing annotations. In contrast, MS/MS-based annotation methods such as MS-DIAL or MZmine, especially those employing in silico libraries, can overcome this limitation but require substantial amounts of data generation. Fragmentation spectra can also vary significantly depending on collision energy, as well as other instrument-specific parameters (Lee et al., 2024). As databases expand, the computational cost and time for annotation increase exponentially, further limiting accessibility to researchers without access to high-performance computing resources. To illustrate the practical consequences of these limitations, we conducted a direct comparison between two widely used annotation platforms: LOBSTAHS and MS-DIAL. This comparison highlights how differences in database structure, spectral libraries, and computational pipelines can significantly influence annotation outcomes, even when applied to identical datasets.

Furthermore, most publicly available lipid databases are built around mammalian, terrestrial or freshwater lipidomes and lack coverage of marine-specific lipid classes. For instance, our study focuses on wax esters, a group of relatively neutral lipids that are produced by the ester bonding of a fatty acid and an alcohol. While wax esters are one of the predominant lipid classes in Calanus copepods, making up to 90% of the total lipid mass (Albers et al., 1996), they are, to our knowledge, notably absent from commonly used databases. We therefore highlight the usage of an “18-series” fragmentation pattern that has been observed in wax esters and β-hydroxy ceramides (Chen et al., 2015; Tsugawa et al., 2017), but has not been implemented in pre-existing lipid databases (Kind et al., 2013; Tsugawa et al., 2020).

Taken together, these challenges in extraction and annotation point to a need for methodological refinement. In this study, we aim to develop and evaluate an extraction protocol suitable for individual chitinous zooplankton and to assess how different annotation strategies influence their interpretation. By addressing both experimental and computational constraints, our goal is to enable more accurate and accessible lipidomic characterization at the scale of individual marine mesozooplankton.

Materials and methods

Sample collection

Copepods from the genus Calanus were collected from Svalbard, Norway aboard the R/V G.O. Sars using a 180 µm mesh size Bongo net during the 2022 Nansen Legacy Polar cod cruise in November, with a smaller subset collected in February 2023. Samples were all identified under an on-deck stereomicroscope and flash frozen in centrifuge tubes at -80 °C using liquid nitrogen. The second subset was collected on a privately chartered boat with a WP2 net, from 180 m to the surface. Collected copepods were kept at -2 °C and brought back to the University Centre in Svalbard (UNIS), after which they were transported by plane to Oslo at 0 °C. This second subset of samples were part of a larger experiment that involved three weeks of lab culture at a pH of 7.9 (control conditions) before being flash frozen for analysis at the University of California, Berkeley.

Lipid extraction & analysis

The experimental design is summarized in Table 1, Figure 2. In brief, “control” groups containing either 1, 4, or 6 individuals underwent a Bligh and Dyer extraction, followed by injection of µL of the extract into the UHPLC for analysis. Two types of “treatment” conditions were tested. Treatment A involved mechanical disruption using a bead-beating protocol and was applied to both individual samples (n = 1) and pooled samples (n = 6). Treatment B involved increasing the UHPLC injection volume to 10 µL and was applied only to individual samples.

Figure 2

Lipids were extracted using a modified Bligh and Dyer protocol adapted from Popendorf et al. (2013). Individual, partially thawed copepods were transferred from cryovials to the inner walls of pre-cleaned glass centrifuge tubes using a clean metal spatula. For Treatment A, approximately 2 mL of combusted glass beads (0.5mm diameter, QIAGEN, Germantown, MD, USA) were added to each tube. A solvent mixture of methanol, dichloromethane (DCM), and phosphate-buffered saline (PBS) in a 2:1:0.8 (v/v/v) ratio was added, along with 10 µL of internal standard (EquiSplash LIPIDOMIX^® Quantitative Mass Spec Internal Standard (Avanti Polar Lipids, Alabaster, AL, USA) diluted 2:1 from stock in methanol). Tubes were placed in an ultrasonic water bath for 10 minutes. Subsequently, additional DCM and PBS (1:1, v/v) were added to induce phase separation, followed by centrifugation at 1600 rpm for 10 minutes at 4 °C. The lower organic phase was collected, evaporated under a stream of nitrogen gas, and reconstituted in 70:30 acetonitrile:isopropanol (ACN: IPA, v/v). Extracts were stored at -80 °C until UHPLC-HRMS analysis.

