A Comparison of Common Mass Spectrometry Approaches for Paleoproteomics
Timothy P. Cleland*
,†and Elena R. Schroeter*
,‡†Museum Conservation Institute, Smithsonian Institution, Suitland, Maryland 20746, United States
‡Department of Biological Sciences, North Carolina State University, Raleigh, North Carolina 27695, United States
ABSTRACT: The last two decades have seen a broad diversity of methods used to identify and/or characterize proteins in the archeological and paleontological record. Of these, mass spectrometry has opened an unprecedented window into the proteomes of the past, providing protein sequence data from long extinct animals as well as historical and prehistorical artifacts. Thus, application of mass spectrometry to fossil remains has become an attractive source for ancient molecular sequences with which to conduct evolutionary studies, particularly in specimens older than the proposed limit of amplifiable DNA detection. However, “mass spectrometry” covers a range of mass-based proteomic approaches, each of which utilize different technology and physical principles to generate unique types of data, with their own
strengths and challenges. Here, we discuss a variety of mass spectrometry techniques that have or may be used to detect and characterize archeological and paleontological proteins, with a particular focus on MALDI-MS, LC−MS/MS, TOF-SIMS, and MSi. The main differences in their functionality, the types of data they produce, and the potential effects of diagenesis on their results are considered.
KEYWORDS: paleoproteomics, extraction-based proteomics, imaging proteomics, diagenesis
■
INTRODUCTIONFrom its inception, the field of mass spectrometry-based proteomics has expanded to incorporate diverse techniques to characterize a broad range of tissue types. As a result, application of mass spectrometry to archeological artifacts and fossil remains has become an attractive alternative source for ancient molecular sequences with which to conduct evolutionary studies, particularly in specimens older than the proposed limit of amplifiable DNA detection.1−6Since thefirst mass spectrometry (MS) detection of osteocalcin from bone up to 53 ka in 2000,7the burgeoningfield of paleoproteomics (i.e., the identification and characterization of proteins preserved in archeological and paleontological substrates) has expanded to include everything from the controversial8−10Mesozoic avian and nonavian dinosaur bone proteins,2−5,11−14to 3.2 Ma egg shell proteins,6 to a diversity of more recent Pleistocene and younger bone, textiles, and other human generated sub- strates.15−23
Historically, early proteomic studies sought to identify fossil proteins by detecting the presence of their constituent amino acids after hydrolysis of fossil tissues with strong acids (e.g.,24−26) (for additional reviews on historical paleoproteo- mics see27−29). Such analyses can provide information about total amino acid profile present in a specimen (i.e., which residues are present and their relative abundances), but are incapable of determining the exact protein in a protein mixture from which the detected amino acids derive, or if those proteins are endogenous to the specimen.30In particular, the complete hydrolysis of the proteome present in a given tissue, without
prior separation with gel electrophoresis or similar,31precludes amino acid analysis from determining the following: (1) the sequence in which the residues were originally ordered in the source protein; (2) which amino acids (or what relative amount of them) derive from which of the proteins that comprise a complex tissue such as bone, as all proteins present in a sample are essentially combined into one profile; and (3) which amino acids are derived from common and ubiquitous contaminating proteins, such as human keratins or fungal and bacterial proteins common in sediment. As a result of these limitations, early studies performed on fossils were forced to assume that, of the entire bone proteome originally present, only collagen I, the most abundant protein in bone,32 could potentially be preserved. Thus, as every third amino acid in collagen I is a glycine,33 these studies established the criterion of an amino acid profile comprised of∼30% glycine to accept its persistence in fossil tissue.34,35This assumption is problematic, because we now know that other proteins can persist into deep time,21 skewing the ∼30% glycine ratio of collagen when hydrolyzed together. Furthermore, amino acid analysis precludes detection of most post-translational modifications (PTMs).36,37 PTMs are alterations to the side chains of the amino acids that occur after the translation of proteins from their source DNA sequence.36When they occur biologically, they help control the structure and function of proteins36,38 and are therefore an important source of biological data. Because the complete Received: September 29, 2017
pubs.acs.org/jpr Cite This:J. Proteome Res.XXXX, XXX, XXX−XXX
© XXXX American Chemical Society A DOI:10.1021/acs.jproteome.7b00703
J. Proteome Res.XXXX, XXX, XXX−XXX
hydrolysis employed in amino acid analyses results in the loss of all but the most robust PTMs (e.g., hydroxyproline34,35), the usefulness of this technique in understanding how these proteins preserve and their changes in association with evolution is limited.
The adoption of soft ionization mass spectrometry by paleoproteomic studies introduced proteomic techniques that overcome some of the shortfalls of many earlier analyses.“Soft ionization” techniques are those that ionize peptides or biomolecules without breaking them apart during the process, allowing them to be measured by MS detectors intact.39The ability to measure intact peptides, as well as subsequently fragment and measure them in a controlled manner during tandem mass spectrometry, makes it possible to identify and validate peptide primary sequences, and thus detect multiple proteins and PTMs from a complex mixture.36,40 Thus, the incorporation of soft ionization techniques in paleoproteomics has made the use of amino acid analyses to detect/identify residual proteins in historical and fossil remains obsolete, as well as other early methods that generate data that are ambiguous and unlocalized compared to our current techno- logical capabilities. For example, although pyrolysis gas chromatography coupled to mass spectrometry (py-GC−MS) has, even recently,41,42 been applied in an attempt to detect preserved proteins in fossil remains, this technique results in the destruction of most amino acids present into their constituent chemical compounds because of the extremely high temperatures employed.43,44 Thus, for the purposes of unambiguous protein identification in complex, heterogeneous samples such as fossil bone, it suffers all the same limitations to identification as amino acid analyses, but to an even broader level of nonspecificity. For these reasons, py-GC−MS is poorly suited for modern paleoproteomic studies (contra41,42), and is more properly applied to studies analyzing biomolecules that are insoluble, which prevents their analysis by soft ionization techniques (e.g., lignin,45cutin,46sporopollenin47).
