Ancient protein
Ancient proteins are the ancestors of modern proteins that survive as molecular fossils. Certain structural features of functional importance, particularly relating to metabolism and reproduction, are often conserved through geologic time. Early proteins consisted of simple amino acids, with more complicated amino acids being formed at a later stage through biosynthesis. Such late-arising amino acids included molecules like: histidine, phenylalanine, cysteine, methionine, tryptophan, and tyrosine. Ancient enzymatic proteins performed basic metabolic functions and required the presence of specific co-factors. The characteristics and ages of these proteins can be traced through comparisons of multiple genomes, the distribution of specific architectures, amino acid sequences, and the signatures of specific products caused by particular enzymatic activities. Alpha and beta proteins (α/β) are considered the oldest class of proteins.[1][2]
Mass spectrometry is one analytical method used to determine the mass and chemical makeup of peptides. Ancestral sequence reconstruction takes place through the collection and alignment of homologous amino acid sequences. These sequences must bear a sufficient amount of diversity to contain phylogenetic signals that resolve evolutionary relationships and allow for further deduction of targeted ancient phenotype. From there a phylogenetic tree can be constructed to illustrate the genetic resemblance between various amino acid sequences and common ancestors. The ancestral sequence is then inferred and reconstructed through maximum likelihood at the phylogenetic node(s). From there, encoding genes are synthesized, expressed, purified, and incorporated into the genome of an extant host organisms. Functionality and product properties are observed and experimentally characterized. Using a greater degree of variance in representative monomeric proteins will increase the overall precision of the results.[1][2]
History
In 1955, Philip Abelson published a short paper[3] that laid out what has become, through several cycles of technical advances, the field of palaeoproteomics or ancient protein research. He was the first to propose that amino acids, and therefore proteins, were present in a fossil bone millions of years old which gave clues about the evolution of very early life forms on our planet. Only a few years later, Hare and Abelson (1968) conducted another pioneering analysis on shells and found out that amino acids degrade or change their internal L to D configuration progressively over time, and that this could thus be used as dating tool, in what is called amino acid dating or amino acid racemization.[4] This dating approach was later shown to be a very capable tool for dating periods extending further back than the limits of radiocarbon at ca. 50,000 years.[5]
Structure and evolution
Ecological and geological events that changed the conditions of Earth's global environment effected the evolution of protein structure. The Great Oxidation Event, triggered by the development of phototrophic organisms like cyanobacteria, resulted in a world-wide increase in oxygen. This pressured various groups of anaerobic prokaryotes, changing the microbial diversity and global metabolome, as well as altering enzyme substrates and kinetics.[1][2]
Certain areas of proteins are more prone to undergo evolutionary change at a rapid rate, while others are unusually tolerant. Essential genes - or sequences of genetic material responsible for protein architecture, structure, catalytic metal co-factor binding centers, or interaction - will experience little change compared to the rest of the genetic material. Portions of this material will be confronted with genetic mutations that affect amino acid sequencing. These mutations laid the ground for other mutations and interactions that had major consequences towards protein structure and function, resulting in proteins with similar sequences serving entirely different purposes.[1][2][6]
Joseph Thornton, an evolutionary biologist, researched steroid hormones and their binding receptors to map their evolutionary relationship. He inserted DNA molecules, equipped with reconstructed amino acid sequences from ancient proteins, into in-vitro cells to make them synthesize ancestral proteins. The team discovered that reconstructed ancestral protein were capable of reconfiguration in response to multiple hormones.[6] Additional studies conducted by other research teams indicate the evolutionary development of greater protein specificity over time. Ancestral organisms required proteins - mainly enzymes - capable of catalyzing a broad range of biochemical reactions to survive with a limited proteostome. Subfunctionalization and gene duplication in multifunctional and promiscuous proteins led to the development of simpler molecules with the ability to perform more specific tasks. Not all studies concur however. Some results suggest evolutionary trends through less-specific intermediates or molecules bearing two high-specificity states or decreased specificity altogether.[7]