The typical sample injection volume for reverse-phase UHPLC-MS/MS on a Vanquish UHPLC with an Accucore C8 column (155 mm x 2.1 mm x 2.6 µm) in tandem with an Orbitrap ID-X mass spectrometer (all from Thermo Scientific, San Jose, CA, USA) was 2uL. For Treatment B, the analytical run was performed with a 5-fold increase in injection volume (10 µL). We used a chromatographic method modified from Hummel et al (Hummel et al., 2011). to include an additional isocratic hold, optimized for wax ester separation based on preliminary analyses using wax ester standards (Figure 3, Supplementary Table 2). Briefly, eluent A was liquid chromatography-grade water with ammonium acetate and acetic acid as additives, and eluent B was 70:30 ACN: IPA with the same additives. The starting gradient was 45/55 (A/B), which held for 1 min before shifting to 1/99 over 17 minutes, followed by an isocratic hold at 1/99 for 13 minutes. During the first minute of the mass spectrometric method, early-eluting small compounds not retained on the column (also colloquially known as the solvent front) were diverted to waste to remove compounds too hydrophilic to be retained that elute at the start of the run. Compounds were then analyzed in positive mode at a resolution of 120,000. MS/MS was analyzed as data-dependent acquisition for compounds greater than 1E + 04 signal intensity. A mass inclusion list was used to acquire MS3 of compounds losing a free fatty acid fragment at the time that triacylglycerols (TAG) and wax esters elute (Kiyonami et al., 2016).

Figure 3

Compounds from all samples were putatively annotated using the LOBSTAHS platform and default lipidomics database. For comparative analysis, the same samples were also processed using MS-DIAL using the default lipidomics database and identical peak detection parameters. Raw UHPLC-MSⁿ data was converted to mzXML files for LOBSTAHS annotation and abf files for MS-DIAL. Only peaks above a threshold of 5E + 4 were picked in both analyses, with a 0.005ppm m/z threshold and a 0.2 min retention time threshold for peak grouping and library matching. The annotation workflow is detailed in the supplementary text of Hwang et al., 2025 (Hwang et al., 2025) and the computational load of workflows are compared in Supplementary Table 3.

Wax esters initially annotated in LOBSTAHS were further verified based on their characteristic “18-series” fragmentation pattern using MS-DIAL’s ability to query custom databases. This pattern involves the progressive loss of water molecules from the fatty acid moiety, transitioning from [RCOOH₂]⁺ to [RCO]⁺ and subsequently to [RCO-H₂O]⁺ (Figure 3; Chen et al., 2015). We created a custom fragmentation library (msp format) containing wax ester species composed of fatty acid and alcohol chains with the following parameters: (1) carbon chain lengths ranging from C₈ to C₃₀, (2) 0–8 double bonds, and (3) 0–4 oxygen modifications. In our assessment of wax ester annotations in our dataset, we focused on compounds containing 20:1 and 22:1 fatty alcohols, which are known to be synthesized de novo by polar copepods (Kattner and Hagen, 1995; Sargent et al., 1977). Additionally, a 36:5 wax ester was included because its unusual composition (8:0/28:5) nevertheless produced a clear 18-series fragmentation pattern. This made it a useful benchmark compound for testing the robustness of our custom fragmentation library across both typical and atypical wax ester species. The reduction of detected features to annotated and high-confidence compounds is summarized in Table 2, Supplementary Figure 1.

Statistics

All statistical analyses were performed in R v4.4.2 or Python 3.12.8. To assess treatment effects on individual wax ester compounds (Table 3), we used one-way analysis of variance (ANOVA) (aov() in R), followed by Tukey’s HSD post-hoc comparisons (TukeyHSD(), α = 0.05). Results are reported as mean peak area ± standard deviation (SD) and standard error (SE), and significance was accepted at p < 0.05. Assessment of treatment effects in overall wax ester abundance (Figure 4A) was conducted via hierarchical clustering using Euclidean distance and Ward’s D2 linkage (hclust(method = “ward.D2”)), which minimizes within-cluster variance and is well suited for quantitative lipidomic data (Murtagh and Legendre, 2014).

