Biological Samples Preparation for Speciation at Cryogenic Temperature using High-Resolution X-Ray Absorption Spectroscopy

Summary

This protocol presents a detailed procedure to prepare biological cryosamples for synchrotron-based X-ray absorption spectroscopy experiments. We describe all the steps required to optimize sample preparation and cryopreservation with examples of the protocol with cancer and phytoplankton cells. This method provides a universal standard of sample cryo-preparation.

Abstract

The study of elements with X-ray absorption spectroscopy (XAS) is of particular interest when studying the role of metals in biological systems. Sample preparation is a key and often complex procedure, particularly for biological samples. Although X-ray speciation techniques are widely used, no detailed protocol has been yet disseminated for users of the technique. Further, chemical state modification is of concern, and cryo-based techniques are recommended to analyze the biological samples in their near-native hydrated state to provide the maximum preservation of chemical integrity of the cells or tissues. Here, we propose a cellular preparation protocol based on cryo-preserved samples. It is demonstrated in a high energy resolution fluorescence detected X-ray absorption spectroscopy study of selenium in cancer cells and a study of iron in phytoplankton. This protocol can be used with other biological samples and other X-ray techniques that can be damaged by irradiation.

Introduction

The study of the cellular biotransformations of essential or toxic elements requires speciation techniques with high sensitivity and should minimize sample preparation steps that are often prone to modification of chemical species.

Physiological elements such as selenium and iron are known to be particularly difficult to speciate due to their complex chemistry, various stabilities of the selenium or iron species, and their low concentration in the ppm (mg/kg) or even sub-ppm range. Thus, the study of the speciation of these elements by XAS can be extremely challenging. Synchrotron XAS and especially high energy resolution fluorescence detected XAS (HERFD-XAS), which allows a very low signal-to-background ratio¹, are available at synchrotron sources to speciate highly diluted elements in complex biological matrices^2,3. Conventional fluorescence-XAS measurements can be performed using an energy resolved solid state detector (SSD) with an energy bandwidth ~150–250 eV, on the CRG-FAME beamline at the European Synchrotron Radiation Facility (ESRF)⁴, while HERFD-XAS measurements need a crystal analyzer spectrometer (CAS), with an energy bandwidth ~1–3 eV, on the CRG-FAME-UHD beamline at the ESRF². Fluorescence photons are discriminated with respect to their energy with electronic or optical processes respectively.

The sample cryo-preparation is essential to preserve structures and maintain compositional chemical integrity, thus allowing analysis close to the biological native state⁵. Moreover, analyses performed at cryogenic temperatures as low as 10 K using liquid helium cryogenic cooling (LN₂), allow radiation damage to slow down and preserve elemental speciation for XAS. Although some reviews on XAS techniques applied to biological samples report the necessity to prepare and analyze samples in cryogenic conditions (e.g., Sarret et al.⁶, Porcaro et al.⁷), none of them clearly describes the related detailed protocol. In this publication, a method for cryo-preparation of cancer cells and plankton microorganisms is described for HERFD-XAS speciation of Se⁸ and Fe⁹ at cryogenic temperature.

Good practice for sample preparation and environment during state-of-the-art XAS spectroscopy measurements require 1) a setup; 2) an analysis procedure that limits the effects of radiation damage as much as possible; and 3) a sample (or model compound reference) as homogeneous as possible with respect to the X-ray photons beam size. The first item is taken into account by performing the acquisition at a low temperature, using a liquid helium cryostat. The second item is dealt with by performing each acquisition on a fresh area of the sample by moving it with respect to the beam. Finally, considering the third condition, samples (pellets) and references (powders) are conditioned in pressed bulk pellets in order to limit porosities and inhomogeneities as much as possible, and to avoid roughness with respect to the beam size on the X-ray probed sample surface. We explain how the protocol deals with all of these points.

We used human prostate cell line PC-3 (high metastatic potential) and ovarian cell line OVCAR-3 (which accounts for up to 70% of all ovarian cancer cases) to investigate the antiproliferative properties towards cancer cells of selenium nanoparticles (Se-NPs), and Phaeodactylum tricornutum diatom as a model species to investigate iron sequestration in phytoplankton.

