Introduction
In 2021, NASA’s Mars 2020 mission landed the Perseverance rover in Jezero Crater on the Red Planet where it is currently collecting samples of regolith and rock from the upper 10 cm of the surface. This first step in the planned NASA-ESA Mars Sample Return Campaign would be followed by the Mars Sample Return (MSR) Program, which would develop and operate the flight projects required to obtain samples directly from Perseverance and return the samples safely to Earth in the 2030s. Upon return to Earth, the samples would be managed by the Sample Receiving Project (SRP), which would provide containment until the samples were assessed to be safe for release or sterilized to mitigate any potential biological hazards.
As currently envisioned, MSR Program-specific objectives include: (a) acquiring and returning to Earth a scientifically selected set of Mars samples for investigation in terrestrial laboratories, (b) ensuring the scientific integrity of those samples (i.e., controlling contamination, managing exposure to scientifically relevant environmental factors, and maximizing sample mechanical integrity), and (c) complying with planetary protection requirements associated with the return of Mars samples to Earth’s biosphere (Beaty et al., Reference Beaty, Grady, McSween, Sefton-Nash and Carrier2019; Kminek et al., Reference Kminek, Meyer, Beaty, Carrier, Haltigin and Hays2022).
Primary SRP objectives include recovering the returned samples safely and securely transporting from the landing site. Additionally, SRP is responsible for designing, constructing, and operating the sample receiving facility (SRF) to contain the samples, while maintaining sample integrity, facilitating the sample safety assessment, and performing time-sensitive science measurements (McCubbin et al., Reference McCubbin, Farley, Harrington, Hutzler and Smith2025). Lastly, the SRP will facilitate the release of the samples from high containment to the broader scientific community and complete the MSR Campaign science objectives. One key function of an SRF would be sterilization both as a standard method for decontaminating lab tools and surfaces that may have contacted the samples directly, as well as a means to safely release samples from containment to external laboratories while the samples are assessed via a standard protocol of tests (McCubbin et al., Reference McCubbin, Farley, Harrington, Hutzler and Smith2025).
To comply with planetary protection requirements, spaceflight missions must limit biological contamination on the outbound leg (known as forward planetary protection, FPP) and protect Earth’s biosphere from the potential for harm that might result from the return leg (known as backward planetary protection, BPP). Approaches to FPP largely focus on minimizing the number of viable organisms that might reach a target body like Mars and proliferate in such a way as to compromise scientific study of pre-biotic chemistry and, possibly, biology beyond Earth. Similarly, approaches to BPP are focused on preventing the release of unsterilized extraterrestrial material that may have the potential to harm ecosystems or populations on Earth. Note that since a sample return mission would deliver only a small amount of material (MSR plans to return approximately 500 grams of Mars material), the key focus for planetary protection is on preventing proliferation of any potential extraterrestrial biology on Earth, not containing hazards (reactive chemistry, toxicity) posed by the material itself.
The current environment in the upper few centimeters of Mars is thought to have a very low likelihood of hosting active biological organisms and bioactive molecules due to high radiation exposure, lower water content, and lower temperature than conditions required for active metabolism (Orosei et al., Reference Orosei, Lauro, Pettinelli, Cicchetti, Coradini, Cosciotti, Di Paolo, Flamini, Mattei, Pajola, Soldovieri, Cartacci, Cassenti, Frigeri, Giuppi, Martufi, Masdea, Mitri, Nenna, Noschese, Restano and Seu2018; Craven et al., Reference Craven, Winters, Smith, Lalime, Mancinelli, Shirey, Schubert, Schuerger, Burgin, Seto, Hendry, Mehta, Benardini and Ruvkun2021). In addition, host-pathogen relationships are evolutionary in nature and Mars-Earth exchange is infrequent, making any pathogens likely non-specific to Earth species, including humans, animals, and plants. Thus, any putative Mars organisms living in the environments sampled would encounter a largely inhospitable environment on Earth, and meteoritic inputs have already delivered untreated (not actively inactivated) Martian material throughout Earth’s history without any apparent ill effects. Nonetheless, due to the paucity of life detection experiments performed on Mars material to date, the risk posed in deliberately delivering such material to Earth as part of a spaceflight mission is non-zero. Since there remains possibility that Mars material could pose a hazard to Earth’s biosphere, NASA is implementing a “safety first strategy” of containing or inactivating (sterilizing) Mars material with the potential to host harmful biology through a process called “breaking-the-chain” (BTC) wherein any hardware contaminated is either contained or sterilized, thus breaking the chain of contact between Earth and Mars (Cataldo et al., Reference Cataldo, Affentranger, Clement, Glavin, Hughes, Hall, Sarli and Szalai2024).
To ensure an effective inactivation process for hypothetical Mars biology, the MSR Program’s Sterilization Team worked to establish inactivation and sterilization processes tailored to specific potential hazards in order to develop a range of options for decontaminating select portions of the returned hardware on the return trip to Earth. While the MSR Program is currently refining its BTC approach and may not choose to implement sterilization, developing methods to confirm inactivation of bioactive molecules may also enable a wider selection of treatment options and parameters for the SRP as well as future restricted-Earth return sample missions.