Figure 4

To evaluate whether pooling samples altered lipidomic profiles, we performed partial least squares–discriminant analysis (PLS-DA) in Python using scikit-learn (PLSRegression) (Figure 5). Peak-area matrices were log₁₀-transformed (zeros replaced with the minimum finite value) and mean-centered by feature. Model performance was assessed by leave-one-out cross-validation (LOOCV) using a logistic classifier trained on the two-component scores. We report classification accuracy, sensitivity, and specificity from out-of-fold predictions. Global model fit was summarized by R²X (variance in X captured), R²Y (variance in Y explained), and Q² (LOOCV predictive ability). Statistical significance of group separation was evaluated with a 1,000-permutation test, where the permutation p-value was defined as the fraction of shuffled models with Q² greater than or equal to the observed Q².

Figure 5

Results

Analysis of single copepods reveals robust recovery of lipids

In the control experiment, where non-bead-beat samples were run with 2 μL injection volumes, single-copepod extractions had detectable peak areas but many compounds were below the typical threshold that was used for peak detection in our peak-picking algorithms and programs (5E + 4; Figure 4). Increasing the injection volume 5-fold, from 2 μL to 10 μL (Treatment B), instead of lowering the peak detection threshold, produced higher signal intensity. We observed a non-linear increase in peak area, approximately 9-fold higher than the control value (Figure 4, Table 3). Under the same parameters, the peak-peaking algorithm XCMS discovered ~8000 peaks in the 10 μL injection samples, compared to ~4300 peaks found in the 2 μL injection samples, showcasing the effects of our signal amplification.

The 4-individual and 6-individual groups produced large, well-resolved peaks, reflecting sufficient analyte concentration and eliminating the need for further method optimization. Notably, the 6-individual control group exhibited an average peak area only ~3.6 times greater than that of the single-individual controls. PLS-DA analysis showed the wider spread of biological variability within individual samples compared to pooled (Figure 5). These results indicated that individual biochemical profiles were well preserved in our analytical workflow, while pooled samples exhibited more cohesive, averaged trends.

Glass beads greatly improve extraction efficiency

Comparing control and bead-beat (Treatment A) samples, pooled samples (n = 6) showed a 2-fold increase in peak area, while individual samples (n = 1) exhibited a greater effect, with a 2.9-fold increase (Figure 4). Extending sonication time would further improve yields for larger sample sets. Paired with the 1.8-fold increase from increased injection volume (Treatment B; volume-corrected), the combined effect of both treatments was estimated to yield a theoretical 3.6~5.2-fold increase in peak area, not accounting for ion suppression effects.

Differences in annotation output across pipelines

Compounds from all control and treatment samples were comparatively identified using two different annotation pipelines; the adduct-hierarchical LOBSTAHS and fragmentation-based MS-DIAL, which have distinct default lipid databases (Supplementary Figure 2). Overall, high confidence annotations within the copepod lipidome revealed a dominance of neutral lipids that mainly function as energy storage such as wax esters and triacylglycerols (TAG); lipids that participate in metabolism such as coprostanol esters and diacylglycerols (DAG); as well as phytoplankton-associated membrane lipids such as betaine-like lipids (BLL) and digalactosyldiacylglycerols (DGDG) (Figure 6). At the compound and lipid subclass level of the LOBSTAHS-annotated lipidome, wax esters and coprostanol esters were the most frequently detected lipid subclasses, whereas in MS-DIAL, TAG and DAG dominated the annotations (Figure 6A). Weighting the feature counts by peak area to reflect relative quantities of lipids did not alter the overall composition of the dominant lipid classes, but it did result in a modest reordering of their relative abundances (Figure 6B).