Protocol

1. Preparation of the human PC-3 and OVCAR-3 cancer cell pellets for selenium speciation

NOTE: The following protocol is adapted from Weekley et al.¹⁰. All steps have to be carried out under a cell culture hood under biosafety level 2 conditions and restrictions, using aseptic techniques.

1. Count the cells using a Malassez cell counting chamber. Seed 150,000–200,000 cells per flask for the PC-3 cell line and 300,000 cells for the OVCAR-3 cell line.
2. Seed cells in T-75 flasks (three flasks per condition in order to have triplicates) in the adequate cell culture media (Table 1) under the laminar flow hood.
3. Leave the cell cultures to grow in an incubator at 37 °C and a humidified atmosphere with 5% CO₂ until they reach 80% confluence. Usually, PC-3 prostate cancer cell lines double after 24 h while OVCAR-3 ovarian cancer cell lines take 72 h. The flasks must be left in the incubator.
4. Meanwhile, dilute the Se-NPs used for treatment to a final concentration that corresponds to the IC20 (inhibitory concentration to achieve 20% cell death) that has been predetermined by the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay for cytotoxicity assessment. These concentrations are specific to each Se-NPs type but also specific to the studied cell line.
   1. Place the nanoparticle stock solution (2 mg/mL of bovine serum albumin [BSA] or chitosan coated Se-NPs) in an ultrasound water bath for 30 min at room temperature (RT).
   2. Dilute the Se-NPs solution by serial dilutions in complete cell culture medium to reach working concentrations. Between each dilution, vortex for 1 min.
5. Prepare a selenium positive control with diluted selenium salt for treatment.
   1. Prepare an aqueous sodium selenite stock solution (1 mg/mL) in a 15 mL polypropylene tube. Mix 1 mg of sodium selenite powder with 1 mL of ultrapure water.
   2. Vortex the stock solution for 1 min.
   3. Dilute the sodium selenite stock solution by serial dilutions in complete cell culture medium to reach working concentrations.
      NOTE: Do not forget to pipet several times before each dilution in order to homogenize the solution.
6. Expose the cancer cells to selenium treatments.
   NOTE: This part of the protocol has to be repeated for each nanoparticle (NP) tested. Here, the NPs are of two types, BSA and chitosan coated Se-NPs, plus the controls. The sodium selenite solution is the positive control and no treatment is the negative control.
   1. Open the three T-75 flasks containing the cells in the laminar flow hood.
   2. Remove the medium and wash the cells gently 2x with 5 mL of warmed (37 °C) phosphate buffered saline (PBS).
   3. Gently, on the bottom-side of the flask and not directly on the cell layer, add 15 mL of the treatment in each flask using an automatic pipette equipped with 25 mL sterile plastic pipettes. Place the lids on the flasks. Do not close the lids tightly.
      NOTE: As described in steps 1.3 and 1.4, treatment solutions must have been resuspended beforehand to be homogenized.
   4. Leave the three flasks horizontally in an incubator at 37 °C and an atmosphere humidified with 5% CO₂ for 24 h.
7. Preparation of the cell pellets
   1. Gently wash the cells 2x with 5 mL of warmed PBS (37 °C) directly in the T75 flasks.
   2. Add 6 mL of warmed cell culture medium (37 °C) on the cell layer in the T-75 flask using a pipette boy equipped with 10 mL sterile plastic pipettes.
   3. Collect the cells by gentle scraping using a cell scraper. Use one cell scraper per flask.
   4. Using a pipette boy and a 10 mL pipette, aspirate the liquid and flush back the cell/medium on the flask surface in order to collect all the cells. Repeat this step several times to dissociate and individualize the cells. Collect the medium containing the cells in a 15 mL polypropylene tube.
   5. Spin down at 250 x g for 5 min at RT. Aspirate the supernatant.