This study is focused on direct mitigation of an extreme example from the suggested scope of potential threats, namely bioactive molecules (per NASA Procedural Requirement 8715.24). While the risk from such molecules and other sub-cellular entities, such as viruses “can be considered to be far lower [than cellular entities] and almost negligible” (Ammann et al., Reference Ammann, Barros, Bennett, Bridges, Fragola, Kerrest, Marshall-Bowman, Raoul, Rettberg, Rummel, Salminen, Stackebrandt and Walter2012), the MSR Program established specific studies to identify robust method for mitigating sub-cellular, extraterrestrial biology. The fundamental assumption in addressing such a broad scope of potential molecules is that potential Mars-based biology must adhere to Earth-based physics, which bounds the physical and biochemical properties (e.g., carbon-based chemistry with the same types of chemical bonds) of Mars-based life. To date, all available knowledge indicates that the fundamental chemistries of life that occur on Earth (e.g., covalent, ionic, and hydrogen bonds between molecules) are also in place on Mars (Farley et al., Reference Farley, Stack, Shuster, Horgan, Hurowitz, Tarnas, Simon, Sun, Scheller, Moore, McLennan, Vasconcelos, Wiens, Treiman, Mayhew, Beyssac, Kizovski, Tosca, Williford, Crumpler, Beegle, Bell, Ehlmann, Liu, Maki, Schmidt, Allwood, Amundsen, Bhartia, Bosak, Brown, Clark, Cousin, Forni, Gabriel, Goreva, Gupta, Hamran, Herd, Hickman-Lewis, Johnson, Kah, Kelemen, Kinch, Mandon, Mangold, Quantin-Nataf, Rice, Russell, Sharma, Siljeström, Steele, Sullivan, Wadhwa, Weiss, Williams, Wogsland, Willis, Acosta-Maeda, Beck, Benzerara, Bernard, Burton, Cardarelli, Chide, Clavé, Cloutis, Cohen, Czaja, Debaille, Dehouck, Fairén, Flannery, Fleron, Fouchet, Frydenvang, Garczynski, Gibbons, Hausrath, Hayes, Henneke, Jørgensen, Kelly, Lasue, Le Mouélic, Madariaga, Maurice, Merusi, Meslin, Milkovich, Million, Moeller, Núñez, Ollila, Paar, Paige, Pedersen, Pilleri, Pilorget, Pinet, Rice, Royer, Sautter, Schulte, Sephton, Sharma, Sholes, Spanovich, St. Clair, Tate, Uckert, VanBommel, Yanchilina and Zorzano2022; Craven et al., Reference Craven, Winters, Smith, Lalime, Mancinelli, Shirey, Schubert, Schuerger, Burgin, Seto, Hendry, Mehta, Benardini and Ruvkun2021). The 95% CO2 atmosphere of Mars is conducive for carbon-based chemistry (Franz et al., Reference Franz, Trainer, Malespin, Mahaffy, Atreya, Becker, Benna, Conrad, Eigenbrode, Freissinet, Manning, Prats, Raaen and Wong2017).
To establish an inactivation process, the MSR Program explored the use of dry heat sterilization with prion-like proteins as the test case for inactivation of proteinaceous material, as an analog for bioactive molecules referenced in NASA requirements (NASA Procedural Requirement 8715.24). Prions are purported to be the most heat-resistant biologically produced, replicating entities described to date. Heat denaturation alone, i.e., unfolding of the protein, is insufficient for prion inactivation, as infectivity is retained after heat denaturation (Marín-Moreno et al., Reference Marín-Moreno, Aguilar-Calvo, Moudjou, Espinosa, Béringue and Torres2019; Langeveld et al., Reference Langeveld, Balkema-Buschmann, Becher, Thomzig, Nonno, Andréoletti, Davidse, Di Bari, Pirisinu, Agrimi, Groschup, Beekes and Shih2021). Degradation of protein primary structure, i.e., breaking of amide bonds in the protein backbone, is thus a more reliable means of ensuring inactivation (Orsini et al., Reference Orsini, Bramanti and Bonaduce2018). Thus, MSR identified amide bond degradation as a primary goal for inactivating bioactive molecules.