Figure 6

DAG, cholesterol esters, and wax esters were frequently co-annotated with ceramides and N-acyl ethanolamine (NAE) (Figure 7). Their simple elemental compositions, consisting primarily of carbon, hydrogen, and oxygen, and their propensity to form [M+NH₄]⁺ adducts contributed to frequent mis-annotation; ceramides and NAE contain a single nitrogen atom and preferentially form [M+H]⁺ adducts.

Figure 7

Discussion

Interpretation of extraction modifications

The increased injection volume of treatment B improved peak detectability without lowering peak-picking thresholds, which would have introduced issues such as lower-quality peaks, peaks with lower signal-to-noise ratios, and non-Gaussian peak shapes that ultimately complicate annotation further down the pipeline. Although higher injection volumes can introduce ion suppression effects, the non-linear increase in peak area suggests that signal amplification outweighed suppression under these conditions. However, we emphasize that the objectives for each alteration differ; injection volume increase is a method to amplify the sensitivity or resolution of measurements (that should not affect post-analysis corrected peak areas to a large degree), while bead-beating results in an “actual” increase in absolute amount of extracted compounds.

The narrower multivariate spreads observed in pooled samples compared to individuals reflects the reduced biological heterogeneity associated with sample averaging. Individual-level variation is expected for actively feeding, mobile meso-zooplankton, and the method captured this variability clearly.

As seen in Supplementary Table 1, Figure 1, this is not the first attempt to extract lipids from chitinous organisms. However, our method has the smallest footprint in resources such as time, extra equipment (i.e., cost efficiency and accessibility), and required sample size. Our method, inspired by bead-beating techniques commonly used in DNA extraction protocols, reduces overall processing time compared to existing protocols and can be integrated directly into the pre-existing lipid extraction workflow without the need for additional extraction steps. While ceramic beads are commonly used in molecular biology, we use glass beads for their proven compatibility with lipidomics work due to their non-reactivity with organics solvents and combustibility.

While homogenization using external equipment such as a Precellys24 bead homogenizer is effective, it requires quantitative transfer between two vessels and can therefore complicate full sample recovery. Other homogenizers that have been used for copepod analysis, such as Potter-Elvehjem homogenizers, are optimal for softer tissue samples and hence could require more initial biomass and/or longer agitation times to obtain an adequate concentration. Most other copepod lipid extraction methods are optimized for derivative analytical methods such as FAME (fatty acid methyl esterification), which differ in scope from our intact lipidome analysis.

Interpretation of annotation behavior

Overall, we showed that annotation is strongly biased depending on the chemoinformatic pipeline utilized. While the dominant lipid classes identified by each algorithm were broadly similar in functional category (energy storage as opposed to energy harvesting, structural, signaling), at the compound and lipid subclass level the annotated lipidomes varied considerably, resulting in distinct profiles that reflect differences in annotation method only.

In certain pipelines, structurally distinct lipids could be incorrectly classified with high annotation confidence, despite limited supporting fragmentation evidence. This likely explains the elevated number of DAG reported in the MS-DIAL-annotated dataset and underscores the need for integrating class-specific fragmentation patterns, such as the “18-series” for wax esters, into existing MS/MS-based annotation platforms to improve compound specificity.

Our results revealed significant divergence between the tools, necessitating a hybridized approach that combines elements from both platforms. However, reconciling output from disparate peak-picking algorithms and annotation strategies adds to the complexity and time required for data processing. While the widely used GNPS (Global Networks Public Social Molecular Networking), especially when paired with MZmine, exemplifies the value of community-curated spectral libraries and network-based annotation (Schmid et al., 2023; Wang et al., 2016), it is mainly focused on terrestrial and mammalian compounds, limiting our ability to comprehensively annotate untargeted marine lipidomes. We advocate for a unified, community-driven effort to develop a comprehensive, marine-specific spectral database, analogous to the centralized reference systems used in genomics (e.g., INSDC (International Nucleotide Sequence Database Collaboration)).