A second apparent evolutionary trend is the global transition away from thermostability for mesophilic protein lineages. The temperature at which various ancient proteins melt was correlated with the optimum growth temperature of extinct or extant organisms. The higher temperatures of the Precambrian affected optimum growth temperatures. Higher thermostability in proteinaceous structures facilitated their survival under more critical conditions. Heterogeneous environments, neutral drift, random adaptations, mutations, and evolution are some of the factors that influenced this non-linear transition and caused fluctuation in thermostability. This led to the development of alternative mechanisms of surviving fluctuating environmental conditions.[7]
Certain ancestral proteins followed alternative evolutionary routes to obtain the same functional outcomes. Organisms that evolved along different pathways developed proteins that performed similar functions. In some cases, changing a single amino acid was enough to provide an entirely new function. Other ancestral sequences became over-stabilized and were incapable of conformational changes in response to shifting environmental stimuli.[6]
Palaeoproteomics
Overview
Paleoproteomics is a relatively young and rapidly growing field of molecular science in which proteomics-based sequencing technology is used to resolve species identification and evolutionary relationships of extinct taxa. While complementary to paleogenomics in application, the study of ancient proteins has the potential to reveal older, more complete phylogenies due to the relative stability of amino acids in proteins as compared to the nucleic acids of DNA.[8] Ancient protein studies can further reveal types and sources of recovered tissues,[9] as well as the developmental stages of fossilized specimens.[10] Paleoproteomics can also be extended to archaeological materials such as textiles, animal skins, food remains, and pottery.
Palaeoproteomics[11] is a neologism used to describe the application of mass spectrometry (MS)-based approaches to the study of ancient proteomes. As with palaeogenomics (the study of ancient DNA, aDNA), it intersects evolutionary biology, archaeology and anthropology, with applications ranging from the phylogenetic reconstruction of extinct species to the investigation of past human diets and ancient diseases.
The field was pioneered when Peggy Ostrom [12] used MALDI-TOF with post source decy to sequence osteocalcin in 50,000 year old bison bone. With the advent of soft-ionization and use of coupled liquid chromatography (LC) and tandem MS systems (i.e. MALDI-TOF-MS and LC-MS/MS), the high-throughput of information about ancient proteins has grown significantly. One reason for this, is that proteins have been shown to be robust molecule in archaeological and other Quaternary samples. In situations where ancient DNA has long since degraded to sub-useful fragments, protein sequencing has helped to answer phylogenetic questions such as the placing of Toxodon sp. Analysis of more complex protein mixtures, are just emerging. Analysis of binding adhesives usually thought to derive from birch resin (Betula L.), have shown to actually be animal glue.[13]
Background
Philip Abelson first characterized the findings of ancient amino acid residues from fossilized materials in 1955, proposing that the peptide bonds of proteins might persist for millions of years.[14] These initial discoveries were limited by available methodologies, and so protein sequencing remained an elusive idea for almost four decades. In 2000, mass spectrometry (MS) revealed the presence of osteocalcin in ancient bone samples[15] and ignited a renewed interest in protein’s potential as a tool for molecular paleontology.[16]
The development of higher resolution instruments further increased the efficiency and depth of ancient protein recovery. In 2012, the first extended fossil bone proteome from a Pleistocene mammoth femur was confidently retrieved and identified,[17] strengthening the future of paleoproteomics research.
Collagen Type I
The analysis of ancient bone proteomes has primarily focused on the identification of collagen type I (COL1), the dominant protein found in mineralized tissues.[8][18][19] Collagen is highly conserved across species[20] and comprises about 90% of organic bone compounds. Fibrillar collagens, of which COL1 is categorized, are thought to have evolved from a common metazoan ancestor,[21] thus contributing to their abundance and importance in the fossil record.
Collagen has also been found to survive much longer than other non-collagenous proteins in fossilized specimens,[22] and the protein remains intact beyond the degradation of ancient DNA (aDNA).[8][19] Its tightly coiled triple-helical structure (consisting of two genetically identical alpha-1 chains and a third genetically distinct alpha-2 chain)[20] and hydrophobic composition[8] also make this protein an excellent candidate for survival, even in temperate and humid climates that support the rapid break down of organic molecules.[20]
The taxonomic resolution of collagen has been thoroughly investigated, and it is known that amino acid substitutions can be resolved to the genus level in most medium and large mammals.[8] Species-level identification is also possible, even in small mammal remains from high thermal climates.[23] It is for these reasons that COL1 remains a key protein in paleoproteomics and phylogenetic investigations.
