Constraining the oxygen requirements for modern microbial eukaryote diversity
Edited by Paul Hoffman, University of Victoria, Victoria, Canada; received March 6, 2023; accepted November 7, 2023
Significance
Modern eukaryotes were present, but marginalized, in ecosystems until no later than ~800 Ma or so, when the biosphere began transitioning to a eukaryote-rich state. Why eukaryotes diversified around this time in Earth history remains unclear. For decades, the most popular proposed driver for this early diversification of eukaryotes has been a contemporaneous increase in atmospheric oxygen (O2). The oxygen levels required to drive such a diversification event, however, remain unknown. To empirically constrain these requirements, we explored the diversity of microbial eukaryotes living along a modern marine oxygen gradient. Our results suggest that only a rise in atmospheric oxygen from below to above 2 to 3% of modern levels would have directly driven the early diversification of modern microbial eukaryotes.
Abstract
Eukaryotes originated prior to the establishment of modern marine oxygen (O2) levels. According to the body fossil and lipid biomarker records, modern (crown) microbial eukaryote lineages began diversifying in the ocean no later than ~800 Ma. While it has long been predicted that increasing atmospheric O2 levels facilitated the early diversification of microbial eukaryotes, the O2 levels needed to permit this diversification remain unconstrained. Using time-resolved geochemical parameter and gene sequence information from a model marine oxygen minimum zone spanning a range of dissolved O2 levels and redox states, we show that microbial eukaryote taxonomic richness and phylogenetic diversity remain the same until O2 declines to around 2 to 3% of present atmospheric levels, below which these diversity metrics become significantly reduced. Our observations suggest that increasing O2 would have only directly promoted early crown-eukaryote diversity if atmospheric O2 was below 2 to 3% of modern levels when crown-eukaryotes originated and then later met or surpassed this range as crown-eukaryotes diversified. If atmospheric O2 was already consistently at or above 2 to 3% of modern levels by the time that crown-eukaryotes originated, then the subsequent diversification of modern microbial eukaryotes was not directly driven by atmospheric oxygenation.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
There is a temporal gap, spanning several hundred million years, separating the earliest eukaryotic fossils from the diversification of eukaryotic fossils with definitively modern traits (1–3) (Fig. 1). While the oldest reputed eukaryotic-grade fossils date to ca. 1.65 billion years ago (Ga) (3)—potentially reflecting stem-eukaryote lineages that diverged prior to the last eukaryote common ancestor, or LECA (4–6)—microbial body fossils and certain molecular clocks predict that modern (that is, crown-) eukaryote lineages began increasing in taxonomic richness (1, 3, 7) and phylogenetic diversity (4, 8–10) no later than ca. 800 Ma (Fig. 1). This latter proposed diversification coincides with the appearance and increase of regular (4-desmethyl) sedimentary steranes, which serve as molecular fossils for extant sterol-producing eukaryote lineages (5, 11, 12) (Fig. 1). Why so much time—that is, ca. 800 million years—separates the earliest eukaryotic fossils from the apparent diversification of crown-eukaryotes remains unclear. Indeed, numerous explanations for this so-called “macroevolutionary lag” have been proposed, from marine phosphorus (11, 13) and nitrate limitation (14) to the absence of animal bodyplans and feeding strategies (15). The most widely cited and hotly debated driver for early eukaryote diversification, however, is increasing marine O2 availability (2, 3, 7, 16–18).
Fig. 1.
The eukaryote crown-group, and therefore the modern eukaryotic cell, most likely emerged by the middle of the Proterozoic Eon (2.5 to 0.539 Ga) (2–4, 19) (Fig. 1). Over the past two decades, it has become increasingly clear that the ocean during the middle of the Proterozoic was only moderately oxic at the surface and predominantly anoxic at depth (29). Furthermore, modern atmospheric O2 levels (pO2) post-date the origin of the eukaryotic cell and were only achieved and sustained relatively recently in geologic time (after 82 to 90% of Earth’s history had elapsed) (19, 25). Debate continues, however, as to whether pO2 was at least 1 to 10% of PAL (17, 22–24, 30) or at most <1% of PAL (16, 18, 26, 27) during the middle Proterozoic (Fig. 1). While there remains no consensus on how low pO2 could have been during this interval, the deep oceans would have remained primarily anoxic across the full range of proposed pO2 levels (29, 31). Indeed, what separates these scenarios is the amount of dissolved O2—both absolutely and temporally—that would have been available to early eukaryotes living in surface waters in direct exchange with the atmosphere.
The question of what controlled the diversification of eukaryotes by ca. 800 Ma has been primarily approached by the tools that revealed this environmental-evolutionary pattern in the first place—that is, correlations between paleoredox reconstructions using geochemical data and numerical modeling (25) and descriptions of the diversity and abundance of eukaryote body and lipid fossils through time (3, 12). As a result, the mechanistic control of O2 availability on microbial eukaryote diversity over geologic time has remained conspicuously unconstrained in these discussions, as additional lines of evidence, mainly modern ecological observations, are required to do so. At the same time, information on how marine O2 levels shape microbial eukaryote diversity in the modern biosphere remains limited (32). Given these knowledge gaps and lack of empirical constraints, we explored the diversity of modern microbial eukaryotes living along the spatial and temporal O2 gradients of Saanich Inlet, British Columbia, Canada.