The last two decades have seen a broad diversity of methods used to identify and/or characterize proteins in the archeological and paleontological record. These have included MS techniques (e.g., time of flight secondary ion mass spectrometry [TOF-SIMS],2,13,48liquid chromatography mass spectrometry [LC−MS],4−6,19,21−23,49,50 matrix assisted laser desorption ionization [MALDI] MS7,17,18,51−55) as well as non- MS methods (e.g., Fourier transform infrared [FT-IR]
spectroscopy,2,12,13 Raman spectroscopy,56 immunological assays2−4,48). Here, we discuss a variety of mass spectrometry techniques that have or may be used to detect and characterize archeological and paleontological proteins, touching on the main differences in their functionality, the types of data they produce, and the potential effects of diagenesis on their results.
Taphonomic Influences Affecting All MS Techniques The geochemical interaction of the bone with the burial environment during entombment/fossilization can produce a variety of diagenetic alterations to the tissue, many of which are still poorly understood. Such changes can be expressed at various levels, from macroscopic osteological deformation,57to molecular sequence alterations that result in PTMs that are diagenetic in origin instead of biological.58,59To date, only a few diagenetic PTMs have been identified despite almost 100 years49,60,61 of protein study on archeological and paleonto- logical remains. This dearth of diagenetic PTM data is primarily caused by both a major reduction in active research on
archeological proteins after the expansion of ancient DNA studies, and technological limitations to studies of ancient proteins in the years prior (e.g., amino acid analysis,62,63 immunological techniques,64−66thin layer and paper chroma- tography),67,68 which largely did not have the capability to identify or differentiate specific PTMs. Further, because many PTMs are labile in acidic conditions,69−71they were historically lost during protein extraction and/or sample preparation for MS analyses. Thus, until recently, deamidation of glutamine and asparagine residues58,59 was the only hypothesized diagenetic PTM characterized. Recent examination of Hol- ocene and Pleistocene remains has expanded the number and types of known diagenetic PTMs,49but the actual diversity of these modifications remains heavily uncharacterized. In addition to diagenetically induced PTMs, in vivo modifications beyond hydroxylated proline have been detected, including labile ones such as glycosylation.49Thus, the possibility that both biological PTMs and diagenetic alterations may be present on fossil proteins must be explicitly accounted for during the bioinformatic processing of any paleoproteomic data; other- wise, peptides that are both preserved and detected by MS may nevertheless remain unidentified.
In addition to post-translational modifications, variation in the spatial distributions of preserved proteins in fossil tissues complicates consistent detection in MS testing.72 Differ- entiation in preservational conditions at multiple spatial scales can influence the types of proteins and/or tissues that preserve throughout a given specimen (Figure 1). For example, at the
microscopic scale, protein content can vary between osteonal and interstitial bone73 making the biochemistry of these adjacent tissues different. During diagenesis, this in vivo chemical variation may lead to differing preservational chemistries, resulting in a motif in which a protein might be detectable in one area of the bone, but not in another. Further, as geochemical conditions within a burial environment can vary within centimeters,74−76 this spatial effect may be amplified across more than a meter of sediment required to bury fossils as large as sauropod femora, leading to“hot spots”of differential preservation or levels of degradation within large specimens, or between bones within a single bone bed.77 Thus, sampling location and/or specimen choice can potentially have a Figure 1. Depiction of different levels of variation in geochemical environments experienced by fossils. At small scales, areas of a bone can be subjected to different microenvironments, leading to intrabone variation in molecular preservation. Within a locality, these variations will be unique to each bone, generating intrabone bed variation in preservation. At the largest scale, differences in depositional environ- ments, or specific regional conditions, can lead to interbone bed (or global) variation in molecular preservation.
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX B
significant impact on success and consistent replication of paleoproteomic results. On the regional and continental scales, differences in environments of deposition (e.g., sandy vs clay deposits) or local mineralization conditions will also affect the quality and types of proteins preserved within bones. Trends in preservational environments for protein recovery remain an active area of research, and more samples from a diversity of depositional environments and ages will elucidate the best possible samples to test.
Mass Spectrometry Approaches
Mass spectrometry is a technique by which molecules or molecular fragments from a sample are ionized, then introduced to a mass detector that precisely measures their mass as a function of their mass-to-charge ratio (m/z) (for a detailed introduction to the ion physics at work in MS-based proteomics, see40). The mass profiles of the ions detected in the sample (or,“mass spectra”) that are output by the machine are then compared with databases of exact masses for known compounds, peptides, or intact proteins (depending on the MS approach being employed) for identification.78MS techniques used to measure proteins preserved within archeological and paleontological remains fall into two broad categories:
extraction-based MS and imaging MS (Figure 2). The dominant approach has been to employ protein extraction reagents to solubilize proteins from fossil tissue for subsequent analysis by MALDI-MS, MALDI-MS/MS, or LC-ESI-MS/MS.
However, recent attempts have been made to use TOF-SIMS as an imaging technique to localize SIMS protein products to the original tissue.13To date, the use of modern MALDI imaging has not been utilized for paleoproteomics studies, but we will also include a discussion of its potential as an MS imaging technique for paleoproteomic applications.
Extraction-Based Techniques
Matrix Assisted Laser Desorption Ionization−MS (MALDI-MS) and Zooarcheology by Mass Spectrometry (ZooMS).MALDI is a soft ionization technique that allows the detection of intact peptides or proteins as singly charged ions.
The maximum size of intact proteins that can be analyzed is dependent on the mass spectrometer (e.g., time-of-flight [TOF] instruments) coupled to the MALDI source.81 In extraction-based MALDI, peptides or proteins resulting from chemical extraction are mixed with a matrix (e.g., α-cyano-4- hydroxycinnamic acid; sinapic acid; 2,5-dihydroxybenzoic acid) and spotted onto a MALDI plate (Figure 3A).81The plate is then typically placed in the vacuum of the mass spectrometer where the spotted samples are irradiated with a laser to ablate and ionize the solid matrix and peptide/protein mixture, followed with analysis by the mass spectrometer, the choice of which is dependent on the paleoproteomic questions being addressed. MALDI sources are typically coupled to either TOF (e.g.,55) or Fourier transform ion cyclotron resonance (FT- ICR) instruments (e.g.,82). TOF instruments can detect a large dynamic range of masses, while FT-ICR have a smaller mass range, but can provide measurements at higher resolution and mass accuracy for analyzing the peptides/proteins.
