PhD School in Chemical and Materials Science
Degraded proteins from paintings, polychromes and
archaeological objects
Candidate:
Sibilla Orsini
Supervisor:
Prof. Ilaria Bonaduce
PhD School in Chemical and Materials Science
Degraded proteins from paintings, polychromes and
archaeological objects
Candidate:
Sibilla Orsini
Supervisor:
Prof. Ilaria Bonaduce
III
Proteins are the building blocks of cells in all living species, and they carry the information necessary for life, replication, defence and reproduction. It has recently been proven that proteins are much longer preserved than DNA, as they have been found in paleontological findings dating back to the Pleistocene. Proteins are also important components of archaeological findings, including bones, textiles and residue of ceramic and vessels, and they are the ingredients of cosmetics thanks to their ability to protect the first layer of the skin from the harsh effects of aging and environment (such as UV rays, pollution and free radicals). Proteins are also fundamental constituents of works-of-art, used as paint binders, as adhesives and varnishes in painting, polychromies and to decorative objects. The study of proteins in artistic, archaeological and paleontological objects provides significant information to understand the evolution of life through the ages, to improve our knowledge of our history, to understand technological developments and artistic manufacture, and to bring essential information to art-historians and conservators.
This PhD thesis aims at studying structural and molecular changes undergone by proteins and proteinaceous materials commonly used in paintings, polychromes and archaeological objects, as an effect of manufacture and ageing. This information is fundamental to design reliable analytical techniques, approaches, methods and models for data analysis, to successfully characterise and indentify proteins in art and archaeological objects.
Many analytical methods for the identification and characterization of proteins from artistic, archaeological and paleontological objects have been proposed [1-4][1-4]. Although, procedures optimized for protein denaturation and extraction from ancient materials and artworks were recently developed, the analysis of
IV
these samples present several drawbacks and the identification of protein is often a difficult task. As an effect of the long term exposure to the changeable and sometimes harsh environments where the object is displayed or stored, degradation phenomena take place. Besides some information are available on the degradation of certain acids -deamidation, hydroxylation, oxidation and carbonylation [5]- very little is known on the degradation of proteins in art and archaeological objects. The main observation relates to the loss of solubility of aged proteins [6], which has been suggested to be related to aggregation, cross-linking and complexation phenomena with pigments [7-10]. If these hypothesis prove to be true, then they might account for several drawbacks that are encountered in the analysis of proteins in degraded samples. Aggregation, cross-linking and complexation with pigment may make the protein binding site unavailable for the stain, may control the degree of flexibility of the polypeptides chains, and may form deleterious changes of the protein conformation, compromising the antigen/antibody interaction, cause loss of solubility, making the protein extraction necessary for chromatographic and mass spectrometric techniques a difficult task and may affect the degree of access of cleavage enzymes in proteomics experiments.
The research presented in this thesis is focused on two main research lines. The first one was dedicated at understating common paths of molecular and structural changes undergone by proteins in artistic and archaeological samples. To this aim a novel analytical approach based on analytical pyrolysis– pyrolysis gas chromatography mass spectrometry in single pyrolysis step (Py/GC/MS) and double pyrolysis step (DSP/GC/MS), evolved gas analysis mass spectrometry (EGA/MS) and thermogravimetric coupled with infrared spectroscopy (TGA and TGA/FTIR) was designed. This approach helped us to:
characterise proteins in samples from artistic and archaeological samples by analytical pyrolysis.
V
Chapter 2: In order to overcome limitations of wet chemical methods arising from the reduced solubility of aged proteins in samples of cultural heritage, a combination of analytical pyrolysis techniques was used to characterise reference materials, paint reconstructions and samples from different historical periods (2nd century BC-20th century AD) and geographical origins, which were collected from paintings and archaeological findings. In particular evolved gas analysis mass spectrometry (EGA/MS), pyrolysis coupled with gas chromatography/mass spectrometry (Py/GC/MS) and double shot pyrolysis/gas chromatography/mass spectrometry (DSP/GC/MS) were used. This analytical approach allowed us to characterise and differentiate the proteinaceous media, investigate their thermal behaviour and evidence changes occurring with ageing. Data clearly indicate that egg, casein and animal glue can be identified and distinguished in a sample of unknown composition using each of the analytical pyrolysis technique used. With time though differences tend to disappear to the extent that extremely degraded samples present pyrolytic profiles extremely similar to each other, irrespective of the nature of the proteins present. The data also indicate that proteins tend to become more thermally stable with ageing, suggesting that extensive intramolecular and intermolecular aggregation, and/or covalent cross-linking occur with time.
Understand the thermal degradation mechanisms of a globular protein (Ovalbumin).
Chapter 3: In this study the thermal degradation of ovalbumin (OVA) under nitrogen atmosphere was investigated. For this scope, a multi instrumental approach based on thermogravimetry (TG), thermogravimetry coupled with infrared spectroscopy (TG/FTIR) and pyrolysis coupled with mass spectrometric detection, i.e. flash pyrolysis-coupled with gas chromatography-mass spectrometry (Py/GC/MS), evolved gas analysis coupled with mass spectrometry (EGA/MS) and double shot pyrolysis-coupled with gas chromatography-mass
VI
spectrometry (DSP/GC/MS), was used. The pyrolysis of a protein involves a combination of several complex mechanisms which produce a very high number of pyrolytic products. The study highlighted that pyrolysis of OVA produces low-molecular weight gasses, such as CO2, H2O, HCNO, NH3 and CO, as main compounds. In addition, a series of organic compounds containing heteroatoms and unsaturations were also identified, whose formation occurred at different temperatures over the pyrolytic process. Among these, cyclic pyrolysis products such as dialkyl substituted 2,5-diketopiperazines (DKPs) and unsaturated-DKPs (un-DKPs) are formed below 350°, and their composition depends on the protein sequence and the reactivity of the amino acid side chains. 3,5-alkyl-3,4-dihydro-2H-pyrrole-2,4-diones (ADPDs) and 3-alkenyl-5-alkyl-pyrrolidine-2,4-diones (AAPDs) were detected for the first time among the pyrolysis products of OVA. AAPDs and ADPDs are formed below 350°C and they are due to homolytic cleavage of the polypeptide chain and cyclization of two neighbouring amino acids. Pyroglutamic acid was also found among the main pyrolysis products of OVA, obtained as pyrolytic product of Glu, which is the most abundant amino acid in OVA. Aromatic compounds, such as pyridine, pyrrole, toluene, alkyl-benzenes and alkyl-pyrroles, phenol and alkyl-phenols, benzeneacetonitrile, benzenepropanenitrile, indole and alkyl-indoles, were detected, produced over a wide range of temperatures. Those produced below 320°C are associated to the pyrolysis of specific amino acid side chains, while at higher temperatures, they are the pyrolysis products of the residual material remaining after condensation reactions, pyrolytic scissions and cyclization reactions.
Evaluate how aggregation by hydrophobic interactions and cross-linking by covalent bonding may affect the thermal degradation mechanisms of proteins.
Chapter 4: Ovalbumin (OVA) has been investigated as a model to study aggregation and cross-linking phenomena using a multi-technique approach.
VII
Aggregation and/or cross-linking reactions has been induced by dissolving the proteins in water at different concentrations, by heating the solutions up to 80°C, or by adding the 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) as cross-linkers.
Thermogravimetric analysis (TG) and TG coupled with infrared spectroscopy (TG-FTIR), evolved gas analysis coupled with mass spectrometry (EGA/MS), double shot pyrolysis coupled with gas chromatography/mass spectrometry (DSP/GC/MS and Py/GC/MS) have been employed to study the protein thermal stability and to characterize the gas compounds evolved during pyrolysis. FTIR spectroscopy was employed to perform the conformational analysis of proteins.
β-sheet formation and the formation of intermolecular aggregates stabilized by weak hydrophobic and hydrogen bonds in OVA are favorite by higher protein concentration and temperatures. The cross-linkers (EDC) induce the formation of covalent bonds randomly distributed in the protein inducing significant portions of random coil structures in OVA. Both these processes increase the thermal stability of proteins.
The cross-linking by EDC gives more thermally stable structures, which thermally degrade at higher temperatures forming mainly aromatic compounds and in less extent DKPs.