Ecological and methodological implications

Both meso-zooplankton and their associated lipids are critical not only to marine ecosystems but also to the global carbon cycle. Notably, Jónasdóttir et al., 2015 demonstrated that the seasonal lipid pump, the transport of carbon to the deep ocean via copepod lipids, rivals the particulate sinking flux at high latitudes. In that paper, they use measurements of copepod prosome length and their relationship to oil sac volume to estimate storage lipid content of C. finmarchicus. A reliable extraction and analysis protocol for intact copepod lipids would allow for a more detailed understanding of the molecular mechanisms driving this carbon sequestration process.

In particular, our approach enables direct detection of whole molecular species, offering compound-level insight into biological processes. Here, we show that the analysis of the intact lipidomes of individual copepods is not only achievable, but also can be easily integrated into pre-existing analysis pipelines, opening a new dimension in this area of research.

Importantly, the efficacy of this specific Bligh and Dyer variation has been well established in phytoplankton and microbial lipidomics (Aveiro et al., 2020; Becker et al., 2018; Bowman et al., 2021; Carriot et al., 2021; Diaz et al., 2021; Laber et al., 2018; Muratore et al., 2022; Popendorf et al., 2013; Rempfert et al., 2023). Expanding its application to mesozooplankton and higher trophic levels enables methodological integration for tracing trophodynamic interactions within marine ecosystems. For example, it may allow us to trace the movement of specific fatty acids from storage lipids in phytoplankton, such as TAG, into zooplankton biomass, including membrane lipids, and other functional lipid classes.

One current approach in zooplankton feeding studies involves gut pigment content analysis, using fluorometric measurements of chlorophyll a and pheophytin to estimate ingestion rates of phytoplankton (Cawley et al., 2021). While fluorometry is straightforward, it is potentially constrained in chemical breadth when used by itself. The addition of intact lipidomics could help quantify the full suite of ingested pigments, such as fucoxanthin, violaxanthin, astaxanthin which would allow to infer phytoplankton community composition of gut content, via tools such as CHEMTAX or phytoclass (Hayward et al., 2023; Mackey et al., 1996).

A major advantage of our method is its suitability for capturing biological heterogeneity. Populations at this trophic level and size scale exhibit high inter-individual heterogeneity due to active feeding and mobility (Seuront and Strutton, 2004). We demonstrate that our single-animal extractions capture this variability. In environmental surveys, where individual variability complicates source attribution, samples can be pooled to reduce standard deviations. Conversely, in controlled culture experiments with biological triplicates, single-copepod analyses preserve individual responses to experimental treatments, enabling more nuanced physiological insights.

Finally, from an analytical standpoint, we have identified key limitations in existing lipid annotation pipelines and implemented improvements to better resolve metabolically important compounds, wax esters. Our application of this protocol to both individual and pooled samples enables analyses across multiple biological scales, from capturing the individual-level variability in lab-based experiments to population- or community-scale patterns in environmental samples. This scalability also enables integrative studies across trophic levels, linking lipid metabolism to broader ecosystem functions. Ultimately, our high-resolution, compound-specific approach offers a powerful tool for elucidating the metabolic responses of chitin-bound organisms to environmental change.

Conclusion

We have presented a robust method for extracting lipids from individuals of trophically important chitin-bound meso-zooplankton, in which lipid metabolism plays a central role. The workflow yields high recovery at both individual and pooled scales and is compatible with existing extraction pipelines.

By identifying annotation biases and improving wax-ester specificity, the analytic workflow strengthens compound-level resolution for complex zooplankton lipidomes. Together, this method establishes a robust and efficient approach for intact lipid analysis in chitinous zooplankton.

References

• 1

  AarflotJ. M.SkjoldalH. R.DalpadadoP.Skern-MauritzenM. (2018). Contribution of Calanus species to the mesozooplankton biomass in the Barents Sea. ICES J. Mar. Sci.75, 2342–2354. doi: 10.1093/icesjms/fsx221

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  AlbersC. S.KattnerG.HagenW. (1996). The compositions of wax esters, triacylglycerols and phospholipids in Arctic and Antarctic copepods: evidence of energetic adaptations. Mar. Chem.55, 347–358. doi: 10.1016/S0304-4203(96)00059-X