The earliest application of modern mass spectrometry to fossil proteins utilized MALDI-MS to detect osteocalcin from various extinct species,7followed years later by identification of the osteocalcin primary sequence in fossils using MALDI-MS/
MS.54,55Whereas precursor mass approaches (i.e., MS or MS1;
peptide massfingerprinting [PMF]) measure the mass of small intact proteins or peptides, tandem mass spectrometry
Figure 2. Flowchart depicting the basic workflow of different mass spectrometry techniques, and the different data produced by each.
Tissue sections and spectral maps shown for TOF-SIMS and MALDI- IMS were adapted from Hness et al. (CC BY 2.0),79and Powers et al.
(CC BY 4.0),80respectively.
Figure 3. Illustration of the different ionization methodologies employed by MS. (A) Matrix assisted laser desorption ionization (MALDI). Extracted proteins mixed with a matrix are spotted on a plate, then ablated and ionized by irradiation with a laser. (B) Electrospray ionization (ESI). Extracted proteins in solution are sprayed through a high-voltage emitter tip. Through a combination of coloumbic repulsion and evaporation of the solvent, the hydrogen ions are transferred to the proteins, producing multiply charged ions. (C) Time offlight-secondary ion mass spectrometry (TOF-SIMS). Tissue sections are bombarded with ions or ion clusters, releasing secondarily ionized peptide fragments. (D) Matrix assisted laser desorption ionization−mass spectrometry imaging (MALDI-MSi). Tissue sections are coated with matrix, allowing peptides from the tissue beneath to be ionized as in (A) above.
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX C
approaches (i.e., sequential mass spectrometers attached together; MS/MS or MS2) allow more complex analyses including peptide fragmentation and analysis. The pattern of peaks generated during fragmentation is compared to theoretical mass spectra of peptides from a protein database, giving confident identification of the sequence and localization of PTMs to specific residues (Figure 4).78
Although MALDI has been successfully coupled with MS/
MS to identify osteocalcin sequences in fossils,54,55the majority of MALDI usage in paleoproteomics has relied on single MS and PMF to determine differences between spe- cies.15,16,51,52,83−85 PMF uses the identification of multiple intact peptide masses, or the identification of specific target peptide masses, to identify proteins within a sample (Figure 2).86 This approach, also called collagen fingerprinting or zooarcheology by mass spectrometry (ZooMS), is generally used to identify the taxonomic origin of indeterminate bone fragments found at archeological excavation localities to the supraspecific level.51,52,83−85
Because single MS approaches are incapable of providing data on the primary sequences of the peptides they detect, PMF studies have some limitations. For example, they are generally only able to determine the taxonomy of a given bone to the supraspecific level, not the species level (with a few exceptions such as the red fox and arctic fox87), because closely related species have not had sufficient time for their collagen I sequences to diverge to a degree that gives species level
information.85Additionally, though mass-shifts that indicate the presence of PTMs can be allowed for in PMF bioinformatic analyses, any observed PTMs cannot be localized to residues within a peptide sequence, precluding confident identification and validation.86 Further, without offline separation or additional sample preparation,66which is not typically part of paleoproteomic MALDI-MS workflows, only the most abundant or most easily ionized peptides in the sample will be detected. Because all peptides in the sample are ionized together, low abundant peptides or those with low ionization efficiency will remain undetected. For bone, this limits most identification to collagen I, regardless of whether there are additional preserved proteins in the sample. Most recently, MALDI-MS has been used as a screening tool for selecting samples for more rigorous analysis by LC−MS/MS.19 As MALDI-MS is more rapid and often less costly than LC−MS/
MS, this application of the technique can be effective in identifying a small, optimal sample set for tandem MS from a large pool of specimens.
Taphonomic Influences on MALDI-MS. Because MALDI-MS relies on the high-resolution detection of either many intact peptide masses (for protein identification)86 or specific intact peptide targets (for taxonomic differentia- tion),15,83,85 the results are heavily influenced by diagenetic loss or alteration of peptides. For example, if a target peptide in a ZooMS analysis is not retained in a sample, then identification of the supraspecific taxon is impossible, regardless of whether Figure 4.Example of an MS2 spectra of collagen I alpha 1 fromCastoroides ohioensis(i.e., spectra of fragments produced by the disassociation of a peptide detected in MS1 during an LC−MS/MS analyses). When a doubly charged peptide is fragmented, it results in two singly charged fragment ions: a b-fragment representing the N-terminal (labeled in blue) and a y-fragment representing the C-terminal (labeled in red). Fragmenting many ions of the same peptide at once results in many different fragmentation points within the sequence. In tandem mass spectrometry analyses, MS2 spectra such as this one are identified by matching them against an in silico database of spectra from known peptides, which include all potential b- and y-fragment ions for each potential match (with and without PTMs). From the Cleland et al.50data set athttp://datadryad.org/resource/doi:10.
5061/dryad.326fv.
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX D
large portions of the protein are otherwise well-preserved.
Nontandem MALDI-MS has also been used to detect deamidation on multiple peptides to evaluate rates of deamidation across the collagen I molecule.88,89 However, beyond detection of evidence for asparagine and glutamine deamidation, single MS of peptides is unable to directly evaluate the nature and position of diagenetic post-translational modifications, or detect peptides that are broken down through backbone cleavage/truncation.49,50Thus, MALDI MS, ZooMS, or PMF are not suitable for detailed characterizations or comparisons of diagenetic and/or preservational motifs.