Saanich Inlet, located on the southern end of Vancouver Island, is a persistently stratified marine fjord connected to Haro Strait via the Satellite and Swanson Channels. An extended shallow-water (~70 m) sill at the mouth of Saanich Inlet, however, separates Saanich Inlet’s deeper waters (maximum depth of ~200 m) from Haro Strait, thereby restricting circulation and promoting bottom water anoxia in Saanich Inlet as high rates of primary productivity and aerobic respiration consume O2. Episodic renewal and oxygenation of Saanich Inlet’s deep basin occurs via influxes of dense, O2-rich, saline water from Haro Strait in response to tidal current and regional upwelling dynamics. Over many decades of research, Saanich Inlet has become a model ecosystem for studying microbial community responses to ocean deoxygenation with exceptional access to geochemical parameter and multi-omic (DNA, RNA, and protein) sequence information supporting time-resolved comparative analyses (33–37). Here, we explored the diversity of planktonic microbial eukaryotes living along the O2 gradients within Saanich Inlet’s water column (Fig. 2). To achieve this goal, we analyzed eukaryote small subunit ribosomal genes (that is, 18S rRNA genes, the DNA sequences encoding the ribosomal RNA that comprises the small subunit of eukaryotic ribosomes) collected from six oxycline-spanning water column depths (10, 100, 120, 135, 150, and 200 m) once a month for the entirety of the year 2015 (Fig. 2).
Fig. 2.
Results
From the six depths sampled monthly throughout 2015, we identified 6,347 amplicon sequence variants (ASVs, functional taxonomic designations equivalent to species or strains), spanning every major eukaryote clade in the PR2 database (40) (Fig. 2). Consistent with previous time-series observations at Saanich Inlet using alternative identification protocols (38, 41), each sampled depth was dominated by sequences associated with the Alveolata and Stramenopiles (Fig. 2), members of the TSAR supergroup (42). Alveolata sequences featured the highest relative abundance (defined here as percent of total 18S rRNA gene sequences) at each depth, peaking at 100 m with an average relative abundance of 68.6%, while Stramenopiles peaked at 135 m with an average relative abundance of 35.5% (Fig. 2). Alveolata sequences had their lowest relative abundance (46.1%) at 200 m, where the relative abundances of the Rhizaria, Opisthokonta, and Excavata all peaked (Fig. 2). To assess the possible contribution of sinking cells to these vertical distribution patterns, we calculated the percent overlap between ASVs at a given depth with those at the depth sampled immediately above, revealing an average overlap of 2.2% with a maximum overlap of 9.5%, showing that the majority of ASVs observed at any given depth are not sourced from the depth sampled immediately above it in the water column (SI Appendix, Tables S4–S6). Overall, in agreement with previous observations from the Saanich-Inlet time series (38, 41), as well as surveys from other stratified marine water columns (32, 43–45), the composition of microbial eukaryote communities at Saanich Inlet clearly shifts as O2 declines with depth and through time (Fig. 2).
To explore changes in alpha (or within-sample) diversity as a function of O2 availability, we applied two commonly used alpha diversity metrics to our samples: 1) observed features (or observed ASVs in this case), and 2) Faith’s phylogenetic diversity (Faith’s PD). Observed ASVs is the total number of ASVs recovered from each sample and is therefore a direct measure of richness. Faith’s PD, unlike observed ASVs, incorporates phylogeny by summing the lengths of the phylogenetic branches uniting the ASVs from a given sample and is therefore expressed in terms of total minimum branch length (46). In other words, Faith’s PD captures the phylogenetic breadth of a sample (that is, how much of the phylogenetic tree a sample covers), with greater branch sums reflecting greater phylogenetic coverage and greater diversity. For each sample, we plotted observed ASVs and Faith’s PD as a function of O2 concentration, revealing similar nonlinear correlations for both metrics (Fig. 3). To explore inflection points for each metric, we ran regression trees (also called breakpoint analyses), which represent a nonparametric machine learning technique that can be used to identify the environmental variable (and the value of that variable) that best divides a dataset into high versus low levels of diversity (47, 48). Statistically, this identification is done by recursively partitioning the dataset based on an exhaustive search across all variables—in this case, both O2 and H2S—for the split that maximizes the homogeneity of the two resulting subsets. These analyses are thus a non-biased test of inflection points in our diversity trends. For both diversity metrics, O2 (not H2S) was determined to be the variable that best split the diversity data, specifically at 2.7% PAL (Fig. 3 and SI Appendix, Figs. S1 and S2). We also identified high-O2, but low-diversity, outliers in each dataset as representing potential dinoflagellate blooms at 10-m depth in May and June of 2015, where dinoflagellate sequences comprised up to 89% of all total 18S rRNA sequences (Fig. 3).
