Evidence of microalgal isotopic fractionation through enrichment of depleted uranium

Resulting from the nuclear fuel cycle, large amounts of depleted uranium (DU) tails are piling up, waiting for possible use or final disposal. To date, the recovery of the residual 235U isotope contained in DU has been conducted only marginally by physical processes. Relative isotope abundances are often mediated by biological processes, and the biologically driven U isotopic fractionation has been previously identified in reducing bacteria. Our results indicate that the cells of two microalgal strains (freshwater Chlamydomonas sp. (ChlGS) and marine Tetraselmis mediterranea (TmmRU)) took up DU from the exposure solutions, inducing U isotopic fractionation with a preference for the fissile 235U isotope over 238U. The n(235U)/n(238U) isotopic fractionation magnitudes (δ235) were 23.6 ± 12.5‰ and 370.4 ± 103.9‰, respectively. These results open up new perspectives on the re-enrichment of DU tailings, offering a potential biological alternative to obtain reprocessed natural-equivalent uranium. Additionally, the findings present implications for identifying biological signatures in the geologic records.

has several uses, both civilian (i.e., radiation shielding of medical equipment or ballast in aircrafts) and military (particularly in ammunition), the estimated world's stock is approximately 1.6 million tons of DU.
After processing, the by-product DU tails present the unique feature that they may be reprocessed and recycled to provide fresh nuclear fuel and reduce the volume of low-level wastes 10 . Several incentives of re-enrichment mainly in Russia 11 and more recently by the U.S. Department of Energyhave been put forward to recover the residual 235 U contained in the DU and produce uranium with 235 U natural contents (0.71 atomic %). However, current re-enrichment technology is only economically viable in centrifuge enrichment plants with spare capacity and low operation costs, and it involves high energy consumption and the associated CO 2 emissions 12 . New technological developments pursuing a significant reduction of the environmental impact and greater U recycling-reprocessing would be desirable objectives considering the continuous increase of energy demand and the pressures upon energy sustainability 13 .
U is a ubiquitous element, present at significant amounts in the Earth's crust 14 . U is not biologically linked with any type of life, yet various mechanisms through which U is biotically processed are common in the environment, for instance: biosorption, bioaccumulation, biomineralization, and biotransformation [15][16][17][18][19] . Consequently, microbial communities can also have dramatic effects on U mobilization/immobilization 20,21 . However, the study of in vivo and biologically mediated U isotope fractionation constitute research areas still to be explored. Traditionally, natural 235 U and 238 U variability, i.e., differential isotopic behaviours, has gone unaddressed and is assumed invariant owed to the small relative differences in mass of the isotopes 22,23 . Driven by the advent of technological advances in analytical measurements, the growing field of isotopic fractionation revealed considerable variations of U isotope ratios in natural settings (e.g., ores, granites, corals, seawater [22][23][24]. Therefore, significant U isotopic fractionation might take place at the Earth´s surface 22 and represent a powerful tool in environmental, geological, marine, life and energy sciences. Biological U isotopic fractionation in nature has been linked to bacteria adept at inducing U(VI) biotic reduction 7,25-27 , as it was recently found that the redox reaction is responsible for the isotopic fractionation and is not related to the U uptake inside the cells 26 . Biologically U(VI) reduction studies resulted in the accumulation of 238 U in the reduced product, except Rademancher et al. 27 found 235 U enrichment in the resultant U(IV). In another framework, human neuron-like cells in vitro achieved an isotopic fractionation of natural U with preferential intracellular uptake of 235 U isotope 28 .
Hence, natural physicochemical processes leading to isotopic fractionation, both mass dependent and independent 29,30 , might take place during U uptake by the cells. This raises the question of whether different microalgal species give rise to U isotopic fractionation. Recently, evidence of U fractionation has been obtained during Chlamydomonas cells' U uptake in a U acid mine drainage medium, suggesting a 235 U enrichment 31,32 . Here, we have studied the U isotopic fractionation during DU uptake in two marine and freshwater Chlorophyta strains. As enrichment of the fissile 235 U is expected in the cellular pellet, DU was used to address the potential in U reprocessing. Changes in the 235 U/ 238 U ratios of the extracellular and cellular U were investigated for 24 days in an extremophile Chlamydomonas (ChlGS strain) isolated from a U mining pond and a marine Tetraselmis (TmmRU strain). These results represent a potential tool for U recycling and reprocessing and may entail implications in the study of U isotopes in natural samples.

