|
Post by Admin on Sept 11, 2023 20:53:34 GMT
Synthesis of the EVA Films and Protein Harvesting The synthesis of the EVA film was carried out under biologically and chemically clean conditions as previously reported by Cucina et al. (20) In particular, a special plastic-like film based on EVA as a binder of ground AG 501 mix-bed cation/anion exchange, C8 and C18 resins (all from Bio-Rad) was prepared. The mixture was made comprising 70% 1–10 μm size ground beads and 30% EVA (the melting temperature was 75 °C). This mixture of melted EVA and Bio-Rad resins was poured into a “Brabender” mixer W30 and extruded via a “Brabender” extruder KE19 (both from Brabender GmbH, Duisburg, Germany) in the form of either a thin film or diskettes. The final thickness of the EVA film was 150–200 μm. The proportion of the various resins in the plastic film was: 35% strong cation, 35% strong anion exchangers, and 15% C8 and 15% C18 hydrophobic resins. EVA foils were manufactured and supplied by SpringStyle Tech Design Ltd (Israel) with proprietary technology.
Protein Sampling by EVA Diskettes Protein sampling by EVA diskettes was carried out at the government archive of Sibiu, Romania, under clean conditions with nitrile gloves, tweezers, clean and sterile covers, and support surfaces. For sampling from large area letters, it is necessary to find places with a higher concentration of proteins. For this navigation, the fluorescence of phenylalanine, tyrosine, and tryptophan under UV illumination was used. Particularly, UV LED for illumination and a digital camera on a smartphone with a special optical filter for fluorescence detection were used. Special software for augmented reality interface was used for quick indication of places with high concentrations of proteins. The fluorescence level at each point was displayed in pseudo-colors on the instrument interface (green, yellow, and red, in order of increasing fluorescence intensity). This made it possible to quickly identify regions for sampling on Dracula’s letter surfaces (Figures 1b, 1c, and 1c). This mobile system and software were made in SpringStyle Tech Design Ltd for quick examination of protein traces’ presence on archaeological and culture heritage samples. To harvest proteins, the EVA diskettes were gently humidified with ultrapure water and then placed on letters in different regions for 60 min (see Figures 1a, 2a, and 3a). Finally, to prevent drying of the EVA films, they were covered with parafilm from the outside. A modern letter written and touched by the authors was used as the reference sample. Protein sampling by EVA diskettes from the modern reference letter and proteomics characterization were carried out in the same way as that of the ancient letters (see the Supporting Information).
|
|
|
Post by Admin on Sept 12, 2023 19:42:31 GMT
Protein Extraction from EVA Diskettes EVA diskette handling and protein extraction protocol were carried out in a dedicated laboratory “clean room” in compliance with protection guidance for ancient samples and adopting all precautions to minimize the effects of contamination from modern proteins, as previously reported. (21) A section (2 mm × 2 mm) of each EVA diskette was cut with a scalpel and put in an Eppendorf microtube. Proteins harvested in EVA films were eluted sequentially with a total of 1.3 mL of volatile buffers (formate at pH 3, followed by ammonia at pH 10, and sequentially by bicarbonate buffer at pH 8 with Rapigest 0.1%) and finally with volatile solvents (acetonitrile). The eluates were dried under vacuum (Concentrator Plus, Eppendorf) and then resuspended in 300 μL of 50mM AMBIC. Proteins were quantified by a fluorometric assay using the Qubit Protein Assay kit with the Qubit 1.0 Fluorometer. (22) Then, about 50 μg of protein extracts (in 50 mM AMBIC solution) were reduced by 38 μg of DTT (3 h, at room temperature) and alkylated by 79 μg of IAA (1 h, in the dark at room temperature). Enzymatic protein digestion was carried out overnight at 37 °C, by 0.98 μg porcine trypsin. Tryptic peptide mixture solutions were dried under vacuum (Concentrator Plus, Eppendorf), re-dissolved in 100 μL of 5% aqueous FA, filtered by ultracentrifugation (750 μL, 0.2 μm Nonsterile Micro-Centrifugal Filters, Sepachrom, Rho, Milan), and analyzed by nano-UHPLC/high-resolution nano-ESI–MS/MS in duplicate. An empty diskette of EVA film was used as the control sample. It was processed and analyzed by proteomics in the same way as that of the EVA diskettes of the letters.