These results explain the pyrolytic behavior of proteinaceous materials found in artistic and archaeological samples, often characterized by an extended degree of covalent cross-linking and it may help to solve several analytical issues in the analysis of proteins in samples from artistic, archaeological and paleontological objects.
Understand the how pigments interact with proteins in samples from artistic objects.
Chapter 5: Thanks to the chemical composition of a paint, the proteins in a paint have a strong tendency to complex with metallic cations from the inorganic
VIII
substance. It is well known, in fact, that pigments can affect the protein identification by amino acid analysis if suitable purification steps are not adopted. In order to evaluate if the protein-pigment interaction can influence the thermal degradation mechanisms of the protein we decided to analyses a series of paint reconstruction with EGA/MS analysis and the results are compared with unpigmented materials.
To fully understand the influence of pigments in the protein identification through proteomics procedures, the effect of pigments is systematically investigated in model paint samples of egg white, casein and animal glue with Pb2+, Hg2+, Cu2+, Fe3+ and Ca2+ and the results are compared with unpigmented samples.
The second search line was focused on the optimization of specific protocol for sample preparation and digestion to the development of data analysis tools that can cope with ancient or damaged samples. Liquid tandem mass spectrometry technique was used to development of a unique method, efficient and applicable to different kinds of samples containing proteinaceous material with high degradation condition. This approach helped us to:
design a novel non-enzymatic digestion protocol by bottom-up proteomic approach to be used for the MS analysis of proteins in samples from artistic and archaeological objects. The results are compare with the most common enzymatic digestion protocols.
Chapter 6 : Chemical hydrolysis assisted by microwave irradiation has been proposed as an alternative method for the analysis of proteins in highly insoluble matrices. In this work, trifluoroacetic acid (TFA) was applied for the first time to the chemical digestion of proteins into unspecific peptides to detect degraded proteins from paintings and polychromes. In order to evaluate the performances of this approach, the input parameters for data analysis had to be optimised.
IX
Number of identified peptides, protein sequence coverage (%) and PSMs obtained with chemical digestion were compared with those obtained with two analytical procedures based on the use of enzymatic digestion with trypsin, which were proven successful in the analysis of samples from cultural heritage objects. The use of TFA allowed the successful identified all proteinaceous materials in all paint samples analysed except for egg proteins in one, extremely degraded sample. On the other hand, the protocol using TFA identified more peptides, more PSM’s and greater sequence coverage in the samples containing caseins, and often also in animal glue, highlighting the great potential of this approach for the easily and rapid digestion of insoluble and degraded proteins from the field of the cultural heritage.
References
1. Dallongeville, S., et al., Proteins in Art, Archaeology, and Paleontology: From Detection to Identification. Chemical Reviews, 2016. 116(1): p. 2-79.
2. Calvano, C.D., et al., Revealing the composition of organic materials in polychrome works of art: the role of mass spectrometry-based techniques. Analytical and Bioanalytical Chemistry, 2016: p. 1-25.
3. Bonaduce, I., et al., Analytical Approaches Based on Gas Chromatography Mass Spectrometry (GC/MS) to Study Organic Materials in Artworks and Archaeological Objects. Topics in Current Chemistry, 2016. 374(1): p. 1-37.
4. Dallongeville, S., et al., Proteins in art, archaeology, and paleontology: from detection to identification. Chemical reviews, 2015. 116(1): p. 2-79. 5. Vinciguerra, R., et al., Proteomic strategies for cultural heritage: From
bones to paintings. Microchem. J., 2016. 126(Supplement C): p. 341-348. 6. Bonaduce, I., et al., Analytical Approaches Based on Gas Chromatography
Mass Spectrometry (GC/MS) to Study Organic Materials in Artworks and Archaeological Objects. Top Curr Chem (J), 2016. 374(1): p. 1-37.
7. Duce, C., et al., Interactions between inorganic pigments and proteinaceous binders in reference paint reconstructions. Dalton Trans., 2013. 42: p. 5975–5984.
X
8. Duce, C., et al., Physico-chemical characterization of protein–pigment interactions in tempera paint reconstructions: casein/cinnabar and albumin/cinnabar. Anal. Bioanal. Chem., 2012. 402(6): p. 2183-2193. 9. Ghezzi, L., et al., Interactions between inorganic pigments and rabbit skin
glue in reference paint reconstructions. Journal of Thermal Analysis and Calorimetry, 2015. 122(1): p. 315-322.
10. Pellegrini, D., et al., Fourier transform infrared spectroscopic study of rabbit glue/inorganic pigments mixtures in fresh and aged reference paint reconstructions. Microchemical Journal, 2016. 124: p. 31-35.
3
Physical, thermal and mechanical treatments in various enviroments, storage conditions, deterioration and aging, and the presence of additives and contaminants are only some examples of the most common degradative factors that can reduce the quality of several type of protein-based materials.
Protein modification is a growing topic for numerous fields of research, such as clinical, biomedical and food chemistry, as well cosmetics, leathers, textiles and artworks, due to their impact on the structural, function and physic-chemical properties of these objects and substances. Their degradation and the loss of their “native” properties are mainly due to two type of changes in the protein content which often are linked between them. The first is the modification of the amino acidic residues, most commonly to oxidation, while the second is the structural modifications of the proteins, including hydrolysis of peptide bonds, aggregation of assemblies, formation of covalent bonds by cross-linking and formation of strong complexes with cations.
The most common issues about protein modifications in “died” systems, such as food (meat, milk and cereals), textile (linen, wool and silk) and artworks (painting, archaeological and paleontological finding), are set out and discussed in the following chapter.
1.1 The most common amino acids modifications
1.1.1 Oxidation
Oxidation of proteins is one of the most investigated change in several fields, such as during the heat treatment by cooking of food, which is consider dramatic harm to human health[1]. In addition, the degradation of ancient artworks, paleontological, archaeological objects and historic silk textiles is the cause of the deterioration of the materials and the difficultly of their preservation [2].
4
Oxidation reactions are also extensive linked to the degradation of collagenous fibers of parchment due to different factors, such as the lipid content, the chemical composition of inks, the temperature and humidity values during the storage of the parchment [3-6].
From the chemical point of view, the oxidation of amino acids can be generated via chemical and biological processes, such as lipid oxidation, oxygen radicals from metabolic processes and metal- or enzyme-catalyzed oxidative reactions. Some of the most common oxidative modifications are the cleavage the peptide bonds, the introduction of carbonyl groups in the amino acidic residues and the formation of compounds more susceptible to intermolecular aggregation and cross-links[7] [8]. The oxidative degradation processes can modify the properties of the proteins, such as hydrophobicity and solubility and they can change the conformation and the susceptibility of the protein to proteolytic enzymes [9]. Several amino acidic residues are susceptible to oxidation process and the most common are described below.
1.1.1.1 The formation of aminomalonic acid
Aminomalonic acid (Ama) is a post-translational modification of amino acid in protein[10]. Ama is a dicarboxylic acid with one methylene group shorter than aspartic acid which is formed by oxidation of amino malonic aldehyde as oxidation degradative product of amino acids in protein. As the example of Cys, Ama might result of oxidative product of dehydrocysteine via β-elimination of sulfur from Cys (Figure 1.1)[10,11]. After the formation of Ama , Cys is not alkylated by iodoacetoamide.
In the field of the cultural heritage the formation of Ama is observed as a modification of Cys, Ser and Phe [12,13]. The data analysis in mass error tolerance mode of LC/MS/MS spectra of peptides digested by enzyme and
5
extracted by painting samples of the 13th century from the Camposanto Monumentale of Pisa identifies Ama as variable amino acid modification [14]. The process was related with the oxidative damage of the protein and it was assigned to different sequences, such as peptides from bovine beta-casein, alpha-S1-casein and collagen alpha 1(I)[14]. The oxidation of amino acids to amino malonic acid was supposed to occur during the curing of the proteinaceous binder and the process increases during the aging of the samples [12,14,15].
Figure 1.1: Proposed pathway for oxidation of Cys[11].