In the present study, we utilized differential scanning calorimetry (DSC), Fourier transform infrared (FTIR) spectroscopy, analytical reversed-phase high-performance liquid chromatography (RP-HPLC), and mass spectrometry (MS) to examine the temperature and time requirements for sufficient inactivation/degradation of a heat-stable prion protein (Sup35NM). Sup35p is a subunit of the translation termination factor of Saccharomyces cerevisiae (Frolova et al., Reference Frolova, Le Goff, Rasmussen, Cheperegin, Drugeon, Kress, Arman, Haenni, Celis, Phllippe, Justesen and Kisselev1994; Stansfield et al., Reference Stansfield, Jones, Kushnirov, Dagkesamanskaya, Poznyakovski, Paushkin, Nierras, Cox, Ter-Avanesyan and Tuite1995). A self-propagating form of Sup35p results in the [PSI +] prion (Paushkin et al., Reference Paushkin, Kushnirov, Smirnov and Ter-Avanesyan1996; King et al., Reference King, Tittmann, Gross, Gebert, Aebi and Wüthrich1997; Glover et al., Reference Glover, Kowal, Schirmer, Patino, Liu and Lindquist1997). Sup35p is composed of a prion propagation domain at the N-terminal end, a highly charged middle domain (M), and a translation termination function domain at the C-terminal end (Kushnirov et al., Reference Kushnirov, Kochneva-Pervukhova, Chechenova, Frolova and Ter-Avanesyan2000). Prion formation from Sup35p has been studied extensively using a shortened form of the protein composed of the N-terminal and M domains (residues 1-253, designated Sup35NM). Sup35NM can adopt a variety of distinct amyloid core structures (Tanaka et al., Reference Tanaka, Collins, Toyama and Weissman2006) and has been shown to propagate as a prion in neuroblastoma cells (Krammera et al., Reference Krammera, Kryndushkinb, Suhrec, Kremmerd, Hofmanne, Pfeifere, Scheibelc, Wicknerb, Schätzla and Vorberg2009). Sup35NM amyloid fibril formation is the result of either β-strands aligned in a helical pattern around the fibril axis (Kishimoto et al., Reference Kishimoto, Hasegawa, Suzuki, Taguchi, Namba and Yoshida2004; Krishnan and Lindquist, Reference Krishnan and Lindquist2005; Frederick et al., Reference Frederick, Michaelis, Caporini, Andreas, Debelouchina, Griffin and Lindquist2017) or formation of an in-register parallel β-sheet along the fibril axis (DePace and Weissman, Reference DePace and Weissman2002; Shewmaker et al., Reference Shewmaker, Wickner and Tycko2006; Frederick et al., Reference Frederick, Michaelis, Caporini, Andreas, Debelouchina, Griffin and Lindquist2017). The most stable forms of Sup35NM and full-length Sup35p transition from polymers to monomers with a T m of 65–70°C (Krammera et al., Reference Krammera, Kryndushkinb, Suhrec, Kremmerd, Hofmanne, Pfeifere, Scheibelc, Wicknerb, Schätzla and Vorberg2009; Alexandrov et al., Reference Alexandrov, Polyanskaya, Serpionov, Ter-Avanesyan and Kushnirov2012). Treatment of Sup35NM fibrils at 100°C for 10 min led to fibril dissociation and subsequent susceptibility to proteolytic digestion (Chen et al., Reference Chen, Rojanatavorn, Clark and Shih2005).
It is postulated that, due to the inherent stability and activity of Sup35MN, the temperature and treatment duration requirements for inactivation of Sup35MN should be the upper limit for sufficient inactivation of heat-stable proteins and thus these conditions would sterilize Mars material.
Methods
Differential Scanning Calorimetry (DSC)
Bovine serum albumin (BSA) was purchased from Sigma-Aldrich. Prion protein Sup35NM, sequence given in (Ghosh et al., Reference Ghosh, Dong, Wall and Frederick2018), was expressed in E. coli by GenScript Biotech Corporation (Piscataway, NJ). BSA (10 mg) or Sup35NM (1.5 mg) was analyzed by DSC (Mettler Toledo Stare System Polymer DSC differential scanning calorimeter with Stare data acquisition) using a 50°C/min temperature ramp to 350 or 400°C, holding 5 sec, and immediately cooling.
Sample preparation for analysis
1 mg of BSA or Sup35NM was dissolved in 1 mL 5% formic acid in HPLC grade water/methanol (1:1). Half of the dissolved BSA sample was sonicated using a sonic dismembrator for 3-15 sec. Samples were filtered for analytical HPLC using Titan3 filters (17 mm, 0.45 μm, Thermo Scientific).
Analytical methods
Fourier transform infrared (FTIR) spectroscopy was performed using a Bruker Vectra 80 Fourier Transform Infrared spectrometer. The BSA and Sup35NM samples were cast from a liquid solution onto KBr plates and examined using a temperature-controlled Diffuse Reflectance attachment (Pike Technologies).
Analytical RP-HPLC was performed on an Agilent 1260 Infinity system equipped with a Regis Evoke c18 column (3 μm, 4.6 mm × 150 mm). The gradient conditions were 2% B for 2 min, 2-98% B in 20 min, and 98% B for 1 min, where A = 0.1% TFA in H2O and B = 0.1% TFA in acetonitrile. The flow rate was 1 mL/min with detection at λ = 220 and 280 nm. Chromatograms with detection at λ = 220 nm are indicative of the protein amide backbone, while chromatograms with detection at λ = 280 nm are indicative of aromatic side chains. Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) was performed on a Bruker Microflex using α-cyano-4-hydroxycinnamic acid matrix (Sigma). Laser power required for sufficient protein and/or peptide ionization without inducing fragmentation are reported for each analysis.
Software
In order to evaluate the amino acid sequences of fragments represented in the mass spectral (MS) data, we wrote a PERL script (Script S1). For this purpose, masses of amino acid residues from Matrix Science were utilized (http://www.matrixscience.com/help/aa_help.html). All the unique fragments, starting from a fragment size of one to all 253 residues, were determined for the sequence representing the prion protein Sup35NM (Table 1). Fragment identification was performed with and without including H2O. Furthermore, monoisotopic and average mass analyses were performed separately. Predicted masses of each fragment were compared to experimental MS data (Table S1). The complete PERL script is given in the Supporting Information.