Fig. 3.
For the purposes of statistical testing, we divided our observed ASVs and Faith’s PD values into the five following redox bins (all expressed in terms of percent PAL O2): 1) >10%; 2) 2.7 to 10%; 3) 1 to 2.7%; 4) <1%; and 5) >1 µM H2S (see SI Appendix, Table S3 for unit conversions). These redox bins were determined primarily by geological considerations—namely the two competing views of mid-Proterozoic pO2 (Fig. 1). However, we divided the traditional 1 to 10% PAL O2 range for the mid-Proterozoic into two distinct bins based on the 2.7% PAL O2 best split recovered from our regression tree analyses. It must be noted that the <1% PAL O2 and the >1 µM H2S bins include samples collected at O2 concentrations below the detection limit of our O2 sensor (Materials and Methods), including samples likely collected under anoxia, especially within the sulfidic (>1 µM H2S) bin. Furthermore, using the results of our breakpoint analyses to define one bin boundary will maximize the chances of yielding statistically significant results for the two bins adjacent to that boundary (but not for higher or lower O2 bins). Indeed, for both observed ASVs and Faith’s PD, all bins below 2.7% PAL O2 differ significantly (Q < 0.05) from each bin above 2.7% PAL O2 and do not differ significantly from each other (Benjamini–Hochberg Q-value, based on P-values from Kruskal–Wallis tests) (Fig. 4 and SI Appendix, Tables S7 and S8). Binning our diversity metrics into three alternative redox bins that are not based on the breakpoint analysis (but again expressed in terms of percent PAL O2)—1) >10%; 2) 1 to 10%; and 3) <1%—yields similar results: The first two bins only differ significantly from the third, when O2 is below 1% PAL (SI Appendix, Fig. S3 and Tables S9 and S10).
Fig. 4.
Oxygen Availability and Microbial Eukaryote Diversity.
Our two diversity metrics—observed ASVs (richness) and Faith’s PD—reveal similar nonlinear relationships with dissolved O2 availability (Fig. 3). Regression tree (or breakpoint) analyses identified a best split, maximizing the homogeneity of the two resulting subsets, of 2.7% PAL O2 (Fig. 3). By binning these diversity data into discrete redox bins, we consistently find that samples collected below the equivalent of 2.7% PAL O2 are significantly (Q < 0.05) lower than the high-O2 bins (>10% and 2.7 to 10% PAL O2), which, in turn, do not significantly differ from each other (Fig. 4). Importantly, looking at the maximum diversity estimates from each bin, relatively high levels of both richness and Faith’s PD are still recovered below our best split of 2.7% PAL O2, although the medians and minimum values of these low-O2 bins are all lower (Fig. 4). Why these diversity metrics shift as O2 declines below 2.7% PAL O2 is unclear. Historically, 1% PAL O2 is treated as the equivalent of the “Pasteur point,” the O2 tension below which facultative aerobes switch to anaerobic energy metabolisms (49). Generally, it is appreciated that the growth efficiency of anaerobic fermenters is 25% of that of aerobes and that anaerobic food chains are necessarily shorter than aerobic ones on bioenergetic grounds (50). While aerobic respiration demonstrably occurs—at least in bacteria—under nanomolar O2 concentrations (51, 52), and therefore well below 2.7% PAL O2, some of the likely anaerobic (and therefore less diverse) communities sampled from the <1% PAL O2 and >1 µM H2S bins may necessarily decrease the overall diversity levels of these bins. The observation that some of the samples collected under <1% PAL O2 and >1 µM H2S still exhibit relatively high levels of Faith’s PD (as seen by the outliers and maximum values) agrees with the conclusion that the specialization to consistently anoxic and euxinic settings is phylogenetically widespread across the eukaryote tree and is not constrained to a few isolated lineages (53) (Figs. 3 and 4). Concurrently, we recovered relatively low-diversity samples from high-O2 settings, although these samples were identified as likely capturing dinoflagellate blooms at 10-m depth in May and June of 2015, where the relative abundance of dinoflagellate sequences ranged from 60 to 89%, thereby driving both diversity metrics down (Fig. 3). Generally, we recovered a positive nonlinear relationship between microbial eukaryote diversity and O2, with significantly lower diversity levels occurring below the calculated inflection point of 2.7% PAL O2 (Fig. 3).
From a Modern Fjord to the Mid-Proterozoic Marine Biosphere.
