Ecology

Widespread occurrence and environmental stimuli responsiveness of trypsin in marine phytoplanktonTo assess whether trypsin occurs broadly in marine phytoplankton and what ecological functions phytoplankton trypsin genes may play, we investigated the occurrence of trypsin genes and environmental stimuli regulating their expression based on PhyloDB, Tara Oceans unigenes and metatranscriptomes datasets. From Tara Oceans unigenes and metatranscriptomes, trypsin homologs were found at all the sampling stations worldwide and in all major phytoplankton phyla (Fig. 1a and Supplementary Fig. 1). The broad phylogenetic representation is corroborated by the prevalence of trypsin in the individual species’ transcriptomes in the PhyloDB database (Fig. 1b), most notably in Bacillariophyta, Dinophyta, Chlorophyta, Cryptophyta and Haptophyta, the major eukaryotic groups of phytoplankton in the ocean. These indicate that trypsin is widely distributed in phytoplankton both taxonomically and geographically, a finding that advances our knowledge on the distribution of this ancient enzyme. Moreover, phylogenetic and structure alignment analysis showed that phytoplankton trypsins are more closely related with bacterial trypsins than metazoan and fungal counterparts, but contain the conserved important residues and structure typical of animal trypsins (Supplementary Figs. 2–4). These observations suggest some complex evolutionary trajectory that might result in functional innovation of phytoplankton trypsin.Fig. 1: Widespread occurrence and environmental nutrient responsiveness of trypsin in global marine phytoplankton.a Wide geographic distribution of trypsin in phytoplankton found in Tara Oceans. Color scale depicts trypsin mRNA abundance. b Wide taxonomic distribution of trypsin in algae found in PhyloDB. c Environmental nutrient drivers of phytoplankton trypsin abundance. Pairwise comparisons of environmental nutrient concentrations are shown with a color gradient denoting Pearson’s correlation coefficient. The trypsin abundance and taxonomic distribution based on the 5–180 µm size fraction from SRF layer from Tara Ocean datasets. Taxonomic trypsin abundance was related to each nutrient factor by partial (geographic distance-corrected) Mantel tests. Edge width corresponds to the Mantel’s r statistic for the corresponding distance correlations, and edge color denotes the statistical significance based on 9999 permutations. Baci Bacillariophyta, Dino Dinophyta, Chlo Chlorophyta, Cryp Cryptophyta, Hapt Haptophyta. Source data are provided as a Source Data file.Full size imageWe found a large amount of trypsin gene duplication, 5 copies to 65 copies in each algal genome we examined6. The evolution of the gene family, in gene sequence and organization relative to other functional domain, need to be treated in a separate paper6, but the rampant gene duplication suggests that trypsin may have important roles in phytoplankton. Moreover, our correlation analysis for trypsin gene expression with environmental parameters in the Tara Oceans metatranscriptomic data showed that the phytoplankton trypsin transcript abundance was correlated with environmental conditions in some taxa, size fractions, and water depths, evidence that trypsin may be important in phytoplankton to adapt to dynamical environmental conditions6. To further explore specific environmental drivers modulating the expression of trypsin, we analyzed distance-corrected dissimilarities of phytoplankton trypsin transcript abundance with environmental nutrient factors using the partial Mantel test. Analyses were restricted to the 5–20 and 20–180 µm size fractions from surface layer as their trypsin appeared to be more responsive to environmental stimuli. As shown in Fig. 1c, trypsin expression in Bacillariophyta, Dinophyta, Chlorophyta, Cryptophyta and Haptophyta was differentially correlated with nutrient availability, most notably in Bacillariophyta and Chlorophyta. Moreover, nitrate and nitrite (NO3, NO3_5m*, and NO3_NO2) and phosphate (PO4) were the strongest correlates of both Bacillariophyta and Chlorophyta trypsin transcript abundances (Fig. 1c). Hence, we posit that trypsin have important functions in the response of phytoplankton to N and P nutrient conditions.Involvement of trypsin in nitrogen and phosphorus nutrient responsesTo gain mechanistic insights into the function of trypsin in phytoplankton, we conducted experiments on the model diatom Phaeodactylum tricornutum. We identified ten trypsin genes from its genome (Supplementary Table 1), and based on qRT-PCR, we observed their growth stage- and condition-specific expression variations (Fig. 2a and Supplementary Fig. 5). Interestingly, one of these genes (PtTryp2) exhibited opposite directions of expression dynamic under N- and P-depleted conditions: downregulated under N-depleted but upregulated under P-depleted condition (Fig. 2a). Furthermore, PtTryp2 transcript increased with increasing cellular N content but decreased with increasing cellular P content (Fig. 2b, c). These results suggest that PtTryp2 is involved in an opposite-direction regulation of responses to nitrogen and phosphorus nutrient status.Fig. 2: Involvement of PtTryp2 in nitrogen and phosphorus nutrient responses.a PtTryp2 expression in P. tricornutum under different growth stages and conditions based on qRT-PCR. Nutrient-replete, HNHP; N-depletion, LNHP; P-depletion, HNLP. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. b Time-course expression patterns of PtTryp2 when P. tricornutum was grown with different forms of nitrogen nutrients. Data are presented as mean values ± SD (n = 2 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. c PtTryp2 expression pattern after phosphorus supplement. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. Source data are provided as a Source Data file.Full size imageTo interrogate the function of PtTryp2 in N and P nutrient responses, we analyzed the physiology of homologous overexpression and CRISPR/Cas9 knockout lines we generated. A PtTryp2-overexpression cell line with C-terminal eGFP fusion (named PtTryp2-OE) was generated, and the expression of OE cell line was confirmed at a protein level through Western blot (Fig. 3a). Because the function of a protein corresponds with its subcellular location, we first examined where PtTryp2 is located inside P. tricornutum cells. By computational simulation, we find PtTryp2 is potentially localized in the chloroplast via the secretory pathway (Supplementary Table 2), in accordance with the fact that chloroplasts contain a rather high number of proteases and are the main location of nutrients assimilation and remobilization7. To obtain experimental verification of the chloroplast localization, we carried out subcellular localization analysis in the OE and OEC cell lines using confocal fluorescence microscopy. Interestingly, results show PtTryp2-eGFP are localized in both the chloroplast and cytoplasmic endoplasmic reticulum (ER), to the exclusion of the nucleus and Golgi apparatus, whereas the fluorescence from the eGFP blank vector control is outspread in the cell instead of being co-localized with chloroplast and ER (Fig. 3b and Supplementary Figs. 6–8). Further analyses show that PtTryp2 lacks the C-terminal -(K/H) DEL sequences, a typical ER-retention signal that prevents ER-resident proteins from being transported to downstream locations of the secretory system8,9. Hence, PtTtryp2 is evidently transported via the ER to the chloroplast, as in the case of the previously documented light-harvesting chlorophyll a/b-binding protein in Euglena10.Fig. 3: Subcellular localization of PtTryp2.a Detection of the expression of GFP-PtTryp2 by Western blot using anti-GFP primary antibody. Left panel, GFP-PtTryp2 fusion protein. Middle panel, GFP protein. GAPDH (on the right) was detected using anti-GAPDH as the control to indicate equal protein quantities loaded to each lane. The GFP-PtTryp2 was confirmed expressed successfully at protein level in OE cell line. All experiments were repeated independently three times, and similar results were obtained. b Confocal micrographs showing subcellular localization of GFP-PtTryp2 in chloroplast (PAF, showing red autofluorescence) and endoplasmic reticulum (ER, showing blue fluorescent stain by ER-Tracker) but not in nucleus (Hoechst 33342, showing blue fluorescent stain). TL merge, merger of the fluorescence images with transmission light image. Scale bar, 10 µm, applies to all images. All experiments were repeated independently three times, and similar results were obtained. Source data are provided as a Source Data file.Full size imagePtTryp2 contains one trypsin domain and two internal repeats 1 (RPT) (Fig. 4a), offering one single target for trypsin mutagenesis. Using an optimized efficient CRISPR/Cas9 gene editing system11, we obtained three PtTryp2 mutants with different mutation characteristics in the trypsin domain (named KO1, KO2, and KO3, respectively; Fig. 4b). As shown in Fig. 4c, compared with the knockout control cell line (KOC), all three PtTryp2-KO lines exhibited a significantly diminished PtTryp2 expression under both nutrient depletion and repletion; conversely, the OE cell line displayed markedly elevated PtTryp2 expression in comparison to the overexpression control cell line (OEC). Moreover, the PtTryp2 expression level in KOC cell lines strongly responded to the ambient N and P level, but consistently showed a constant and low expression pattern in KO lines (Fig. 4d). These results verified that KO cell lines with the loss of PtTryp2 function, and OE with enhanced function of PtTryp2, can be used for subsequent functional analyses of PtTryp2.Fig. 4: Mutation generations of PtTryp2 and characters of mutants.a Schematic presentation of PtTryp2 protein. The target site (vertical arrow) for CRISPR/Cas9-based knockout is located within the conserved functional domain (green pentagon), with PAM motif shown in orange font. Red rectangle on the left depicts signal peptide; RPT: internal repeat 1; b Alignment of partial PtTryp2 sequences of the CRISPR/Cas9-generated mutants showing frameshift indels compared to wild type. The frequency by which the sequence was detected within the same colony is indicated in parenthesis. Font color coding: Black, WT sequence; Orange, functional domain containing target for CRISPR/Cas9; Purple, PAM sequence; Blue, Inserted bases; Red dashes, deleted bases. c PtTryp2 expression patterns of knockout and overexpression mutants under different conditions. FC fold change. Data are presented as mean values ± SD (n = 3 biologically independent samples). d PtTryp2 expression of knockout mutants exhibited no response to ambient N and P fluctuation. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. e Growth curves of different PtTryp2 mutants under different N and P conditions. Nutrient conditions in c–e are indicated by HNHP (Nutrient-replete), LNHP (N-depleted, P-replete), HNLP (N-replete, P-depleted), and LNLP (Nutrient-depleted). Data are presented as mean values ± SD (n = 3 biologically independent samples). Source data are provided as a Source Data file.Full size imageMoreover, we observed the growth physiology of different PtTryp2 mutants across different nutrient conditions. As shown in Fig. 4e and Supplementary Fig. 9, both of the knockout and overexpression of PtTryp2 resulted in decreases in the exponential growth rates (days 1–4) and maximum cell density across different N and P culture conditions. Taken together, these results demonstrate that both elevation and reduction of PtTryp2 expression result in cell growth repression, evidence that PtTryp2 has a crucial role in modulating cell growth in response to different N and P conditions.