1.1.1.2 Oxidation of Pro
The amino acid content of collagenous materials is rich in Pro, which is often found partially hydroxylate in the organism. The hydroxylation process undergoes in vivo organisms at the endoplamic reticulum by collagen propyl
4-NH2 SH O OH NH2 SH O OH NH2 S O OH dehydrocysteine Cys NH2 C O OH O OH Aminomalonic acid
6
hydroxylated or propyl 3-hydroxylase in presence of oxygen, ascorbic acid, α-ketoglutarate and Fe2+[16]. The oxidation product of Pro is its post-translational metabolite, as 4-hydroxy proline and 3-hydroxyproline (Figure 1.2). The ratio of 4-hydroxyproline to 3-hydroxyproline in collagen proteins is about 100:1[16]. Hydroxyproline is a common non-proteogenic amino acid and it is one of the most abundant amino acid in collageneous materials from mammalian (33% Gly, 10% Ala, 14% Pro, 11% Hyp), such as tissue, glue, gelatin and bone [17,18]. In marine organisms the amount of Hyp is more changeable than mammalia because the hydroxylation of Pro is major influenced by the habitat of the organism [19,20].
The detection, identification and quantification of Hyp in collagenous material is often discussed in several fields. In food proteins, the amount of 4-hydroxyproline can be measured to establish the protein quality of meat and meat products [21,22]. The quantification of Hyp was calculated to estimate the collagen content in biological tissues employed in biomedical and functional food products [23,24]. The content of Hyp as also evaluated in collagen to confirm the purity of the synthesis of collagen hydrolysate in skin and hair cosmetic[25]. From the field of the cultural heritage the detection of Hyp in the chromatographic profile of the amino acids obtained by gas chromatographic (GC) and liquid chromatography (LC) techniques of a sample establishes with certainty a collagenous based material [26-30]. By analytical pyrolysis coupled with mass spectrometry (Py/GC/MS) derivative pyrolysis products of Pro and Hyp, such as pyrrole, alkyl-pyrrole and cyclic pyrolysis products (dialkylsubstituted 2,5-diketopirperazines), are consider markers of collagenous material [31-33]. In proteomic analysis the detection of Hyp as variable modification of Pro, increases the reliability of the identified peptides from collagen for the identification of the material and the animal sources [34].
7 1.1.1.3 Oxidation of Lys
5-Hydroxylysine(Hyl) is a natural post-translational product of Lys residue in collagen and collagen-like proteins [35] which is formed by lysyl oxidase in cell [36] and the mechanism is reported in Figure 1.2 [37]. The amount of hydroxylation of Lys in collagen is highly variable respect to Hyp[36] and its quantification was used as an alternative method to Hyp to determine the amino acidic content of the connective tissue in meat and its quality[38].
Hyl is highly reactive with other compounds[39,40]. In cell, Hyl can be glycosylated with the addition of galactose and glucose [37,41], while outside the cell an enzymatic oxidative deamidation of Hyl produces a reactive aldehydic residue, which is involved in the formation of a series of non-enzymatic condensation reactions to form extensive covalent intra- and inter-molecular cross-links (collagen fibrils)[36].
In the analysis of the amino acidic content of parchment and leather after laser cleaning test, the loss of basic amino acids, including Lys and Hyl is observed as oxidative degradation processes based on heat and light due to the formation of derived-products and cross-links structures [6,42,43].
By proteomic analysis in the field of the Cultural Heritage, the addition of hydroxylysine as variable modification in the data analysis of the MS/MS spectra was tested to improve the detection of high degraded proteinaceous materials from funerary objects found in the Xiaohe Cemetery [44]. In these samples, hydroxylysine was identified in peptides from collagen of bovine, the material used as adhesive in the funerary objects. The study highlighted the origin of hydroxylysine by aging process of the samples, even if hydroxylysine from Lys was found with lower frequency than the hydroxyproline from Pro [44]. The glycosylation of Lys was detected in the archaeological samples[45]. Data analysis of LC/MS/MS spectra of peptides in mass error tolerance mode identified several glycosylated Lys residues at peptide C termine, such as
8
galactosyl-hydroxylysine and glucosylgalactosyl-hydroxylysine, in proteomic analysis of cranial bones found at the Ziegler archaeological site[45].
Figure 1.2: Hydroxylation reactions of Pro and Lys.
Aminoadipic acid from Lys is obtained as oxidative degradation of α-aminoadipic semialdehydes (AAS) as an oxidation product of Lys [46]. In vivo system the formation of aminoadipic acid is caused by septicemia, while in archaeological samples the presence of aminoadipic acid is associated with the decomposition of the organism after died or with infiltration of biogenic taphonomic factors during the prolonged deposition of the sample in soil [46,47]. This modification was revealed in peptides of collagen alpha 1(I) extracted from archaeological mammoth bones [47].
HN O OH enzyme O2 Pro HN OH O OH 4-hydroxyproline + HN OH O OH 3-hydroxyproline H2N H2N O OH Lys enzyme O2 H2N NH2 O HO HO 5-hydroxylysine
9
In food matrices, the presence of AAA has never been detected. The carbonyl group of AAS may react with the amino moiety of an amino acidic residue to generate a covalent bond via Schiff base[48].
1.1.1.4 Oxidation of Met
Met oxidizes to create sulfoxide and disulfoxide compounds. Oxidation of methionine leads to a mixture of the S- and R- epimers of methionine sulfoxide[49-51], which then can degrade with the formation of Met disulfoxide, as is shown in Figure 1.3. The oxidation of methionine is another relevant degradative process of the aging, which is found during the analysis of archaeological and paleontological samples and proteinaceous materials in paintings [52-59].
Figure 1.3: Oxydation of Met.
1.1.1.5 Oxidation of Cys
Cys residue in protein can occur as free sulphydryl form or as oxidize cystine, by the formation of a disulphyde bridge. Denaturations of protein by high temperature, pressure, mechanical shear or exposition to air/water, imply substantial structural changes which give available and reactive free sulphydryl group and disulphyde bridge. The disruption of S-S bridge in the native
H2N S O HO Met H2N S O HO O Methionine sulfoxide O2 H2N S O HO O Methionine disulfoxide O2 O
10
conformation of protein leads to the formation of extensive intermolecular cross-links by S-S bridge and the oxidation of free sulphydryl group. This process induces irreversible chemical cross-linking of proteins and it can be usually observe during cooking processes of food and gelation process [60,61].
1.1.1.6 Oxidation of Trp and other amino acids
During the study of ancient parchment scrolls and model paint films of egg white and casein from milk oxidation of Trp was founded as variable amino acid modification due to the aging process of the samples [62,63]. The oxidation process of Trp generated by radicals involves initial abstraction of hydrogen atom from the position 1 of the indole ring of Trp to generate indolyl radical. The indolyl radicals can successively react with other perxoxyl radicals and another Trp residue to generate a multiple reaction pathways wich can lead to degradative compounds, such as hydroperoxides, alcohols, N-formylkynurenine and kynurenine[64,65] (Figure 1.4). Laser-based analysis (LIF) of model paint films of egg white and casein from milk reveals the presence of photooxidation products of Trp, such as N-formylkynurenine (NFK) and kynurenine (emissions at approximately 435 nm)[66-68]. Specific oxidation markers of Trp, such as kynurenin, were also detected by MALDI/TOF/MS analysis of archaeological textiles[69] and wool[70].
11
Figure 1.4: Products of Trp oxidation [71,72].
The aromatic side chains of Phe, Trp and Tyr are highly susceptible to radical oxidation processes [72,73]. Tyr in egg yolk and casein can produce dityrosine (emission at 405 nm) and Phe can produce 3,4-dihydroxyphenylalanine (DOPA) (emission at 480 nm) [66-68].
Several studies reveal that their oxidation products are the mainly responsible of the discoloration of wool. Data analysis of the in-solution heating of wool fibers during the dyeing treatment displays the formation of oxidative products of Tyr
12
and Trp. Moreover, the calculus of the oxidation score provides a relative evaluation of the levels of oxidative damage in the fibers and reveals an increase of the oxidation score by exposition of the samples to UV irradiation [69].