Non-collagenous proteins
The remaining 10% of organic bone molecules are non-collagenous proteins (NCPs). The most abundant NCP, osteocalcin, is a bone and dentin protein involved in bone assembly, often used as a marker for the bone formation process. Preserved osteocalcin was first detected via mass spectrometry (MALDI-MS) in 10,000 year-old bison bone and a 53,000-year-old walrus bone,[15] revealing phylogenetic reconstruction potential beyond the temporal limits of aDNA.
More advanced proteomic techniques have enabled the investigation of additional NCPs present in the bone extracellular matrix. Though type I collagen is the longest lived protein identified in fossilized bone specimens, the identification and sequencing of NCPs may allow for a greater taxonomic resolution than collagen-based methods.[18][22]
Other proteins
Proteomic analysis has also been applied to other fossilized and ancient materials. The examination of damaged artifacts through the sequencing of their keratin peptides[24] has allowed researchers to discriminate between horn and hoof remains of important species used at archeological sites. The keratin of textiles and animal skins worn by Ötzi, the Iceman, were also identified using peptide mass fingerprinting (PMF) from the ancient samples and from reference species.[25] Immune response proteins have illuminated the presence of infections and diseases in multiple studies of mummified human remains.[18] Additionally, the identification of egg proteins, caseins, whey globulins, and other proteinaceous materials used as binders in the paint of historical artworks has allowed for a better understanding of proper conservation methods.[18]
Sequencing methods
Analysis of a fossil sample begins with demineralization of the bone/tooth mineral matrix. Trypsin is commonly used to digest the protein residues into peptides which can then be purified and analyzed.[19]
Peptide mass fingerprinting
Peptide mass fingerprinting (PMF) is an analytical technique that can be applied to the digested protein mixture. The masses of the unknown peptides can be detected with a mass spectrometer like MALDI (matrix-assisted laser desorption/ionization) or ESI (electrospray ionization), combined with a mass analyzer,[8] and then compared to masses of peptides that are predicted to derive from known proteins.[19]
Liquid Chromatography-Tandem Mass Spectrometry
The inclusion of liquid chromatography (LC) solvents can enhance peptide electrospray ionization.[19] When combined with mass spectrometry, this allows for a much greater number of peptide ions to be analyzed with improved fragment spectra.[8] Liquid chromatography-mass spectrometry LC-MS can then be performed for peptide mass fingerprinting; however, liquid chromatography-tandem mass spectrometry (LC-MS/MS) is most often used in the case of ancient proteome analysis due to the nature of these complex samples.
De novo sequencing
At the present time, protein databases remain limited to particular taxa, so novel protein sequences that differ greatly from those available will not be identified via the protein search engines.[8] Manual interpretation through de novo sequencing remains a viable solution until these databases become more robust, and this technique will allow for identification of amino acid substitutions not previously reported.[26]
A hybrid solution of error-tolerant search algorithms that use protein sequence databases while allowing amino acid substitutions may similarly enable the identification of novel single amino acid polymorphisms (SAPs).[19]
Dinosaur collagen
A 2007 paleontology study reported the alleged discovery of endogenous collagen peptides in 68 mya Tyrannosaurus rex fossils.[27] This claim purported survival beyond experimental decay rates,[16] leading to controversy in the emerging field. The same team again reported finding similar collagen peptide sequence matches in 2009 from 80 mya hadrosaur fossils belonging to Brachylophosaurus canadensis.[28]
Subsequent studies have reanalyzed the original T. rex sequence data to infer that the sample was predominantly laboratory contaminants, soil bacteria, and bird-like hemoglobin and collagen;[29] the former protein is typically only seen in relatively recent samples.[17][22] Another exceptionally preserved hadrosaur from the Hell Creek Formation (USA), yielded none of the previous findings despite extensive testing, and only the presence of protein breakdown products were detected.[30]
Further experimentation demonstrated that contamination from other specimens present in the T. rex lab cannot be ruled out. Every peptide that was considered unique to both dinosaurs in the 2009 study could be matched to modern ostrich with much greater confidence than could be placed on their own, unique identifications.[31]
While there have been several methods described to support the authenticity of paleoproteomics, including immunological or amino acid composition and racemization data, both of these approaches have limitations and are known to yield false-positive reactions in fossils.[32] Great care must be taken to rule out contamination by determining whether sequences differ from those of all extant taxa present in the laboratory environments.[31] Deamidation has also been proposed as an effective method for distinguishing between endogenous and contaminating NCPs, when extraction protocols may permit for this evaluation.[19] This kind of research has almost the same goals with projects focused on studying the impact of differential therapeutic treatments on the hippocampus proteome of depressed mice but also has tremendously different sets of instruments.