PtTryp2 represses nitrogen assimilation and metabolismTranscriptomic data show that PtTryp2 knockout led to the upregulation of most of the nitrogen assimilation and metabolism genes under both N-depleted and replete conditions (Fig. 5a). The transcriptomic data are confirmed to be reproducible based on the correlation analysis of housekeeping genes (Supplementary Fig. 10 and Supplementary Table 3). Notably, the expression fold change of most N assimilation and metabolism genes under N-depleted, P-replete (LNHP) versus nutrient repete (HNHP) conditions were moderated in the PtTryp2 knockout mutant compared to that in its control (KOC), with the exception of GOGAT, which exhibited larger response to the nutrient changes in KOC (Fig. 5a). All these indicate that the inactivation of PtTryp2 enhanced N assimilation and metabolism to mitigate cell stress and reduce overall transcriptomic swing from N-depletion. Under replete conditions (HNHP), substantial transcriptional reprogramming and a significant increase in nitrate uptake rate and cellular N content was observed in the knockout mutants (KO1, KO2 and KO3) (Fig. 5b). The physiological changes were reversed in the overexpression cell lines: a decline in nitrate uptake rate and cellular N content was noted in PtTryp2-OE (Fig. 5c). All the results demonstrate that PtTryp2 functions as a repressor of nitrogen assimilation and metabolism.Fig. 5: Transcriptomic and physiological evidence that PtTryp2 directly represses nitrogen assimilation and metabolism.a PtTryp2 knockout resulted in upregulation of major nitrate-uptake and N-metabolism genes in PtTryp2 knockout (KO1) and control (KOC) under N-depleted (LNHP), P-depleted (HNLP), and nutrient-replete conditions (HNHP). NRT nitrate transporter, NR nitrate reductase, NiR nitrite reductase, GS glutamine synthetase, GOGAT glutamate synthase, GDH glutamate dehydrogenase, 2OG 2-Oxoglutarate; b NO3− uptake rate and cellular N content, increasing dramatically in PtTryp2-KO under HNHP, but decreasing remarkably under HNLP. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. c NO3− uptake rate and cellular N content, decreasing remarkably in PtTryp2-overexpressing P. tricornutum under HNHP, but increasing under HNLP. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. d Venn diagram showing the number of N-depletion induced DEGs in PtTryp2-KO1 and KOC. In parentheses, total number of DEGs; red font, upregulated; green font, downregulated. e Log2 fold changes (FC) of N-depletion induced differential gene expression in PtTryp2-KO1 against that in KOC. Most data points (93.37%) are distributed in 1,3 quadrants, indicating the same direction of change. Source data are provided as a Source Data file.Full size imageIn addition, when comparing N-depleted with N-replete conditions, 646 differentially expressed genes (DEGs) were identified in the blank vector control (KOC) but only 187 in PtTryp2-KO1, considerably fewer in the knockout mutant (Fig. 5d). Besides, the magnitude of change was smaller in PtTryp2-KO1 than in KOC for the majority (73%) of the DEGs (Fig. 5e). It is thus evident PtTryp2 in the wild type functions as an amplifier of general metabolic response to N-starvation by repressing nitrogen assimilation and metabolism. Notably, the PtTryp2-KO-promoted and PtTryp2-OE-repressed NO3− uptake patterns observed under nutrient repletion were reversed under P-depletion, indicating that PtTryp2’s roles in N and P signaling are not separated, but rather the protein might mediate the cross-talk between N and P signaling.Besides the direction of action (repression or promotion) shown above, the function of PtTryp2 involves another layer of regulation: the direction of its own expression changes. We find that PtTryp2 expression decreased under N-depletion and increased after N-supplement. Under this two-level regulatory scheme, PtTryp2 is a repressor of N uptake and assimilation genes and a promoter of N starvation-responsiveness in general metabolic pathways per se; yet its own expression decreases under N-limitation to upgrade N-uptake and assimilation under N depletion, and increases under N richness to prevent excessive N-uptake and assimilation; meanwhile, the decreased expression of PtTryp2 actually dampens the dynamic swing in the metabolic landscape in response to N-starvation. This PtTryp2-based regulatory mechanism might enable cells to swiftly respond to fluctuating N availability and cellular demand in order to finetune N responses so that N acquisition is optimized.
PtTryp2 promotes P starvation-induced genes and Pi uptakeAs shown above, PtTryp2 expression is downregulated under N-deficiency to release PtTryp2’s repressing effects on N-starvation response and to promote N uptake, thereby the cells achieve N homeostasis, and an opposite expression pattern of PtTryp2 was observed under P-deficiency, suggesting a N-P coregulation. However, the role of PtTryp2 in P-starvation responses and P homeostasis still needs to be unraveled. Toward that goal, we examined the effects of PtTryp2 inactivation on the expression changes of P starvation-induced genes and the inhibitory regulator of P signaling (SPX), which in plants is a typical P starvation response mechanism12. Consistently, most of Pi transporters (PTs) and alkaline phosphatase (APs) exhibited upregulation to P starvation response in KOC, but most of SPX genes showed downregulation (Fig. 6a).Fig. 6: Transcriptomic and physiological evidence that PtTryp2 positively modulates P starvation-induced genes during Pi starvation.a PtTryp2 knockout resulted in a reverse regulation of most P starvation-induced genes relative to that in control (KOC) under N-depleted (LNHP), P-depleted (HNLP), and nutrient-replete (HNHP) conditions. b PtTryp2 knockout caused decreases in Pi uptake and cellular P content under nutrient-replete condition (HNHP) but caused increases under N-depleted condition (LNHP). Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. c PtTryp2 knockout caused increases in Pi uptake rate and cellular P content under HNHP and LNHP. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. d Venn diagram showing the number of P-depletion induced DEGs in PtTryp2-KO1 and KOC. In parentheses, total number of DEGs; red font, upregulated; green font, downregulated. e Log2 fold changes (FC) of P-depletion induced differential gene expression in PtTryp2-KO1 against that in KOC. Most data points (95.69%) are distributed in 1,3 quadrants, indicating the same direction of change. Source data are provided as a Source Data file.Full size imageInterestingly, under P-depletion, PtTryp2 knockout downregulated the expression of most of PTs and APs, but upregulated most of the SPX genes (Fig. 6a), revealing PtTryp2’s role in WT to promote P-starvation responses. Consistent with gene transcription, PtTryp2 knockout lowered Pi uptake rate and cellular P content under the nutrient-replete condition (Fig. 6b), whereas an increase was noted in the overexpression cell line PtTryp2-OE (Fig. 6c). Based on RNA-seq, remarkably more DEGs were found for the P-depleted versus nutrient-replete comparison in PtTryp2-KO1 (1501) than that in KOC (277) (Fig. 6d). Besides, in PtTryp2-KO1, the majority of these DEGs (77.25%) exhibited greater fold changes than that in KOC (Fig. 6e). These results indicate that PtTryp2 upregulation in the wild type would dampen metabolic reprogramming in responses to P-limitation, and PtTryp2 downregulation would prevent cells from over P accumulation after P supplement, as opposed to the response to N-depletion. All these findings are indicative that PtTryp2 in the WT functions to upregulate the P starvation-induced genes and restrict general metabolic reconfiguration in response to P-limitation, a mechanism to maintain P homeostasis. Similar to that the PtTryp2-KO-promoted and PtTryp2-OE-repressed NO3− uptake patterns were reversed under P-depletion, the PtTryp2-KO-repressed Pi uptake pattern was reversed under N-depletion (Fig. 6b), implying that PtTryp2 might mediate the cross-talk between N and P signaling. The PtTryp2-OE-promoted Pi uptake pattern was not reversed under N-depletion, however, because N-depletion downregulated the expression of PtTryp2, resulting in the PtTryp2 expression pattern between OEC and OE similar to that under nutrient repletion.