MS Analysis MS data were acquired via a Thermo Fisher Scientific Orbitrap Fusion Tribrid (Q-OT-qIT) mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) coupled online with a Thermo Scientific Dionex UltiMate 3000 nano-Liquid chromatography system (Sunnyvale, CA). nLC-nESI MS and MS/MS analyses were carried out using the instrument parameters reported in Cucina et al. (21) Full scans of peptide precursors were performed in high resolution (i.e., 120 K resolution @ 200 m/z), whereas tandem MS of those precursors with charge state 2–4 was carried out in the ion trap (low-resolution acquisition). To avoid cross-contamination with other biological samples, all solvents were prepared freshly, and ancient samples were not processed or analyzed in one batch with modern references. In addition, to avoid carryover during nLC-MS/MS runs, three to five blank runs were performed before each analysis using the same gradient program. Spectra acquired in the last blank run were searched by PEAKS software against the Swiss-Prot database without species origin restrictions and using the same parameters as the archaeological samples.
|
|
|
Post by Admin on Sept 13, 2023 20:20:20 GMT
Database Search All MS data were merged and processed using two different search engines, the PEAKS X de novo sequencing software (v. 10.0, Bioinformatics Solutions Inc., Waterloo, ON, Canada) and the MaxQuant (MQ) software 1.6.17.0 (https://www.maxquant.org/). Raw MS data were searched against the Swiss-Prot database restricted to different taxonomies separately. Particularly, the following taxonomies were investigated: (i) “Human” (20,386 entries, release July 2022); (ii) “Bacteria” (340,707 entries, release January 2023); (iii) “Viruses” (17,957 entries, release February 2023); (iv) “Fungi” (36,956 entries, release January 2023); (v) “Insecta” (10,986 entries, release February 2023); and (vi) “Viridiplantae” (42,802 entries, release February 2023). The common Repository of Adventitious Proteins (c-RAP; www.thegpm.org/crap/) contaminant database was also enabled as background in all the database searches. The first step of database search was carried out using the following parameters: (a) tryptic peptides with a maximum of two missed cleavage sites; (b) cysteine carbamidomethylation as a fixed modification; and (c) oxidation of methionine, the transformation of N-terminal glutamine and N-terminal glutamic acid residue to pyro-glutamic acid form, the deamidation of asparagine and glutamine, and the N-terminal protein acetylation as variable modifications. Then, to improve peptide identification, databases were also searched investigating the following PTMs, as variable modifications: (i) oxidation, di-oxidation, formation of kynurenine, and formation of oxo-lactone, for tryptophan residues; (ii) oxidation, di-oxidation, and formation of dopaquinone, for tyrosine residues; (iii) di-oxidation of methionine; and (iv) trioxidation of cysteine. The precursor mass tolerance threshold was set to 10 ppm, and the max fragment mass error was set to 0.6 Da. Peptide spectral matches (PSMs) were validated using a Target Decoy PSM Validator node based on q-values at a false discovery rate (FDR) ≤ 0.1%. PEAKS score thresholds for PSMs were set to achieve, for each database search, FDR values for PSMs, peptide sequences, and proteins identified below the 0.1% value. In the MaxQuant software, the match type was “match from and to,” the decoy mode was “revert,” the PSM, Protein, and Site decoy fraction FDR were set at 0.01 as the threshold for peptide and protein identifications. The minimum score for modified and unmodified peptides was set at 40. All the other parameters were set as default. A protein was considered identified if a minimum of two peptides (including at least one unique peptide) were matched. Finally, to be sure of the species assigned by the software to each protein identified, all the identified peptides underwent both the BLASTp (Basic Local Alignment Search Tool for protein) searches through the NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and the “Tryptic Peptide Analysis” of the open-source web application Unipept (https://unipept.ugent.be/) to check the taxon-specificity, validate species identifications, and rule out conserved peptides between species. The MS data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository with the data set identifier < PXD041350>. (23) Calculation of the Level of Deamidation and Other Modifications An estimation of the percentage of deamidation for each sample was calculated with the aid of a freely available command-line script for Python 2.x (https://github.com/dblyon/deamidation), which uses the MaxQuant “evidence.txt” file. (24) The calculations were done separately for potentially original peptides and potential contaminant peptides as previously reported. (4) Analogously, estimation of the percentage of the other amino acid modifications investigated was obtained by applying the same model of the deamidation script, separately for potentially original and potentially contaminant peptides (see the Supporting Information).