   6. Rinse the cells by resuspending them in 5 mL of PBS.
   7. Spin down at 250 x g for 5 min at RT. Aspirate the supernatant and repeat steps 1.6.5 and 1.6.6 2x to get rid of all remaining traces of the treatment.
   8. Resuspend the cells in 1 mL of PBS and transfer them into a 1.5 mL polypropylene tube. Finally, spin them down at 250 x g for 5 min at RT. Figure 1 represents the cell pellet obtained after centrifuging and discarding the supernatant.
   9. Gently remove all the PBS supernatant using a 200 µL pipette.
      NOTE: Be careful to not touch and damage the cell pellet. Ideally, the pellet should be 2–3 mm high or thick, with a diameter of 3–6 mm. For the PC-3 cell line, 9 x 10⁶–10 x 10⁶ cells are required for the assay. For the OVCAR-3 cell line, 8 x 10⁶–9 x 10⁶ cells are required (Figure 1). Some cells will likely be lost during the subsequent buffer rinsing steps. The higher the cell number, the better the pellet will be.
8. Flash-freezing of the cell pellets
   1. Plunge the bottom part of the 1.5 mL tube containing the pellet in liquid nitrogen (LN₂) to the level of the cell pellet.
   2. In a glove box fully purged with an inert atmosphere such as nitrogen gas, collect the cell pellet by gently tapping the 1.5 mL tube on the flat surface of a LN₂-cooled copper block (Figure 2). The pellet is then collected without cell loss.
      NOTE: Use cryoprotection equipment, lab coat, closed shoes, cryo-gloves, and a face shield or safety goggles for LN₂ handling.
   3. If needed, a LN₂ cooled needle can be used to help to take out the pellet.
      NOTE: The resulting frozen cell pellet can fragment. These steps must be performed in the space of a few minutes and rely on the researcher's dexterity and speed.
   4. The frozen pellet dimensions must fit the cryostat sample holder. Using the equipment described, if the cell pellet is larger than 2 mm in height and slightly smaller than the hole of the sample holder used for XAS, the sample can be used directly. If not, or if a flat surface for the analysis is desired, a bulk cylindrical pellet must be prepared, still keeping the pellet in its frozen state. If the cell pellet is fragmented the following steps can be used ideally in a glove box fully purged with an inert atmosphere.
9. Preparation of the bulk frozen cylindrical pellets
   1. Immerse all the parts of a manual hydraulic press that will be in contact with the sample in LN₂ (Figure 3A; 3- or 5-mm diameter pellet dies, mold, piston/wire pulling).
   2. Transfer the rapidly frozen pellet fragments into the mold loaded with the appropriate pellet dies (Figure 3B).
   3. Quick pelletize with the hydraulic press. Use of 1.5 tons for a 5 mm diameter pellet or 0.5 tons for a 3 mm diameter pellet is sufficient (Figure 3C).
      NOTE: After each pellet preparation with the hydraulic press, all the materials need to be thawed and dried properly using nitrogen gas to avoid any problems with the remaining frost and humidity.
   4. Collect the bulk pellet of frozen cells quickly and place it in an adequate cryotube for storage.
   5. Store it back into LN₂ tank for long-term storage.
   6. Transfer and mount the cryo-pellets for analysis. The pellet stored in a LN₂ cooled cryotube is transferred to a 77 K precooled cryostat sample holder in liquid N₂ (Figure 4, as used for a CRG-FAME-UHD beamline). This step should be performed as quickly as possible but can take a few minutes. Then transfer the sample holder to the cryostat and maintain it at 10 K during the whole XAS analysis.
      NOTE: Steps 1.9.3 and 1.9.6 are difficult because they need to be done quickly in order to avoid any frost formation that will block the piston and the mold. An alternative is to perform these steps under a plastic tent fully purged with nitrogen gas. The LN₂-cool copper mass allows the pellet to be kept frozen.