Results
We performed two independent experiments, each with one of the strains of interest, the Chlamydomonas (ChlGS) and Tetraselmis (TmmRU) Chlorophyta strains, in different media supplemented with DU. ChlGS is an extremophile isolated from an acid U mine tailings pond, tolerant up to 25 mg U L −1 and other metals, and artificially selected for U uptake 33 . The isotopic ratio n( 235 U)/n( 238 U) with a value of 0.007375 ± 0.000013 found in the mine water (see Supplementary Fig. S1) was far above the consensus natural abundance 235 U/ 238 U [0.007198-0.007202] 34 , suggesting a possible enrichment process in the U mine pond. Conversely, TmmRU is a seawater strain, only previously exposed to naturally occurring trace U 35 and subsequently selected for U tolerance. ChlGP cell replicates were exposed to 4 mg L −1 DU ~ 0.0050 atomic 235 U freshwater stock solution and TmmRU to 2 mg L −1 DU ~ 0.0022 atomic 235 U marine stock solution. The analytical procedure and sample resin purification validation were accomplished by the analyses of procedure control solutions (certified IRMM-053 material) between samples during the measure sessions to correct the bias induced during the inductively coupled plasma mass spectrometry (ICP-MS) measurements. The average n( 235 U)/n( 238 U) values found in the control solution was 0.007112 ± 0.000024 (σ, n = 34) (Fig. S2), these values were used to calculate the discriminatory factor of each sample (for definition, see SI Material and Methods).
Within each bioassay, twelve independent trials were conducted, and the cellular pellet incorporated from 6.1 to 78.5 µg U for ChlGS and from 0.57 to 3.22 µg U for TmmRU (Table 1). ChlGS cells exhibited an increased U uptake performance until day 24, virtually the entire U amount in dissolution. Conversely, TmmRU cells incorporated 1.5-5% of the U present in the exposure solution. DU stock solutions were isotopically characterized before the exposition, and the average n( 235 U)/n( 238 U) ratio values found for the stock marine and freshwater solutions were 0.005058 ± 0.000083 (σ, n = 3) and of 0.002172 ± 0.000023 (σ, n = 3), respectively. The n(235U)/n(238U) ratios were determined in the cellular pellets and the extracellular solutions for each independent trial at different times: 3, 12 and 24 days (Figs 1, 2 and Table 1). The IRMM-053 material bracketed the samples for mass bias correction. No significant difference in the n( 235 U)/n( 238 U) ratio between days was found in either bioassay (P > 0.5).
The δ 235 (U) values, equation (S3), of each independent experiment were calculated from the n( 235 U)/n( 238 U) ratio between the cellular pellet and the supernatant (Table 1). For both microalgal species tested during this investigation, we observed a significant shift in U isotopic composition towards lighter values of δ 235 . Magnitudes of U removal from the dissolution by cells uptake were of δ 235 (U) pellet-supernatant ≈ 22.2-25.5‰ (ChlGS) and 305.2-445.6‰ (TmmRU). Meanwhile, both studies found microalgal pellets to be considerably enriched in the isotopically light 235 U isotope, whilst the supernatants showed a depletion in 235 U. This implies an isotopic fractionation towards a lighter composition in the cell pellet in comparison to the supernatant. Additionally, these  Table 1. Results from the U isotopic fractionation bioassays in Chlamydomonas strain (ChlGS) and Tetraselmis strain (TmmRU). ¥δ235(‰) values were calculated of the respecting sample with the obtained n(235U)/n(238U) ratio of the cellular pellet relative to the supernatant. differences represent a down-blending of the isotope 235 U in the cellular adjoin media of the isotope 235 U, which necessarily was consumed in the process (Table 1).