Recognizing that breakpoint analyses can be influenced to some extent by the specifics of the dataset (47), we generalize our results for the discussion of the Proterozoic biosphere to the fact that Saanich Inlet microbial diversity rises substantially at around 2 to 3% PAL O2. While we interpret our results from Saanich Inlet as capturing fundamental ecological relationships between O2 availability and microbial eukaryote diversity that can be applied to both the deep past and near future (50), it needs to be stressed that Saanich Inlet is not a “mid-Proterozoic ocean analogue,” and that non-uniformitarian variables, such as nutrient availability and animal predation, also contribute to modern levels of microbial eukaryote diversity (see SI Appendix for discussion and details). However, as an established model system for exploring the control of dissolved O2 availability on microbial community structure (33–37), Saanich Inlet allows us to isolate the control of dissolved O2 on microbial eukaryote diversity for the purposes of testing hypotheses of how ancient O2 levels would have directly impacted early eukaryote diversity. While previous studies also focused on microbial eukaryote diversity in Saanich Inlet (38, 41), as well as other O2-deficient marine water columns (32, 43–45, 54), here we use gene diversity data collected along a modern marine O2 gradient to explicitly reconstruct the diversity of microbial eukaryotes living under the range of O2 levels proposed for the Proterozoic atmosphere.
Proterozoic Oxygen Levels and Crown-Eukaryote Diversity.
Crown-group eukaryotes apparently diversified in terms of both richness and phylogenetic diversity no later than 800 Ma, several hundreds of millions of years after the eukaryote lineage entered the fossil record (Fig. 1). If a rise in pO2 facilitated this diversification of eukaryotes, a corollary is that low pO2 values prior to this diversification must have kept early eukaryote diversity low. Minimum pO2 estimates for the mid-Proterozoic remain debated and have not even been reliably constrained within an order of magnitude (17, 18) (Fig. 1). Despite this uncertainty, if mid-Proterozoic pO2 was below 1% PAL, as some argue (16, 18, 25, 27), then our diversity estimates suggest that microbial eukaryote taxonomic richness and phylogenetic diversity would have both been significantly reduced relative to the potential diversity above 2 to 3% PAL O2 (Fig. 4). Likewise, if pO2 rose from <1% PAL to ≥2 to 3% PAL, then a significant increase in both richness and phylogenetic diversity would be predicted. Indeed, the Fe isotope composition of Proterozoic ironstones (18) has been recently interpreted to signal a rise in pO2 from levels below 1% PAL between 900 and 750 Ma to levels greater than 1%, in tandem with paleontological evidence for eukaryote diversification (Fig. 1). Similarly, if mid-Proterozoic pO2 levels were consistently between 1 and 2% PAL, which has also been argued (24), then eukaryote diversity would have also been significantly reduced until pO2 levels reached or exceeded 2 to 3% PAL O2. As an alternative to these scenarios, if mid-Proterozoic pO2 was at least 4 to 8% PAL (22, 30), or perhaps even 24% PAL (17), then our results suggest that modern levels of eukaryote richness and phylogenetic diversity were already permitted—at least by the O2 content of the atmosphere—well before the diversification of eukaryotes no later than ca. 800 Ma (55). While the pO2 content of the mid-Proterozoic atmosphere remains to be definitively determined, our results provide the empirical constraints necessary for interpreting how these different pO2 estimates would have directly impacted—or not impacted—microbial eukaryote diversification in the Proterozoic Eon.
Conclusions
Overall, our results predict that if pO2 levels in the mid-Proterozoic were at least 2 to 3% PAL O2 (Scenario 1, Fig. 1), and that crown-eukaryotes were already living in air-saturated surface environments at this time, then early eukaryote taxonomic richness and phylogenetic diversity were not directly restricted by O2 availability (17, 22, 30). Accordingly, the early diversification of the eukaryote crown-group, which happened no later than 800 Ma or so (Fig. 1), could not have been directly driven by rising pO2. However, if the mid-Proterozoic atmosphere contained <2 to 3% PAL O2 (18, 24–27) (Scenario 2, Fig. 1), and crown-eukaryotes already lived in air-saturated waters at this time, then a late-Proterozoic pO2 rise to ≥2 to 3% PAL O2 would have indeed permitted crown-eukaryote diversification with respect to both taxonomic richness and phylogenetic diversity. Discerning between these possibilities requires constraining the O2 concentrations that the earliest crown-eukaryotes lived under following their emergence in the Proterozoic Eon.
Materials and Methods
Sampling and Collection.
Saanich Inlet is a persistently anoxic fjord on the southeast coast of Vancouver Island, British Columbia, Canada. Between January and December 2015, as part of a historic 10-year time series experiment, six depths (10, 100, 120, 135, 150, and 200 m) from Station 3 (48°35.5 N, 123°30.3 W, 227 m deep) were sampled each month for microbial community profiling. 10 L of water from each depth were prefiltered through 2.7 µm glass fiber filters onto 0.22 µm Sterivex (Millipore) filters. Filtered biomass was soaked in lysis buffer then frozen immediately in liquid nitrogen. Filters were stored at −80 °C until further analysis.
For geochemical measurements, 16 depths (10, 20, 40, 60, 75, 85, 90, 97, 100, 110, 120, 135, 150, 165, 185, and 200 m) from Station 3 were sampled monthly using a series of 12L GO-FLO bottles, as described by Michiels et al. (37). Briefly, CTD profiles [pressure (SBE 29), conductivity (SBE 4C), temperature (SBE 3F), and oxygen (SBE 43)] were generated using the SBE25 Sealogger CTD (SBE). The SBE 43 has an O2 detection limit of 1 to 2 µM (56). Oxygen concentrations expressed as percentage of air saturation (100% atmospheric saturation = 100% of PAL) were calculated using the recorded temperature and salinity associated with each oxygen measurement (see SI Appendix, Table S1 for original measurements and conversions to percent PAL O2).