2. Preparation of the plankton cells for iron speciation

NOTE: Synthetic seawater used for this protocol was prepared by adding ultrapure water sea salts, morpholino propanesulfonic acid, ammonium nitrate, sodium nitrate, sodium metasilicate pentahydrate, sodium phosphate monobasic, vitamin stock, trace metal stock, antibiotic stock, and HEPES buffer at pH = 7.9. The concentrations of each component are indicated in the Table of Materials. Details on the culture can be found in Sutak et al.¹¹

1. Centrifuge a P. tricornutum diatom culture in a 50 mL polypropylene tube at 1,100 x g for 6 min and transfer the pellet into a flask with fresh medium.
2. Let the culture grow in synthetic seawater in a growth chamber for 24 h.
3. Centrifuge the plankton culture for 6 min at 1,100 x g and transfer the pellet into fresh medium containing Fe-citrate (1 µM in the final culture). Incubate for 72 h.
4. Prepare the plankton cells
   1. Centrifuge the cells for 3 min at 1,100 x g and rinse with medium without iron.
   2. Transfer the pellet into a 1.5 mL polypropylene tube, wash with ultrapure water, and centrifuge 2x for 3 min at 1,100 x g. Remove the supernatant.
   3. Plunge the bottom part of the 1.5 mL polypropylene tube containing the pellet in LN₂ as described in step 1.8.
   4. Transfer the frozen cells in a LN₂ frozen molder and quickly press using a XRF pellet press (Figure 3 and Figure 7) as described in step 1.9.
   5. Remove the pellet and store it in a LN₂ tank for long-term storage.
   6. Follow step 1.9.6 (see Figure 4 for the transfer into the cryostat sample holder).

3. Reference compound preparation and measurement

NOTE: Reference compounds (solid or liquid) representative of the species and expected to be found in the biological system must be prepared and analyzed by XAS for comparison with experimental biological samples. Reference spectra can be also found in databases^12,13,14 and can be used provided that the measurement conditions (e.g., spectral resolution) were similar to the experimental samples.

1. For references in aqueous solution, weigh initial compounds in powder or liquid form under anaerobic condition or inert atmosphere. Prepare redox-sensitive references in an inert atmosphere (i.e., in an anaerobic glove box or in a Schlenk line, see Figure 5 and Figure 6) to avoid any oxido-reductive reaction and to preserve the chemical integrity of the compound, which can be highly reactive and unstable in ambient air). Here, all the selenium references were prepared using a Schlenk line and degasified ultrapure water (Figure 6). Liquid Fe^III-malate and ferritine were prepared at ambient air.
2. Mix with appropriate solution in order to reach a 1% wt final concentration of the studied element (i.e., selenium or iron) for collection in fluorescence mode.
   NOTE: Ultrapure water is the most frequently used solution to prepare XAS references. Addition of glycerol (15%–20%) is required for liquids measured using standard XAS fluorescence to prevent artifacts arising from X-ray diffraction by ice crystals and preserved quality of the XAS signal collected with the solid-state detector. However, for HERFD-XAS, glycerol is not mandatory. Using a crystal analyzer spectrometer instead of a solid-state detector, the energy is selected with an extremely high-resolution and the diffraction induced by ice crystals does not impact the data collection of the studied element¹.
3. Preserve and store the prepared solution in a sealed Schlenk balloon (Figure 5) or in a 1.5 mL polypropylene tube until measurement in the same conditions as the samples.
4. For solid references, weigh selenium or iron model compound powders at ambient air or in a glove box if the species are redox-sensitive. Here, solid Fe^II references (Fe^II-acetate and vivianite) were prepared in a gaseous N₂ glove box while Fe^III (Fe(OH)₃, and Fe^III-phosphate) were prepared at ambient air.
5. Weigh high purity boron nitride powder and mix with the model compounds powders in order to reach a 1% wt final concentration of the studied element (Figure 7A).
6. Grind to a fine powder using a mortar for at least 15 min.
7. Place the powder mixture in the molder to prepare pellets with the hydraulic press (Figure 7B, similar to Figure 3A for the sample pellet).
8. Close the molder with the piston (Figure 7C, similar to Figure 3B for the sample pellet).
9. Press to obtain the bulk pellet (Figure 7D, similar to Figure 3C for the sample pellet).
10. Store the prepared bulk pellets in the glove box until they are analyzed in the same conditions as the samples.
    NOTE: The bulk pellet sometimes sticks to the pistons, depending on the powder compacity, for example. Removing it softly without breaking it then can be difficult. In that case, the following procedure using polyimide (e.g., Kapton) disks has to be used.
11. Put a drop of ethanol on each side of the pistons that will be in contact with the powder (Figure 8A).
12. Prepare two polyimide disks with a diameter slightly smaller than the pistons. The polyimide thickness needs to be 10–25 µm (thinner will be difficult to manipulate, thicker will absorb too much of the X-ray photons during analysis).
13. Put the disks on the pistons. They will stick by capillary action (Figure 8B).
14. Mount the molder with the piston, add the powder, and put in the second piston (Figure 8C).
15. Press (see step 3.11) and remove the compressed bulk pellet from the pistons.
    NOTE: The solid references can be mounted on the cryostat-specific sample holder like the samples (see step 1.8.6). The liquid references can be mounted on the cryostat-specific sample holder. See Figure 9A for a CRG-FAME cryostat sample holder. The same procedure can be applied using a CRG-FAME-UHD sample holder, shown in Figure 4D, using the following procedure.
16. Mounting of liquid reference on the cryostat sample holder.
    1. Set the polyimide (e.g., Kapton) tape (25 µm thick) in order to seal one side of the hole on the sample holder at RT (Figure 9B).
    2. Using a syringe, fill the cavity slowly with the solution until it reaches the top. Avoid bubbles. The reservoir must be filled (Figure 9C, left).
    3. Seal the other side of the cavity with the tape (Figure 9C, right).
    4. Plunge the sealed sample holder into LN₂ (Figure 9D, T₀–T₃).
    5. A thermic reaction induces LN₂ bubbling. Wait until bubbling stops, signaling the end of the reaction (Figure 9D, T₃–T₅).
    6. Transfer the sample holder at 77 K into the cryostat of the beamline before starting data collection (Figure 9D, T₆) or storing in the LN₂ Dewar.
       NOTE: The frozen solution is well spread in the cavity and forms a uniform and homogeneous pellet (Figure 9D, T₇).