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 3

  AveiroS. S.MeloT.FigueiredoA.DominguesP.PereiraH.MaiaI. B.et al. (2020). The polar lipidome of cultured emiliania huxleyi: A source of bioactive lipids with relevance for biotechnological applications. Biomolecules10, 1434. doi: 10.3390/biom10101434

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  BeckerK. W.CollinsJ. R.DurhamB. P.GroussmanR. D.WhiteA. E.FredricksH. F.et al. (2018). Daily changes in phytoplankton lipidomes reveal mechanisms of energy storage in the open ocean. Nat. Commun.9. doi: 10.1038/s41467-018-07346-z

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  BoothD. T.KiddellK. (2007). Temperature and the energetics of development in the house cricket (Acheta domesticus). J. Insect Physiol.53, 950–953. doi: 10.1016/j.jinsphys.2007.03.009

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  BowmanJ. S.Van MooyB. A. S.LowensteinD. P.FredricksH. F.HanselC. M.GastR.et al. (2021). Whole community metatranscriptomes and lipidomes reveal diverse responses among antarctic phytoplankton to changing ice conditions. Front. Mar. Sci.8. doi: 10.3389/fmars.2021.593566

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  CarriotN.PaixB.GreffS.ViguierB.BriandJ.-F.CulioliG. (2021). Integration of LC/MS-based molecular networking and classical phytochemical approach allows in-depth annotation of the metabolome of non-model organisms - The case study of the brown seaweed Taonia atomaria. Talanta225, 121925. doi: 10.1016/j.talanta.2020.121925

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 8

  CawleyG. F.DécimaM.MastA.PrairieJ. C. (2021). The effect of phytoplankton properties on the ingestion of marine snow by Calanus pacificus. J. Plankton Res.43, 957–973. doi: 10.1093/plankt/fbab074

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  Cequier-SánchezE.RodríguezC.RaveloÁ.G.ZárateR. (2008). Dichloromethane as a solvent for lipid extraction and assessment of lipid classes and fatty acids from samples of different natures. J. Agric. Food Chem.56, 4297–4303. doi: 10.1021/jf073471e

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  ChenJ.GreenK. B.NicholsK. K. (2015). Characterization of wax esters by electrospray ionization tandem mass spectrometry: double bond effect and unusual product ions. Lipids50, 821–836. doi: 10.1007/s11745-015-4044-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11

  DiazB. P.KnowlesB.JohnsC. T.LaberC. P.BondocK. G. V.HaramatyL.et al. (2021). Seasonal mixed layer depth shapes phytoplankton physiology, viral production, and accumulation in the North Atlantic. Nat. Commun.12, 6634. doi: 10.1038/s41467-021-26836-1

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 12

  HansenB. H.DegnesK.ØverjordetI. B.AltinD.StørsethT. R. (2013). Metabolic fingerprinting of arctic copepods Calanus finmarchicus, Calanus glacialis and Calanus hyperboreus. Polar Biol.36, 1577–1586. doi: 10.1007/s00300-013-1375-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13

  HaywardA.PinkertonM. H.Gutierrez-RodriguezA. (2023). phytoclass: A pigment-based chemotaxonomic method to determine the biomass of phytoplankton classes. Limnol Oceanogr Methods21, 220–241. doi: 10.1002/lom3.10541

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  HummelJ.SeguS.LiY.IrgangS.JueppnerJ.GiavaliscoP. (2011). Ultra performance liquid chromatography and high resolution mass spectrometry for the analysis of plant lipids. Front. Plant Sci.2. doi: 10.3389/fpls.2011.00054

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  HunterJ. E.FredricksH. F.BehrendtL.AlcolombriU.BentS. M.StockerR.et al. (2021). Using high-sensitivity lipidomics to assess microscale heterogeneity in oceanic sinking particles and single phytoplankton cells. Environ. Sci. Technol.55, 15456–15465. doi: 10.1021/acs.est.1c02836

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  HwangJ.HaywardA.SofenL. E.PitzK. J.ChavezF. P.EdwardsB. R. (2025). Daily microbial rhythms of the surface ocean interrupted by the new moon—a lipidomic study. ISME Commun.5, ycaf044. doi: 10.1093/ismeco/ycaf044