Discussion
Our results demonstrate that the two Chlorophyta microalgal strains studied fractionate DU. Each strain displayed a different enrichment factor but both reflect a strong fractionation of the light 235 U isotope in the cellular pellet. The marine strain TmmRU, despite presenting higher cell volume and organic weight 36,37 , presented a lower U uptake rate. However, the enrichment factor was significantly higher than in the freshwater ChlGS, despite being exposed to a lower 235 U content DU (~0.0022 235 U). The similar U isotopic composition found in the cellular pellets of the different replicates of each strain, irrespective of the culture time and the total U incorporated by the cells, raises the issues of preferential U uptake paths. The pathways leading to U uptake by the cells are not well documented 18 , and the joint complexity derivate of the simultaneously of several processes 28,38redox reactions, ligand exchange, diffusion, adsorptionmake interpretations of U fractionation origin extremely complex. Furthermore, U speciation and redox state may influence the fractionation process. Whatever the route, as previously described by Baselga-Cervera et al. 39 , U is bound to the outer wall and transported across the cell wall and membrane, becoming distributed inside the cell. These data suggest that U carried by the cellular pellet accomplishes isotopic partitioning, resulting in 235 U being concentrated by the cells and the surrounding environment being enriched in 238 U. Consistent with our results, the isotopic ratio n( 235 U)/n( 238 U) value found in the U mine water suggest a possible enrichment process. Only microbial life has been detected in this pond 39,40 , and therefore biomass does not pass to higher trophic levels. Thus, 235 U, as one of the lighter isotopes of U, could preferentially enter cells. When cells break, U enriched in 235 U might be liberated to the water. The remaining U enriched in 238 U bound to the cell wall might sediment in the bed of the pond. Additionally, bacteria that might be present in this pond and can contribute to this result. Reductive bacteria can induce U fractionation; the reaction products (U(IV)) are enriched in 238 U, rendering the residual dissolved U enriched in 235 U 25,41 . Combined effects of different microorganisms may have led to this result.
Microbial isotopic behaviour in elements with higher mass numbers, such as U, is typically poorly studied compared to light elements because of the need for more sophisticated and precise analyses. Recent evidence in the field has demonstrated U isotopic fractionation mediated by bacteria and neuron-like human cells. Biotic reduction studies with metal-reducing bacterial isolates show an enrichment in the heavier 238 U isotope into the solid U(IV) byproduct 26,42,43 that is not dependent on microbial sorption. Conversely, the isotopic behaviour displayed by neuron-like human cells showed a preferential 235 U incorporation 28 . Paredes et al. 28 suggested two possible isotopic fractionation processes based on the enrichment direction and U bioaccumulation: a mass-dependent "zero-point energy" mechanism 29 and the mass-independent "nuclear field shift". In the case of our study, as the two processes are not mutually exclusive and operate in the same fractionation direction, we cannot determine the precise contribution of each mechanism proposed. High-precision determination of other U isotopes, such as 234 Ucommonly concentrated during 235 U enrichmentcould provide insights into the fractionation mechanism at the cellular level and the proposed biological preference for lighter isotopes, but it is complex due to the abundances limitation [natural 234 U abundance is comprehended between 0.000050-0.000059]. Reported isotope fractionation ratios for 235 U/ 238 U during U reduction by bacteria have ranged from −0.31 to −0.99‰ 26,42,43 , and 0.38 ± 0.13‰ in neuron-like line cells 29 . We found that the cellular pellet was enriched in 235 U relative to the supernatant with DU by 23.6 ± 12.5‰ and 370.4 ± 103.9‰ for the ChlGS and TmmRU strains, respectively. This fractionation behaviour is consistent in its direction with that observed in neuron-like cells, but the fractionation factors are significantly higher. One potential application of the observed microalgae-induced U isotopic variation is for DU waste re-enrichment along the U fuel cycle for nuclear fission energy. The current DU stockpiles worldwide would render one-third of natural-equivalent U after several cycles of re-enrichment, reducing DU tail and U-mine production. Currently, only centrifuge separation and gaseous diffusion have operated at commercial scale, even though several enrichment processes have been demonstrated historically and in the laboratory [44][45][46][47] . Both enrichment processes present important drawbacks: water reactivity by-products rendering hazardous compounds highly corrosive, as well as significant amounts of energy consumption, considerable costs and generate DU as a low-level waste product 48 . Reprocessing U tails with current techniques, despite its potential, is not economically and energetically feasible and does not ease the problem of final tail disposition 49 . Particularly interesting would be to develop a viable biological process to enrich DU, recovering natural-equivalent U