PtTryp2 coordinately regulate N and P uptake and mediates N-P cross-talkGiven the PtTryp2-mediated cross-talk between N and P signaling in P. tricornutum implied in the results presented above, we were tempted to investigate the nature and the mechanism the cross-talk. Here, we uncover Pi and NO3− antagonistic interactions in P. tricornutum, which resemble that in land plants to achieve an optimal N-P nutrient balance13,14. In wild-type (WT) P. tricornutum, we observed a significant repression of NO3− uptake under P starvation and a significant repression of Pi uptake rate under N starvation. Consequently, cellular N content decreased under the P-depleted condition, and cellular P content decreased under the N-depleted condition, relative to nutrient-replete conditions (Fig. 7a, b). In accordance, the transcription of N assimilation and metabolism genes was repressed by P deficiency, and that of P starvation-induced genes was repressed by N limitation (Supplementary Fig. 11). Moreover, transcriptomic results demonstrated that PtTryp2 knockout led to the magnification of Pi and NO3− antagonistic interaction (Supplementary Fig. 11), linking PtTryp2 inactivation to the disruption of the N-P homeostasis. Taken together, our data reveal that PtTryp2’s function operates in opposite directions for N and P responses, but in a coordinated manner, consistent with a role to coregulate N and P signaling.Fig. 7: Illustration that PtTryp2 coordinately regulates N and P acquisition under fluctuating nutritional conditions.a NO3− uptake and cellular N content repressed under HNLP in wild-type cells (WT). Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. b Pi uptake and cellular P content repressed under LNHP in wild-type cells (WT). Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. c Time-course expression of PtTryp2 showed co-varied with the N/P nutrient ratio. Moreover, PtTryp2 expression fluctuated less at the N/P ratio of 16:1 compared to other N/P ratios. The 4 h after nutrient addition represents nutrient-repletion and 72 h nutrient-depletion. Data are presented as mean values ± SD (n = 3 biologically independent samples). d The cellular N/P ratio was significantly elevated by the inactivation of PtTryp2. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. e The cellular N/P ratio was significantly decreased by the overexpression of PtTryp2. Data are presented as mean values ± SD (n = 3 biologically independent samples). The comparisons between the averages of the two groups were evaluated using the one-tailed Student’s t test. The p values with significance (p ≤ 0.05) are shown. f Hypothetical model depicting the role of PtTryp2 in balancing N and P acquisition. Under N-depletion, PtTryp2 expression is downregulated to promote N-starvation responses and repress P-starvation responses. In contrast, under P-depletion, PtTryp2 expression is upregulated to reinforce P-starvation responses and lessen N-starvation response. By this feedback loop, optimal N-P uptake is achieved to maintain stoichiometric homeostasis. Upregulated genes and enhancement processes are shown in red, downregulated genes and weakened processes colored green. The black arrows depict transcriptional activation. Black bar at line’s end depicts inhibitory regulation. The gray arrows depict possible but unverified interaction between PtTryp2 and the existing P regulating cascade SPX-PHR or an equivalent of the N regulating cascade known in plants (SPX-NLP where NLP stands for NIN-like protein, a transcription factor). Source data are provided as a Source Data file.Full size imageTo further illustrate this, we have carried out PtTryp2 expression pattern analysis across different N/P nutrient stoichiometric ratio conditions, and found that PtTryp2 expression co-varied with the N/P nutrient ratio (Fig. 7c). The time-course analysis showed that PtTryp2 expression fluctuated less under different N or P conditions at the N/P ratio of 16:1 compared to other N/P ratios. The N/P nutrient ratio of 16:1 is considered balanced stoichiometry (Redfield ratio) and appears to be optimal for P. tricornutum growth (Supplementary Fig. 12), as previously documented15, suggesting that at this nutrient stoichiometry there is no need for a significant change in PtTryp2 expression to maintain N/P balance, but other N:P nutrient ratios deviating from 16:1 caused changes in PtTryp2 expression to maintain N/P balance. Moreover, the extent of change in PtTryp2 expression varied between cultures with different levels of N:P nutrient ratios, and between 4 and 72 h after culture inoculation from N- and P-depletion-acclimated parent culture into the experimental nutrient conditions. At 72 h PtTryp2 expression level increased with the degree of P stress (the higher the N:P ratio, the more P stressed the cultures were), except for the N:P = 1:1 condition, an extreme N-limited condition that seemed to cause PtTryp2 expression not to respond according to the general trend. Overall, all these data indicate that PtTryp2 responds strongly to the variability of the N:P ratio. Correspondingly, the cellular N/P ratio under nutrient-repletion also seems to be influenced by PtTryp2 expression level: the cellular N/P ratio was significantly elevated by PtTryp2 knockout, but conversely, was significantly decreased by the overexpression of PtTryp2 (Fig. 7d). Evidently, PtTryp2 serves to coordinate N and P uptake and metabolism to dampen the amplitude of N:P ratio changes that occur when the P. tricornutum cells experience fluctuations in nutrient conditions16,17. That is, PtTryp2 in P. tricornutum acts like an amplitude reducer of the N-P seesaw to achieve the N and P stoichiometric homeostasis (Fig. 7f).As critical nutrients for phytoplankton and plants, the balance and homeostasis of N and P are crucial to the growth of the organisms. For plants, nutrient supply in the soil is highly variable; therefore, to achieve optimal and coordinated utilization of N and P, integration of N and P signaling into an integrated network is required18. Recent studies have revealed the critical components of the network in the model plants Arabidopsis thaliana and Oryza sativa12,19,20,21. Similarly, phytoplankton in the ocean face remarkable environmental nutrient variations, and N and P nutrients are often limited22,23. Although the respective responses to N and P deficiencies have been extensively studied in phytoplankton24,25, an integrative signaling pathway of N-P nutrition cross-talk has remained unknown until now. It is striking to find that trypsin, rather than homologs of plant NRT1.1 and NIGT114,19, mediates and regulates the nitrate-phosphate signaling cross-talk.The two-level model of PtTryp2 function (Fig. 7f), including the direction of PtTryp2 action and the direction of PtTryp2 expression changes, demonstrate that PtTryp2 functions by shifting the setpoints, by tuning its own expression level, at which N signaling or P signaling is triggered in response to environmental nutrient fluctuations so that cells commit to appropriate responses. However, much of the mechanics in the regulatory cascade, from environmental nutrient sensing, PtTryp2-mediated signaling, to the regulation of the effectors such as N- and Pi-transporters and assimilatory genes, remains to be elucidated. Although the interplay between N and P nutrition based on SPX-NLP-NIGT1 and SPX-PHR-NIGT1 cascades, respectively have been uncovered in plants12,19, how PtTryp2 interacts with the SPX-PHR cascade26 and whether a SPX-NLP cascade or other regulatory cascades exist and interact with PtTryp2 for P and N nutrient regulation in phytoplankton remain to be addressed.As an initial attempt, we have performed transcriptional regulatory interaction analysis based on the Inferelator algorithm27 to predict the potential co-regulated genes in the PtTryp2-dependent regulatory cascade. Consequently, a set of 1034 genes co-regulated with PtTryp2 were identified, including 10 transcription factors (Supplementary Table 4), 10 N metabolism and assimilation genes, and a P responsive gene (Supplementary Fig. 13). Moreover, the functional enrichment of the gene set showed that PtTryp2 is possibly involved in post-transcriptional regulation, intracellular signal transduction pathway and kinase-based phosphorus metabolism and recycle pathway (Supplementary Fig. 14). The results hint on a potentially complex regulatory network that requires much more transcriptomes derived from more growth conditions than just the N and P conditions used in this study and other experimental approaches to unravel.We used the potential co-regulated gene list identified in this study in a comparative analysis with the published co-regulatory analysis datasets that contained hundreds of public RNA-seq datasets: DiatomPortal28 and PhaeoNet29. Interestingly, based on the DiatomPortal dataset, the PtTryp2 was found in the Phatr_hclust_0381 hierarchical cluster that consists of 10 genes, which has been identified as the GO term of ubiquitin-dependent protein catabolism. In terrestrial plants, the ubiquitination and degradation of SPX4 was found to mediate the nitrate-phosphate interaction signaling pathway by enabling the release of PHR2 and NLP3 into the nucleus to activate the expression of both phosphate- and nitrate-responsive genes12,19. In addition, we found 120 genes that were common in our gene list and PhaeoNet, some of which are transcription factors.Taken together, our analyses showed that the deletion and overexpression of PtTryp2 simultaneously impacted nitrogen and phosphorus uptake, nitrogen and phosphorus contents of the cell, and the N:P ratio. The simultaneous impact on N and P in opposite directions suggests that this protein either directly regulates the N and P uptake machinery or is close to the direct regulator, e.g., functioning through the ubiquitination and degradation of the direct regulators as in terrestrial plants. Furthermore, it is conceivable that one or more intermediate relays between PtTryp2 and the direct regulator would make it extremely challenging, if not impossible, to exert such precise and coordinated bidirectional regulation on N and P. To understand the mechanics of the regulatory mechanism, co-immunoprecipitation and Chromation immunoprecipitation sequencing are underway in our laboratory to experimentally identify the potential proteins and DNAs interacting with PtTryp2. Further studies on multiple fronts surrounding trypsin and its regulatory pathway are required for gaining an in-depth understanding of the interplay between N and P nutrition in phytoplankton and how phytoplankton will adapt to the potentially more variable and skewed N-P environment in the Anthropocene oceans. More