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17

  JónasdóttirS. H.VisserA. W.RichardsonK.HeathM. R. (2015). Seasonal copepod lipid pump promotes carbon sequestration in the deep North Atlantic. Proc. Natl. Acad. Sci.112, 12122–12126. doi: 10.1073/pnas.1512110112

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18

  KattnerG.HagenW. (1995). Polar herbivorous copepods - different pathways in lipid biosynthesis. ICES J. Mar. Sci.52, 329–335. doi: 10.1016/1054-3139(95)80048-4

  • CrossRef
  • Google Scholar

• 19

  KindT.LiuK.-H.LeeD. Y.DeFeliceB.MeissenJ. K.FiehnO. (2013). LipidBlast in silico tandem mass spectrometry database for lipid identification. Nat. Methods10, 755–758. doi: 10.1038/nmeth.2551

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20

  KiyonamiR.PeakeD. A.YokoiY.MillerK. (2016). Increased throughput and confidence for lipidomics profiling using comprehensive HCD MS2 and CID MS2/MS3 on a tribrid orbitrap mass spectrometer. Thermo Fisher Scientific. 10, 648.

  • Google Scholar

• 21

  LaberC. P.HunterJ. E.CarvalhoF.CollinsJ. R.HunterE. J.SchielerB. M.et al. (2018). Coccolithovirus facilitation of carbon export in the North Atlantic. Nat. Microbiol.3, 537–547. doi: 10.1038/s41564-018-0128-4

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22

  LeeJ.TantilloD. J.WangL.-P.FiehnO. (2024). Predicting collision-induced-dissociation tandem mass spectra (CID-MS/MS) using ab initio molecular dynamics. J. Chem. Inf. Model.64, 7470–7487. doi: 10.1021/acs.jcim.4c00760

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23

  LundyM. E.ParrellaM. P. (2015). Crickets Are Not a Free Lunch: Protein Capture from Scalable Organic Side-Streams via High-Density Populations of Acheta domesticus. PloS One10, e0118785. doi: 10.1371/journal.pone.0118785

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  MackeyM. D.MackeyD. J.HigginsH. W.WrightS. W. (1996). CHEMTAX - A program for estimating class abundances from chemical markers: Application to HPLC measurements of phytoplankton. Mar. Ecol. Prog. Ser.144, 265–283. doi: 10.3354/meps144265

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  MatyashV.LiebischG.KurzchaliaT. V.ShevchenkoA.SchwudkeD. (2008). Lipid extraction by methyl-tert-butyl ether for high-throughput lipidomics. J. Lipid Res.49, 1137–1146. doi: 10.1194/jlr.D700041-JLR200

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26

  MayorD. J.SommerU.CookK. B.ViantM. R. (2015). The metabolic response of marine copepods to environmental warming and ocean acidification in the absence of food. Sci. Rep.5, 13690. doi: 10.1038/srep13690

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  MuratoreD.BoysenA. K.HarkeM. J.BeckerK. W.CaseyJ. R.CoeselS. N.et al. (2022). Complex marine microbial communities partition metabolism of scarce resources over the diel cycle. Nat. Ecol. Evol.6, 218–229. doi: 10.1038/s41559-021-01606-w

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 28

  MurtaghF.LegendreP. (2014). Ward’s hierarchical agglomerative clustering method: which algorithms implement ward’s criterion? J. Classif31, 274–295. doi: 10.1007/s00357-014-9161-z

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29

  PopendorfK. J.FredricksH. F.Van MooyB. A. S. (2013). Molecular ion-independent quantification of polar glycerolipid classes in marine plankton using triple quadrupole MS. Lipids48, 185–195. doi: 10.1007/s11745-012-3748-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  RempfertK. R.KrausE. A.NothaftD. B.DildarN.SpearJ. R.SepúlvedaJ.et al. (2023). Intact polar lipidome and membrane adaptations of microbial communities inhabiting serpentinite-hosted fluids. Front. Microbiol.14. doi: 10.3389/fmicb.2023.1198786