Microbial Community Profiling.
DNA was extracted from both the prefilters and Sterivex filters in accordance with Wright et al. (57). Extracted DNA was quantified using the picogreen assay (Invitrogen). DNA from Sterivex samples was sent to the Joint Genome Institute for 18S rRNA amplicon sequencing on the Illumina MiSeq platform1, using universal primers targeting the v4 region 565F CCAGCASCYGCGGTAATTCC and 948R ACTTTCGTTCTTGATYRA (58) of the 18S rRNA gene (59). DNA from prefilters was sent to the Integrated Microbiome Resource (IMR) center at the University of Dalhousie and amplified with 565F CCAGCASCYGCGGTAATTCC and a modified 948R primer ACTTTCGTTCTTGATYRATGA (60).
Sequences from prefilter and Sterivex filters were processed individually using the Quantitative Insights Into Microbial Ecology 2 (QIIME 2) software package (61). Denoising quality, chimera check, and clustering were performed using the Divisive Amplicon Denoising Algorithm 2 (DADA2) plugin tool and denoise-paired instruction (62). Prefilter and Sterivex filter datasets were merged using the QIIME2 VSEARCH plugin, which clustered ASVs at 100% identity (63). For taxonomic assignment of the resulting ASVs, a Naïve Bayes classifier was pre-trained on the PR2 database (version 14.2.0) (40), and ASVs were assigned using the QIIME2 classify-sklearn plugin (https://docs.qiime2.org/2022.2/data-resources/). All taxonomic data were visualized in R and Excel. For alpha (that is, “within-sample”) diversity measures, all samples were subsampled to the lowest coverage depth, and standard indices (observed features, Faith’s phylogenetic diversity) were calculated and visualized in QIIME2. To isolate the control of dissolved O2 concentrations on the diversity of microbial eukaryotes, 18S sequences assigned to the following complex multicellular lineages (64) were systematically excluded from our analyses: Metazoa, Streptophyta, Florideophyceae, Phaeophyceae, Ascomycota, and Basidiomycota.
Statistical Analyses.
To test for differences in observed ASVs and Faith’s phylogenetic diversity as a function of redox regime, QIIME2 performed Kruskal–Wallis tests, a nonparametric ANOVA (65) on all redox bins (K = 5), as well as on two groups at a time (K = 2) to identify differences between individual redox regimes. The false discovery rate (FDR) method of Benjamini and Hochberg (66) was then used to correct for multiple comparisons in the pairwise tests. All results are presented in SI Appendix, Tables S7–S10. Regression tree analyses were run using the R package rpart with O2 and H2S levels included as predictor variables.
Data, Materials, and Software Availability
18S rRNA amplicon data have been deposited in BioProject (PRJNA1008712) (67). All other data are included in the manuscript and/or SI Appendix.
Acknowledgments
Daniel Brady Mills was supported by the Agouron Institute Geobiology Postdoctoral Fellowship Program (#AI-F-GB53.19.2) and the Deutsche Forschungsgemeinschaft (DFG project OR 417/7-1). This work was also supported by the Natural Sciences and Engineering Research Council discovery grant 04867 awarded to Sean Crowe. Sequencing was performed and funded by the United States Department of Energy Joint Genome Institute. Daniel Brady Mills acknowledges feedback from Ömer Coskun, Virginia Edgcomb, Warren Francis, and William Orsi, assistance with bioinformatics from Jenifer Spence, and support from William Orsi. We also thank all individuals involved in the sampling and processing of Saanich Inlet samples during the year 2015, including the crew of the Marine Science Vessel Strickland, Chris Payne and Lora Pakhomova, members of the Crowe and Hallam Labs, and all those who assisted before and after sampling days.
Author contributions
D.B.M., S.J.H., E.A.S., and S.A.C. designed research; D.B.M. and R.L.S. performed research; R.L.S., T.R.S., S.J.H., E.A.S., and S.A.C. contributed new reagents/analytic tools; D.B.M., R.L.S., T.R.S., E.A.S., and S.A.C. analyzed data; and D.B.M., R.L.S., T.R.S., S.J.H., E.A.S., and S.A.C. wrote the paper.
Competing interests
The authors declare no competing interest.
Supporting Information
Appendix 01 (PDF)
- Download
- 1.27 MB
References
1
L. A. Riedman, P. M. Sadler, Global species richness record and biostratigraphic potential of early to middle Neoproterozoic eukaryote fossils. Precambrian Res. 319, 6–18 (2018).
2
P. A. Cohen, R. B. Kodner, The earliest history of eukaryotic life: uncovering an evolutionary story through the integration of biological and geological data. Trends Ecol. Evol. 37, 246–256 (2021), https://doi.org/10.1016/j.tree.2021.11.005.