Brazil’s government risks fuelling the climate and biodiversity crisis by offsetting the fertilizer shortage resulting from Russia’s invasion of Ukraine this year (J. Liu et al. Nature 604, 425 (2022); S. Osendarp et al. Nature 604, 620–624; 2022). To produce an alternative fertilizer, it plans to mine up to 12 million tonnes annually of rhodoliths taken from an area in the South Atlantic that is roughly the size of the United Kingdom (see go.nature.com/3yhiyio).A full list of co-signatories to this letter appears in Supplementary Information.
Competing Interests
The author declares no competing interests. More

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript. More

Moraes, G. J., Mcmurtry, J. A., Denmark, H. A. & Campos, C. B. A revised catalog of the mite family Phytoseiidae. Zootaxa 434, 1–494 (2004).Article 

Google Scholar 
Fraulo, A. B. & Liburd, O. E. Biological control of twospotted spider mite, Tetranychus urticae, with predatory mite, Neoseiulus californicus, in strawberries. Exp. Appl. Acarol. 43, 109–119 (2007).PubMed 
Article 

Google Scholar 
Kuştutan, O. & Cakmak, I. Development, fecundity, and prey consumption of Neoseiulus californicus (McGregor) fed Tetranychus cinnabarinus Boisduval. Turk. J. Agric. For. 33, 19–28 (2009).
Google Scholar 
Kishimoto, H. et al. Occurrence of Neoseiulus californicus (Acari: Phytoseiidae) on citrus in Kyushu district, Japan. J. Acarol. Soc. Japan 16, 129–137 (2007).Article 

Google Scholar 
Albayrak, T., Yorulmaz, S., İnak, E., Toprak, U. & Van Leeuwen, T. Pirimicarb resistance and associated mechanisms in field-collected and selected populations of Neoseiulus californicus. Pestic. Biochem. Phys. 180, 104984 (2022).CAS 
Article 

Google Scholar 
Abdellah, A., Abdelaziz, Z., Philipe, A., Serge, K. & Abdelhamid, E. M. Seasonal trend of Eutetranychus orientalis in Moroccan citrus orchards and its potential control by Neoseiulus californicus and Stethorus punctillum. Syst. Appl. Acarol. 26, 1458–1480 (2021).
Google Scholar 
Vidrih, M., Turnšek, A., Rak Cizej, M., Bohinc, T. & Trdan, S. Results of the single release efficacy of the predatory mite Neoseiulus californicus (McGregor) against the two-spotted spider mite (Tetranychus urticae Koch) on a hop plantation. Appl. Sci. 11, 118 (2021).CAS 
Article 

Google Scholar 
Jiang, C. X., Chen, L., Huang, T. T., Mumtaz, M. & Li, Q. Neoseiulus californicus (Acari: Phytoseiidae) shows good predation potential when reared on an artificial diet supplemented with Tetranychus cinnabarinus. Syst. Appl. Acarol. 26, 2229–2246 (2021).
Google Scholar 
Katayama, H. et al. Density suppression of the citrus red mite Panonychus citri (Acari: Tetranychidae) due to the occurrence of Neoseiulus californicus (McGregor) (Acari: Phytoseiidae) on Satsuma mandarin. Appl. Entomol. Zool. 41, 679–684 (2006).Article 

Google Scholar 
Zhu, R., Guo, J. J., Yi, T. C., Xiao, R. & Jin, D. C. Preying potential of predatory mite Neoseiulus californicus to broad mite Polyphagotarsonemus latus. J. Plant Prot. 46, 465–471 (2019) ([In Chinese]).
Google Scholar 
Silva, D. E. et al. Impact of vineyard agrochemicals against Panonychus ulmi (Acari: Tetranychidae) and its natural enemy, Neoseiulus californicus (Acari: Phytoseiidae) in Brazil. Crop Prot. 123, 5–11 (2019).CAS 
Article 

Google Scholar 
Sato, M. E., Da Silva, M. Z., De Souza Filho, M. F., Matioli, A. L. & Raga, A. Management of Tetranychus urticae (Acari: Tetranychidae) in strawberry fields with Neoseiulus californicus (Acari: Phytoseiidae) and acaricides. Exp. Appl. Acarol. 42, 107–120 (2007).PubMed 
Article 

Google Scholar 
De Souza-Pimentel, G. C. et al. Biological control of Tetranychus urticae (Tetranychidae) on rosebushes using Neoseiulus californicus (Phytoseiidae) and agrochemical selectivity. Rev. Colombi. Entomol. 40, 80–84 (2014).
Google Scholar 
Elith, J. & Leathwick, J. R. Species distribution models: Ecological explanation and prediction across space and time. Annu. Rev. Ecol. Evol. Syst. 40, 677–697 (2009).Article 

Google Scholar 
Peterson, A. T. & Shaw, J. Lutzomyia vectors for cutaneous leishmaniasis in southern Brazil: ecological niche models, predicted geographic distribution, and climate change effects. Int. J. Parasitol. 33, 919–931 (2003).PubMed 
Article 

Google Scholar 
Peterson, A. T. & Soberón, J. Species distribution modeling and ecological niche modeling: Getting the Concepts Right. Nat. Conserv. 10, 102–107 (2012).Article 

Google Scholar 
Phillips, S. J., Anderson, R. P. & Schapire, R. E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 190, 231–259 (2006).Article 

Google Scholar 
Stockwell, D. & Peters, D. P. The GARP modelling system: problems and solutions to automated spatial prediction. Int. J. Geogr. Inf. Sci. 13, 143–158 (1999).Article 

Google Scholar 
Beaumont, L. J., Hughes, L. & Poulsen, M. Predicting species distributions: use of climatic parameters in BIOCLIM and its impact on predictions of species’ current and future distributions. Ecol. Model. 186, 251–270 (2005).Article 

Google Scholar 
Arslan, E. S. & Örücü, Ö. K. MaxEnt modelling of the potential distribution areas of cultural ecosystem services using social media data and GIS. Environ. Dev. Sustain. 23, 2655–2667 (2021).Article 

Google Scholar 
Hutchinson, G. E. Concluding remarks. Cold Spring Harb. Symp. Quant. Biol. 22, 415–427 (1957).Article 

Google Scholar 
Soberon, J. & Peterson, A. T. Interpretation of models of fundamental ecological niches and species distributions areas. Biodivers. Inf. 2, 1–10 (2005).
Google Scholar 
Ab Lah, N. Z., Yusop, Z., Hashim, M., Salim, J. M. & Numata, S. Predicting the habitat suitability of Melaleuca cajuputi based on the MaxEnt Species Distribution Model. Forests 12, 1449 (2021).Article 

Google Scholar 
Ali, H. et al. Expanding or shrinking? range shifts in wild ungulates under climate change in Pamir-Karakoram mountains, Pakistan. PLoS ONE 16, e0260031 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Boral, D. & Moktan, S. Predictive distribution modeling of Swertia bimaculata in Darjeeling-Sikkim Eastern Himalaya using MaxEnt: current and future scenarios. Ecol. Process. 10, 1–16 (2021).Article 

Google Scholar 
Kamyo, T. & Asanok, L. Modeling habitat suitability of Dipterocarpus alatus (Dipterocarpaceae) using MaxEnt along the Chao Phraya River in Central Thailand. Forest Sci. Technol. 16, 1–7 (2020).ADS 
Article 

Google Scholar 
Barber, R. A., Ball, S. G., Morris, R. K. A. & Gilbert, F. Target-group backgrounds prove effective at correcting sampling bias in Maxent models. Divers. Distrib. 28, 128–141 (2022).Article 

Google Scholar 
Pearson, R. G., Raxworthy, C. J., Nakamura, M. & Peterson, A. T. Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. J. Biogeogr. 34, 102–117 (2007).Article 

Google Scholar 
Elith, J. et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29, 129–151 (2006).Article 

Google Scholar 
Comino, E., Fiorucci, A., Rosso, M., Terenziani, A. & Treves, A. Vegetation and Glacier Trends in the area of the Maritime Alps Natural Park (Italy): MaxEnt application to predict habitat development. Clim. 9, 54 (2021).Article 