Future perspectives
Paleoproteomics is still a young field, with most complex proteomes only being discovered in the last decade. Current proteomic methods greatly suffer from the fact that it is not a true form of sequencing, relying on probability-matching against expected results. While several MS methods are being employed to increase the robustness of retrievable data, these techniques also increase the sensitivities to contamination.[8]
Other associated fields
Zooarcheology uses mass spectrometry and protein analyses to determine the evolutionary relationship between different animal species due to differences in proteinaceous mass, for instance collagen. Techniques such as shotgun proteomics allows researchers to identify proteomes and the exact sequences of amino acids within different kinds of proteins. These sequences can be compared to other organisms within different clades to determine their evolutionary relationships within the phylogenetic tree. Proteins are also more preserved in fossils than DNA, allowing researchers to recover proteins from the enamel of 1.8 million year old animal teeth and mineral crystals of 3.8 million year old eggshells.[33]
Applications and products
Combined genome and protein sequencing research has allowed for scientists to further piece together narratives of archaic environmental conditions and past evolutionary relationships. Research into the thermostability of protein structures permits predictions of past global temperatures. Ancestral sequence reconstruction further reveals the origins of human ethanol metabolism and the evolution of various species. An example of this would be the identification and differentiation of Denisovan hominids from modern Homo sapiens sapiens through amino acid variants in collagen obtained from the former's teeth.[1][33]
The study of ancient proteins has not only helped to determine the evolutionary history of viral proteins but facilitated the development of new drugs.[6]
Benefits and limitations
Understanding protein function and evolution provides new methods of engineering and controlling evolutionary pathways to produce useful templates and byproducts - more specifically, proteins with high thermostability and broad substrate specificity.[7]
Multiple limitations as well as possible sources of error must be taken under consideration, and possible solutions or alternatives put into place. The statistical construction of ancient proteins is unverifiable and will not have identical amino acid sequences to ancestral proteins. Reconstruction can also be affected by multiple factors including: mutations; turnover rates - as prokaryotic species are more prone to genetic change than their eukaryotic counterparts, making it harder to determine their proteomic past; amino acid distribution; and limited resources of fully sequenced genomes and amino acid sequences of extant species.[1][7] Ancestral protein reconstruction also assumes that certain homologous phenotypes actually existed within ancient proteinaceous populations when in fact the data recovered is but an estimate consensus of the total pre-existing diversity. Inadequate taxonomic sampling can lead to inaccurate phylogenetic trees due to long branch attraction.[1] Proteins can also get degraded over time into small fragments and have modern proteins incorporated into them - making identification difficult or inaccurate. Last but not least, fossilized remnants contain minuscule amounts of proteins that can be used for further study and identification and actually provide less information regarding evolutionary patterns compared to genome sequences.[33]
Additional concerns regarding ancestral sequence reconstruction (ASR) method would lie in the underlying bias in thermostability due to the usage of maximum-likelihood in obtaining data. This makes ancient proteins appear more stable than they actually were. Using alternative reconstruction methods - for instance the Bayesian method that incorporates and averages over the level of uncertainty - could provide a comparable reference regarding ancestral stability. However, this method performs poor reconstructions and may not accurately reflect actual conditions.[7]
See also
References
- Garcia, Amanda K.; Kaçar, Betül (2019-04-02). "How to resurrect ancestral proteins as proxies for ancient biogeochemistry". Free Radical Biology and Medicine. 140: 260–269. doi:10.1016/j.freeradbiomed.2019.03.033. ISSN 0891-5849. PMID 30951835.