3
H. Agić, “Origin and early evolution of the eukaryotes: Perspectives from the fossil record” in Prebiotic Chemistry and the Origin of Life, A. Neubeck, S. McMahon, Eds. (Springer International Publishing, 2021), pp. 255–289.
4
S. M. Porter, Insights into eukaryogenesis from the fossil record. Interface Focus 10, 20190105 (2020).
5
J. J. Brocks et al., Lost world of complex life and the late rise of the eukaryotic crown. Nature 618, 767–773 (2023).
6
S. M. Porter, L. A. Riedman, Frameworks for interpreting the early fossil record of eukaryotes. Annu. Rev. Microbiol. 77, 173–191 (2023).
7
P. A. Cohen, F. A. Macdonald, The Proterozoic record of eukaryotes. Paleobiology 41, 610–632 (2015).
8
L. W. Parfrey, D. J. G. Lahr, A. H. Knoll, L. A. Katz, Estimating the timing of early eukaryotic diversification with multigene molecular clocks. Proc. Natl. Acad. Sci. U.S.A. 108, 13624–13629 (2011).
9
L. Eme, S. C. Sharpe, M. W. Brown, A. J. Roger, “On the age of eukaryotes: Evaluating evidence from fossils and molecular clocks” in The Origin and Evolution of Eukaryotes, P. J. Keeling, E. V. Koonin, Eds. (Cold Spring Harbor Perspectives in Biology, 2014), pp. 165–180.
10
H. C. Betts et al., Integrated genomic and fossil evidence illuminates life’s early evolution and eukaryote origin. Nat. Ecol. Evol. 2, 1556–1562 (2018).
11
J. J. Brocks et al., The rise of algae in Cryogenian oceans and the emergence of animals. Nature 548, 578–581 (2017).
12
G. D. Love, J. Alex Zumberge, “Emerging patterns in Proterozoic lipid biomarker records” in Elements in Geochemical Tracers in Earth System Science (Cambridge University Press, 2021).
13
C. T. Reinhard et al., The impact of marine nutrient abundance on early eukaryotic ecosystems. Geobiology 18, 139–151 (2020).
14
J. Kang, B. Gill, R. Reid, F. Zhang, S. Xiao, Nitrate limitation in early Neoproterozoic oceans delayed the ecological rise of eukaryotes. Sci. Adv. 9, eade9647 (2023).
15
N. J. Butterfield, Early evolution of the Eukaryota. Palaeontology 58, 5–17 (2015).
16
E. J. Bellefroid et al., Constraints on Paleoproterozoic atmospheric oxygen levels. Proc. Natl. Acad. Sci. U.S.A. 115, 8104–8109 (2018).
17
D. E. Canfield et al., Petrographic carbon in ancient sediments constrains Proterozoic Era atmospheric oxygen levels. Proc. Natl. Acad. Sci. U.S.A. 118, e2101544118 (2021).
18
C. Wang et al., Strong evidence for a weakly oxygenated ocean–atmosphere system during the Proterozoic. Proc. Natl. Acad. Sci. U.S.A. 119, e2116101119 (2022).
19
D. B. Mills et al., Eukaryogenesis and oxygen in Earth history. Nat. Ecol. Evol. 6, 520–532 (2022).
20
T. M. Gibson et al., Precise age of Bangiomorpha pubescens dates the origin of eukaryotic photosynthesis. Geology 46, 135–138 (2018).
21
D. B. Cole et al., On the co-evolution of surface oxygen levels and animals. Geobiology 319, 55 (2020).
22
S. Zhang et al., Sufficient oxygen for animal respiration 1,400 million years ago. Proc. Natl. Acad. Sci. U.S.A. 113, 1731–1736 (2016).
23
T. M. Lenton, S. J. Daines, Biogeochemical transformations in the history of the ocean. Ann. Rev. Mar. Sci. 9, 31–58 (2017).
24
X.-M. Liu et al., A persistently low level of atmospheric oxygen in Earth’s middle age. Nat. Commun. 12, 351 (2021).
25
T. W. Lyons, C. T. Reinhard, N. J. Planavsky, The rise of oxygen in Earth’s early ocean and atmosphere. Nature 506, 307–315 (2014).
26
N. J. Planavsky et al., Low mid-Proterozoic atmospheric oxygen levels and the delayed rise of animals. Science 346, 635–638 (2014).
27
D. B. Cole et al., A shale-hosted Cr isotope record of low atmospheric oxygen during the Proterozoic. Geology 44, 555–558 (2016).
28
P. F. Hoffman et al., Snowball Earth climate dynamics and Cryogenian geology-geobiology. Sci. Adv. 3, e1600983 (2017).
29
D. E. Canfield, A new model for Proterozoic ocean chemistry. Nature 396, 450–453 (1998).
30
S. Zhang et al., The oxic degradation of sedimentary organic matter 1400 Ma constrains atmospheric oxygen levels. Biogeosciences 14, 2133–2149 (2017).