Liquid Chromatography Tandem Mass Spectrometry (LC−MS/MS).LC−MS/MS first separates peptides within an extract using liquid chromatography, and then applies these separated peptides to a coupled MS system for robust and rigorous analyses using tandem mass spectrometry (i.e., MS/
MS, or MS2; Figure 4). Peptides in solution are separated by their hydrophobicity on a reverse-phase (i.e., from hydrophilic to hydrophobic) column, and are then sequentially introduced to the mass spectrometer via electrospray ionization (ESI, or nanoESI at very low flow rates). This effectively reduces the complexity of peptide mixtures and allows the detection of lower-abundance peptides that are masked during analyses by more abundant ones. While this approach is still concentration dependent (i.e., the most abundant peptides are still more likely to be fragmented than less abundant ones), the separation of the peptides allows for detection of a wider concentration range of proteins than MALDI-MS. Additionally, different types of stationary phase media (i.e., the solid substrate that peptides interact with during separation) can be used to enhance separation or preferentially separate certain types of peptides from others.90,91
ESI is the softest ionization technique, allowing the best retention of PTMs and the least chance of peptide breakdown during ionization.81In ESI, high voltage is applied to the LC solvent containing the sample, generating a spray through afine emitter tip. Through Coulombic repulsion, droplets become smaller and smaller, transferring protons to the peptides generating multiply charged ions92(Figure 3B). The ability to produce multiply charged peptide ions is a benefit of ESI over MALDI; because mass spectrometers measure ions by the ratio of their mass to their charge, multiple charges allow mass spectrometers with limited mass ranges to detect a broader range of peptide sizes.92ESI has been coupled to many types of tandem mass spectrometers, including ion traps, Orbitraps, quadrupole-TOF instruments, triple quadrupole instruments, and FT-ICR instruments, showing its utility in peptide analyses.
At a similar time as the first applications of MALDI-MS to fossils, the first ESI-LC−MS/MS analysis was applied to mammoth remains.93As a result of LC separation prior to mass analyses and continued improvements to extraction and mass spectrometry, the detection of as many as 126 proteins from a single fossil protein extraction has been achieved in LC−MS/
MS paleoproteomic experiments,21 but efforts continue to characterize proteins beyond collagen I in bone, as well as the types of diagenetic PTMs that occur or which in vivo PTMs can persist on fossil proteins.49,50,94 Additionally, although it has not yet been utilized in paleoproteomics, exploration of methods to concentrate phosphorylated peptides (e.g., from osteopontin) or glycosylated peptides (e.g., from collagen I)95−97 utilized in modern proteomics will help future MS analyses concentrate and detect more in vivo modifications that may preserve on fossil proteins.
Taphonomic Influences on LC−MS/MS. Because LC− MS/MS generates and measures fragments across a peptide, this technique is able to localize diagenetic modifications (e.g., deamidation) that may be missed by MALDI-MS ap- proaches.49,50 Further, recent advances in LC−MS/MS have led to recognition of diagenetic PTMs that were previously unknown as a result of harsh extraction methods and/or the use of amino acid analyses.49,98Additionally, because of the large number of MS2 events that occur in an LC−MS/MS analysis, both the modified and unmodified form of the same peptide can be detected, supporting the idea that the modification observed is a result of diagenesis or loss of biological PTMs.
The large number of MS2 events also provides an opportunity to detect unknown diagenetic modifications.
Despite the benefits of using LC−MS/MS for paleoproteo- mics, this technique also has its own challenges when applied to fossils. In particular, during ESI, diagenetically induced compounds present in fossils can interfere with the ionization of peptides that coelute from the LC column (i.e., ion suppression), rendering any preserved peptides undetectable in MS even if they are present.99Additionally, the specific reagents and protocols used for the chemical extraction of protein from tissues for LC−MS/MS (or any extraction-based MS technique) can cause variation in the portion of the proteome that is recovered,100 an effect that might be magnified in degraded specimens in which much of the original protein content has already been lost.
Mapping Techniques
Time of Flight-Secondary Ion Mass Spectrometry (TOF-SIMS).TOF-SIMS is a mass spectrometry technique that analyzes the surface of a sample and is capable of producing ultrafine spatial resolution of organic components found on that surface.101A cut section, or a small sample of a specimen, is introduced to a vacuum where its surface is ablated with ions (e.g., Ga+) or ion clusters (e.g., Bi3+), releasing secondarily ionized protein components from the bombarded sam- ple101−103(Figure 3C). SIMS is not a soft ionization technique, and it generally results in fragments of proteins as small as individual amino acids or amino acid components,104,105 although recent innovations to TOF-SIMS methodology have allowed for the ionization of peptides, similar to those found in MALDI imaging, and even subsequent tandem MS analyses.
These include coating the surface of samples with silver or gold (metal-assisted SIMS, or MetA SIMS)101,106 or a matrix (matrix-enhanced SIMS, or ME-SIMS),104or using extra-large ion cluster beams (e.g., Ar2000+).105 Despite the total loss of sequence data associated with conventional TOF-SIMS approaches, this technique can provide spatial distributions of the molecular information it does retain after ionization at ultrafine (submicron) scales.101,107 Thus, TOF-SIMS analyses can provide high resolution distribution maps of numerous amino acids or compounds within a tissue section that can then be compared within samples or across specimens.
Thefirst application of TOF-SIMS to fossil material was on Shuvuuia deserti (Alvaresauridae; Late Cretaceous108) beta keratin fibers in 1999.109 Since then, TOF-SIMS has been used to detect the protein breakdown products in a variety of fossil specimens, including claw sheath from Rahonavis (Paraves;110 Upper Cretaceous111),112 collagen and blood in dinosaur bone2,14and Cretaceous bird cartilage,13and melanin in fish eye,113 feathers, squid ink, and amphibian skin.114 To date, TOF-SIMS analyses applied to fossils have not yet utilized
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX E
the metal-assisted,106 matrix-enhanced,104 or ion cluster105 innovations that may allow the detection of intact peptides/
proteins by this method; as a result, the largest protein fragment assigned to a fossil using TOF-SIMS is an amino acid trimer,13 and single amino acid results dominate most paleoproteomic studies. Like amino acid analyses, this approach only retains the most robust post-translational modifications (e.g., hydroxylation of proline2) on their constituent amino acids.
It is important to note that although TOF-SIMS is often presented as a nondestructive technique (e.g., “minimally destructive”,114 “quasi-non-destructive”115) which is an attrac- tive feature for paleoproteomic studies, as well-preserved fossils can be extremely scientifically valuable and irreplaceable, this can be easily misunderstood to suggest that fossil specimens may be evaluated by TOF-SIMS intact, without sectioning or removal of material. In fact, destructive sampling/sectioning is still required to obtain a small piece of the specimen that can be placed within a vacuum for analyses. Rather, the nondestructive nature of TOF-SIMS lies in that the surface of the sections are themselves are only minimally altered by ablation with ions, in proportion to the intensity of the process (i.e., “sputtering intensity”),115,116 unlike extraction-based methods that physi- cally and chemically destroy the bone samples to retrieve preserved protein. Nevertheless, the ability to repeatedly analyze small sections of tissue is a very attractive feature of this technique for analyzing specimens for which a sufficient amount of tissue cannot be sacrificed for extract-based methodology.