1.1.2 Deamidation
Deamidation is a non-enzymatically hydrolytic modification of Asn and Gln to Asp and Glu, respectively[74] (Figure 1.5). Under physiological condition (pH 7.4) deamidation of pentapeptides has a half-life for Asn of 6-507 days and 96-3409 days for Gln, but the rate of the process increase dramatically with the environment treatment and it is strongly influenced by the primary structure and the three-dimensional structure of the protein[75,76]. The Asp deamidation reaction involves formation of 5-membered cyclic imide intermediate, which subsequent hydrolytic ring opening and the formation of L-aspartic acid and isopimaric acid in a ratio of about 3:1[75]. The formation of cyclic imide is strongly influenced by the neighboring amino acid and the three-dimensional structure of the protein. Moreover, the formation of 6-membered cyclic imide in Gln is entropically less favorable and this is the reason because Gln deamidation is slower than Asp[75]. The schematic mechanism of glutamine deamidation is reported in Figure 1.5. The nearby amino acid residues can catalysed the deamidation reaction and for example, the -Asn-Gly- sequence is especially prone to deamidation [77,78]. The rate of deamidation is influenced also by a series of factors such as steric hindrance, charged residues in the vicinity of the deamidation site, pH, temperature and secondary, tertiary and quaternary structure. Moreover deamidation is competitive with hydrolysis and cyclization reaction. Deamidation at Asn takes place more rapidly than Gln [14,56,79,80]. Deamidation causes a mass shift of +0.984 Da and introduce negative charges in the primary structure of the protein. The mechanism increase the solubility of the protein chains, and contribute to degradation of the protein with the
13
disruption of the protein structure[80]. Deamidation acts in vivo processes and it is abundantly documented in protein biological aging by several traces of Asn and Gln deamidation. In archaeological studies, the percentage of Gln deamidation from the soluble fraction of collagen from bones is correlated with the thermal age and the chronological age of the object [81]. The results indicate that burial environment will bear influence the percentage of Gln deamidation rather than chronological age. The primary mechanism of Gln deamidation in archaeological collagen is the direct side-chain hydrolysis and it is influenced by the pH of the soil. Alkalinity has a greater effect than acidity and deamidation is catalysed by OH- ions. Acidic pH acts as buffer due to the dissolution of bone apatite [81]. Deamidation of Gln and Asp is also considered the major modification during aging of proteinaceous binders, such as casein, collagen and proteins from egg, in mural paintings [14], panel paintings[82] and the wool textiles [83]. Extensive deamidation of Gln and Asp has been proposed as a deterioration marker in collagenous and noncollagenous proteins [81].
Figure 1.5: General mechanism of glutamine deamidation to form glutamic acid and isoglutamic acid via a glutarimide intermediate [84].
H2N NH O NH O Gln NH3 N O O N H Glutarimide intermediate H2O H2O HO NH O NH O HN NH O OH O
Glutamic acid (alpha- Glu)
14
1.2 The most common modifications of the
three-dimensional rearrangement of the proteins
1.2.1 Glycation
Glycation is a non-enzymatic reaction between a carbohydrate compounds, such as glucose and glucose degradation products of sugar, honey, natural juice and gums, and the side chains of a protein, in particular Lys and Arg residues [85,86] under the name of Maillard reaction. The reaction comprises a cascade of chemical reactions between a carbonyl compounds of a carbohydrate and the amino group of an ɛ-N-lysine at high temperatures with the formation of un brown insoluble product [86,87]. The first step reaction is the formation of a reversible Schiff’s base that rearranges into a stable glycation products called Amadori rearrangement products. The final step of the Maillard reaction is the formation of an advanced glycation end products (AGEs)[87]. AGEs are considered cross-linked species between two macromolecules and they change the properties of the material.
The Maillard reaction products are extensive evidenced and quantified during heating and storage of food [86]. N-γ-carboxymethyllysin was found in milk and other foods as oxidative degradation of Amadori compounds, imidazolinone was found as the degradation product of Arg in bakery products and roasted coffee while argpyrimidine was found as free amino acids in beer[86,88]. The Maillard reaction might easily occur during the manufacture of milk powder by a mixture of glucose and casein [89-91]. The presence of Maillard reaction products are determined by spectroscopic analysis. The IR spectrum of the Maillard reaction product between casein and glucose highlights the changes of the amide I, II and III bands of casein. There is the appearances in the mid-IR spectrum of C=O, C=N
15
and C-N signals, from Amadori compound, base Schiff and pyrazines, respectively. Moreover, also the physical and chemical properties of the protein, such as reducing power, chelating activity and resistance to tryptic hydrolysis, change[89].
In the field of the cultural heritage, laser induced fluorescence (LIF) analysis of parchment glue assigns the presence of cross-linkages between the side chains of Lys and Arg with the free sugar, such as ribose, with the formation of pentosidine (imidazo (4,5 pyridinium)) by Maillard reaction[66-68].
1.2.2 Non-covalent interactions
Different types of non-covalent interactions stabilize the secondary structure, the α-helix and β-sheet, the tertiary structure, the chain fold and quaternary structures in several proteins[92]. Moreover many proteins interact with other to form complexes and assemblies of two or more proteins[93]. The van der Waals interactions act between atoms of protein and comprise a series of contribution, such as Keesom, Debye, and London contributions[94]. Electrostatic interactions occur between amino acid residues with positive and negative charge depending on the pH of the environment and their isoelectric point[95]. Hydrophobic effects arise from the repulsion between water molecules and the nonpolar protein residue. Hydrogen bonds are formed between a proton from an electronegative donor atom and electronegative acceptor atom[96].
1.2.3 Cross-linking
During the aging proteins from paintings, polychromes and archaeological objects reduce their solubility in the most common extraction solvent [97,98]. The reduce of solubility is related to denaturation of the secondary, tertiary and quaternary structure of the protein and cross-linking processes by formation of
16
covalent bonds. Numerous studies conducted on ancient and degraded materials, such as Py/GC/MS analysis of fresh collagen and archaeological bones, highlighted the presence of one or more degradative processes of the collagenous proteins during the aging with the formation of new interactions and/or covalent bonds[99]. These processes change the pyrolytic profile of the collagenous protein and increase of brittleness and fragility of material [99]. The formation of inter-chains hydrogen bonds between the amino group of Gly and the carbonyl group of adjacent Pro and covalent bonds between chains stabilize collagen proteins and change the chemical-physico properties of the macromolecules [100]. These inter- and intra- chains interaction and bonds retard the rate of dissolution, alter the size distribution of fragments and leads to a tailing off of the final part of the decay curve, prolonging the survival of the protein[101].
A series of thermogravimetric and infrared spectroscopy studies on model paint samples reveal that aggregation and cross-linking reactions are catalysed by metals from the pigment and aging [102-104]. HgS promotes cross-linking of casein and ovalbumin and the phenomena become more significant during the aging [103]. Pigments based on Cu, Fe, Ca and Pb disrupt the protein-protein interactions and lead to covalent cross-linking, such as by the dityrosines and disulphide bridges, and the formation of intramolecular interactions [104,105]. The intramolecular interactions between metal and proteins change the secondary structure of the proteins, which become aggregates during the aging by an increase of intramolecular beta-sheets [102].
1.2.4 Mineralization
The mineralization process is comprised of primary mineral deposition of collagenous matrix on the calcification front and subsequent slow and progressive deposition called secondary mineralization[106]. The more extensive
17
cross-linking of mineralized collagen may involve an inorganic components of bone (apatite) via calcium bridge. There are several important mineralized tissues such as bones [107,108] and tooth (enamel, dentin and cementum)[109,110]. The mineralization of these matrices is guided by noncollagenous proteins, such as dentin sialophosphoprotein in dentin which plays a fundamental rule in the nucleation of hydroxylapatite into dentin matrix collagen[109].
1.2.5 Peptide bond cleavage
Internal amide bonds of proteins can easily hydrolyze, resulting in the reduction of the protein to polypeptides and free amino acids [101,111]. The hydrolytic reaction by peptide bond cleavage is catalysed by acidic and basic treatments[112] and metals[102,103,105].
Mass spectrometry analysis of modern and medieval grape seeds of Vinis vinifera L. determine the high level of semi-tryptic peptides in the medieval samples than the unalterated peptides of the modern specimens [113]. Error tolerance searches of several paint models and sample from the Monumental Cemetery of Pisa identify semi-tryptic cleavages as a effect of the protein degradation [14]. The cleavage of the peptide bonds due to the aging introduces the semi-tryptic search mode during the analysis of ancient and degraded proteinaceous materials, such as proteomic analysis of archaeological human dental calculus to characterize diseases and dietary of extinct populations [114,115]. The semi-tryptic search mode is used as in paleoproteomics analysis for the identification of proteins from bones found at the Ziegler site[45].