Table 1. Putative Sup35NM fragment identification
Parent Sequence (253 residues) = MSDSNQGNNQQNYQQYSQNGNQQQGNNRYQGYQAYNAQAQPAGGYYQNYQGYSGYQQGGYQQYNPDAGYQQQYNPQGGYQQYNPQGGYQQQFNPQGGRGNYKNFNYNNNLQGYQAGFQPQSQGMSLNDFQKQQKQAAPKPKKTLKLVSSSGIKLANATKKVGTKPAESDKKEEEKSAETKEPTKEPTKVEEPVKKEEKPVQTEEKTEEKSELPKVEDLKISESTHNTNNANVTSADALIKEQEEEVDDEVVND.
Results
Bovine serum albumin (BSA) was used to establish the method parameters and feasibility. BSA is not known to be thermally resistant to degradation but is a readily available protein. Once the test methods were established, they were applied to the prion protein Sup35NM, which was available in much smaller quantities.
Differential scanning calorimetry (DSC) was utilized to characterize protein structural transitions by measuring the change in molecular energy upon heating at a constant rate (50°C/min). DSC analysis of BSA following the first heat cycle showed a broad endothermic transition at ∼75°C and a sharper transition at ∼215°C (Figure 1). The transition at ∼75°C was most likely due to protein denaturation as prior DSC analyses of BSA showed protein denaturation at 64-69°C (Giancola et al., Reference Giancola, De Sena, Fessas, Graziano and Barone1997; Michnik, Reference Michnik2003). The transition at ∼215°C was likely due to protein degradation (breaking of amide bonds), in agreement with prior pyrolysis results (Bikaki, Reference Bikaki2019). Following a cooling cycle, a second heat cycle was performed to verify the permanence of chemical alterations made to the protein by the first heat cycle. As expected, the second heat cycle showed relatively little activity compared to the first heat cycle, suggesting that the alterations of BSA made at ∼215°C were permanent.
Figure 1. DSC analysis of BSA. The blue line indicates initial heating of the sample with the two thermal transitions corresponding to protein denaturation and degradation of the primary protein structure, respectively. The brown line shows reheating of the sample where no further transitions occur indicating permanent degradation of the sample during the first heat cycle. The gray line corresponds to the cooling period between the two heating cycles.
For further elucidation of the thermal degradation, BSA was examined using FTIR spectroscopy (Figure 2). FTIR reveals the loss of water and amide bonds during denaturation and pyrolysis. The protein samples were thermally processed using a temperature controlled, diffuse reflectance cell in an FTIR. The diffuse reflectance infrared Fourier transform method applied herein provides sensitive, quantitative analysis (Fuller and Griffiths, Reference Fuller and Griffiths1980; Herrick et al., Reference Herrick, Dyer, Guy, Lee, Soules, Anderson, Chen and Lee2002; Chalmers and Griffiths, Reference Chalmers, Griffiths, Chalmers and Griffiths2002; Averett and Griffiths, Reference Averett and Griffiths2006). Thermal processing up to 500°C was examined. The loss of water, disruption of secondary structure, and decomposition of the protein backbone (amide bonds) are revealed. The infrared absorption spectrum bands for amides are amide I (1600–1800 cm–1), amide II (1470–1570 cm–1), amide III (1250–1350 cm–1), and amide A (3300–3500 cm–1). Due to overlapping bands from char formation, the degradation is most clearly revealed in the loss of the amide II and the amide III bands. After heating BSA to 400°C no proteinaceous material was detected (Figure 2). Amide functional group loss indicated decomposition of the protein backbone. The protein amide I and amide II peaks were not present in the 400°C spectra after 7 min at 50 °C/min ramp rate. The peaks observed do not match the peak position or shape of protein amide bonds.
Figure 2. Infrared monitoring of BSA pyrolysis as a function of increasing temperature. The loss of water and amide functional groups are noted by decreased spectral intensities. The temperature increase is from 30°C (light blue line) to 100°C, 150°C, 200°C (darker blue lines), 250°C (yellow line), 300°C, 350°C, 400°C, 450°C, and 500°C (progressively darker red lines).
BSA was examined by RP-HPLC to determine the retention time of the intact protein and to allow for quantification of the protein. The intact protein was found to elute at 12.96 min (Figure 3). The peak eluting at 1.84–2.02 min in the λ = 220 nm HPLC analysis contained buffer components.
Figure 3. RP-HPLC analysis of BSA with monitoring at (top) λ = 220 nm and (bottom) λ = 280 nm.
To facilitate detection of intact BSA after heat treatment to induce degradation, a standard curve was constructed by analyzing BSA concentrations ranging from 0.156 to 10 μg and plotting the amount of protein injected versus the area of the integrated peak eluting at 12.96 min (Figure 4). The limit of detection was at 0.156 μg.
Figure 4. BSA standard curve generated using species eluting at 12.96 min by HPLC.