from tailings waiting for final disposal without high energy demand and the associated carbon footprint. Hypothetically, multiple stages of microalgal bioaccumulation of DU, cell harvesting and U resuspension into the next higher step would enrich the DU to the desired amount, up to the 235 U natural content or even above. The predicted advantages of this biological process are the reduced cost and low energy requirements, turning DU in a resource that even the U-tail problem decreases only marginally. Further experimentation may resolve the expected value range for U fractionation mediated by microalgal bioaccumulation and the efficiency of the enrichment process. In addition, other microalgal species might display a different isotopic behaviour, considering the significant differences exhibited by the two Chlorophyta species. Our findings might also have implications for the identification of biotic signatures using isotope tools to study ancient microbial life and understand global uranium flux. Significant uncertainties remain regarding isotopic signatures owing to the ambiguity in the interpretation of the signs. Development of analytical techniques has opened the possibility of studying small U isotope natural variation. Uranium is the heaviest element for which natural variations have been reported 50 . U isotopic fractionation in nature occurs, with opposite directions, in both anoxic/euxinic and oxic environments and can be associated with chemical transformations such as adsorption, speciation or redox chemistry 22 . Microorganisms probably have a part in the U cycling and deposit formation in nature [51][52][53] . The large isotope fractionation that occurs during microalgal U uptake suggests the role of microalgae in the conservative behaviour of U and a tight range of isotope composition in modern oceans. For geological implications, recent studies of biotic redox transformation of U with metal-reducing bacteria induced U isotopic fractionation enriching for 238 U in the reaction products 25,26 , contradicting previous studies 27 and consistent with environmental studies and U-reduced depositional samples 22,41,43,53,54 . U isotopic fractionation mediated by bacterial enzymatic reduction raises as a tangible biosignature in the rock record for specific metabolic groups and onto the timing emergence of specific metabolisms. In oxic sedimentary environments, sample fractionation occurs towards lighter isotope composition, and intriguingly, banded-iron formation samples present the lightest U composition studied 22 . The banded-iron formations' lighter values could indicate microbial phototroph co-precipitation by adsorption of iron and U, supporting the previously suggested microbial implication 55 . Thus, the preferential accumulation of the fissile 235 U isotope in sediments might be a proxy for the activity and presence of microalgae. U fractionation biological fingerprints in ancient sedimentary rocks would provide insights into ancient microbial activity and establish temporal constraints. Additionally, our findings might provide insights into the Oklo phenomenon on the assumption of a microbial contribution in the initiation of the natural nuclear reactors 40,56,57 . Several factors probably contribute to these unique chain reactions. Two thousand million years ago -the same time proposed for Oklo's event 58-61 -235 U made up approximately 3% of the natural U, a condition that makes possible the starting of a fission reaction. However, a U-rich mineral deposit needs to be formed to obtain a critical mass. Most likely, the presence of increasing oxygen in the Earth's atmosphere enabled the U flux and subsequent concentration in U ore bodies 62 . The direction and magnitude of the observed microalgal U isotopic fractionation during bioaccumulation could support the biological hypothesis of the origin of the natural reactors.

Methods
Microalgal species and culture conditions. We used two Chlorophyta, a freshwater environmental extremophile, Chlamydomonas saelices nov. sp. (strain ChlSG) isolated from a uranium mining pond in Saelices el Chico, Salamanca, Spain (previously described in depth by García-Balboa et al. 39 and Baselga-Cervera 31 ), and the marine Tetraselmis mediterranea (Lucksch) R.E. Norris, Hori & Chihara (strain TmmRU) isolated in the east-coastal waters of Sardinia (Italy). Both strains were selected for U tolerance and placed in the Spanish Algae Bank (Spanish acronym BEA) with the access numbers BEA/D04-12 and BEA/IDA/0062 for ChlSG and TmmRU, respectively 63,64 . Strains were propagated asexually in cell culture flasks (Greiner; Bio-One Inc., Longwood, NJ, USA) with BG-11 medium (Sigma-Aldrich Chemie, Taufkirchen, Germany)) for ChlGS and Guillard's F/2 medium (Sigma-Aldrich Chemie, Taufkirchen, Germany) for TmmRU (media were prepared according to the manufacturers' directions). Strains were sub-cultured by serial transfers to fresh medium once every 20 days to ensure mid-log-exponential growth, and the axenicity was regularly tested. Culture conditions were continuous illumination at 80 µmol m −2 s −1 over the wavelength range from 400 to 700 nm and a temperature of 22 °C.
Prior to experiments, both microalgal cultures of ChlGS and TmmRU were analysed to ensure the absence of U before exposure to DU. No U was detected in either of the cultures.