Google Scholar 
Wang, R. L. et al. Prediction of the potential distribution of the predatory mite Neoseiulus californicus McGregor in China using MaxEnt. Glob. Ecol. Conserv. 29, e01733 (2021).Article 

Google Scholar 
Bertolino, S. et al. Spatially explicit models as tools for implementing effective management strategies for invasive alien mammals. Mamm. Rev. 50, 187–199 (2020).Article 

Google Scholar 
Raffini, F. et al. From nucleotides to satellite imagery: approaches to identify and manage the invasive Pathogen Xylella fastidiosa and its insect vectors in Europe. Sustainability 12, 4508 (2020).CAS 
Article 

Google Scholar 
Tang, J. T., Li, J. H., Lu, H., Lu, F. P. & Lu, B. Q. Potential distribution of an invasive pest, Euplatypus parallelus, in China as predicted by Maxent. Pest Manag. Sci. 75, 1630–1637 (2019).CAS 
PubMed 
Article 

Google Scholar 
Chang, Y. et al. Predicting dynamics of the potential breeding habitat of Larus saundersi by MaxEnt model under changing land-use conditions in wetland nature reserve of Liaohe Estuary, China. Remote Sens. 14, 552 (2022).ADS 
Article 

Google Scholar 
Freeman, B. G., Lee-Yaw, J. A., Sunday, J. M. & Hargreaves, A. L. Expanding, shifting and shrinking: The impact of global warming on species’ elevational distributions. Glob. Ecol. Biogeogr. 27, 1268–1276 (2018).Article 

Google Scholar 
Smeraldo, S. et al. Generalists yet different: distributional responses to climate change may vary in opportunistic bat species sharing similar ecological traits. Mamm. Rev. 51, 571–584 (2021).Article 

Google Scholar 
Pörtner, H. O. et al. Climate Change 2022: The Physical Science Basis. Working Group II contribution to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, 15. https://www.ipcc.ch/report/ar6/wg3/ (2022).Ahmed, S. E. et al. Scientists and software–surveying the species distribution modelling community. Divers. Distrib. 21, 258–267 (2015).Article 

Google Scholar 
Tognelli, M. F., Roig-Juñent, S. A., Marvaldi, A. E., Flores, G. E. & Lobo, J. M. An evaluation of methods for modelling distribution of Patagonian insects. Rev. Chil. Hist. Nat. 82, 347–360 (2009).Article 

Google Scholar 
Pangga, I., Salvacion, A., Hamor, N. & Yap, S. Maximum entropy (MaxEnt) modeling of the potential distribution of Aspidiotus rigidus Reyne (Hemiptera: Diaspididae) in the Philippines. Philipp. Agric. Sci. 104, 1–7 (2021).
Google Scholar 
Zhou, R. B. et al. Projecting the potential distribution of Glossina morsitans (Diptera: Glossinidae) under climate change using the MaxEnt model. Biol. 10, 1150 (2021).Article 

Google Scholar 
Soberon, J. & Peterson, A. T. Interpretation of models of fundamental ecological niches and species’s distribtional areas. Biodivers. Inform. 2, 1–10 (2005).Article 

Google Scholar 
Soberon, J. M. Niche and area of distribution modeling: a population ecology perspective. Ecography 33, 159–167 (2010).Article 

Google Scholar 
Soberon, J. M. & Nakamura, M. Niches and distributional areas: concepts, methods and assumptions. P. Natl. Acad. Sci. USA 106, 19644–19650 (2009).ADS 
CAS 
Article 

Google Scholar 
Zhang, Y. X., Ji, J., Chen, X., Lin, J. Z. & Chen, B. L. The effect of temperature on reproduction and development duration of Neoseiulus (Amblyseius) californicus (Mcgregor). Fujian J. Agric. Sci. 27, 157–161 (2012) ([In Chinese]).
Google Scholar 
Neto, M. P., Reis, P. R., Zacarias, M. S. & Silva, R. A. Influence of rainfall on mite distribution in organic and conventional coffee systems. Coffee Sci. 5, 67–74 (2010).
Google Scholar 
Hu, Z., Gui, L. Y., Hua, D. K. & Luo, J. Effect of simulated rainfall on laboratory population dynamics of Tetranychus cinnabarinus. J. Environ. Entomol. 38, 936–941 (2016) ([In Chinese]).
Google Scholar 
Lawler, J. J. Climate change adaptation strategies for resource management and conservation planning. Ann. N. Y. Acad. Sci. 1162, 79–98 (2009).ADS 
PubMed 
Article 

Google Scholar 
www.carbonbrief.org/cmip6-the-next-generation-of-climate-models-explained.Gotoh, T., Yamaguchi, K. & Mori, K. Effect of temperature on life history of the predatory mite Amblyseius (Neoseiulus) californicus (Acari: Phytoseiidae). Exp. Appl. Acarol. 32, 15–30 (2004).PubMed 
Article 

Google Scholar 
Yuan, X. P., Wang, X. D., Wang, J. W. & Zhao, Y. Y. Effects of brief exposure to high temperature on Neoseiulus californicus. Ying Yong Sheng Tai Xue Bao 26, 853–858 (2015) ([In Chinese]).PubMed 

Google Scholar 
Zhang, G. H. et al. Intraspecific variations on thermal susceptibility in the predatory mite Neoseiulus barkeri Hughes (Acari: Phytoseiidae): responding to long-term heat acclimations and frequent heat hardenings. Biol. Control 121, 208–215 (2018).Article 

Google Scholar 
Phillips, S. J., Dudík, M. & Schapire, R. E.[Internet] Maxent software for modeling species niches and distributions (Version 3.4.1). url: http://biodiversityinformatics.amnh.org/open_source/maxent/. Accessed 17 March 2022.R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. url: https://www.R-project.org/ (2021).Seyedizadeh, S., Ghane-Jahromi, M., Sedaratian-Jahromi, A. & Faraji, F. Discovery of the predatory mite Neoseiulus californicus (Acari: Phytoseiidae) in some rose greenhouses in Iran and describing variation in spermathecal calyx shape. Pers. J. Acarol. 6, 67–70 (2017).
Google Scholar 
Fang, X. D., Nguyen, V. L., Ouyang, G. C. & Wu, W. N. Survey of phytoseiid mites (Acari: Mesostigmata, Phytoseiidae) in citrus orchards and a key for Amblyseiinae in Vietnam. Acarologia 60, 254–267 (2020).Article 

Google Scholar 
Greco, N. M., Tetzlaff, G. T. & Liljesthröm, G. G. Presence–absence sampling for Tetranychus urticae and its predator Neoseiulus californicus (Acari: Tetranychidae; Phytoseiidae) on strawberries. Int. J. Pest Manag. 50, 23–27 (2004).Article 

Google Scholar 
Beaulieu, F. & Beard, J. J. Acarine biocontrol agents Neoseiulus californicus sensu Athias-Henriot (1977) and N. barkeri Hughes (Mesostigmata: Phytoseiidae) redescribed, their synonymies assessed, and the identity of N. californicus (McGregor) clarified based on examination of types. Zootaxa 4500, 451–507 (2018).Kawashima, M. & Jung, C. Effects of sheltered ground habitats on the overwintering potential of the predacious mite Neoseiulus californicus (Acari: Phytoseiidae) in apple orchards on mainland Korea. Exp. Appl. Acarol. 55, 375–388 (2011).PubMed 
Article 

Google Scholar 
Koller, M., Knapp, M. & Schausberger, P. Direct and indirect adverse effects of tomato on the predatory mite Neoseiulus californicus feeding on the spider mite Tetranychus evansi. Entomol. Exp. Appl. 125, 297–305 (2007).Article 

Google Scholar 
Ohno, S. et al. Geographic distribution of phytoseiid mite species (Acari: Phytoseiidae) on crops in Okinawa, a subtropical area of Japan. Entomol. Sci. 15, 115–120 (2012).Article 

Google Scholar 
Tixier, M. S., Otto, J., Kreiter, S., Dos Santos, V. & Beard, J. Is Neoseiulus wearnei the Neoseiulus californicus of Australia? Exp. Appl. Acarol. 62, 267–277 (2014).PubMed 
Article 

Google Scholar 
Vacacela Ajila, H. E. et al. Supplementary food for Neoseiulus californicus boosts biological control of Tetranychus urticae on strawberry. Pest Manag. Sci. 75, 1986–1992 (2019).CAS 
PubMed 
Article 

Google Scholar 
Xu, X. N., Wang, B. M., Wang, E. D. & Zhang, Z. Q. Comments on the identity of Neoseiulus californicus sensu lato (Acari: Phytoseiidae) with a redescription of this species from southern China. Syst. Appl. Acarol. 18, 329–344 (2013).
Google Scholar 
Pringle, K. L. & Heunis, J. M. Biological control of phytophagous mites in apple orchards in the Elgin area of South Africa using the predatory mite, Neoseiulus californicus (McGregor) (Mesostigmata: Phytoseiidae): a benefit-cost analysis. Afr. Entomol. 14, 113–121 (2006).
Google Scholar 
Tai, Y. W. et al. R package ‘corrplot’: Visualization of a Correlation Matrix. url: https://github.com/taiyun/corrplot (2021).Muscarella, R. et al. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol. Evol. 5, 1198–1205 (2014).Article 