4. HERFD-XAS: measuring procedure

1. Optimize all the crystals from the CAS in Bragg conditions with respect to the fluorescence line energy of interest. This procedure can be done with a concentrated reference compound.
2. Calibrate the incident monochromatic beam energy using a reference for which the energy position of the absorption edge is known (typically a pure metallic foil). This reference can be positioned in a second step after the sample, in double transmission mode, in order to be able to check the calibration for each obtained spectra and eventually align them posttreatment.
3. Position the sample in a liquid helium cryostat for biological samples following the appropriate procedure described in step 1.9.6.
4. Close the experimental hutch following the synchrotron safety rules.
5. Perform the XAS acquisition on an energy range that covers the selected absorption edge.
6. After each spectral acquisition, move the sample at least twice the beam size in the direction of the motion before starting a new acquisition.
   NOTE: In this case, measurements were performed using Ge(440) crystals to select the Fe K_a1 line and using Ge(844) crystals to select the Se K_a1 line. The beam size on the sample was 200 x 100 µm² (H x V Full Width Half Maximum). The sample motion between each acquisition was 500 µm in both directions, in order to probe structural homogeneity and to analyze a new area free from previous possible radiation damage each time. Individual spectra are compared and merged if they are superimposable within the noise. The measurement is stopped for a given sample when the quality of the merged spectrum is correct. Then data analysis can be performed using any software dedicated to XAS analysis, especially those that allow a linear combination fitting of normalized spectra, for example Athena and Artemis (Demeter group of software)¹⁵, SixPack¹⁶, or Viper¹⁷. Complete and detailed explanations of XAS analysis falls outside the scope of this protocol and can be found in recent papers^18,19, some of them more specifically devoted to biological samples²⁰. Selenium XANES reference spectra are already gathered in the SSHADE spectra database²¹.

Representative Results

The main aims of these preparations were to investigate the interaction between selenium nanoparticles (Se-NPs) and cancer cells, and iron binding and sequestration in phytoplankton.