31
C. T. Reinhard, N. J. Planavsky, S. L. Olson, T. W. Lyons, D. H. Erwin, Earth’s oxygen cycle and the evolution of animal life. Proc. Natl. Acad. Sci. U.S.A. 113, 8933–8938 (2016).
32
M. T. Duret et al., Size-fractionated diversity of eukaryotic microbial communities in the Eastern Tropical North Pacific oxygen minimum zone. FEMS Microbiol. Ecol. 91, fiv037 (2015).
33
D. A. Walsh et al., Metagenome of a versatile chemolithoautotroph from expanding oceanic dead zones. Science 326, 578–582 (2009).
34
A. K. Hawley, H. M. Brewer, A. D. Norbeck, L. Paša-Tolić, S. J. Hallam, Metaproteomics reveals differential modes of metabolic coupling among ubiquitous oxygen minimum zone microbes. Proc. Natl. Acad. Sci. U.S.A. 111, 11395–11400 (2014).
35
S. Louca et al., Integrating biogeochemistry with multiomic sequence information in a model oxygen minimum zone. Proc. Natl. Acad. Sci. U.S.A. 113, E5925–E5933 (2016).
36
M. Torres-Beltrán et al., A compendium of geochemical information from the Saanich Inlet water column. Sci. Data 4, 170159 (2017).
37
C. C. Michiels et al., Rates and pathways of N2 production in a persistently anoxic fjord: Saanich Inlet, British Columbia. Front. Mar. Sci. 6, 4897 (2019).
38
W. Orsi, Y. C. Song, S. Hallam, V. Edgcomb, Effect of oxygen minimum zone formation on communities of marine protists. ISME J. 6, 1586–1601 (2012).
39
M. Torres-Beltrán et al., Sampling and processing methods impact microbial community structure and potential activity in a seasonally anoxic fjord: Saanich Inlet, British Columbia. Front. Mar. Sci. 6, 132 (2019).
40
L. Guillou et al., The Protist Ribosomal Reference database (PR2): A catalog of unicellular eukaryote small sub-unit rRNA sequences with curated taxonomy. Nucleic Acids Res. 41, D597–604 (2013).
41
M. Torres-Beltrán, T. Sehein, M. G. Pachiadaki, S. J. Hallam, V. Edgcomb, Protistan parasites along oxygen gradients in a seasonally anoxic fjord: A network approach to assessing potential host-parasite interactions. Deep Sea Res. Part 2, Top. Stud. Oceanogr. 156, 97–110 (2018).
42
F. Burki, A. J. Roger, M. W. Brown, A. G. B. Simpson, The new tree of Eukaryotes. Trends Ecol. Evol. 35, 43–55 (2020).
43
V. Edgcomb et al., Protistan microbial observatory in the Cariaco Basin, Caribbean. I. Pyrosequencing vs Sanger insights into species richness. ISME J. 5, 1344–1356 (2011).
44
W. Orsi et al., Protistan microbial observatory in the Cariaco Basin, Caribbean. II. Habitat specialization. ISME J. 5, 1357–1373 (2011).
45
R. De la Iglesia et al., Distinct oxygen environments shape picoeukaryote assemblages thriving oxygen minimum zone waters off central Chile. J. Plankton Res. 42, 514–529 (2020).
46
D. P. Faith, Conservation evaluation and phylogenetic diversity. Biol. Conserv. 61, 1–10 (1992).
47
T. Hastie, J. Friedman, R. Tibshirani, The Elements of Statistical Learning (Springer, New York, NY, 2009).
48
C. Strobl, J. Malley, G. Tutz, An introduction to recursive partitioning: Rationale, application, and characteristics of classification and regression trees, bagging, and random forests. Psychol. Methods 14, 323–348 (2009).
49
W. F. Martin, A. G. M. Tielens, M. Mentel, Mitochondria and Anaerobic Energy Metabolism in Eukaryotes: Biochemistry and Evolution (Walter de Gruyter GmbH & Co KG, 2020).
50
T. Fenchel, B. J. Finlay, Ecology and Evolution in Anoxic Worlds (Oxford University Press, 1995).
51
D. A. Stolper, N. P. Revsbech, D. E. Canfield, Aerobic growth at nanomolar oxygen concentrations. Proc. Natl. Acad. Sci. U.S.A. 107, 18755–18760 (2010).
52
J. Berg et al., How low can they go? Aerobic respiration by microorganisms under apparent anoxia. FEMS Microbiol. Rev. 46, fuac006 (2022), https://doi.org/10.1093/femsre/fuac006.
53
M. Müller et al., Biochemistry and evolution of anaerobic energy metabolism in eukaryotes. Microbiol. Mol. Biol. Rev. 76, 444–495 (2012).
54
J. M. Bernhard, K. R. Buck, M. A. Farmer, S. S. Bowser, The Santa Barbara Basin is a symbiosis oasis. Nature 403, 77–80 (2000).