Next, BSA was heated to 400°C and the amount of intact protein and the presence of degradation products were examined by RP-HPLC. After heating, the sample was dissolved in 1 mL 5% formic acid in HPLC grade water/methanol (1:1) and either not sonicated or sonicated to potentially enhance solubilization of materials. The protein amide bond backbone was monitored at λ = 220 nm, which provides insight into the amount of intact protein before and after treatment. No peak eluting at 12.96 min was visible in the chromatograms after heating to 400°C (Figure 5), indicative of protein degradation. RP-HPLC of BSA following 400°C treatment was repeated using sonication to potentially enhance solubilization of the charred material, but no peak corresponding to intact BSA (retention time = 12.96 min) was visible (Figure 5).
Figure 5. RP-HPLC analysis of the protein amide bond backbone of BSA, untreated (top panel), heated to 400°C and not sonicated (middle panel), and heated to 400°C followed by sonication (bottom panel) with monitoring at λ = 220 nm.
RP-HPLC analysis of BSA was also monitored at λ = 280 nm, which allows for visualization of aromatic species, providing insight into the amino acid side-chains of Phe, Tyr, and Trp and their aromatic breakdown products. As described above, the heat-treated sample was analyzed before and after sonication to enhance solubilization of the charred materials. The resulting chromatograms did not show any species eluting at the same retention time as the intact protein (Figure 6). Expansion of the relative absorbance (Figure 6, middle and bottom panels y-axis) indicated that even small amounts of the intact protein (with a retention time of ∼12.9–13.0 min) were not present. Sonication did enhance solubilization of additional constituents with retention times between 12.7 and 15.1 min (Figure 6, compare middle and bottom panels). Based on the differences in the absorbances, it can be concluded that relatively small amounts of peptide fragments containing aromatic substituents are present after BSA is heated to 400°C.
Figure 6. RP-HPLC analysis of aromatic constituents (side chain aromatic groups and breakdown products) of BSA, untreated (top panel), heated to 400°C and not sonicated (middle panel), and heated to 400°C followed by sonication (bottom panel) with monitoring at λ = 280 nm.
MALDI-MS analysis of untreated BSA indicated two heterogeneous species centered at 66,003.6 and 32,862.0 m/z (Figure 7). The molecular weight of BSA, based on amino acid sequence, is 66,430.3 Da, while the commercial supplier reports a molecular weight of ∼66,000 Da. Thus, the observed species represent BSA [M+H]+ and [M+2H]+, respectively.
Figure 7. MALDI-TOF MS analysis of intact, untreated BSA. Laser power = 65%.
Further examination of the untreated BSA focusing on the lower mass range indicated smaller molecular weight species present below 1000 m/z (Figure 8). The corresponding laser powers are provided to demonstrate the usage of laser power sufficient for ionization but not high enough to induce fragmentation.
Figure 8. MALDI-TOF MS analysis of intact, untreated BSA, in the 1000–6000 m/z range (top panel, with laser power = 77%) and in the 500–5000 m/z range (bottom panel, laser power = 44.6%).
MALDI-TOF MS of BSA following 400°C treatment was performed to determine if any intact BSA was present (66,003.6 m/z). The resulting mass spectra did not show any species at the same m/z as the intact protein (Figure 9). The MALDI-TOF MS analysis focusing on lower m/z values for the 400°C treated BSA showed multiple constituents in the range of 650–3200 m/z (Figure 10).
Figure 9. MALDI-TOF MS analysis of 400°C BSA sample before sonication (high m/z range). Laser power = 50.2%.
Figure 10. MALDI-TOF MS analysis of 400°C BSA sample before sonication in the lower mass ranges. The top panel laser power = 37.2% and bottom panel laser power = 49.3%.
MALDI-TOF MS of BSA following 400°C treatment was repeated after sonication to enhance solubilization of the charred materials to determine if any intact BSA was present (66,003.6 m/z). The resulting mass spectrum did not show any signal at the same m/z as the intact protein (data not shown). Mass spectra at lower m/z ranges indicated species in the range of 500–3200 m/z (data not shown), in similar fashion to the non-sonicated sample (Figure 10). A notable difference is that the sonicated sample contained species corresponding to 957.2, 1261.3, and 1565.2 m/z that were not observed in the non-sonicated sample. These species may correspond to the additional peaks observed in the RP-HPLC analysis of the sonicated sample at λ = 280 nm (see discussion above and Figure 6). Nonetheless, the overall results indicated that sonication of samples was not beneficial, as the degradation of the intact protein was readily apparent in both the sonicated and non-sonicated samples.
DSC analyses were performed on a sample of Sup35NM. For the first cycle the protein was heated from 25°C to 400°C at a rate of 50°C/min (Figure 11). Two broad endothermic transitions were apparent, where through the lower temperatures the protein likely denatured, and then experienced the same pyrolysis reactions (at ∼240°C) as observed in the DSC of BSA (Figure 1). The higher transition (at ∼340°C) was most likely further degradation of pyrolysis products (formation of simple molecules and aromatics). As with BSA (Figure 1), a second heat cycle was performed on the Sup35NM sample to determine if the chemical alterations to the protein made by the first heat cycle were permanent or if the protein quickly converted back to the original form. The sample was returned to 25°C and heated to 200°C at a rate of 50°C/min. Analysis of the second heat cycle showed relatively little activity compared to the first heat cycle (data not shown), suggesting that the alterations of Sup35NM made at the temperatures of ∼240°C and ∼340°C were either longer lived or permanent.