HERFD-XANES spectra of the selenium in the initial state (BSA Se-NPs) and in cells incubated in nutritive medium (BSA Se-NPs after 24 h incubation) are shown in Figure 10. Results showed that selenium in the initial Se-NPs was present as both Se⁽⁰⁾ and selenite-like forms, whereas after interactions with PC-3 cells the selenium in cells was mainly present as Se⁽⁰⁾, thus demonstrating a change of selenium species in cells. For iron, HERFD-XANES spectra of references showed distinct edge positions depending on the iron oxidation state, with reduced species of iron shifted to low energy values (Figure 11). Two successive spectra collected on the same position of a diatom pellet were similar, indicating that beam damage was limited between two acquisitions when using a He-cryostat at 10 K. Furthermore, spectra from different positions of the diatom pellet (here three positions) were identical, demonstrating that the sample pellet was homogeneous, and that the spectra could be averaged to obtain a better signal-to-noise ratio spectrum (averaged spectrum). The averaged spectrum for diatoms shows edge features mostly corresponding to Fe(III). Linear combination fitting of the reference spectra was then performed to quantify the proportion of the iron species as described in Sarret et al.⁶.

Figure 1: Typical cell pellets. The 1.5 mL polypropylene tubes (OVCAR-3 cells left, PC-3 cells right) containing 8 x 10⁶ cells and 1 x 10⁷ cells, respectively. Credit: Caroline Bissardon. Please click here to view a larger version of this figure.

Figure 2: Preparation in anaerobic glove box. (A) Anaerobic glove box containing a (B) 1.5 mL polypropylene tube rack (blue) entirely immersed in liquid LN₂. The liquid LN₂ is not presented in the picture for clarity. Please click here to view a larger version of this figure.

Figure 3: Schema of cell pelleting. (A) Sample pellet in the press before loading. (B) Sample during loading. (C) Sample bulk pellet after loading. Please click here to view a larger version of this figure.

Figure 4: Mounting of the cryo-sample(s). Mounting in the sample holder specific to BM16 XAS beamline at ESRF. Please click here to view a larger version of this figure.

Figure 5: Anoxic preparation of reference solution. Examples of selenium reference preparations stored in Schlenk balloons. Please click here to view a larger version of this figure.

Figure 6: Setup for anoxic preparation of reference solutions. Setup of the Schlenk ramp with degasified ultrapure water in the bottle on the right. Please click here to view a larger version of this figure.

Figure 7: Preparation in solid reference pellet. Preparation of pressed cylindrical pellets for powder compounds either in ambient air or under the glove box. (A) Weight of the reference powder. (B) The powder in the cylindrical hole of the press support. (C) Closed press support. (D) Pressing. Please click here to view a larger version of this figure.

Figure 8: Preparation of a solid reference pellet using polyimide disks. Preparation of pressed cylindrical pellet for sticky or brittle powder compounds. (A) A drop of ethanol on each side of the pistons that will be in contact with the powder will allow the polyimide disks on the pistons to stick to them (B). The molder is mounted with the piston. Then, the powder can be deposited, and the molder can be closed with the second piston (C). Please click here to view a larger version of this figure.

Figure 9: Preparation steps for the cryostat sample holder for a liquid reference or sample. For picture clarity, photos were taken outside the glove box and without hazardous substances. Here, we used ddH₂O water. If necessary, this preparation can be made under a glove box or inert atmosphere (N₂) plastic tent. (A) Sample holder. (B) Place polyimide tape on the curved surface to seal the hole. (C) Slowly inject the reference solutions drop by drop using a syringe until the cavity is filled. No air bubbles must remain. Seal the other hole with polyimide tape. (D) Plunge in LN₂. Please click here to view a larger version of this figure.

Figure 10: Se K-edge typical HERFD-XANES analysis. Se K-edge HERFD-XANES of BSA-coated Se-NPs, as received and left 24 h in cell culture medium, and PC-3 cell exposed to BSA-coated Se-NPs. Spectra are shifted for clarity. Please click here to view a larger version of this figure.

Figure 11: Fe K-edge typical HERFD-XANES analysis. Example of the spectra collected during our experiment on the iron speciation in plankton. The six upper spectra correspond to the references/standards. The superimposed spectra (in red and black) correspond to two spectra collected on the same position of the pellet. The spectra labeled pos#1, pos#2, and pos#3 were collected on three different positions of the sample pellet. The spectrum called Diatoms Average corresponds to the averaged spectra collected on various positions of the pellet and is the sample for which the speciation needs to be determined. Please click here to view a larger version of this figure.