Google Scholar 
Araujo, M. B., Pearson, R. G., Tuiller, W. & Erhard, M. Validation of species–climate impact models under climate change. Glob. Change Biol. 11, 1504–1513 (2005).ADS 
Article 

Google Scholar 
Allouche, O., Tsoar, A. & Kadmon, R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. 43, 1223–1232 (2006).Article 

Google Scholar  More

The Sargasso Sea was identified as the spawning area of the European eel (Anguilla anguilla) 100 years ago, and numerous subsequent surveys have verified that eel larvae just a week old are regularly recorded there. However, no adult eels or eel eggs have ever been found, leaving room for alternative hypotheses on the reproduction biology of this enigmatic species. Chang et al.1 theorize about an area along the Mid-Atlantic Ridge as a potential spawning ground. The main argument for this hypothesis was that the chemical signature found in eel otoliths would indicate that early stage larvae had been exposed to a volcanic environment, such as the one present along the Mid-Atlantic Ridge. Since this correlation was solely based on a mis-interpretation of cited literature data, no new, conclusive information to pinpoint the Mid-Atlantic Ridge as an additional or even alternative spawning area was presented by Chang et al.For more than 100 years, the life history of Atlantic eels remains a matter of scientific debate. In a recent paper by Chang and colleagues, published in Scientific Reports (Sci Rep 10, 15981 (2020)), it is hypothesized that the spawning areas of the European eel (Anguilla anguilla) and the American eel (A. rostrata) are located along the Mid-Atlantic Ridge at longitudes between 50° W and 40° W1. This area lies outside the Sargasso Sea, which has so far been widely assumed to be the spawning region of both species since the beginning of the twentieth century2. The Danish researcher Johannes Schmidt collected eel leptocephali 30 mm long or less, some as short as 9 mm, all south of 30° N and west of 50° W3,4. Since then, Schmidt’s assumption was supported by a number of investigations that found recently hatched European eel larvae ( More

Satellite tracking has yielded key information about the movements and behaviour of marine vertebrates in ways that were previously logistically impossible34. In the current study, we tagged the first 18 angler-caught ABFT in Skagerrak, and tracked their movements for up to one year. Despite the majority of tags detaching prematurely, our data provides new insights regarding the migration behaviour and habitat use of this species, both locally within the Nordic region and more widely throughout the northeast Atlantic and western Mediterranean Sea. Most fish (N = 9) left Skagerrak via the Norwegian Trench, heading north before exiting into the Atlantic. In addition, the two tags which remained deployed for approximately one full year showed a return migration into the Skagerrak from the northern North Sea and southern Norwegian Sea regions, re-entering north of the British Isles and through the Norwegian Trench. No fish exited or re-entered through the English Channel or the southern North Sea. These observations of entry/exit from the Skagerrak are similar to migration behaviour inferred from historical commercial fishery data in the region during the 1950s–1960s16,19. These historical records also demonstrated that some individuals migrated from the southern Norwegian Sea into the Skagerrak, Kattegat and Øresund, before leaving the area several weeks later, potentially indicating exploratory feeding on herring and mackerel, abundant in the area during this time of year. Our new tagging results confirm this behaviour among at least some of the ABFT migrating to these areas.The migration patterns revealed by our tagging study exposes tuna entering and exiting the Skagerrak, Kattegat and Øresund to targeted exploitation by regional commercial fishing vessels. Presently, these vessels catch ABFT under a Norwegian quota (315 tonnes in 2021) but additional countries in the region may acquire a quota in the future. Moreover, the relatively narrow size distribution of tunas caught indicates that this migratory behaviour may only be performed by a limited number of year classes35, meaning that the continued long-term migration of ABFT to these waters is highly dependent on recruitment and survival of younger year classes. These younger year classes, perhaps once they reach a certain size, could then also undertake a migration to Skagerrak–Kattegat–Øresund. However, the combination of local exploitation pressures, and the presently limited number of year classes found in Skagerrak could result in ABFT migrating into Skagerrak–Kattegat–Øresund being a short-lived phenomenon if those year classes are subject to a large yearly fishing mortality (both regionally within the Nordic region, and more generally throughout the population range) and no younger year classes appear. Additionally, currently there is no scientifically-derived estimates of ABFT abundance for this region. We suggest to monitor the size distribution and abundance of ABFT in Scandinavian waters in the coming years to (1) confirm that visiting ABFT consist of only a few year classes, and clarify if younger year classes begin to appear, (2) evaluate how the numbers migrating to the region annually may change over time (e.g., under different levels of exploitation, or in relation to environmental factors).While most of our tagged ABFT went north after exiting the Skagerrak, one individual turned south into the south-central North Sea before eventually leaving through the northern part of the North Sea. The region to which it migrated in the North Sea is congruent with earlier commercial catches and sightings in this region, including the Dogger Bank vicinity15,16. Although the exact routes that tagged individuals followed were not identical, no individuals used the shortest route to reach the Atlantic: from the Skagerrak through the North Sea to the English Channel, and further south to the Bay of Biscay and other southern regions. Migration along a northerly route probably reflects a trade-off between the potential for higher energetic gain from more abundant food and higher energy resources, and the longer migration distance. This could suggest that ABFT either follow the food, or simply follow the same route by which they came through learned behaviour.Three tags remained attached long enough to explore long-term migration patterns and showed widely different behaviours. One fish crossed the Atlantic and utilized areas near the Grand Banks, crossing the ICCAT management boundary between the Western and Eastern stocks of ABFT (the 45° meridian), while the other two fish remained in the eastern Atlantic. The area west of Ireland, the Bay of Biscay and the area west of Portugal appear to be important feeding areas when the fish are not in Skagerrak or the Norwegian Sea. These results reflect interconnected seascapes for foraging through the NE Atlantic. Connecting foraging grounds off Ireland and the Bay of Biscay, which was previously suggested by Ref.24 is further corroborated by one of the fish tagged in this study, which passed over the Irish continental shelf when returning to Skagerrak in 2018.Depth and temperature useWithin ICES Area 3a, ABFT were predominantly roaming the upper water column, with most observations in the upper 100 m. However, some ABFT did dive to much deeper depths, with the maximum depth recorded being 520 m, showing that they can use the majority of the depth range available in the area (max. depth in the Norwegian Trench is app. 725 m, but represents a relatively small area). The behaviour likely reflects foraging, as ABFT were also observed by both the scientific tagging crews and the anglers to actively chase prey fish, like garfish and mackerel, at the surface during the tagging operations. The temperature ranges recorded varied between 7 and 17 °C. Both the depths and temperatures recorded are well within the thermal and depth limits reported in the literature for ABFT36.SpawningABFT have been shown to successfully spawn at temperatures above 20 °C at night30,31, and to display a distinct dive pattern thought to represent courtship and spawning behaviour29. When matching this described behaviour with the data from fish 34859 in the Mediterranean Sea, almost identical behavioural patterns were detected on specific days (Fig. 4). In total, seven days aligned with temperatures above 20 °C and oscillatory movement past the thermocline. All detected spawning events occurred west of Sardinia, where fishing for mature ABFT has been conducted for centuries37.In light of the recently proposed third spawning area in the Slope Sea of northeast United States38 and other proposed areas outside the Mediterranean19, it is relevant to look for similar temperature and behavioural patterns for fish 34840, which did not enter the Mediterranean Sea, and instead stayed in the eastern Atlantic. We found that this fish did not display a similar oscillatory behaviour, and the temperature experienced during the alleged spawning period (June–July) was above 20 °C only once (20.4 °C on 11 July). In this period, the fish was on the continental shelf west of Ireland, likely feeding and not spawning. Due to the size of the fish (247 cm CFL), reflecting a likely age of 14–16 years (matching the strong 2003 cohort), and the assumption that all eastern ABFT above five years and western ABFT above eight years are mature, we find it unlikely that this fish was immature. As such, these observations may suggest that this fish skipped spawning in 2018. Fish 34861 surfaced on 25 April and the tag was not recovered. The transmitted data does not allow for a detailed analysis of potential spawning behaviour for this fish. It did however, display 6 days where maximum temperatures from the transmitted dataset reached 20 °C (observations from 15. March to 20 April, with temperatures ranging from 20 to 20.6 °C). Given the lack of detailed behaviour and the fact that this time is well outside the normal spawning time for Mediterranean ABFT, we propose that this ABFT did not spawn in that period. However, the documentation of spawning depends on the general applicability of the temperature limits and nightly spawning behaviour30,31. More studies documenting spawning behaviour will be needed to corroborate if this pattern is consistent among locations and stocks. We also suggest more studies with longer lasting tags to elucidate if skipped spawning is a common behaviour and if fish skip one or more consecutive spawning seasons. Skipped spawning has been demonstrated in many fish species, including both freshwater and marine fish39, and likely reflects physiological condition40. If a considerable proportion of the adult population skips spawning every season, current population models, which assume annual spawning by all adult fish, should be modified to more accurately reflect population egg production and reproductive output. Current population modelling may be even further challenged if the proportion of adults that skip spawning varies over time, perhaps depending on environmental conditions. However, we acknowledge that only one of two fish followed through the spawning season appeared to skip spawning, and therefore caution against broad general interpretations. More studies are needed to verify that skipped spawning is a common behaviour, and if so, to estimate just how common that behaviour is.
Return migrationIn exploited fish populations, large adults are hypothesized to be important components of the spawning population because they contribute more to recruitment than smaller individuals due to a variety of maternal effects including higher fecundity, better quality of eggs and differences in spawning behaviour (e.g. time, location)41. Although such effects remain to be documented for ABFT, it may be prudent to conserve these large individuals as a precautionary measure, to maximize their potential contributions to reproduction and recruitment.In order to protect these fish, new knowledge about their movements and distribution is required. Data from ABFT deployed with long-term electronic tags suggests that after spawning in the Gulf of Mexico, the fish return to the feeding grounds where they were initially tagged, indicating a return feeding migration7. The same has been observed more recently from ABFT tagged in Ireland24, and other large highly migratory fish species (e.g., swordfish, Xiphias gladius42). In the current study, both ABFT that retained the tag for one year also returned to the same area, suggesting a similar seasonal return feeding migration. We also note that ABFT appeared to perform recurrent visits to the Norwegian Sea, Ireland and the Bay of Biscay on their way from Nordic waters and upon their return to the latter. Hence, we hypothesize that large ABFT in Nordic waters generally return to the same feeding area the following year, given suitable habitat features (e.g., food and temperature conditions), and follow a similar migration route as they do so. More studies are nonetheless needed to confirm this hypothesis, given few long-term deployments in the current study. For a deeper understanding of behavioural repeatability, and if/when shifts in the behaviour occur, it will be necessary to follow the same fish over multiple years. Such studies would also act as a highly valuable indicator of survival, independent of stock assessment-derived mortality estimates, and could be used to estimate the local abundance of larger ABFT43. Thus, a promising avenue for future research would be to deploy long-lasting ( > 5 to 10 years) acoustic tags and use existing infrastructure from networks such as the European Tracking Network to track these large fish over the next decade44. Given that ABFT appear to return to the area annually, we suggest that Skagerrak is a promising area for the future deployment and retrieval of PSATs and other long-lasting tags, because of the relatively easy access to locate and recover detached floating tags, given that the area is reachable from land within a few hours by boat. Retrieving PSATs that have detached from animals enables scientists to access full datasets (in the present case with 5 s resolution, rather than the much coarser and variable resolution typically transmitted). This much higher resolution enables much more detailed analysis, as shown in our analysis of spawning behaviour. Additionally, floating Pop-off Data Storage Tags (PDST) tags may also be a prominent and less costly avenue forward as the geographical region is densely populated and contains many sandy beaches and highly visited coastal areas, giving ample opportunity for tag recovery. Previous studies with floating DSTs in this area have shown remarkably high return rates45.The evidence that ABFT have returned to Nordic waters following many years of rarity or absence, and our findings that at least some individuals return to the same site for feeding in consecutive years, raises new questions about the mechanisms that underlie habitat discovery—or the return to previously used habitats—by highly migratory fish species. How individuals or entire schools have discovered this region again as a suitable feeding area after an absence of more than 50 years is unclear. In light of the positive stock development in the last 1–2 decades22 and modelling studies showing suitable habitat in the area46, density-dependent foraging and exploratory behaviour for new feeding areas may be a prominent hypothesis for their return, potentially accompanied by complex social learning interactions among individuals within the population47,48. New tagging data which documents the use of new or formerly occupied habitats will be essential for understanding these processes and how they might be affected by human pressures (e.g., exploitation, climate change). Such data can help to parameterize and validate advanced conceptual models of group movement behaviour, collective memory and habitat use49,50,51, as well as to inform modern stock assessment models used for management.
Tag deploymentFollowing recommendations from experienced taggers previously operating in the Mediterranean, most fish were tagged in the water alongside the boat. All these tags surfaced prematurely, while two (out of three) tags deployed on tunas brought on board the tagging boat surfaced after approximately one year. Depending on the conditions at sea, tagging along the side of the boat may not be as precise as on-board tagging, and the quality of the tag anchoring cannot be properly assessed. We therefore suggest that tagging on-board a boat is superior to tagging in the water alongside the boat for the deployment of long-lasting tags. This was also suggested in Ref.24. Furthermore, on-board tagging makes biological sampling fast and feasible, as opposed to tagging in the water alongside the boat. However, our advice is limited by a small sample size, making it difficult to draw formal conclusions; more studies are necessary to assess the best method to tag large ABFT. More