Figure 11. DSC analysis of a 3.44 mg sample of Sup35NM run from 25°C to 400°C. The blue line indicates initial heating of the sample with the two thermal transitions corresponding to protein denaturation and degradation of the primary protein structure, respectively.
The diffuse reflectance infrared spectra of Sup35NM during thermal processing up to 350°C was examined (Figure 12). The amount of remaining amide was assessed by the loss of the amide II group and was found to be less than 0.01 μg in a mg sample or better than 5-log reduction using a 7-min ramp to 350°C and a 1-min hold.
Figure 12. Infrared monitoring of Sup35NM pyrolysis as a function of increasing temperature. The loss of water and amide functional groups are noted by decreased spectral intensities. The temperature increase is from 30°C (dark blue line) to 100°C, 150°C, 200°C (light blue lines), 250°C, 300°C, and 350°C (progressively darker red lines).
Sup35NM was examined by RP-HPLC at λ = 220 nm. The intact protein was found to elute at 10.33 min (Figure 13). The peaks eluting at 1.84–2.02 min and 3.259 min were produced by the buffer (as verified by an injection of buffer alone) and excluded from further analyses.
Figure 13. HPLC analysis of untreated Sup35NM with monitoring at λ = 220 nm.
A standard curve was constructed for Sup35NM by testing concentrations ranging from 0.781 to 30 μg and plotting the amount of protein injected versus the area of the integrated peak at 10.333 min. The relationship between concentration and peak area was linear over this range (Figure 14). The limit of detection was at 0.781 μg.
Figure 14. Sup35NM standard curve generated using species eluting at 10.33 min by HPLC.
Sup35NM was heated to 350°C or 400°C and the amount of intact protein and the presence of degradation products were examined by RP-HPLC. After heating, the sample was dissolved in 1 mL 5% formic acid in HPLC grade water/methanol (1:1). HPLC analysis of Sup35NM following 350°C treatment showed no constituent eluting at 10.33 min (Figure 15, middle panels), indicative of the absence of intact protein. In similar fashion, RP-HPLC analysis of Sup35NM following 400°C treatment showed no apparent intact protein (Figure 15, bottom panels). It is important to note the y-axes were expanded for the middle and bottom panels monitoring absorbance at λ = 280 nm (right column) to facilitate visualization of weak signals.
Figure 15. RP-HPLC analysis of Sup35NM protein untreated (top panels), heated to 350°C (middle panels), and heated to 400°C (bottom panels). The panels on the left show data with monitoring at λ = 220 nm and the panels on the right show data with monitoring at λ = 280 nm.
As discussed above, monitoring at λ = 220 nm provides insight into the protein amide bond backbone. Monitoring at this wavelength (Figure 15, left panels) demonstrated the complete decomposition of intact protein after heating to either 350 or 400°C. Monitoring at λ = 280 nm provides insight into aromatic groups, such as amino acid side chains or breakdown intermediates. Two peaks were observed after heating to 350°C that are absent after heating to 400°C (at 5 and 8 min) (Figure 15, right panels). This result is indicative of further decomposition of aromatic substituents at the higher temperature and/or continued breakdown of intermediates. As above, the y-axes in the middle and bottom panels for the data derived at λ = 280 nm are different from the top panel to highlight breakdown products present at relatively low levels.
MALDI-MS analysis of intact Sup35NM revealed a predominant species at 29,962.2 m/z (Figure 16). This species represents intact Sup35NM based on the literature sequence (Ghosh et al., Reference Ghosh, Dong, Wall and Frederick2018). The lower mass species (19,257.4, 16,354.8, and 12,789.8 m/z) (Figure 16) are likely non-prion by-products from the E. coli expression system, common artifacts for weakly expressed His-tagged proteins.
Figure 16. MALDI-TOF MS analysis of intact, untreated Sup35NM (prion protein). Laser power = 71.5%.
MALDI-TOF MS analysis of Sup35NM after heating to 350°C was performed to determine if any intact Sup35NM was present (29,962.2 m/z). Mass spectrometry analysis is more sensitive than RP-HPLC analysis and thus small amounts of intact protein may be revealed by MALDI-TOF MS that were not visible by RP-HPLC. The resulting mass spectrum did not show any signals at the same m/z as the intact protein (Figure 17). Mass spectra at lower m/z ranges revealed species of 643.7, 2248.1, and 4608.6 m/z (Figure 18).
Figure 17. MALDI-TOF MS analysis of 350°C Sup35NM (prion protein) sample (high m/z range). Laser power = 71.5%.
Figure 18. MALDI-TOF MS analysis of 350°C Sup35NM (prion protein) sample (lower m/z range). Laser power = 44.6% (top panel) and 42.8% (lower panel).
MALDI-TOF MS of Sup35NM following 400°C treatment was examined to determine if any intact Sup35NM was present (29,962.2 m/z). The resulting mass spectrum did not show any signal at the same m/z as the intact protein (data not shown). Mass spectra at lower m/z ranges displayed species between 2252.0 and 4565.0 m/z, as well as smaller species between 654.2 and 866.4 m/z (Figure 19).