Products	PC3 cell line	OVCAR3 cell line
ATCC modified RPMI 1640 medium	445 mL	394.5 mL
Foetal bovine serum	50 mL	100 mL
Penicillin-Streptomycin	5 mL	5 mL
Bovine insulin	-	500 µL

Table 1. Cell culture media preparation. The cells are cultured in American Type Culture Collection (ATCC) modified RPMI 1640 medium supplemented with 10% of fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS) for PC-3 prostate cancer cells and with 20% of FBS and 1% penicillin-streptomycin and 0.01 mg/mL bovine insulin for OVCAR-3 ovarian cancer cells.

Discussion

This protocol was used to study the chemical form of selenium and iron in biological samples by X-ray absorption spectroscopy. It focuses on the cryo-preparation and storage of biological samples and references compounds, as well as on the HERFD-XAS measurements.

Cryo-preparation and storage
The cryo-preparation of the bulk biological sample pellets allows preservation of the chemical integrity of the species present in the samples. This is crucial, because speciation changes have been observed when using freeze-drying or air-drying for preparation⁶. This protocol should be used as soon as the beamline selected for measurements is equipped with a helium cryostat.

For reference compounds, it is necessary to work in an anoxic atmosphere when using redox-sensitive elements to preserve the oxidation state. This can then be checked with variations in spectral edge position as shown for ferrous and ferric reference compounds (Figure 11).

This sample preparation can be performed before the synchrotron or laboratory analysis. In this case, the frozen pellet or reference should be transferred into a cryotube and stored in a LN₂ tank. This LN₂ storage is mandatory to provide a protective inert environment and avoid changes in the chemical species, particularly for reactive iron or seleno-compounds. Preferably, samples should be prepared just before measurement or stored a few days prior to analysis. It is best to not store samples for a long period of time (i.e., months).

Cryo-analysis
Analysis at low temperatures, ideally at 10 K, is highly advocated in order to slow down damage induced by the intense X-ray beam, particularly the formation of free radicals from the hydrolysis of water, which can damage the protein matrix and create photoreduction of transition metal ions or photoreduction of elements such as sulfur²². It is best to take into account these possible changes in the chemical species through fast acquisition of the XAS spectrum. The sample should also be scanned to expose a nonirradiated area for each spectrum. As demonstrated by the spectra collected on the three different positions of the phytoplankton pellet (Figure 11), such a strategy reveals powerful spectra, which can then be averaged to obtain a better signal-to-noise ratio.

Future applications
In our work, we used a beamline dedicated to HERFD-XAS measurements (FAME-UHD, ESRF, France) but this protocol for biological sample preparation can be applied on any standard XAS beamline equipped with a cryostat while using other types of detectors such as multielement Ge Solid-State-Detector or Silicon-Drift Detector. The proposed workflow can be also applied to any other chemical elements or any biological material expected to be studied by X-ray absorption spectroscopy.