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  RinaudoM. (2006). Chitin and chitosan: Properties and applications. Prog. Polym Sci.31, 603–632. doi: 10.1016/j.progpolymsci.2006.06.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32

  SargentJ. R.GattenR. R.McIntoshR. (1977). Wax esters in the marine environment - their occurrence, formation, transformation and ultimate fates. Mar. Chem.5, 573–584. doi: 10.1016/0304-4203(77)90043-3

  • CrossRef
  • Google Scholar

• 33

  SchmidR.HeuckerothS.KorfA.SmirnovA.MyersO.DyrlundT. S.et al. (2023). Integrative analysis of multimodal mass spectrometry data in MZmine 3. Nat. Biotechnol.41, 447–449. doi: 10.1038/s41587-023-01690-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34

  SeurontL.StruttonP. G. (2004). Using Multiagent Systems to Develop Individual-Based Models for Copepods: Consequences of Individual Behavior and Spatial Heterogeneity on the Emerging Properties at the Population Scale, in: Handbook of Scaling Methods in Aquatic Ecology (United Kingdom: CRC Press), 543–562. doi: 10.1201/9780203489550-42

  • CrossRef
  • Google Scholar

• 35

  SieburthJ.SmetacekV.LenzJ. (1978). Pelagic ecosystem structure: Heterotrophic compartments of the plankton and their relationship to plankton size fractions. Limnol Oceanogr23, 1256–1263. doi: 10.4319/lo.1978.23.6.1256

  • CrossRef
  • Google Scholar

• 36

  SturtH. F.SummonsR. E.SmithK.ElvertM.HinrichsK. U. (2004). Intact polar membrane lipids in prokaryotes and sediments deciphered by high-performance liquid chromatography/electrospray ionization multistage mass spectrometry - New biomarkers for biogeochemistry and microbial ecology. Rapid Commun. Mass Spectrom18, 617–628. doi: 10.1002/rcm.1378

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37

  SumnerL. W.AmbergA.BarrettD.BealeM. H.BegerR.DaykinC. A.et al. (2007). Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI). Metabolomic Off. J. Metabolomic Soc.3, 211–221. doi: 10.1007/s11306-007-0082-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38

  TsugawaH.IkedaK.TakahashiM.SatohA.MoriY.UchinoH.et al. (2020). A lipidome atlas in MS-DIAL 4. Nat. Biotechnol.38, 1159–1163. doi: 10.1038/s41587-020-0531-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  TsugawaH.IkedaK.TanakaW.SenooY.AritaM.AritaM. (2017). Comprehensive identification of sphingolipid species by in silico retention time and tandem mass spectral library. J. Cheminform9, 19. doi: 10.1186/s13321-017-0205-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40

  Van MooyB. A. S.FredricksH. F. (2010). Bacterial and eukaryotic intact polar lipids in the eastern subtropical South Pacific: Water-column distribution, planktonic sources, and fatty acid composition. Geochim Cosmochim Acta74, 6499–6516. doi: 10.1016/j.gca.2010.08.026

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  Van MooyB. A. S.MoutinT.DuhamelS.RimmelinP.Van WambekeF. (2008). Phospholipid synthesis rates in the eastern subtropical South Pacific Ocean. Biogeosciences5, 133–139. doi: 10.5194/bg-5-133-2008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  WangM.CarverJ. J.PhelanV. V.SanchezL. M.GargN.PengY.et al. (2016). Sharing and community curation of mass spectrometry data with Global Natural Products Social Molecular Networking. Nat. Biotechnol.34, 828–837. doi: 10.1038/nbt.3597

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  WangZ.CaoM.LamS. M.ShuiG. (2023). Embracing lipidomics at single-cell resolution: Promises and pitfalls. TrAC Trends Anal. Chem.160, 116973. doi: 10.1016/j.trac.2023.116973

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  WoodP. L.WoodM. D.KunigelisS. C. (2023). Pilot lipidomics study of copepods: investigation of potential lipid-based biomarkers for the early detection and quantification of the biological effects of climate change on the oceanic food chain. Life13, 2335. doi: 10.3390/life13122335

  • Pubmed Abstract
  • CrossRef
  • Google Scholar