Taphonomic Influences on TOF-SIMS.Because the harsh ionization of TOF-SIMS breaks peptides apart,106,107 any modifications to proteins caused by diagenesis at the peptide level (e.g., backbone breakage, deamidation, preferential loss of specific regions of the protein molecule)49cannot be observed using this technique. However, the excellent spatial resolution capabilities of TOF-SIMS can document areas of protein loss in fossil remains by mapping areas in which no compounds consistent with proteins are detected. This spatial distribution information will expand our knowledge of “hot spots” and variation in protein loss and preservation, even if we are unable to directly evaluate the changes to the proteins themselves.
One additional difficulty of TOF-SIMS is its inability to differentiate surface contamination from endogenous signal. As a result, extra care must be taken to compensate for possible surface contamination by utilizing controls to identify the contaminating molecules (e.g., adjacent sediments or additional negative controls).
MALDI Imaging Mass Spectrometry (MALDI-IMS;
MSi). MALDI-IMS, or simply MSi, combines the mapping capabilities of TOF-SIMS with MALDI ionization method- ology. Unlike extraction-based MALDI-MS, MSi analyzes a matrix-coated tissue section, which is placed inside a vacuum and irradiated with a laser to generate ions that can be measured by a coupled MS103,107(Figure 3D). After collecting a rasterized distribution of spectra across the tissue corresponding to intact peptide masses, peptide ion distribu- tions can be mapped on the surface, similar to TOF-SIMS. To identify peptides corresponding to the ion masses detected in MS, tandem MS can be applied to the MSi;117 however, the most common approach is to correlate peptide masses generated in MSi against high resolution LC−MS/MS data of proteins extracted from identical/adjacent tissue (e.g.,117−119).
Thus, combining MSi with LC−MS/MS data has the potential
to provide complementary information on preservational motifs of proteins within fossils, MSi elucidating spatial distributions that are lost during tissue homogenization for LC−MS/MS, and LC−MS/MS effectuating peptide sequencing and PTM localization not obtainable during a single MS event.
MALDI mass spectrometry imaging (MSi) has not yet been applied to fossil remains; however, MSi can provide extensive spatial distribution information that to date has only been detected by antibodies from fossil remains (e.g.,120). Further, as application of MSi expands within the field of modern biology,103 it will become a more ideal technique for paleoproteomic investigations than the harsh ionization of TOF-SIMS. Coupling the soft ionization of MALDI to spatial mapping has already allowed for fine spatial detection of peptides and proteins from a diversity of extant tissues, including highly mineralized tissues such as antler118 and bone.121 It may in the future provide information on the optimal areas of fossil tissue to conduct selective microsampling (through laser microdissection73) to maximize target proteins and PTMs detection from fossil samples. The ability of this technique to localize peptides, sequence information, and post- translational modifications within tissues, as well as detect variation in patterns of preservation and of these features across fossil tissue microstructure, will open new avenues of research into the process of diagenesis and preservation that we have been unable to explore using past methods.
Taphonomic Influences on MSi.Like TOF-SIMS, spatial variation in areas of protein loss and/or preservation will have a major influence on MSi analyses, potentially resulting in differential success depending on where specimens are sectioned. However, when sectioning is successful, hot spots of preservation within the tissue microstructure that would otherwise remain unknown may become evident. Because MSi retains intact molecular ions that are lost with TOF- SIMS,106,107 this technique may also be able to show distributional changes of diagenetic PTMs (e.g., deamidation, protein backbone cleavage) that are inaccessible by any other MS method currently applied to fossils, as well as more definitively differentiate surface contamination (e.g., keratin) from endogenous signal. As with LC−MS/MS and MALDI- MS, additional external controls (e.g., adjacent sediments or additional negative controls) should be employed to allow for a direct way to detect contamination added as a result of taphonomy, sample handling, or storage conditions.
Future investigations of fossil remains by MSi will likely reveal additional technical difficulties like the previously described methods; however, it is a method that has extraordinary potential for understanding protein diagenesis in deep time.
■
CONCLUSIONSPaleoproteomics is a nascentfield that is continuing to expand in scope and type of questions it addresses. Continued application of extraction based methods will expand our knowledge of what types of PTMs can preserve within the fossil record, as well as what alterations may be caused by diagenetic processes. The continued application of TOF-SIMS and novel usage of MSi on fossil tissues will help us fully develop and robustly test models of protein preservation at unmatched spatial scales.
However, application of correct mass spectrometry techni- ques for the hypothesis being tested is critical to the utility of the data produced for evolutionary and taphonomic studies.
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX F
Archeologists and paleontologists who want to pursue paleoproteomic questions must remain vigilant in following emerging trends in biological mass spectrometry and move away from antiquated approaches for analyzing fossil proteins, always being mindful that the destruction of irreplaceable samples should be balanced by the long-term value of the data produced. We argue that because all MS analyses are, to some extent, destructive, but fossils are nonrenewable, techniques that provide data that is both ambiguous and unlocalized compared to our current capabilities, such as amino acid analyses and py-GC−MS, should be rendered obsolete, or relegated to the status of secondary confirmation when there is sufficient sample available. Additionally, beyond the need to move past obsolete techniques, the field of paleoproteomics should continually re-examine its hypotheses as new technologies become availablefor as our ability and capacity to detect evolves, the challenges to our assumptions and the questions we ask should evolve with them.
■
AUTHOR INFORMATION Corresponding Authors*E-mail:[email protected].
*E-mail:[email protected].
ORCID
Timothy P. Cleland: 0000-0001-9198-2828
Elena R. Schroeter: 0000-0003-4314-2976 Notes
The authors declare no competingfinancial interest.