18
References
[1] F. Alisson-Silva, K. Kawanishi and A. Varki, Molecular Aspects of Medicine, 51, (2016) 16.
[2] Vilaplana F, Nilsson J, Sommer DVP and K. S., Analytical and Bioanalytical Chemistry, 407, (2015) 1433.
[3] A. Možir, M. Strlič, T. Trafela, I.K. Cigić, J. Kolar, V. Deselnicu and G. de Bruin, Applied Physics A, 104, (2011) 211.
[4] B. Kanngießer, O. Hahn, M. Wilke, B. Nekat, W. Malzer and A. Erko, Spectrochimica Acta Part B: Atomic Spectroscopy, 59, (2004) 1511. [5] A. Možir, L. Gonzalez, I. Kralj Cigić, T.J. Wess, I. Rabin, O. Hahn and M.
Strlič, Analytical and bioanalytical chemistry, 402, (2012) 1559.
[6] K.M. Axelsson, R. Larsen, D.V.P. Sommer and R. Melin, Studies in Conservation, 61, (2016) 46.
[7] M.N. Lund, M. Heinonen, C.P. Baron and M. Estévez, Molecular Nutrition & Food Research, 55, (2011) 83.
[8] D. Kyung Kang, J. Jeong, S.K. Drake, N.B. Wehr, T.A. Rouault and R.L. Levine, J. Biol. Sci., 278, (2003) 14857.
[9] S.P. Wolff and R.T. Dean, Biochemical Journal, 234, (1986) 399.
[10] S.D. Copley, E. Frank, W.M. Kirsch and T.H. Koch, Analytical Biochemistry, 201, (1992) 152.
[11] D. Kyung Kang, J. Jeong, S.K. Drake, N.B. Wehr, T.a. Rouault and R.D. Levine, J. Biol. Che., 278, (2003) 14857.
[12] A. Casoli, M.P. Colombini, R. Fuoco, F. Modugno and G. Palla, Metodologie integrate per l'analisi dei dipinti, Progetto Libreria, Padova, 2003, p.
[13] I. Bonaduce, M. Cito and M.P. Colombini, Journal of Chromatography A, 1216, (2009) 5931.
[14] G. Leo, I. Bonaduce, A. Andreotti, G. Marino, P. Pucci, M.P. Colombini and L. Birolo, Analytical Chemistry, 83, (2011) 2056.
[15] M.P. Colombini and F. Modugno, Journal of Separation Science, 27, (2004) 147.
[16] G. Wu, F.W. Bazer, R.C. Burghardt, G.A. Johnson, S.W. Kim, D.A. Knabe, P. Li, X. Li, J.R. McKnight, M.C. Satterfield and T.E. Spencer, Amino Acids, 40, (2011) 1053.
[17] H.N. Poinar and B.A. Stankiewicz, Proceedings of the National Academy of Sciences, 96, (1999) 8426.
[18] A.M. Child, Studies in Conservation, 40, (1995) 19.
[19] P. Szpak, Journal of Archaeological Science, 38, (2011) 3358.
[20] K.A. Piez and J. Gross, The Journal of biological chemistry, 235, (1960) 995.
19
[21] M.C. Messia, T. Di Falco, G. Panfili and E. Marconi, Meat Science, 80, (2008) 401.
[22] J.A. Belloni, M.S. Di Nezio, M.F. Pistonesi and M.E. Centurión, Talanta, 89, (2012) 526.
[23] K. Hofman, B. Hall, H. Cleaver and S. Marshall, Analytical Biochemistry, 417, (2011) 289.
[24] M.L. Colgrave, P.G. Allingham and A. Jones, Journal of Chromatography A, 1212, (2008) 150.
[25] G.Y. Li, S. Fukunaga, K. Takenouchi and F. Nakamura, International Journal of Cosmetic Science, 27, (2005) 101.
[26] A. Andreotti, I. Bonaduce, M.P. Colombini, G. Gautier, F. Modugno and E. Ribechini, Analytical chemistry, 78, (2006) 4490.
[27] S. Wei, Q. Ma and M. Schreiner, Journal of Archaeological Science, 39, (2012) 1628.
[28] A. Casoli, P.C. Musini and G. Palla, Journal of Chromatography A, 731, (1996) 237.
[29] D. Bersani, P.P. Lottici, A. Casoli and D. Cauzzi, Journal of Cultural Heritage, 9, (2008) 97.
[30] L.R. Čukovska, B. Minčeva – Šukarova, A. Lluveras-Tenorio, A. Andreotti, M.P. Colombini and I. Nastova, Journal of Raman Spectroscopy, 43, (2012) 1685.
[31] G. Chiavari, G.C. Galletti, G. Lanterna and R. Mazzeo, Journal of Analytical and Applied Pyrolysis, 24, (1993) 227.
[32] G. Chiavari, D. Fabbri, G.C. Galletti and R. Mazzeo, Chromatographia, 40, (1995) 594.
[33] D. Fabbri, A. Adamiano, G. Falini, R. De Marco and I. Mancini, Journal of Analytical and Applied Pyrolysis, 95, (2012) 145.
[34] Y. Kumazawa, Y. Taga, K. Iwai and Y.-i. Koyama, Journal of Agricultural and Food Chemistry, 64, (2016) 6051.
[35] M.H. Schweitzer, E.R. Schroeter and M.B. Goshe, Analytical Chemistry, 86, (2014) 6731.
[36] M. Yamauchi and M. Sricholpech, Essays In Biochemistry, 52, (2012) 113. [37] M. Yamauchi and M. Sricholpech, Essays Biochem., 52, (2012) 113. [38] G.C. Zarkadas, C.D. Karatzas and C.G. Zarkadas, Journal of Agricultural
and Food Chemistry, 44, (1996) 2563.
[39] G. Krishnamoorthy, S. Sadulla, P.K. Sehgal and A.B. Mandal, Journal of Cleaner Production, 42, (2013) 277.
[40] G. Krishnamoorthy, S. Sadulla, P.K. Sehgal and A.B. Mandal, Journal of Hazardous Materials, 215-216, (2012) 173.
20
[41] R.C. Hill, M.J. Wither, T. Nemkov, A. Barrett, A. D’Alessandro, M. Dzieciatkowska and K.C. Hansen, Molecular and Cellular Proteomics, 14, (2015) 1946.
[42] W. Kautek, S. Pentzien, M. Röllig, P. Rudolph, J. Krüger, C. Maywald-Pitellos, H. Bansa, H. Grösswang and E. König, Journal of Cultural Heritage, 1, (2000) S233.
[43] M. Cooper, S. Sportun, A. Stewart, M. Vest, R. Larsen and D. Poulsen, The Conservator, 24, (2000) 71.
[44] H. Rao, Y. Yang, I. Abuduresule, W. Li, X. Hu and C. Wang, Journal of Archaeological Science, 53, (2015) 148.
[45] R.C. Hill, M.J. Wither, T. Nemkov, A. Barrett, A. D'Alessandro, M. Dzieciatkowska and K.C. Hansen, Mol Cell Proteomics, 14, (2015) 1946. [46] D.R. Sell, C.M. Strauch, W. Shen and V.M. Monnier, Biochem. J., 404,
(2007) 269.
[47] E. Cappellini, L.J. Jensen, D. Szklarczyk, A. Ginolhac, R.A.R. da Fonseca, T.W. Stafford, S.R. Holen, M.J. Collins, L. Orlando, E. Willerslev, M.T.P. Gilbert and J.V. Olsen, Journal of Proteome Research, 11, (2012) 917. [48] M. Utrera, J.-G. Rodríguez-Carpena, D. Morcuende and M. Estévez,
Journal of Agricultural and Food Chemistry, 60, (2012) 3917.
[49] E.R. Stadtman, H. Van Remmen, A. Richardson, N.B. Wehr and R.L. Levine, Biochimica et Biophysica Acta (BBA) - Proteins and Proteomics, 1703, (2005) 135.
[50] W. Vogt, Free Radical Biology and Medicine, 18, (1995) 93.