- Ma, Bin-Guang; Chen, Lei; Ji, Hong-Fang; Chen, Zhong-Hua; Yang, Fu-Rong; Wang, Ling; Qu, Ge; Jiang, Ying-Ying; Ji, Cong (2008-02-15). "Characters of very ancient proteins". Biochemical and Biophysical Research Communications. 366 (3): 607–611. doi:10.1016/j.bbrc.2007.12.014. ISSN 0006-291X. PMID 18073136.
- Abelson, Philip (1955). Paleobiochemistry: organic constituents of fossils. Carnegie Institution of Washington, Yearbook, No. 54. pp. 107–109.
- Hare, P.E.; Abelson, P.H. (1968). "Racemization of amino acids in fossil shells". Carnegie Inst. Wash. Yearb. 66: 526–528.
- Penkman, Kirsty (2010-05-01). "Amino acid geochronology: its impact on our understanding of the Quaternary stratigraphy of the British Isles". Journal of Quaternary Science. 25 (4): 501–514. Bibcode:2010JQS....25..501P. doi:10.1002/jqs.1346. ISSN 1099-1417. S2CID 130512257.
- Michalowski, Jennifer (2017-03-03). "Scientists dig up proteins from the past". Science News. Retrieved 2019-07-05.
- Wheeler, Lucas C; Lim, Shion A; Marqusee, Susan; Harms, Michael J (2016-06-01). "The thermostability and specificity of ancient proteins". Current Opinion in Structural Biology. New constructs and expression of proteins • Sequences and topology. 38: 37–43. doi:10.1016/j.sbi.2016.05.015. ISSN 0959-440X. PMC 5010474. PMID 27288744.
- Buckley, Michael (2019), Lindqvist, Charlotte; Rajora, Om P. (eds.), "Paleoproteomics: An Introduction to the Analysis of Ancient Proteins by Soft Ionisation Mass Spectrometry", Paleogenomics: Genome-Scale Analysis of Ancient DNA, Population Genomics, Cham: Springer International Publishing, pp. 31–52, doi:10.1007/13836_2018_50, ISBN 978-3-030-04753-5, retrieved 2021-03-20
- Buckley, Michael; Harvey, Virginia L.; Chamberlain, Andrew T. (July 2017). "Species identification and decay assessment of Late Pleistocene fragmentary vertebrate remains from Pin Hole Cave (Creswell Crags, UK) using collagen fingerprinting". Boreas. 46 (3): 402–411. doi:10.1111/bor.12225.
- Sawafuji, Rikai; Cappellini, Enrico; Nagaoka, Tomohito; Fotakis, Anna K.; Jersie-Christensen, Rosa Rakownikow; Olsen, Jesper V.; Hirata, Kazuaki; Ueda, Shintaroh (June 2017). "Proteomic profiling of archaeological human bone". Royal Society Open Science. 4 (6): 161004. Bibcode:2017RSOS....461004S. doi:10.1098/rsos.161004. ISSN 2054-5703. PMC 5493901. PMID 28680659.
- Cappellini, Enrico; Collins, Matthew J.; Gilbert, M. Thomas P. (2014-03-21). "Unlocking Ancient Protein Palimpsests". Science. 343 (6177): 1320–1322. Bibcode:2014Sci...343.1320C. doi:10.1126/science.1249274. ISSN 0036-8075. PMID 24653025. S2CID 32042460.
- Ostrom, Peggy H; Schall, Michael; Gandhi, Hasand; Shen, Tun-Li; Hauschka, Peter V; Strahler, John R; Gage, Douglas A (March 2000). "New strategies for characterizing ancient proteins using matrix-assisted laser desorption ionization mass spectrometry". Geochimica et Cosmochimica Acta. 64 (6): 1043–1050. Bibcode:2000GeCoA..64.1043O. doi:10.1016/S0016-7037(99)00381-6.
- Bleicher, Niels; Kelstrup, Christian; Olsen, Jesper V.; Cappellini, Enrico (November 2015). "Molecular evidence of use of hide glue in 4th millennium BC Europe". Journal of Archaeological Science. 63: 65–71. doi:10.1016/j.jas.2015.08.012.
- Abelson, Philip (1955). "Paleobiochemistry: organic constituents of fossils". Yearbook. 54: 107–109 – via Carnegie Institution of Washington.