■
ACKNOWLEDGMENTSWe would like to thank M. Schweitzer for helpful comments on earlier drafts of this manuscript. This project was supported by internal funding at the Smithsonian Institution (TPC) and external funding was provided to ERS through an NSF INSPIRE (EAR1344198) grant and an Arnold O. Beckman Postdoctoral Fellowship Award.
■
(1) Asara, J. M.; Garavelli, J. S.; Slatter, D. A.; Schweitzer, M. H.;REFERENCES Freimark, L. M.; Phillips, M.; Cantley, L. C. Interpreting Sequences from Mastodon and T. rex.Science2007,317(5843), 1324−1325.(2) 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.; Cottrell, J. S.; Horner, J. R.; Cantley, L. C.;
Kalluri, R.; Asara, J. M. Biomolecular Characterization and Protein Sequences of the Campanian HadrosaurB. canadensis.Science2009, 324(5927), 626−631.
(3) Schweitzer, M. H.; Zheng, W.; Cleland, T. P.; Bern, M. Molecular analyses of dinosaur osteocytes support the presence of endogenous molecules.Bone2013,52, 414−423.
(4) Cleland, T. P.; Schroeter, E. R.; Zamdborg, L.; Zheng, W.; Lee, J.
E.; Tran, J. C.; Bern, M.; Duncan, M. B.; Lebleu, V. S.; Ahlf, D. R.;
Thomas, P. M.; Kalluri, R.; Kelleher, N. L.; Schweitzer, M. H. Mass Spectrometry and Antibody-Based Characterization of Blood Vessels fromBrachylophosaurus canadensis.J. Proteome Res. 2015,14, 5252− 5262.
(5) Schroeter, E. R.; DeHart, C. J.; Cleland, T. P.; Zheng, W.;
Thomas, P. M.; Kelleher, N. L.; Bern, M.; Schweitzer, M. H. Expansion for the Brachylophosaurus canadensis Collagen I Sequence and Additional Evidence of the Preservation of Cretaceous Protein. J.
Proteome Res.2017,16(2), 920−932.
(6) Demarchi, B.; Hall, S.; Roncal-Herrero, T.; Freeman, C. L.;
Woolley, J.; Crisp, M. K.; Wilson, J.; Fotakis, A.; Fischer, R.; Kessler, B.
M.; Jersie-Christensens, R. R.; Olsen, J. V.; Haile, J.; Thomas, J.;
Marean, C. W.; Parkington, J.; Presslee, S.; Lee-Thorp, J.; Ditchfield, P.; Hamilton, J. F.; Ward, M. W.; Wang, C. M.; Shaw, M. D.; Harrison, T.; Dominguez-Rodrigo, M.; MacPheel, R. D. E.; Kwekason, A.; Ecker, M.; Horwitz, L. K.; Chazan, M.; Kroger, R.; Thomas-Oates, J.;
Harding, J. H.; Cappellini, E.; Penkman, K.; Collins, M. J. Protein sequences bound to mineral surfaces persist into deep time.eLife2016, 5, e17092.
(7) Ostrom, P. H.; Schall, M.; Gandhi, H.; Shen, T.-L.; Hauschka, P.
V.; Strahler, J. R.; Gage, D. A. New strategies for characterizing ancient proteins using matrix-assisted laser desorption ionization mass spectrometry.Geochim. Cosmochim. Acta2000,64(6), 1043−1050.
(8) Buckley, M.; Walker, A.; Ho, S. Y. W.; Yang, Y.; Smith, C.;
Ashton, P.; Oates, J. T.; Cappellini, E.; Koon, H.; Penkman, K.;
Elsworth, B.; Ashford, D.; Solazzo, C.; Andrews, P.; Strahler, J.;
Shapiro, B.; Ostrom, P.; Gandhi, H.; Miller, W.; Raney, B.; Zylber, M.
I.; Gilbert, M. T. P.; Prigodich, R. V.; Ryan, M.; Rijsdijk, K. F.; Janoo, A.; Collins, M. J. Comment on ″Protein Sequences from Mastodon and Tyrannosaurus rex Revealed by Mass Spectrometry″.Science2008, 319(5859), 33−33.
(9) Pevzner, P. A.; Kim, S.; Ng, J. Comment on″Protein Sequences from Mastodon and Tyrannosaurus rex Revealed by Mass Spectrometry″.Science2008,321(5892), 1040−1040.
(10) Buckley, M.; Warwood, S.; van Dongen, B.; Kitchener, A. C.;
Manning, P. L. A fossil protein chimera; difficulties in discriminating dinosaur peptide sequences from modern cross-contamination.Proc. R.
Soc. London, Ser. B2017,284(1855), 20170544.
(11) Asara, J. M.; Schweitzer, M. H.; Freimark, L. M.; Phillips, M.;
Cantley, L. C. Protein Sequences from Mastodon and Tyrannosaurus Rex Revealed by Mass Spectrometry.Science2007,316(5822), 280−
285.
(12) Lee, Y.-C.; Chiang, C.-C.; Huang, P.-Y.; Chung, C.-Y.; Huang, T. D.; Wang, C.-C.; Chen, C.-I.; Chang, R.-S.; Liao, C.-H.; Reisz, R. R.
Evidence of preserved collagen in an Early Jurassic sauropodomorph dinosaur revealed by synchrotron FTIR microspectroscopy. Nat.
Commun.2017,8, 14220.
(13) Jiang, B.; Zhao, T.; Regnault, S.; Edwards, N. P.; Kohn, S. C.; Li, Z.; Wogelius, R. A.; Benton, M. J.; Hutchinson, J. R. Cellular preservation of musculoskeletal specializations in the Cretaceous bird Confuciusornis.Nat. Commun.2017,8, 14779.
(14) Bertazzo, S.; Maidment, S. C. R.; Kallepitis, C.; Fearn, S.;
Stevens, M. M.; Xie, H.-n. Fibres and cellular structures preserved in 75-million-year-old dinosaur specimens. Nat. Commun. 2015, 6 (7352), 1−8.
(15) O’Connor, S.; Solazzo, C.; Collins, M. Advances in identifying archaeological traces of horn and other keratinous hard tissues.Stud.
Conserv.2015,60(6), 393−417.