Cole, C. T., Anderson, J. E., Lindroth, R. L. & Waller, D. M. Rising concentrations of atmospheric CO2 have increased growth in natural stands of quaking aspen (Populus tremuloides). Glob. Change Biol. 16, 2186–2197 (2010).Article 

Google Scholar 
Urrutia-Jalabert, R. et al. Increased water use efficiency but contrasting tree growth patterns in Fitzroya cupressoides forests of southern Chile during recent decades. J. Geophys. Res. Biogeosci. 120, 2505–2524 (2015).Article 

Google Scholar 
Cienciala, E. et al. Increased spruce tree growth in Central Europe since 1960s. Sci. Total Environ. 619–620, 1637–1647 (2018).PubMed 
Article 

Google Scholar 
Mathias, J. M. & Thomas, R. B. Disentangling the effects of acidic air pollution, atmospheric CO2, and climate change on recent growth of red spruce trees in the Central Appalachian Mountains. Glob. Change Biol. 24, 3938–3953 (2018).Article 

Google Scholar 
Körner, C. et al. Carbon flux and growth in mature deciduous forest trees exposed to elevated CO2. Science 309, 1360–1362 (2005).PubMed 
Article 

Google Scholar 
Klein, T. et al. Growth and carbon relations of mature Picea abies trees under 5 years of free-air CO2 enrichment. J. Ecol. 104, 1720–1733 (2016).CAS 
Article 

Google Scholar 
Norby, R. J. & Zak, D. R. Ecological lessons from free-air CO2 enrichment (FACE) experiments. Annu. Rev. Ecol. Evol. Syst. 42, 181–203 (2011).Article 

Google Scholar 
Peñuelas, J., Canadell, J. G. & Ogaya, R. Increased water-use efficiency during the 20th century did not translate into enhanced tree growth. Glob. Ecol. Biogeogr. 20, 597–608 (2011).Article 

Google Scholar 
IPCC. Climate Change 2013: The Physical Science Basis (eds Stocker, T. F. et al.) (Cambridge Univ. Press, 2013).Dong, N. et al. Rising CO2 and warming reduce global canopy demand for nitrogen. New Phytol. https://doi.org/10.1111/nph.18076 (2022).Finzi, A. C., Allen, A. S., DeLucia, E. H., Ellsworth, D. S. & Schlesinger, W. H. Forest litter production, chemistry, and decomposition following two years of free-air CO2 enrichment. Ecology 82, 470–484 (2001).
Google Scholar 
Liberloo, M. et al. Elevated CO2 concentration, fertilization and their interaction: growth stimulation in a short-rotation poplar coppice (EUROFACE). Tree Physiol. 25, 179–189 (2005).PubMed 
Article 

Google Scholar 
Hungate, B. A. et al. Nitrogen cycling during seven years of atmospheric CO2 enrichment in a scrub oak woodland. Ecology 87, 26–40 (2006).PubMed 
Article 

Google Scholar 
Ainsworth, E. A. & Long, S. P. What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytol. 165, 351–372 (2005).PubMed 
Article 

Google Scholar 
Liberloo, M. et al. Coppicing shifts CO2 stimulation of poplar productivity to above-ground pools: a synthesis of leaf to stand level results from the POP/EUROFACE experiment. New Phytol. 182, 331–346 (2009).CAS 
PubMed 
Article 

Google Scholar 
McCarthy, H. R. et al. Re-assessment of plant carbon dynamics at the Duke free-air CO2 enrichment site: interactions of atmospheric [CO2] with nitrogen and water availability over stand development. New Phytol. 185, 514–528 (2010).CAS 
PubMed 
Article 

Google Scholar 
Dawes, M. A. et al. Species-specific tree growth responses to 9 years of CO2 enrichment at the alpine treeline. J. Ecol. 99, 383–394 (2011).
Google Scholar 
Luo, Y. Q. et al. Progressive nitrogen limitation of ecosystem responses to rising atmospheric carbon dioxide. Bioscience 54, 731–739 (2004).Article 

Google Scholar 
Keenan, T. F. et al. Increase in forest water-use efficiency as atmospheric carbon dioxide concentrations rise. Nature 499, 324–327 (2013).CAS 
PubMed 
Article 