55
L. K. Eckford-Soper, K. H. Andersen, T. F. Hansen, D. E. Canfield, A case for an active eukaryotic marine biosphere during the Proterozoic era. Proc. Natl. Acad. Sci. U.S.A. 119, e2122042119 (2022).
56
N. P. Revsbech et al., Determination of ultra-low oxygen concentrations in oxygen minimum zones by the STOX sensor. Limnol. Oceanogr. Methods 7, 371–381 (2009).
57
J. J. Wright, S. Lee, E. Zaikova, D. A. Walsh, S. J. Hallam, DNA extraction from 0.22 μM Sterivex filters and cesium chloride density gradient centrifugation. J. Vis. Exp. 31, 1352 (2009).
58
T. Stoeck et al., Multiple marker parallel tag environmental DNA sequencing reveals a highly complex eukaryotic community in marine anoxic water. Mol. Ecol. 19, 21–31 (2010).
59
Rivers, iTag Amplicon Sequencing for Taxonomic Identification at JGI (Joint Genome Institute, 2016).
60
R. Piredda et al., Diversity and temporal patterns of planktonic protist assemblages at a Mediterranean Long Term Ecological Research site. FEMS Microbiol. Ecol. 93, fiw200 (2017).
61
E. Bolyen et al., Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37, 852–857 (2019).
62
B. J. Callahan et al., DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581–583 (2016).
63
T. Rognes, T. Flouri, B. Nichols, C. Quince, F. Mahé, VSEARCH: A versatile open source tool for metagenomics. PeerJ 4, e2584 (2016).
64
A. H. Knoll, The multiple origins of complex multicellularity. Annu. Rev. Earth Planet. Sci. 39, 217–239 (2011).
65
M. Hall, R. G. Beiko, 16S rRNA gene analysis with QIIME2. Methods Mol. Biol. 1849, 113–129 (2018).
66
W. Haynes, “Benjamini–Hochberg method” in Encyclopedia of Systems Biology, W. Dubitzky, O. Wolkenhauer, K.-H. Cho, H. Yokota, Eds. (Springer, New York, NY, 2013), pp. 78–78.
67
R. L. Simister, Oxygen requirements of early eukaryote ecosystems. Sequence Read Archive (SRA). https://www.ncbi.nlm.nih.gov/sra/PRJNA1008712. Deposited 23 August 2023.
Information & Authors
Information
Published in
Cover image: Pictured is an adult male black-footed cat (Felis nigripes) resting in a hollow termite mound in South Africa. Jiaqing Yuan et al. assembled a high-quality reference genome of the small black-footed cat. The genome revealed an evolutionary history featuring rapid evolution of sensory genes associated with nocturnal hunting and metabolic genes associated with the cat's high metabolic rate. The analysis also suggested recent inbreeding events, likely caused by habitat fragmentation and genetic isolation. Inbreeding may heighten the cat's risk of the fatal disease amyloidosis, and genome analysis could aid conservation efforts, according to the authors. See the article by Yuan et al. e2310763120. Image credit: Alexander Sliwa (Cologne Zoo, Cologne, Germany).
Classifications
Copyright
Copyright © 2024 the Author(s). Published by PNAS. This article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).
Data, Materials, and Software Availability
18S rRNA amplicon data have been deposited in BioProject (PRJNA1008712) (67). All other data are included in the manuscript and/or SI Appendix.
Submission history
Received: March 6, 2023
Accepted: November 7, 2023
Published online: January 2, 2024
Published in issue: January 9, 2024
Keywords
Acknowledgments
Daniel Brady Mills was supported by the Agouron Institute Geobiology Postdoctoral Fellowship Program (#AI-F-GB53.19.2) and the Deutsche Forschungsgemeinschaft (DFG project OR 417/7-1). This work was also supported by the Natural Sciences and Engineering Research Council discovery grant 04867 awarded to Sean Crowe. Sequencing was performed and funded by the United States Department of Energy Joint Genome Institute. Daniel Brady Mills acknowledges feedback from Ömer Coskun, Virginia Edgcomb, Warren Francis, and William Orsi, assistance with bioinformatics from Jenifer Spence, and support from William Orsi. We also thank all individuals involved in the sampling and processing of Saanich Inlet samples during the year 2015, including the crew of the Marine Science Vessel Strickland, Chris Payne and Lora Pakhomova, members of the Crowe and Hallam Labs, and all those who assisted before and after sampling days.
Author Contributions
D.B.M., S.J.H., E.A.S., and S.A.C. designed research; D.B.M. and R.L.S. performed research; R.L.S., T.R.S., S.J.H., E.A.S., and S.A.C. contributed new reagents/analytic tools; D.B.M., R.L.S., T.R.S., E.A.S., and S.A.C. analyzed data; and D.B.M., R.L.S., T.R.S., S.J.H., E.A.S., and S.A.C. wrote the paper.
Competing Interests
The authors declare no competing interest.
Notes
This article is a PNAS Direct Submission.
Authors
Metrics & Citations
Metrics
Citation statements
Altmetrics
Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.