(16) Solazzo, C. Follow-up on the characterization of peptidic markers in hair and fur for the identification of common North American species. Rapid Commun. Mass Spectrom. 2017, 31 (17), 1375−1384.
(17) Welker, F.; Collins, M. J.; Thomas, J. A.; Wadsley, M.; Brace, S.;
Cappellini, E.; Turvey, S. T.; Reguero, M.; Gelfo, J. N.; Kramarz, A.;
Burger, J.; Thomas-Oates, J.; Ashford, D. A.; Ashton, P. D.; Rowsell, K.; Porter, D. M.; Kessler, B.; Fischer, R.; Baessmann, C.; Kaspar, S.;
Olsen, J. V.; Kiley, P.; Elliott, J. A.; Kelstrup, C. D.; Mullin, V.;
Hofreiter, M.; Willerslev, E.; Hublin, J.-J.; Orlando, L.; Barnes, I.;
MacPhee, R. D. E. Ancient proteins resolve the evolutionary history of Darwin’s South American ungulates.Nature2015,522(7554), 81−84.
(18) Welker, F.; Hajdinjak, M.; Talamo, S.; Jaouen, K.; Dannemann, M.; David, F.; Julien, M.; Meyer, M.; Kelso, J.; Barnes, I.; Brace, S.;
Kamminga, P.; Fischer, R.; Kessler, B. M.; Stewart, J. R.; Pääbo, S.;
Collins, M. J.; Hublin, J.-J. Palaeoproteomic evidence identifies archaic hominins associated with the Châtelperronian at the Grotte du Renne.
Proc. Natl. Acad. Sci. U. S. A.2016,113(40), 11162−11167.
(19) Welker, F.; Smith, G. M.; Hutson, J. M.; Kindler, L.; Garcia- Moreno, A.; Villaluenga, A.; Turner, E.; Gaudzinski-Windheuser, S.
Middle Pleistocene protein sequences from the rhinoceros genus
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX G
Stephanorhinus and the phylogeny of extant and extinct Middle/Late Pleistocene Rhinocerotidae.PeerJ2017,5, e3033.
(20) Cappellini, E.; Gilber, M. T. P.; Geuna, F.; Fiorentino, G.; Hall, A.; Thomas-Oates, J.; Ashton, P. D.; Ashford, D. A.; Arthur, P.;
Campos, P. F.; Kool, J.; Willerslev, E.; Collins, M. J. A multidisciplinary study of archaeological grape seeds. Naturwissenschaften 2010, 97, 205−217.
(21) Cappellini, E.; Jensen, L. J.; Szklarczyk, D.; Ginolhac, A.; da Fonseca, R. A. R.; Stafford, T. W.; Holen, S. R.; Collins, M. J.;
Orlando, L.; Willerslev, E.; Gilbert, M. T. P.; Olsen, J. V. Proteomic Analysis of a Pleistocene Mammoth Femur Reveals More than One Hundred Ancient Bone Proteins.J. Proteome Res.2012,11(2), 917−
926.
(22) Orlando, L.; Ginolhac, A.; Zhang, G.; Froese, D.; Albrechtsen, A.; Stiller, M.; Schubert, M.; Cappellini, E.; Petersen, B.; Moltke, I.;
Johnson, P. L. F.; Fumagalli, M.; Vilstrup, J. T.; Raghavan, M.;
Korneliussen, T.; Malaspinas, A.-S.; Vogt, J.; Szklarczyk, D.; Kelstrup, C. D.; Vinther, J.; Dolocan, A.; Stenderup, J.; Velazquez, A. M. V.;
Cahill, J.; Rasmussen, M.; Wang, X.; Min, J.; Zazula, G. D.; Seguin- Orlando, A.; Mortensen, C.; Magnussen, K.; Thompson, J. F.;
Weinstock, J.; Gregersen, K.; Roed, K. H.; Eisenmann, V.; Rubin, C.
J.; Miller, D. C.; Antczak, D. F.; Bertelsen, M. F.; Brunak, S.; Al- Rasheid, K. A. S.; Ryder, O.; Andersson, L.; Mundy, J.; Krogh, A.;
Gilbert, M. T. P.; Kjaer, K.; Sicheritz-Ponten, T.; Jensen, L. J.; Olsen, J.
V.; Hofreiter, M.; Nielsen, R.; Shapiro, B.; Wang, J.; Willerslev, E.
RecalibratingEquusevolution using the genome sequence of an early Middle Pleistocene horse.Nature2013,499(7456), 74−78.
(23) Sawafuji, R.; Cappellini, E.; Nagaoka, T.; Fotakis, A. K.; Jersie- Christensen, R. R.; Olsen, J. V.; Hirata, K.; Ueda, S. Proteomic profiling of archaeological human bone.R. Soc. Open Sci.2017,4(6), 161004.
(24) Abelson, P. H. Organic constituents of fossils.Mem. - Geol. Soc.
Am.1957,67, 87−92.
(25) Bada, J. L.; Kvenvolden, K. A.; Peterson, E. Racemization of Amino Acids in Bones.Nature1973,245(5424), 308−310.
(26) Grupe, G.; Turban-Just, S. Amino acid composition of degraded matrix collagen from archaeological human bone. Anthropologischer Anzeiger1998,56, 213−226.
(27) Cappellini, E.; Collins, M. J.; Gilbert, M. T. P. Unlocking Ancient Protein Palimpsests.Science2014,343(6177), 1320−1322.
(28) Vinciguerra, R.; De Chiaro, A.; Pucci, P.; Marino, G.; Birolo, L.
Proteomic strategies for cultural heritage: From bones to paintings.
Microchem. J.2016,126(Supplement C), 341−348.
(29) Dallongeville, S.; Garnier, N.; Rolando, C.; Tokarski, C. Proteins in Art, Archaeology, and Paleontology: From Detection to Identification.Chem. Rev.2016,116(1), 2−79.
(30) Armstrong, W. G.; Halstead, L. B.; Reed, F. B.; Wood, L. Fossil Proteins in Vertebrate Calcified Tissues.Philos. Trans. R. Soc., B1983, 301(1106), 301−343.
(31) Fountoulakis, M.; Lahm, H.-W. Hydrolysis and amino acid composition analysis of proteins.Journal of Chromatography A1998, 826(2), 109−134.