Figure 19. MALDI-TOF MS analysis of 400°C Sup35NM (prion protein) sample (lower m/z range). Laser power = 54.8% (top panel) and 72.4% (bottom panel).
To determine the sequences of the observed fragments following heat treatment of Sup35NM, a PERL script was developed that scanned the known sequence of Sup35NM, where all the unique fragments starting from a sequence size of one to all 253 residues were determined. The masses of each observed fragment were compared to predicted fragments from Sup35NM within +1 or -1 Dalton mass unit from the observed fragment (Tables 1 and S1).
Discussion
The goals of the present study were to examine potential methodologies to (a) easily detect and quantify the presence of intact proteins, and (b) detect protein breakdown intermediates. This approach would be used to evaluate the overall efficacy of heat inactivation of potential bioactive molecules on MSR flight hardware to define an in-flight BTC process or a terrestrial sterilization measures to enable safe handling of extraterrestrial material. Heat sterilization of two proteins (BSA and the prion protein Sup35NM) served as the test cases representing a standard and a difficult-to-degrade target, respectively. The complete breakdown of both protein targets, as demonstrated by RP-HPLC and MALDI-TOF MS analyses, upon heating to 350–400°C for 5 sec demonstrate the temperature and time requirements for sufficient degradation of the proteins examined herein. By analogy, these heating protocols may be used in the MSR Campaign or other extraterrestrial sample return efforts as they would be predicted to induce the complete removal of potentially hazardous, intact bioactive molecules.
DSC was utilized to determine the temperature at which the proteins potentially degraded and to examine the reversibility of the process. The most stable forms of Sup35NM transition from polymers to monomers with a T m of 65–70°C (Krammera et al., Reference Krammera, Kryndushkinb, Suhrec, Kremmerd, Hofmanne, Pfeifere, Scheibelc, Wicknerb, Schätzla and Vorberg2009). The DSC data for Sup35NM (Figure 11) suggested that protein denaturation (including disruption of β-sheet formation) and initial pyrolysis occurred up to ∼240°C while more complete decomposition occurred at ∼340 °C. Based on the DSC results temperatures of 350°C and 400°C were chosen to study Sup35NM inactivation.
RP-HPLC was utilized to: (a) provide a distinct signature for the intact protein, (b) determine the detection range for the intact protein, and (c) quantify the degradation of the intact protein. Mass spectrometry was utilized to: (a) define a distinct signature for the intact protein and any unexpected contaminants, (b) analyze the degradation of the intact protein, and (c) identify the masses of peptide fragments produced during the degradation process to determine the extent of breakdown. A recent study examined the heat degradation of Sup35NM but did not provide any molecular insight to the process as only SDS-PAGE analysis of the intact protein and Thioflavin T based fibrillization assays were utilized (Seto et al., Reference Seto, Hirsch, Schubert, Chandramowlishwaran and Chernoff2022).
Treatment of BSA at 400°C resulted in no intact BSA as evaluated by MALDI-TOF MS and RP-HPLC analyses. The 400°C BSA sample contained species in the range of m/z 647.3, 860.5, 2168.9, 2372.1, 2540.8, 2691.6, and 3122.3 prior to sonication. A similar distribution of m/z species was observed after sample sonication (524.3, 568.5, 648.1, 845.3, 957.2, 1261.3, 1565.2, 2299.6, 2540.7, 2779.9, 2937.7, and 3113.2). These species are consistent with the masses of small peptides observed by Bikaki following 220°C BSA treatment (Bikaki, Reference Bikaki2019).
Data derived from heat-treated BSA demonstrated that sonication did not substantially improve the recovery of species. Thus, experiments with Sup35NM did not include sonication of samples. It was previously described that prion proteins can be inactivated at ≥200°C (Appel et al., Reference Appel, Wolff, von Rheinbaben, Heinzel and Riesner2001). Treatment of Sup35NM at 350°C or 400°C showed no intact Sup35NM as evaluated by MALDI-TOF MS and HPLC analyses. Similar breakdown products were observed at both temperatures (species in the range of 4608.6, 2248.1, and 643.7 m/z following 350°C treatment and species in the range of 4565.0, 2252.0, 866.4, and 654.2 m/z following 400°C treatment), indicating that certain regions of the protein may have higher resistance to thermal degradation.
Peptide mapping indicated that the fragments observed in the mass spectra could have corresponded to intact peptides derived from the parent protein (Table 1). However, the initial peptide mapping was based on peptides having intact +H3N and COO- termini. Prior studies have shown that when peptides and proteins are heated at temperatures in the range utilized here there is substantial loss of NH3, H2O, and CO2 (Kasarda and Black, Reference Kasarda and Black1968; Schaberg et al., Reference Schaberg, Wroblowski and Goertz2018). Thus, the peptide fragments observed here do not have +H3N and COO- termini and thus do not correspond to intact, unmodified peptides. Additionally, heat causes degradation of individual amino acids within proteins (Deb-Choudhury et al., Reference Deb-Choudhury, Haines, Harland, Clerens, van Koten and Dyer2014) and aromatic amino acids can undergo decomposition to multiple species (Kato et al., Reference Kato, Kurata and Fujimaki1971; Jie et al., Reference Jie, Yuwena, Jingyan, Zhiyong, Ling, Xi and Cunxin2008). Subsequent peptide mapping where H2O is not included in the fragment mass showed greater deviation between the observed and calculated fragment masses then when H2O was included, and in one case no calculated mass was found (Table 1). These results suggest that the fragments observed by mass spectrometry do not correspond to unmodified peptide fragments from the parent Sup35NM.