Exposure experiments to DU. Here, we studied the U fractionation behaviour of two microalgal species artificially selected for U recovery 63,64 : the freshwater extremophile Chlamydomonas (strain ChlGS) isolated from a well-studied U acid mine pond 31,39 and the marine Tetraselmis (strain TmmRU) isolated from the Mediterranean sea only sparsely exposed to naturally occurring trace 235 U. Two different sets of experiments were performed with two different strains ChlGS and TmmRU. In both bioassays, twelve replicates were established in cell culture flasks with 20 mL of depleted uranium medium with approximately 2 × 10 6 cells and 0.5 × 10 6 cells per mL for ChlGS and TmmRU, respectively. In the case of ChlGS, the depleted uranium medium consisted in a preparation of 4 mg L -1 of U with an isotopic relation n( 235 U)/n( 238 U) of 0.005058 ± 0.000083 in bi-distilled water sterilized and enriched with BG-11. In the TmmRU bioassay, the depleted uranium medium consisted of a preparation of 2 mg L -1 of U with an isotopic ratio n( 235 U)/n( 238 U) of 0.002172 ± 0.000023 in bi-distilled water sterilized and enriched with F/2. Our experiments were performed in aqueous systems, oxidizing conditions and below Σ pH 7-8, conditions were the mobile U(VI) is predominantly found as uranyl ions (UO 2+ 2 ) or hydroxyl complexes 18,65,66 . pH was selected at the onset of the microalgal fractionation experiments to have predominantly soluble U species, so ligands of the cell walls would attract U cations though chemical sorption 67 . The U concentration for TmmRU was obtained from the dose-response curve and represented the higher dose before the IC 50 values for growth inhibition. In ChlGS strain, the U concentration used was selected based upon previous U uptake experiments 33 . At different times along the cells' growth curve (3, 12 and 24 days), four independent cultures were centrifuged at 4000 rpm for 15 minutes, and the two phases obtained (supernatant and pellet) of each replicate were frozen. Throughout the experiments, due to the media alkalization induced by the cells' photosynthetic activity, some U precipitation could take place.
The experimental media were supplied by the Energy, Environmental and Technological Research Center (Spanish acronym CIEMAT). All the samples were sent to CIEMAT for isotopic analysis and total U analyses. The isotopic relation results were obtained by averaging the data obtained in three analyses.
Uranium fractionation characterization. Algal pellets (mg) and filtered supernatants (5 mL) were treated by microwave acid digestion with 5 mL of 65% HNO 3 and 30% H 2 O 2 (4:1 v/v). For pellet samples, before microwave treatment, the sample-reagent mixture was sonicated. After digestion, the sample solution was evaporated almost to dryness, and the residue was dissolved to 5 mL with 3 M HNO 3 . One millilitre of this solution was brought to a final volume of 10 mL with MilliQ water to quantify the U by quadrupole-based ICP-MS (Q-ICP-MS) using the external calibration and internal standard method. The remaining volume was used for isotopic analysis after purifying U with UTEVA resin according to revised protocols reported by Carter et al. 68 and Weyer et al. 22 , summarized as follows. The sample solution was loaded onto a previously washed and pre-conditioned column. The column was rinsed with 2 × 5 mL of 3 M HNO 3 , 6 mL of 9 M HCl and 5 × 3 mL of 5 M HCl. Uranium was eluted in 5 × 3 mL of 0.02 M HCl. The collected uranium fraction was evaporated and the residue re-dissolved in 2 mL of concentrated HNO 3 , and the uranium was again evaporated. This last residue was dissolved in HNO 3 2%. Matrix separation and U concentration were checked in the fractions collected by Q-ICP-MS (iCAP Q, Thermo Scientific). Procedural blanks of the entire sample treatment procedure, including digestion and purification, were performed in every sample batch.
Due to the high uranium content, purified samples were diluted with HNO 3 2% before isotope ratio measurements to a concentration that matched that of the isotopic standard IRMM-053.
Isotope ratios were determined in a double-focusing sector field ICP-MS (Element 2, Thermo Scientific) equipped with a single collector and using a desolvating system (Cetac Aridus II) for sample introduction. This unit stabilized the ion beam and provided a 5-fold enhancement in sensitivity (10 × 10 6 cps per ng.g −1 ). Measurements were carried out under optimized conditions of the overall system to obtain the maximum accuracy and precision in isotopic measurements (for detail information of the procedure validation, see SI Material and Methods). For data acquisition, standard bracketing was used placing every sample between two isotopic standards (IRMM-053) and washing with HNO 3 2% before every sample and standard (see experimental settings in Supplementary Table S1).
Potential isotope fractionation due to chromatographic extraction was evaluated by measuring the 235 U/ 238 U ratio in the isotopic CRM as well as in the DU before and after passing through the column using the same elution protocol as in the samples.

Data Availability
All data generated or analysed during this study are included in this published article (and its Supplementary Information Files).