[51] N. Brot and H. Weissbach, Archives of Biochemistry and Biophysics, 223, (1983) 271.
[52] C. Wadsworth and M. Buckley, Rapid Communications in Mass Spectrometry, 28, (2014) 605.
[53] L. Rampazzi, L. Campo, F. Cariati, G. Tanda and M.P. Colombini, Archaeometry, 49, (2007) 559.
[54] G. Corso, M. Gelzo, A. Chambery, V. Severino, A.D. Maro, F.S. Lomoriello, O. D'Apolito, A.D. Russo, P. Gargiulo, C. Piccioli and P. Arcari, Journal of Separation Science, 35, (2012) 2986.
[55] I.M. Velsko, K.A. Overmyer, C. Speller, L. Klaus, M.J. Collins, L. Loe, L.A.F. Frantz, K. Sankaranarayanan, C.M. Lewis, J.B.R. Martinez, E. Chaves, J.J. Coon, G. Larson and C. Warinner, Metabolomics, 13, (2017) 1270. [56] M. Buckley, A. Walker, S.Y.W. Ho, Y. Yang, C. Smith, P. Ashton, J.T.
Oates, E. Cappellini, H. Koon, K. Penkman, B. Elsworth, D. Ashford, C. Solazzo, P. Andrews, J. Strahler, B. Shapiro, P. Ostrom, H. Gandhi, W. Miller, B. Raney, M.I. Zylber, M.T.P. Gilbert, R.V. Prigodich, M. Ryan, K.F. Rijsdijk, A. Janoo and M.J. Collins, Science, 319, (2008) 33.
21
[57] S.M. Stevens, S. Wolverton, B. Venables, A. Barker, K.W. Seeley and P. Adhikari, Analytical and Bioanalytical Chemistry, 396, (2010) 1491. [58] A. Lluveras-Tenorio, R. Vinciguerra , E. Galano, C. Blaensdorf, E.
Emmerling, M.P. Colombini, L. Birolo and I. Bonaduce, PLOS one, (2017) 1.
[59] I.D. van der Werf, C.D. Calvano, F. Palmisano and L. Sabbatini, Analytica Chimica Acta, 718, (2012) 1.
[60] R.W. Visschers and H.H.J. de Jongh, Biotechnology Advances, 23, (2005) 75.
[61] A.C. Alting, R.J. Hamer, C.G. de Kruif and R.W. Visschers, Journal of Agricultural and Food Chemistry, 48, (2000) 5001.
[62] I.D. van der Werf, C.D. Calvano, G. Germinario, T.R.I. Cataldi and L. Sabbatini, Microchemical Journal, 134, (2017) 146.
[63] I. Osticioli, A. Nevin, D. Anglos, A. Burnstock, S. Cather, M. Becucci, C. Fotakis and E. Castellucci, Journal of Raman Spectroscopy, 39, (2008) 307.
[64] E. Fuentes-Lemus, E. Dorta, E. Escobar, A. Aspee, E. Pino, M.L. Abasq, H. Speisky, E. Silva, E. Lissi, M.J. Davies and C. Lopez-Alarcon, RSC Advances, 6, (2016) 57948.
[65] I. Dalle-Donne, A. Scaloni, D. Giustarini, E. Cavarra, G. Tell, G. Lungarella, R. Colombo, R. Rossi and A. Milzani, Mass Spectrometry Reviews, 24, (2005) 55.
[66] A. Nevin, S. Cather, D. Anglos and C. Fotakis, Analytica Chimica Acta, 573-574, (2006) 341.
[67] A. Nevin, D. Comelli, G. Valentini, D. Anglos, A. Burnstock, S. Cather and R. Cubeddu, Analytical and Bioanalytical Chemistry, 388, (2007) 1897. [68] A. Nevin, S. Cather, D. Anglos and C. Fotakis, in J. Nimmrichter, W.
Kautek and M. Schreiner (Eds.), Lasers in the Conservation of Artworks: LACONA VI Proceedings, Vienna, Austria, Sept. 21–25, 2005, Springer Berlin Heidelberg, Berlin, Heidelberg, 2007, p. 399.
[69] C. Solazzo, S. Clerens, J.E. Plowman, J. Wilson, E.E. Peacock and J.M. Dyer, Journal of Cultural Heritage, 16, (2015) 896.
[70] S.D. Bringans, J.M. Dyer, J.E. Plowman and W.G. Bryson, Textile Research Journal, 76, (2006) 288.
[71] S.W. Taylor, E. Fahy, J. Murray, R.A. Capaldi and S.S. Ghosh, J. Biol. Sci., 278, (2003) 19587.
[72] J.M. Dyer, S.D. Bringans and W.G. Bryson, Photochemistry and Photobiology, 82, (2006) 551.
[73] J.M. Dyer, S.D. Bringans and W.G. Bryson, Photochemical & Photobiological Sciences, 5, (2006) 698.
22
[75] R. Bischoff and H.V.J. Kolbe, Journal of Chromatography B: Biomedical Sciences and Applications, 662, (1994) 261.
[76] N.E. Robinson, Z.W. Robinson, B.R. Robinson, A.L. Robinson, J.A. Robinson, M.L. Robinson and A.B. Robinson, The Journal of Peptide Research, 63, (2004) 426.
[77] O.V. Krokhin, M. Antonovici, W. Ens, J.A. Wilkins and K.G. Standing, Analytical Chemistry, 78, (2006) 6645.
[78] H.T. Wright, Protein Engineering, Design and Selection, 4, (1991) 283. [79] N. Araki and M. Moini, Rapid Communications in Mass Spectrometry,
25, (2011) 3396.
[80] C. Solazzo, J. Wilson, J.M. Dyer, S. Clerens, J.E. Plowman, I. von Holstein, P. Walton Rogers, E.E. Peacock and M.J. Collins, Analytical Chemistry, 86, (2014) 567.
[81] N.L. van Doorn, J. Wilson, H. Hollund, M. Soressi and M.J. Collins, Rapid Communications in Mass Spectrometry, 26, (2012) 2319.
[82] I.D. van der Werf, C.D. Calvano, R. Laviano, A. Simonetti and L. Sabbatini, Microchemical Journal, 106, (2013) 87.
[83] N. Araki and M. Moini, Rapid Commun Mass Spectrom. , 25, (2011) 3396.
[84] X. Li, C. Lin and P.B. O'Connor, Anal Chem, 82, (2010) 3606. [85] A. Karpowicz, Studies in Conservation, 26, (1981) 153. [86] T. Henle, Kidney International, 63, (2003) S145.
[87] S.H. Assar, C. Moloney, M. Lima, R. Magee and J.M. Ames, Amino Acids, 36, (2009) 317.
[88] M.A. Glomb, D. Rösch and R.H. Nagaraj, Journal of Agricultural and Food Chemistry, 49, (2001) 366.
[89] F.-L. Gu, J.M. Kim, S. Abbas, X.-M. Zhang, S.-Q. Xia and Z.-X. Chen, Food Chemistry, 120, (2010) 505.
[90] L. Jiménez-Castaño, M. Villamiel, P.J. Martín-Álvarez, A. Olano and R. López-Fandiño, Food Hydrocolloids, 19, (2005) 831.
[91] G. Ge Pan and L.D. Melton, Journal of Agricultural and Food Chemistry, 55, (2007) 10036.
[92] J. Janin, R. Bahadur and P. Chakrabarti, Quarterly Reviews of Biophysics, 41, (2008) 133.
[93] S. Jones and J.M. Thornton, Proceedings of the National Academy of Sciences, 93, (1996) 13.
[94] Roth CM, Neal BL and L. AM., Biophysical Journal, 70, (1996) 977. [95] F.B. Sheinerman, R. Norel and B. Honig, Current Opinion in Structural
Biology, 10, (2000) 153.
[96] R.E. Hubbard and M. Kamran Haider, in eLS, John Wiley & Sons, Ltd, 2001.
23
[97] M.P. Colombini, A. Andreotti, I. Bonaduce, F. Modugno and E. Ribechini, Accounts of chemical research, 43, (2010) 715.
[98] G. Ranalli, G. Alfano, C. Belli, G. Lustrato, M.P. Colombini, I. Bonaduce, E. Zanardini, P. Abbruscato, F. Cappitelli and C. Sorlini, Journal of Applied Microbiology, 98, (2005) 73.