Google Scholar 
Luo, Y. Q., Hui, D. F. & Zhang, D. Q. Elevated CO2 stimulates net accumulations of carbon and nitrogen in land ecosystems: a meta-analysis. Ecology 87, 53–63 (2006).PubMed 
Article 

Google Scholar 
Kikuzawa, K. A cost-benefit analysis of leaf habit and leaf longevity of trees and their geographical pattern. Am. Nat. 138, 1250–1263 (1991).Article 

Google Scholar 
Wright, I. J. et al. The worldwide leaf economics spectrum. Nature 428, 821–827 (2004).CAS 
PubMed 
Article 

Google Scholar 
Luo, T. X. et al. Summer solstice marks a seasonal shift in temperature sensitivity of stem growth and nitrogen-use efficiency in cold-limited forests. Agric. For. Meteorol. 248, 469–478 (2018).Article 

Google Scholar 
Rossi, S. et al. Conifers in cold environments synchronize maximum growth rate of tree-ring formation with day length. New Phytol. 170, 301–310 (2006).PubMed 
Article 

Google Scholar 
Bauerle, W. L. et al. Photoperiodic regulation of the seasonal pattern of photosynthetic capacity and the implications for carbon cycling. Proc. Natl Acad. Sci. USA 109, 8612–8617 (2012).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Jarvis, P. & Linder, S. Constraints to growth of boreal forests. Nature 405, 904–905 (2000).CAS 
PubMed 
Article 

Google Scholar 
Sullivan, P. F., Ellison, S. B., McNown, R. W., Brownlee, A. H. & Sveinbjörnsson, B. Evidence of soil nutrient availability as the proximate constraint on growth of treeline trees in northwest Alaska. Ecology 96, 716–727 (2015).PubMed 
Article 

Google Scholar 
Dodd, A. N. et al. Plant circadian clocks increase photosynthesis, growth, survival, and competitive advantage. Science 309, 630–633 (2005).CAS 
PubMed 
Article 

Google Scholar 
Jackson, S. D. Plant responses to photoperiod. New Phytol. 181, 517–531 (2009).CAS 
PubMed 
Article 

Google Scholar 
Chapin III, F. S., Matson, P. A. & Mooney, H. A. Principles of Terrestrial Ecosystem Ecology (Springer-Verlag, 2002).Hikosaka, K. Leaf canopy as a dynamic system: ecophysiology and optimality in leaf turnover. Ann. Bot. 95, 521–533 (2005).CAS 
PubMed 
Article 

Google Scholar 
Jiang, M. K. et al. The fate of carbon in a mature forest under carbon dioxide enrichment. Nature 580, 227–231 (2020).CAS 
PubMed 
Article 

Google Scholar 
Guerrieri, R. et al. Disentangling the role of photosynthesis and stomatal conductance on rising forest water-use efficiency. Proc. Natl Acad. Sci. USA 116, 16909–16914 (2019).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Mathias, J. M. & Thomas, R. B. Global tree intrinsic water use efficiency is enhanced by increased atmospheric CO2 and modulated by climate and plant functional types. Proc. Natl Acad. Sci. USA 118, e2014286118 (2021).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Finzi, A. C. et al. Increases in nitrogen uptake rather than nitrogen-use efficiency support higher rates of temperate forest productivity under elevated CO2. Proc. Natl Acad. Sci. USA 104, 14014–14019 (2007).CAS 
PubMed 
PubMed Central 
Article 

Google Scholar 
Soulé, P. T. & Knapp, P. A. Radial growth rate increases in naturally occurring ponderosa pine trees: a late-20th century CO2 fertilization effect? New Phytol. 171, 379–390 (2006).PubMed 
Article 

Google Scholar 
Linares, J. C. & Camarero, J. J. From pattern to process: linking intrinsic water-use efficiency to drought-induced forest decline. Glob. Change Biol. 18, 1000–1015 (2012).Article 

Google Scholar 
Kagawa, A., Sugimoto, A. & Maximov, T. C. 13CO2 pulse-labelling of photoassimilates reveals carbon allocation within and between tree rings. Plant Cell Environ. 29, 1571–1584 (2006).CAS 
PubMed 
Article 

Google Scholar 
Epron, D. et al. Pulse-labelling trees to study carbon allocation dynamics: a review of methods, current knowledge and future prospects. Tree Physiol. 32, 776–798 (2012).CAS 
PubMed 
Article 

Google Scholar 
Wiley, E. & Helliker, B. A re-evaluation of carbon storage in trees lends greater support for carbon limitation to growth. New Phytol. 195, 285–289 (2012).CAS 
PubMed 
Article 

Google Scholar 
Rocha, A. V., Goulden, M. L., Dunn, A. L. & Wofsy, S. C. On linking interannual tree ring variability with observations of whole-forest CO2 flux. Glob. Change Biol. 12, 1378–1389 (2006).Article 

Google Scholar 
Zweifel, R. et al. Link between continuous stem radius changes and net ecosystem productivity of a subalpine Norway spruce forest in the Swiss Alps. New Phytol. 187, 819–830 (2010).CAS 
PubMed 
Article 

Google Scholar 
Kong, G. Q., Luo, T. X., Liu, X. S., Zhang, L. & Liang, E. Y. Annual ring widths are good predictors of changes in net primary productivity of alpine Rhododendron shrubs in the Sergyemla Mountains, southeast Tibet. Plant Ecol. 213, 1843–1855 (2012).Article 

Google Scholar 
Teets, A. et al. Linking annual tree growth with eddy-flux measures of net ecosystem productivity across twenty years of observation in a mixed conifer forest. Agric. For. Meteorol. 249, 479–487 (2018).Article 

Google Scholar 
Luo, T. X., Li, M. C. & Luo, J. Seasonal variations in leaf δ13C and nitrogen associated with foliage turnover and carbon gain for a wet subalpine fir forest in the Gongga Mountains, eastern Tibetan Plateau. Ecol. Res. 26, 253–263 (2011).CAS 
Article 

Google Scholar 
Kobe, R. K., Lepczyk, C. A. & Iyer, M. Resorption efficiency decreases with increasing green leaf nutrients in a global data set. Ecology 86, 2780–2792 (2005).Article 

Google Scholar 
Vergutz, L., Manzoni, S., Porporato, A., Novais, R. F. & Jackson, R. B. Global resorption efficiencies and concentrations of carbon and nutrients in leaves of terrestrial plants. Ecol. Monogr. 82, 205–220 (2012).Article 

Google Scholar 
Holmes, R. L. Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bull. 43, 69–78 (1983).
Google Scholar 
Cook, E. R. & Kairiukstis, L. A. Methods of Dendrochronology: Applications in the Environmental Sciences (Kluwer Academic Publishers, 1990).Editorial Board of Vegetation Map of China, Chinese Academy of Sciences. Vegetation Atlas of China (Science Press, 2001). More

Fossil remains from the Family Bovidae, such as antelopes and buffalo, are frequently used to reconstruct past environments1,2,3. Bovids reflect distinct ecological adaptations in terms of diet, habitat, water dependence, and seasonal migrations that vary according to their respective ecological niches. Widespread cooling in the late Miocene led to a major adaptive radiation of the bovids, and increasingly they began to exploit more open environments4,5,6. Thus, by approximately 4 Ma, bovids came to dominate the African fauna, replacing the previously abundant suids7,8,9. The current distribution of bovids extends across the African continent in myriad environments that differ significantly in proportions of wood and grass cover.The importance of bovid remains to paleoanthropological research was established initially by Broom10,11 and Wells and Cooke12. This dependence has been expanded and now ranges from paleodietary studies and evolutionary trends to hominin behavioral patterns13,14,15. In addition, several studies have demonstrated that changes in the relative abundance of bovid taxa reflected in fossil assemblages are indicative of fluctuations in environmental conditions, as bovids appear to be particularly responsive to environmental changes16,17,18.Bovid teeth, in particular isolated teeth, make up a majority of the southern African fossil record. Thus, bovid teeth, coupled with their ecological tendencies, are important sources of information for reconstructing the paleoenvironments associated with the fossil hominins. Taxonomic identification of fossil bovid teeth, however, is often problematic; biasing factors such as age and degree of wear complicate identifications and often result in considerable overlap in the shape and size of teeth. Traditionally, researchers rely upon modern and fossil comparative collections to identify isolated bovid teeth. However, researchers are somewhat limited by travel and the specific type and number of bovids housed at each institution. Here, we present B.O.V.I.D. (Bovidae Occlusal Visual IDentification) which is a repository of images of the occlusal surface of bovid teeth (~3900). The purpose of the database is to allow researchers to visualize a large sample of teeth from different tribes, genera, and species. The sample includes the three upper and three lower molars in multiple states of wear from the seven most common tribes in the southern African fossil record and the twenty most common species from those tribes. This design will help researchers see the natural variation that exists within a specific tooth type of a taxon and, with the current sample, help taxonomically identify extant and fossil teeth with modern counterparts. More