One concern is that prion proteins may retain activity even if broken down to smaller fragments. It has been reported that the disordered N-terminal domain of mammalian prion protein (PrP) may engage in liquid-liquid phase separation (Polido et al., Reference Polido, Kamps and Tatzelt2021). The N-terminal domains of PrP that exhibited this behavior were in the size range from N1 = 110 residues and N2 = 90 residues (Polido et al., Reference Polido, Kamps and Tatzelt2021). The species observed herein following heat treatment of Sup35NM are much smaller. The largest fragment (4608.6 Da) is ∼42 residues (assuming an average amino acid molecular weight of 110 Da). Thus, retention of Sup35nm activity appears unlikely based on the small size of fragments. However, the present studies indicate that in addition to the breaking of the peptide backbone at amide bonds there is also fragmentation at other susceptible bonds and/or decomposition of the amino acid side chains. Attempts to sequence the fragments through MS2 analyses failed (data not shown), presumably because the molecular masses of the observed peptide fragments do not correspond to expected fragment sizes if only amide bonds are broken. This suggests side chains are also being compromised and/or alternative (non-amide) bonds in the backbone are being broken. Further supporting this conclusion is the RP-HPLC analysis at λ = 280 nm of Sup35NM which indicated that aromatic substituents were further degraded by increased heat treatment from 350 to 400°C (Figure 15).
Interactions between aromatic residues (primarily Tyr) drive Sup35NM oligomer formation (Ohhashi et al., Reference Ohhashi, Ito, Toyama, Weissman and Tanaka2010). In addition, Sup35NM amyloid formation is highly dependent on the Gln/Asn-rich region in the N-terminal domain (King et al., Reference King, Tittmann, Gross, Gebert, Aebi and Wüthrich1997; Glover et al., Reference Glover, Kowal, Schirmer, Patino, Liu and Lindquist1997; Inoue et al., Reference Inoue, Kishimoto, Hirao, Yoshida and Taguchi2001). The temperatures utilized here for Sup35NM decomposition will result in Gln and Asn side chains undergoing deamidation (Sohn, Reference Sohn1996; Pace et al., Reference Pace, Wong, Zhang, Kao and Wang2013; Bhanuramanand et al., Reference Bhanuramanand, Ahmad and Rao2014; Kato et al., Reference Kato, Nakayoshi, Kurimoto and Oda2019). Overall, side chains within the amyloid fibril core greatly impact the stability of the cross-β structures (Mahmoudinobar et al., Reference Mahmoudinobar, Urban, Su, Nilsson and Dias2019). It can ultimately be concluded that self-assembly of fragments produced by heat degradation of prion proteins is highly unlikely. It should be noted that 42-residue fragments of β-amyloid 1-42 (Aβ1-42) are known to self-aggregate and form toxic plaques (Chen et al., Reference Chen, Xu, Yan, Zhou, Jiang, Melcher and Xu2017). However, this aggregation requires intact side chains and as such, the combination of heat-induced fragmentation of the Sup35NM backbone along with heat-induced destruction of side chains results in a product unlikely to self-associate.
One may also consider that a higher temperature treatment (i.e., 500°C) or additional time may further break down the peptides based on predicted protein pyrolysis kinetics (Figure 20) (Bach and Chen, Reference Bach and Chen2017). In the present study, the DSC temperature was held for 5 sec, and while there is certainly an opportunity for a longer heating time, the need for protective measures such as active cooling systems on a space flight system would necessarily increase with the total energy applied.
Figure 20. Arrhenius prediction for protein pyrolysis (destruction of primary structure) kinetics. Data adapted from Bach and Chen (Bach and Chen, Reference Bach and Chen2017).
The present study has demonstrated that heat treatment at 400°C for 5 sec results in substantial decomposition of the prion protein Sup35NM. In addition to the disruption of the protein backbone, side chain residues are also compromised. The impact of treatment on both the backbone and side chains of the protein provide confidence that fragments resulting from heat treatment cannot self-associate to create potentially pathogenic entities. The methodology utilized herein provided detailed insight into the degradation process and can be applied to a variety of treatment strategies to ensure that planetary protection measures undertaken by endeavors such as the MSR Campaign meet key objectives both in scientific integrity and safety.
Acknowledgments
This study was supported by the Institute for Human Health & Disease Intervention (I-HEALTH) at FAU and by Tyson Harrison at the Analytical Chemistry Department of Nelson Labs, LLC. A portion of this work was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). Government sponsorship is also acknowledged.
Competing interests
The authors declare no conflicts of interest. The decision to implement Mars Sample Return will not be finalized until NASA’s completion of the National Environmental Policy Act process. This document is being made available for planning and information purposes only.
Open access