- Ostrom, Peggy H; Schall, Michael; Gandhi, Hasand; Shen, Tun-Li; Hauschka, Peter V; Strahler, John R; Gage, Douglas A (March 2000). "New strategies for characterizing ancient proteins using matrix-assisted laser desorption ionization mass spectrometry". Geochimica et Cosmochimica Acta. 64 (6): 1043–1050. Bibcode:2000GeCoA..64.1043O. doi:10.1016/S0016-7037(99)00381-6.
- Nielsen-Marsh, Christina (2002-06-01). "Biomolecules in fossil remains: Multidisciplinary approach to endurance". The Biochemist. 24 (3): 12–14. doi:10.1042/bio02403012. ISSN 0954-982X.
- Cappellini, Enrico; Jensen, Lars J.; Szklarczyk, Damian; Ginolhac, Aurélien; da Fonseca, Rute A. R.; Stafford, Thomas W.; Holen, Steven R.; Collins, Matthew J.; Orlando, Ludovic; Willerslev, Eske; Gilbert, M. Thomas P. (2011-12-14). "Proteomic Analysis of a Pleistocene Mammoth Femur Reveals More than One Hundred Ancient Bone Proteins". Journal of Proteome Research. 11 (2): 917–926. doi:10.1021/pr200721u. ISSN 1535-3893. PMID 22103443.
- Giuffrida, Maria Gabriella; Mazzoli, Roberto; Pessione, Enrica (2018-05-08). "Back to the past: deciphering cultural heritage secrets by protein identification". Applied Microbiology and Biotechnology. 102 (13): 5445–5455. doi:10.1007/s00253-018-8963-z. hdl:2318/1670455. ISSN 0175-7598. PMID 29737392. S2CID 13691178.
- Cappellini, Enrico; Prohaska, Ana; Racimo, Fernando; Welker, Frido; Pedersen, Mikkel Winther; Allentoft, Morten E.; de Barros Damgaard, Peter; Gutenbrunner, Petra; Dunne, Julie; Hammann, Simon; Roffet-Salque, Mélanie (2018-06-20). "Ancient Biomolecules and Evolutionary Inference". Annual Review of Biochemistry. 87 (1): 1029–1060. doi:10.1146/annurev-biochem-062917-012002. ISSN 0066-4154. PMID 29709200.
- Buckley, Michael (2015-05-07). "Ancient collagen reveals evolutionary history of the endemic South American 'ungulates'". Proceedings of the Royal Society B: Biological Sciences. 282 (1806): 20142671. doi:10.1098/rspb.2014.2671. ISSN 0962-8452. PMC 4426609. PMID 25833851.
- Rodriguez-Pascual, Fernando; Slatter, David Anthony (December 2016). "Collagen cross-linking: insights on the evolution of metazoan extracellular matrix". Scientific Reports. 6 (1): 37374. Bibcode:2016NatSR...637374R. doi:10.1038/srep37374. ISSN 2045-2322. PMC 5120351. PMID 27876853.
- Wadsworth, Caroline; Buckley, Mike (2014-03-30). "Proteome degradation in fossils: investigating the longevity of protein survival in ancient bone: Proteome degradation in fossils". Rapid Communications in Mass Spectrometry. 28 (6): 605–615. doi:10.1002/rcm.6821. PMC 4282581. PMID 24519823.
- Buckley, Michael; Harvey, Virginia L; Orihuela, Johanset; Mychajliw, Alexis M; Keating, Joseph N; Milan, Juan N Almonte; Lawless, Craig; Chamberlain, Andrew T; Egerton, Victoria M; Manning, Phillip L (2020-10-01). Teeling, Emma (ed.). "Collagen Sequence Analysis Reveals Evolutionary History of Extinct West Indies Nesophontes (Island-Shrews)". Molecular Biology and Evolution. 37 (10): 2931–2943. doi:10.1093/molbev/msaa137. ISSN 0737-4038. PMC 7530613. PMID 32497204.