[99] A. Adamiano, D. Fabbri, G. Falini and M. Giovanna Belcastro, Journal of Analytical and Applied Pyrolysis, 100, (2013) 173.
[100] M. Buckley and M.J. Collins, 2011, 1, (2011).
[101] M.J. Collins, M.S. Riley, A.M. Child and G. Turner-Walker, Journal of Archaeological Science, 22, (1995) 175.
[102] C. Duce, E. Bramanti, L. Ghezzi, L. Bernazzani, I. Bonaduce, M.P. Colombini, A. Spepi, S. Biagi and M.R. Tine, Dalton Transactions, 42, (2013) 5975.
[103] C. Duce, L. Ghezzi, M. Onor, I. Bonaduce, M.P. Colombini, M.R. Tine and E. Bramanti, Analytical and bioanalytical chemistry, 402, (2012) 2183. [104] J. Romero-Pastor, N. Navas, S. Kuckova, A. Rodríguez-Navarro and C.
Cardell, Journal of Mass Spectrometry, 47, (2012) 322.
[105] L. Ghezzi, C. Duce, L. Bernazzani, E. Bramanti, M.P. Colombini, M.R. Tiné and I. Bonaduce, Journal of Thermal Analysis and Calorimetry, 122, (2015) 315.
[106] M. Saito, K. Fujii and K. Marumo, Calcified Tissue International, 79, (2006) 160.
[107] R.D.A.M. Alves, J.A.A. Demmers, K. Bezstarosti, B.C.J. van der Eerden, J.A.N. Verhaar, M. Eijken and J.P.T.M. van Leeuwen, Journal of Proteome Research, 10, (2011) 4725.
[108] X. Jiang, M. Ye, X. Jiang, G. Liu, S. Feng, L. Cui and H. Zou, Journal of Proteome Research, 6, (2007) 2287.
[109] M. Jágr, A. Eckhardt, S. Pataridis and I. Mikšík, European Journal of Oral Sciences, 120, (2012) 259.
[110] C.R. Salmon, D.M. Tomazela, K.G.S. Ruiz, B.L. Foster, A.F.P. Leme, E.A. Sallum, M.J. Somerman and F.H.J. Nociti, J Proteomics., 91, (2013). [111] M.J. Collins, C.M. Nielsen–Marsh, J. Hiller, C.I. Smith, J.P. Roberts, R.V.
Prigodich, T.J. Wess, J. Csapò, A.R. Millard and G. Turner–Walker, Archaeometry, 44, (2002) 383.
[112] J.L. Bada, Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 333, (1991) 349.
[113] E. Cappellini, M.T.P. Gilbert, F. Geuna, G. Fiorentino, A. Hall, J. Thomas-Oates, P.D. Ashton, D.A. Ashford, P. Arthur, P.F. Campos, J. Kool, E. Willerslev and M.J. Collins, Naturwissenschaften, 97, (2010) 205.
[114] C. Warinner, J. Hendy, C. Speller, E. Cappellini, R. Fischer, C. Trachsel, J. Arneborg, N. Lynnerup, O.E. Craig, D.M. Swallow, A. Fotakis, R.J.
24
Christensen, J.V. Olsen, A. Liebert, N. Montalva, S. Fiddyment, S. Charlton, M. Mackie, A. Canci, A. Bouwman, F. Rühli, M.T.P. Gilbert and M.J. Collins, Scientific Reports, 4, (2014) 7104.
[115] M. Mackie, J. Hendy, A.D. Lowe, A. Sperduti, M. Holst, M.J. Collins and C.F. Speller, STAR: Science & Technology of Archaeological Research, 3, (2017) 58.
Analytical pyrolysis of proteins in samples from
artistic and archaeological objects
This chapter is based on: S.Orsini, F.Parlanti, I.Bonaduce, J Anal Appl Pyrolysis, 2017, 643-657
27
2.1 Introduction
Identification of proteins in samples from cultural heritage is often a difficult task [1,2]. From a technical point of view, paintings, polychromies, archaeological and paleontological objects are complex systems that often consist of multiple heterogeneous layers, in which pigments, fillers and other inorganic matter are mixed with organic materials. Egg yolk and/or egg white, casein, gelatin, animal glue and collagen are among the most common proteinaceous materials that can be encountered in this type of objects [1,3]. Research has evidenced that, as an effect of the long term exposure to the changeable and sometimes harsh environments where the object is displayed or stored, degradation phenomena take place. These include deamidation, hydroxylation, oxidation and carbonylation, partial hydrolysis, aggregation, cross-linking, formation of complexes with other organic binders and inorganic pigments and fillers [4-10]. Many analytical methods for the identification and characterisation of proteins in samples from artistic, archaeological and paleontological objects have been presented [1,2,11], based on staining, immunological, spectroscopic, chromatographic, and proteomics techniques.
Staining techniques are usually performed on paint cross-sections. The localisation of the proteins by stains is based on the use of dyes that may react with primary amines, or interact with specific portions of the proteins, to produce coloured and fluorescent products [12-21]. The localisation and identification of proteins by immunological methods are based on the antigen/antibody interaction. The antibody often requires to interact with 6-10 amino acids of the antigen to form the complex [21-24]. The specificity and sensibility of the immune techniques depend on the antibody availability and on their multiple epitopes. Spectroscopic techniques used to identify and localise binders are based on Infrared and Raman spectroscopy. The simultaneous
28
presence of other organic and inorganic materials can strongly compromise the success of these techniques [25-32]. Chromatographic techniques using high-performance liquid chromatography (HPLC) and gas chromatography coupled to mass spectrometry (GC/MS) permit the identification of the material based on the determination of the amino acid composition evaluated after hydrolysis [3,33-38]. The thermal decomposition of the sample by pyrolysis coupled with GC/MS (Py/GC/MS) generates pyrolytic profiles which are useful for the identification of the materials [11,39]. Despite the potential of this technique, a relatively few examples in the literature discuss the identification of proteins in highly degraded samples, especially those from paintings and polychromies [39-45]. Proteomics techniques have been more recently adopted in the field of cultural heritage to identify proteinaceous materials by enzymatic digestion and peptides characterisation [46-54].
Each of these techniques has shown the potential of being able to successfully detect, and in some cases, identify, proteins in selected samples, but all of them present more or less drawbacks, which are seldom discussed and rarely linked to degradation phenomena.
The development of a robust and reliable analytical approach for the identification of proteinaceous materials in samples from artistic, archaeological and paleontological objects, as well as understanding its limitations, requires a better understanding of the physico-chemical changes undergone by proteins with time. Given the loss of solubility observed as a consequence of ageing [38,55-60], wet chemical methods may present insuperable difficulties, which may be, in part, overcome by analytical pyrolysis.
In this paper, we present a study, based on analytical pyrolysis, aimed at characterising and better understanding the ageing of proteinaceous materials, and how this reflects on the identification of proteins in paintings and
29
archaeological objects. Pure reference materials, paint reconstructions and a set of samples from paintings and archaeological objects spanning from the 20th century AD to the 2nd century BC were investigated using evolved gas analysis-mass spectrometry (EGA/MS), pyrolysis-gas chromatography-analysis-mass spectrometry (Py/GC/MS) and double shot pyrolysis-gas chromatography-mass spectrometry (DSP/GC/MS).
2.2 Experimental
2.2.1 Reference materials and samples
Casein (CAS), egg white (EGW) and animal glue (GLU) were purchased from Bresciani srl (Milan, Italy) and were used as reference materials [4].
Easel and mural painting samples from artworks of different geographical origin and ages and one archaeological sample were investigated. Some of the samples are paint preparations which have been applied on a wooden support, and covered by paint layers. Others belong to mural paintings, which have, by definition, been more exposed to degrading environmental factors than easel paintings preparations. One sample is from an Egyptian mummy.
The sample code, the typology of the artwork and archaeological object, its age and geographical origin, as well as the nature of the proteinaceous materials present, as identified by GC-MS analysis[61] are reported in Table 2.1.
The samples (50–100 μg) were subjected to extraction with ammonia solution (NH3 2.5M), followed by purification on C18 tips, microwave-assisted acidic hydrolysis, derivatisation of the hydrolysed solution with MTBSTFA, and GC/MS analysis [61].