(32) Hauschka, P. V.; Wians, F. H. Osteocalcin-hydroxyapatite interaction in the extracellular organic matrix of bone.Anat. Rec.1989, 224(2), 180−188.
(33) Kadler, K. E.; Baldock, C.; Bella, J.; Boot-Handford, R. P.
Collagens at a glance.J. Cell Sci.2007,120(12), 1955−1958.
(34) Schweitzer, M. H.; Johnson, C.; Zocco, T. G.; Horner, J. R.;
Starkey, J. R. Preservation of Biomolecules in Cancellous Bone of Tyrannosaurus rex. Journal of Vertebrate Paleontology 1997, 17 (2), 349−359.
(35) Harbeck, M.; Grupe, G. Experimental chemical degradation compared to natural diagenetic alteration of collagen: implications for collagen quality indicators for stable isotope analysis. Archaeological and Anthropological Sciences2009,1(1), 43−57.
(36) Mann, M.; Jensen, O. N. Proteomic analysis of post-translational modifications.Nat. Biotechnol.2003,21(3), 255−61.
(37) Seo, J.; Lee, K.-J. Post-translational modifications and their biological functions: Proteomic analysis and systematic approaches.
Biochemistry and Molecular Biology2004,37(1), 35−44.
(38) Witze, E. S.; Old, W. M.; Resing, K. A.; Ahn, N. G. Mapping protein post-translational modifications with mass spectrometry.Nat.
Methods2007,4(10), 798−806.
(39) Murray, K. K.; Boyd, R. K.; Eberlin, M. N.; Langley, G. J.; Li, L.;
Naito, Y. Definitions of terms relating to mass spectrometry (IUPAC Recommendations 2013).Pure Appl. Chem.2013,85(7), 1515−1609.
(40) Savaryn, J. P.; Toby, T. K.; Kelleher, N. L. A researcher’s guide to mass spectrometry-based proteomics. Proteomics2016,16, 2435−
2443.
(41) Kaal, J.; López-Costas, O.; Martínez Cortizas, A. Diagenetic effects on pyrolysis fingerprints of extracted collagen in archaeological human bones from NW Spain, as determined by pyrolysis-GC−MS.
Journal of Archaeological Science2016,65, 1−10.
(42) Saitta, E. T.; Rogers, C.; Brooker, R. A.; Abbott, G. D.; Kumar, S.; O’Reilly, S. S.; Donohoe, P.; Dutta, S.; Summons, R. E.; Vinther, J.
Low fossilization potential of keratin protein revealed by experimental taphonomy.Palaeontology2017,60(4), 547−556.
(43) Chiavari, G.; Galletti, G. C. Pyrolysis-gas chromatography/mass spectrometry of amino acids.J. Anal. Appl. Pyrolysis1992,24, 123−
127.
(44) Boon, J. J.; Leeuw, D. Amino acid sequence information in proteins and complex proteinaceous material revealed by pyrolysis- capillary gas chromatography−low and high resolution. J. Anal. Appl.
Pyrolysis1987,11, 313−327.
(45) Schellekens, J.; Buurman, P.; Pontevedra-Pombal, X. Selecting parameters for the environmental interpretation of peat molecular chemistry - A pyrolysis-GC−MS study.Org. Geochem.2009,40(6), 678−691.
(46) Gupta, N. S.; Pancost, R. D. Biomolecular and physical taphonomy of angiosperm leaf during early decay: Implications for fossilization.Palaios2004,19(5), 428−440.
(47) al Sandouk-Lincke, N. A.; Schwarzbauer, J.; Hartkopf-Fröder, C.; Volk, H.; Fuentes, D.; Young, M.; Littke, R. The effect of different pyrolysis temperatures on organic microfossils, vitrain and amber−A comparative study between laser assisted- and Curie Point-pyrolysis− gas chromatography/mass spectrometry.J. Anal. Appl. Pyrolysis2014, 107, 211−223.
(48) Schweitzer, M. H.; Suo, Z.; Avci, R.; Asara, J. M.; Allen, M. A.;
Arce, F. T.; Horner, J. R. Analyses of Soft Tissue from Tyrannosaurus rex Suggest the Presence of Protein.Science2007,316(5822), 277− 280.
(49) Cleland, T. P.; Schroeter, E. R.; Schweitzer, M. H. Biologically and diagenetically derived peptide modifications in moa collagens.
Proc. R. Soc. London, Ser. B2015,282(1808), 20150015.
(50) Cleland, T. P.; Schroeter, E. R.; Feranec, R. S.; Vashishth, D.
Peptide sequences from the first Castoroides ohioensis skull and the utility of old museum collections for palaeoproteomics. Proc. R. Soc.
London, Ser. B2016,283(1832), 20160593.
(51) Buckley, M.; Collins, M.; Thomas-Oates, J.; Wilson, J. C.
Species identification by analysis of bone collagen using matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom.2009,23(23), 3843−3854.
(52) Richter, K. K.; Wilson, J.; Jones, A. K. G.; Buckley, M.; van Doorn, N.; Collins, M. J. Fish ’n chips: ZooMS peptide mass fingerprinting in a 96 well plate format to identify fish bone fragments.
Journal of Archaeological Science2011,38(7), 1502−1510.
(53) Nielsen-Marsh, C. M.; Ostrom, P. H.; Gandhi, H.; Shapiro, B.;
Cooper, A.; Hauschka, P. V.; Collins, M. J. Sequence preservation of osteocalcin protein and mitochondrial DNA in bison bones older than 55 ka.Geology2002,30(12), 1099−1102.
(54) Ostrom, P. H.; Gandhi, H.; Strahler, J. R.; Walker, A. K.;
Andrews, P. C.; Leykam, J.; Stafford, T. W.; Kelly, R. L.; Walker, D. N.;
Buckley, M.; Humpula, J. Unraveling the sequence and structure of the protein osteocalcin from a 42 ka fossil horse. Geochim. Cosmochim.
Acta2006,70(8), 2034−2044.
DOI:10.1021/acs.jproteome.7b00703 J. Proteome Res.XXXX, XXX, XXX−XXX H