Materials

Name	Company	Catalog Number	Comments
Ammonium nitrate	Sigma-Aldrich	A3795	NH4NO3, 2.66 mg/L of milliQ water
Anaerobic chamber	Coy Laboratory, USA		equipped with Anaerobic Monitor (CAM-12)
Antibiotic stock	Sigma-Aldrich	A0166 for ampicillin, S9137 for streptomycin sulfate	1 mL/L of milliQ water (ampicillin sodium and streptomycin sulfate, 100 mg/mL)
Boron nitride powder	Sigma-Aldrich	255475
Cell counting chamber			Neubauer or Malassez
Cell scraper
Dulbecco's Phosphate Buffered Saline (DPBS)	GIBCO	14190-094	Without Calcium, Magnesium, Phenol Red
Eppendorf tubes			0.5 mL and 1.5 mL
Falcon tubes			15 mL and 50 mL
Ferric citrate Fe/citrate = 1/20	Sigma-Aldrich	F3388	aqueous solution of FeCl3 50 mM and Na-citrate 1M pH 6.5
Fetal Bovine Serum	GIBCO	A31604-02	Performance Plus, certified One Shot format, US origin
Flasks	Sigma-Aldrich	Z707503	TPP 150 cm2 area
Growth chamber	Sanyo	Sanyo MLR-352	at 20 °C and under a 12:12 light (3,000 lux) dark regime
HEPES buffer	Sigma-Aldrich	H4034	1 g/L of milliQ water HEPES
High grade serous, OVCAR-3	ATCC, Rockville, MD	HTB-161	Storage temperature: liquid nitrogen vapor temperature
Incubator			Incubator at 37°C, humidified atmosphere with 5% CO2
Insulin solution from bovine pancreas	Sigma-Aldrich	I0516	10 mg/mL insulin in 25mM HEPES, pH 8.2, BioReagent, sterile-filtered, suitable for cell culture
Manual hydraulic press	Specac, USA
Marine diatom Phaeodactylum tricornutum	Roscoff culture collection	RCC69	http://roscoff-culture-collection.org/rcc-strain-details/69
Morpholinepropanesulfonic acid	Sigma-Aldrich	M3183	MOPS, 250 mg/L of milliQ water (pH 7.3)
Optical microscope
PC-3	ECCAC, Salisbury, UK	90112714	Storage temperature: liquid nitrogen vapor temperature
Penicillin-Streptomycin	Sigma-Aldrich	P4333	Solution stabilized, with 10,000 units penicillin and 10 mg streptomycin/mL, sterile-filtered, BioReagent, suitable for cell culture
Pipette-boy			25mL-, 10mL-, and 5mL sterile plastic-pipettes
Plankton culture products, Mf medium: Sea salts	Sigma-Aldrich	S9883	40g/L of milliQ water. Composition: Cl- 19.29 g, Na+ 10.78 g, SO42- 2.66 g, Mg2+ 1.32 g, K+ 420 mg, Ca2+ 400 mg, CO32- /HCO3- 200 mg, Sr2+ 8.8 mg, BO2- 5.6 mg, Br- 56 mg, I- 0.24 mg, Li+ 0.3 mg, F- 1 mg
Plastic tweezers	Oxford Instrument	AGT 5230
RPMI MEDIUM 1640 (ATCC Modification)	GIBCO	A10491-01	Solution with 4.5 g/L D-glucose, 1.5 g/L Sodium Bicarbonate, 110 mg/L (1 mM) Sodium Pyruvate, 2.388 g/L (10 mM) HEPES buffer and 300 mg/L L-glutamine for research use
Selenium nanoparticles (Se-NPs), BSA coated, 2 mg/mL	NANOCS Company, USA	Se50-BS-1	BSA stabilized Se-NPs solution. Average size about 30 nm. Stored at 4°C in the dark, protected from the light.
Selenium nanoparticles (Se-NPs), Chitosan coated, 2 mg/mL	NANOCS Company, USA	11. Se50-CS-1	Chitosan stabilized Se-NPs solution. Average size about 30 nm. Stored at 4°C in the dark, protected from the light.
Sodium metasilicate pentahydrate	Sigma-Aldrich	71746	Na2SiO3.5H2O, 22.8 mg/L of milliQ water
Sodium nitrate	Sigma-Aldrich	S5022	NaNO3, 75 mg/L of milliQ water
Sodium phosphate monobasic	Sigma-Aldrich	S5011	NaH2PO4, 15 mg/L of milliQ water
T-75 flasks
Tissue culture hood
Trace metal stock	Sigma-Aldrich	M5005, Z1001, M1651, C2911, 450243, 451193, 229857	1 mL/L of milliQ water (MnCl2.4H2O 200 mg/L, ZnSO4.7H2O 40 mg/L, Na2MoO4.2H2O 20mg/L, CoCl2.6H2O 14 mg/L, Na3VO4.nH2O 10 mg/L, NiCl2 10 mg/L, H2SeO3 10 mg/L)
Trypan Blue Solution (0.4%)	GIBCO	15250061
Trypsin-EDTA (0.05%), phenol red	GIBCO	25300-054
Vitamin stock	Sigma-Aldrich	T1270 for thiamine, B4639 for biotin, V6629 for B12	1 mL/L of milliQ water (thiamine HCl 20 mg/L, biotin 1 mg/L, B12 1 mg/L)
Water bath 37°C