- Solazzo, Caroline; Wadsley, Marc; Dyer, Jolon M.; Clerens, Stefan; Collins, Matthew J.; Plowman, Jeffrey (2013-12-15). "Characterisation of novel α-keratin peptide markers for species identification in keratinous tissues using mass spectrometry: Novel α-keratin peptide markers for species identification". Rapid Communications in Mass Spectrometry. 27 (23): 2685–2698. doi:10.1002/rcm.6730. PMID 24591030.
- Hollemeyer, Klaus; Altmeyer, Wolfgang; Heinzle, Elmar; Pitra, Christian (2008-09-30). "Species identification of Oetzi's clothing with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry based on peptide pattern similarities of hair digests". Rapid Communications in Mass Spectrometry. 22 (18): 2751–2767. Bibcode:2008RCMS...22.2751H. doi:10.1002/rcm.3679. PMID 18720427.
- Cappellini, E.; Collins, M. J.; Gilbert, M. T. P. (2014-03-20). "Unlocking Ancient Protein Palimpsests". Science. 343 (6177): 1320–1322. Bibcode:2014Sci...343.1320C. doi:10.1126/science.1249274. ISSN 0036-8075. PMID 24653025. S2CID 32042460.
- Schweitzer, M. H.; Suo, Z.; Avci, R.; Asara, J. M.; Allen, M. A.; Arce, F. T.; Horner, J. R. (2007-04-13). "Analyses of Soft Tissue from Tyrannosaurus rex Suggest the Presence of Protein". Science. 316 (5822): 277–280. Bibcode:2007Sci...316..277S. doi:10.1126/science.1138709. ISSN 0036-8075. PMID 17431179. S2CID 42153727.
- Schweitzer, M. H.; Zheng, W.; Organ, C. L.; Avci, R.; Suo, Z.; Freimark, L. M.; Lebleu, V. S.; Duncan, M. B.; Vander Heiden, M. G.; Neveu, J. M.; Lane, W. S. (2009-04-30). "Biomolecular Characterization and Protein Sequences of the Campanian Hadrosaur B. canadensis". Science. 324 (5927): 626–631. Bibcode:2009Sci...324..626S. doi:10.1126/science.1165069. ISSN 0036-8075. PMID 19407199. S2CID 5358680.
- Bern, Marshall; Phinney, Brett S.; Goldberg, David (2009-09-04). "Reanalysis of Tyrannosaurus rex Mass Spectra". Journal of Proteome Research. 8 (9): 4328–4332. doi:10.1021/pr900349r. ISSN 1535-3893. PMC 2738754. PMID 19603827.
- Manning, Phillip L.; Morris, Peter M.; McMahon, Adam; Jones, Emrys; Gize, Andy; Macquaker, Joe H. S.; Wolff, George; Thompson, Anu; Marshall, Jim; Taylor, Kevin G.; Lyson, Tyler (2009-10-07). "Mineralized soft-tissue structure and chemistry in a mummified hadrosaur from the Hell Creek Formation, North Dakota (USA)". Proceedings of the Royal Society B: Biological Sciences. 276 (1672): 3429–3437. doi:10.1098/rspb.2009.0812. ISSN 0962-8452. PMC 2817188. PMID 19570788.
- Buckley, Michael; Warwood, Stacey; van Dongen, Bart; Kitchener, Andrew C.; Manning, Phillip L. (2017-05-31). "A fossil protein chimera; difficulties in discriminating dinosaur peptide sequences from modern cross-contamination". Proceedings of the Royal Society B: Biological Sciences. 284 (1855): 20170544. doi:10.1098/rspb.2017.0544. ISSN 0962-8452. PMC 5454271. PMID 28566488.
- Lendaro, Eugenio; Ippoliti, Rodolfo; Bellelli, Andrea; Brunori, Maurizio; Zito, Romano; Citro, Gennaro; Ascenzi, Antonio (November 1991). "On the problem of immunological detection of antigens in skeletal remains". American Journal of Physical Anthropology. 86 (3): 429–432. doi:10.1002/ajpa.1330860308. ISSN 0002-9483. PMID 1746647.
- Warren, Matthew (2019-06-26). "Move over, DNA: ancient proteins are starting to reveal humanity's history". Nature. 570 (7762): 433–436. Bibcode:2019Natur.570..433W. doi:10.1038/d41586-019-01986-x. PMID 31243383.