The amino acidic composition of artistic and archaeological samples are reported in Table 2.2 and the results are compared with reference materials, such as casein (CAS), egg white (EGW) and collagen (GLU).
30
Table 2.1: Sample code, geographical origin and dating of artistic and archaeological specimens. Sample
code
Sample origin Geographical origin
Historical period
Description GC/MS
IDA
EP1997 reconstruction Paint Italy 1997
preparation (animal glue and gypsum) of
a paint reconstruction on
wood panel
collagen
EP15thAD Panel painting Italy century AD 15th
easel painting on wood panel. The sample was collected
from the paint preparation
collagen
MP2ndBC Mural painting Greece 2nd century
BC
wall painted decorations from a
tomb
collagen
M1stAD Archaeological Egypt 1st century
AD
the sample consists in a piece of skin tissues still attached
to the linen of an embalmed Egyptian
mummy
collagen
MP4thBC Mural painting Turkey 4th BC painted decorations of the internal walls
of an house
casein
P6thAD polychromy Afghanistan 6th century AD polychromy on a clay sculpture egg
MP2ndAD Mural painting Pakistan 2ndcentury
AD
painted decorations of the external walls
of civil building
unknown
The quantitative percentage content of 11 amino acids (Pro, Hyp, Asp, Glu, Ala, Phe, Gly, Ile, Leu, Ser, and Val) of reference materials, artistic and archaeological specimens was subjected to multivariate statistical analysis using a principal components analysis (PCA) method. The data were analysed using a large amino acid reference database, obtained with the same analytical procedure (Figure 2.1). All samples were analysed as powders. Samples of bigger dimension, as those from reference materials, were homogenised with an agate mortar
31
previous analysis. When paint samples were available as flakes, these were pulverised with the aid of a scalpel.
Table 2.2: Amino acidic composition (percentage%) of reference materials, artistic and archaeological specimens.
Sample Ala Gly Val Leu Ile Ser Pro Phe Asp Glu Hyp
CAS 5.0 3.0 7.6 11.9 6.6 5.8 11.5 5.9 8.5 22.2 0.0 EGW 7.7 4.8 7.7 11.0 6.7 10.3 5.7 6.4 12.6 15.0 0.0 GLU 12.3 29.4 3.9 4.7 2.5 3.8 12.4 2.8 6.6 9.9 7.7 EP15thAD 12.3 17.5 5.2 8.4 5.2 2.9 18.5 3.7 6.2 8.6 11.4 MP2ndBC 6.2 27.1 8.1 6.9 4.3 4.5 9.4 3.5 11.7 13.0 5.4 MP4thBC 9.1 24.7 9.3 16.7 8.9 3.8 5.4 4.6 6.8 10.7 0.0 P6thAD 5.5 5.5 10.0 13.8 8.9 2.4 15.9 5.4 10.6 22.0 0.0
MP2ndAD yes yes yes yes yes yes yes yes yes yes no
1.2.2 Analysis
EGA/MS The instrumentation consists of a micro-furnace Multi-Shot Pyrolyzer EGA/Py-3030D (Frontier Lab) coupled with a gas chromatograph 6890 Agilent Technologies (Palo Alto, USA) equipped with a deactivated and uncoated stainless steel transfer tube (UADTM-2.5N, 0.15 mm i.d. × 2.5 m length, Frontier Lab).
32
Figure 2.1 : PCA score plot of painting samples.
The GC was coupled with a 5973 Agilent Mass Selective Detector (Palo Alto, USA) single quadrupole mass. A program temperature was chosen for the micro-furnace chamber: initial temperature 50 °C; 10 °C/min up to 700 °C. Analyses were performed under a helium flow (1 ml/min) with a split ratio 1:20. The micro-furnace interface temperature was kept at 100 °C higher than the micro-furnace temperature until the maximum value of 300 °C. The inlet temperature was 280 °C. The chromatographic oven was kept at 300 °C. The mass spectrometer was operated in EI positive mode (70 eV, scanning m/z 50–600). The MS transfer line temperature was 300 °C. The MS ion source temperature was kept at 230 °C and the MS quadrupole temperature at 150 °C. Samples, ranging from 30 to 500 µg, were placed into a stainless steel cup and inserted into the micro-furnace. The amount of sample used depended on the sample nature: samples relatively rich in organic materials, such as reference materials, and easel paintings were smaller in size compared to those coming from mural paintings. The sample
Animal glue Egg Casein
EP15
thAD
MP2
ndBC
P6
thAD
MP4
thAD
Fi
rs
t c
om
po
ne
nt
Second component
33
underwent a thermal decomposition in inert atmosphere (He) over the chosen heating range, and evolved gaseous compounds were transferred to the mass spectrometer and directly ionised and analysed as a function of time.
Py/GC/MS The instrumentation consists of a micro-furnace Multi-Shot Pyrolyzer EGA/Py-3030D (Frontier Lab) coupled to a gas chromatograph 6890 Agilent Technologies (USA) equipped with an HP-5MS fused silica capillary column (stationary phase 5 % diphenyl–95 % dimethyl-polysiloxane, 30 m × 0.25 mm i.d., Hewlett Packard, USA) and with a deactivated silica pre-column (2 m × 0.32 mm i.d., Agilent J&W, USA). The GC was coupled with an Agilent 5973 Mass Selective Detector operating in electron impact mode (EI) at 70 eV. Samples (30 µg for reference materials, easel paintings and polychromies and 100 µg for mural paintings and archaeological residues) were placed into a stainless steel cup and inserted into the micro-furnace. Stainless steel cups, after use were emptied and flame cleaned. Each sample was placed in a clean sample cup, which was previously analysed without sample, in order to ensure the absence of contaminants. The pyrolysis temperature was set at 600 °C for 1 min and interface temperature was 180 °C. The split/splitless injector was used with at 1:10 split ratio and it was interfaced with a liquid nitrogen cryogenic trap (Micro Jet Cryo-Trap MJT-1035E, Frontier Lab). The use of the micro Jet cryo-trap was not necessary but improved the quality of the peaks in the first region of the chromatograms, where highly volatile molecules were eluted. Chromatographic conditions were as follows: initial temperature 40°C, 2 min isothermal;10°C/min up to 140°C; 6°C/min up to 280°C; 10 C/min up to 300 C, 30 min isothermal. Carrier gas: He (purity 99.995%), constant flow 1.2 ml/min.
A sample of GLU was analysed in triplicates by Py/GC/MS in order to estimate the reproducibility that can be expected when the same, homogenous, sample is analysed with the same technique. Extracted ion chromatograms were obtained for selected m/z: m/z 67–pyrrole, m/z 91-toluene m/z 107-phenol derivatives,
34
m/z 117-indole and benzeneacetonitrile, m/z 131-methyl-indole, m/z 70 and 154-DKPs, m/z 186-diketodipyrrole (Figure 2.6). Areas where integrated, and normalised for their sum. The relative standard deviations (RSD) of the normalised areas and retention times were calculated, and resulted below 20% and 1%, respectively.
The relative standard deviations (RSD) of the normalised areas and retention times are presented in Table 2.3.
Table 2.3: Relative Standard Deviation (RSD) of retention time and normalised area of characteristic compounds of animal glue reference sample analysed by Py/GC/MS in triplicates.
Compound RSD time RSD relative area
pyrrole 1.0% 4.2% toluene 0.9% 5.8% diethylpyrazine 0.3% 16.6% benzeneacetonitrile 0.0% 14.5% indole 0.3% 12.4% diketodipyrrole 0.1% 1.7% Cyclo(Pro-Gly) 0.7% 12.7% Cyclo(Pro-Pro) 0.2% 11.5% Cyclo(Pro-Hyp) 0.5% 1.3%
DSP/GC/MS The analysis were conducted with the instrument illustrated in Py/GC/MS section above. Double shot pyrolysis (DSP) entails the pyrolysis of the same sample at two different temperatures[62-65]. Each sample (from 100 to 30 µg) was placed into a stainless steel cup and inserted twice into the micro-furnace, to be pyrolysed once at 350°C and subsequently at 550°C (pyrolysis duration 1 minute). Gases evolved at the two different temperatures were eluted in the chromatographic column and detected by mass spectrometry.