|
|
|
Post by Admin on Sept 15, 2023 21:32:05 GMT
Results As reported above, ancient proteins undergo a series of complex diagenetic reactions (i.e., chemical modifications of amino acids, chemical degradation, and molecular breakdown) that alter their sequence and chemical structure. (9) Some substrates (such as bone, dental calculus, and eggshell) may preserve, from degradation driven by external factors, endogenous proteins better than others. On the contrary, because of the poor screen effect, diagenetic effects may be more extensive in substrates such as animal tissues, manuscripts, or paints. Therefore, identification of proteins may result in more challenges or even failure. In light of this issue, in the present study, we considered not only those proteins identified with a minimum of two peptides (including at least a unique peptide) but also proteomic data interpreted at the peptide level (i.e., not considering also the proteins from which these peptides come). In this way, all the “original peptides” identified by Max Quant were used to perform a meta-proteomics analysis by Unipept as reported in the “Materials and Methods” section to also achieve a global vision of the taxonomic distribution of all peptides. Globally, the approach adopted here allowed for the identification, as potential endogenous original components of the investigated letters, of 575 human-related peptides (Table S1), 692 peptides from Bacteria (Table S2), 389 peptides from Viruses (Table S3), 436 peptides from Fungi (Table S4), 301 peptides from Insecta (Table S5), and 394 peptides from Viridiplantae (Table S6). Instead, 165 peptides were related to proteins of the C-Rap database and therefore classified as potential contaminants (see Table S7). A modern reference letter was analyzed and processed in the same way as that used for the ancient samples. The raw data were analyzed against all the databases described in the main manuscript, and all the PTMs observed in the sample were searched. It was possible to identify: (i) 359 peptides by searching the Human database; (ii) 401 peptides by searching the Bacteria database; (iii) 237 peptides by searching the Virus database; (iv) 293 peptides by searching the Fungi database; (vi) 137 peptides by searching the Insecta database; and (vii) 299 peptides by searching the Viridiplantae database. Overall, about merely 20 peptides for each database search were in common with the ancient letters. Peptides found in both ancient letters and modern reference letters are marked with an asterisk in Tables S1, S2, S3, S4, S5, and S6 and considered as modern contaminants. The complete results of the modern reference letter are reported in Table S9. Level of Deamidation of Asn and Gln Residues and Other Chemical Modifications As reported in the “Introduction” section, proteomic studies of ancient proteins in Cultural Heritage require the authentication of identified peptides and proteins to discriminate between those that genuinely originate from the sample under analysis and contaminant ones. Discrimination of the original endogenous proteins from modern contaminants may be performed by assessing the extent of degradation of proteins, taking into account that ancient proteins exhibit specific patterns of damage and modifications. In this respect, we investigated the level of deamidation (i.e., the removal of an amide group) of glutamine and asparagine residues, which are transformed into glutamic and aspartic acids, respectively, and result in a +0.98 Da mass shift. (27) Although the deamidation rate may be affected by different environmental factors and the inherent properties of proteins, (28) in almost all studies carried out up to date, it has been observed that it is generally much higher in ancient molecules than in modern ones. (29) Consequently, it is one of the proposed markers of age in many archaeological and paleontological studies. In the present study, the deamidation level of asparagine and glutamine residues was calculated for all the types of peptides classified as “original” (i.e., peptides related to “Humans,” “Bacteria,” “Viruses,” “Fungi,” “Insecta,” and “Viridiplantae”). These results were compared with the deamidation level of those peptides classified as “contaminants” due to their belonging to the proteins of the c-RAP database. Figure 4 shows that the deamidation level of the “original peptides” ranges from 49 to 57% for the asparagine residues and from 33 to 61% for the glutamine residues. On the contrary, “contaminant peptides” present a deamidation level ranging from 2 to 6% for the asparagine residues and from 0.2 to 8% for the glutamine residues. These results highlight that “original peptides” show a similar deamidation level, which is about 8–28 times higher for the asparagine residues and from 5 to <100 times higher for the glutamine residues with respect to the corresponding deamidation level of “contaminant peptides.” Moreover, other DCMs may affect the sequence of the ancient proteins. These forms of random, spontaneous, and non-enzymatic alterations are mainly related to oxidative stress and damage that modify the structure of chromophoric amino acids such as tyrosine (Tyr) and tryptophan (Trp) and other amino acids such as cysteine (Cys) and methionine (Met). (30−33) Consequently, the level of the oxidation products at tryptophan, tyrosine, cysteine, and methionine residues was calculated for both the “original” and “contaminant” peptides. Overall, the comparison of the level of the oxidation products of the above-reported amino acids in original and in contaminant peptides confirms the trend already observed for the deamidation, with the original ancient peptides having a much higher level of oxidation than the contaminant ones (see the Supporting Information and Figure S1 for details).
|
|