Elsevier

Bone

Volume 52, Issue 1, January 2013, Pages 414-423
Bone

Original Full Length Article
Molecular analyses of dinosaur osteocytes support the presence of endogenous molecules

https://doi.org/10.1016/j.bone.2012.10.010Get rights and content

Abstract

The discovery of soft, transparent microstructures in dinosaur bone consistent in morphology with osteocytes was controversial. We hypothesize that, if original, these microstructures will have molecular features in common with extant osteocytes. We present immunological and mass spectrometry evidence for preservation of proteins comprising extant osteocytes (Actin, Tubulin, PHEX, Histone H4) in osteocytes recovered from two non-avian dinosaurs. Furthermore, antibodies to DNA show localized binding to these microstructures, which also react positively with DNA intercalating stains propidium iodide (PI) and 4′,6′-diamidino-2-phenylindole dihydrochloride (DAPI). Each antibody binds dinosaur cells in patterns similar to extant cells. These data are the first to support preservation of multiple proteins and to present multiple lines of evidence for material consistent with DNA in dinosaurs, supporting the hypothesis that these structures were part of the once living animals. We propose mechanisms for preservation of cells and component molecules, and discuss implications for dinosaurian cellular biology.

Highlights

► Multiple lines of evidence support endogeneity of osteocyte-like microstructures in two dinosaurs. ► We show the first binding of bone-specific monoclonal antibody to ‘cells’ of these dinosaurs. ► Four independent lines of evidence support the presence of a component chemically consistent with DNA. ► We propose a novel mechanism for the preservation of these materials over geological time.

Introduction

The recovery of original tissues, cells and molecules from vertebrate remains dating to the Mesozoic has wide-ranging implications for a number of disciplines. We have previously reported the preservation of microstructures consistent in location and morphology with blood vessels, osteocytes, and fibrous bone matrix in multiple fossils, including dinosaurs [1], [2], and we have provided evidence that at least some original molecular fingerprints remain associated with these structures [3], [4], [5], [6]. Here, we present molecular analyses of the soft tissue structures consistent with osteocytes, recovered from two dinosaurs whose handling history and geological setting are well defined, and show that these structures demonstrate characteristics of bone cells. Applying the rules of parsimony and consideration of all data, the conclusion that these are remnants of the cells of the once living organisms is supported.

The evolution of bone as a tissue and the endoskeleton it forms allowed the radiation of vertebrates into virtually every ecological niche. Bone provides a ready reservoir of ions vital to homeostasis, permits a wide variation in organismal size, and allows efficient locomotion in every medium by providing effective resistance to muscle action as well as protection of vital organs [7], [8]. Bone comprises an organic phase dominated by collagen I, a hydroxyapatite mineral phase [9], [10], and, in most bone, blood vessels and populations of cells. The organic phase is controlled by these cells, with osteoblasts producing bone de novo and osteocytes maintaining and, along with osteoclasts, remodeling bone [7], [11]. Thus, osteocytes and osteoblasts control the expression, secretion and composition of the organic phase of bone. They also function to detect mechanical stress and compensate for this stress via remodeling [7], [12].

Bone as a tissue is first evidenced in the fossil record in ostracoderms; jawless, heavily armored fish of the Devonian (~ 417–354 million years ago) [7], [13], [14]. Even then, it is recognized as bone by the presence of osteocyte lacunae and radiating, interconnected canaliculi which, in life, are mineral-free openings housing cells and filopodia, respectively [15], [16]. The lacuno-canalicular network (LCN) in fossil bone thus reflects the presence of once-living cells in the mineralized bone matrix [11], [15], [17].

Osteocytes, terminally differentiated cells in the osteoblast lineage, are recognizable in all tetrapod taxa by their unique morphology. As they become encased and immobilized in the bone matrix, cell morphology changes [18], and dendritic processes or filopodia—long, narrow extensions of the cell membrane—radiate outward in three dimensions, connecting with other osteocytes, osteoblasts, or bone lining cells [7], [19] via gap junctions through which chemical signals, nutrients and wastes are exchanged [20].

Osteocytes outnumber other cells in bone (osteoblasts and osteoclasts) by at least a factor of 10 [7], [11], [21], yet the multiple functions of these abundant cells are still incompletely defined [22], [23]. Osteocytes contribute to bone maintenance and remodeling, and it has recently been shown that these cells have a vital role in bone mechanosensing through lateral projections that attach the filopodia to the canalicular wall [24], [25], [26], enabling immediate cellular responses to changing stresses on bone [11], [25], [27], [28].

The typical osteocyte lacunae with radiating canaliculi identified in dinosaur and other fossil bone are usually arranged in the lamellae of osteonal bone [29], [30], [31], [32]. It has been assumed that, in fossils, these lacunae are empty after death because the occupying cells have degraded. Alternatively, the lacunae may be filled with exogenous mineral deposits during fossilization. However, as early as the 1960s, work by Pawlicki [17], [33], [34], [35] demonstrated the presence of three-dimensional bodies consistent in morphology with vertebrate osteocytes. Our research built upon these discoveries, and showed that removing the mineral phase of fossil bone released soft and transparent, cell-like microstructures with internal structure and pliable filopodia, consistent with osteocytes, from dinosaur and other fossils [36], [37], [38], [39]. Similar microstructures have recently been identified in fossil bone by other investigators [40], [41], [42].

The cellular morphology of osteocytes is conserved across vertebrate taxa, so much so that osteocytes are considered to be “fully defined by their location within the bone matrix and their typical stellate morphology” [15] (bold italics added). However, fossils millions of years old have been subjected to many biogeochemical interactions within the burial environment, so simple morphology is insufficient to identify these structures as components produced by the once-living organism. Indeed, it has been suggested that these microstructures may be biofilm ‘morphs’ of original cells [43], but there is no direct evidence that biofilm will form in osteocyte lacunae, nor that biofilm would retain this three-dimensional structure after demineralization, if it did. Furthermore, the data presented here and elsewhere [38], [39] are not consistent with a biofilm source (see Discussion). Nevertheless, when claiming that something as delicate as cell bodies may persist over the course of geological time, chemical confirmation is needed.

After demineralization of bone from many fossil vertebrates, including multiple dinosaurs, we have previously reported the presence of four apparently original, still soft components. These include fibrous and pliable bone matrix, transparent, flexible blood vessel-like structures, translucent microstructures with dendritic processes consistent with osteocytes, and intravascular material, either as red, round microstructures or amorphous red material [37], [2]. We then conducted in-depth chemical and molecular analyses on two dinosaurs, Tyrannosaurus rex (MOR 1125) and Brachylophosaurus canadensis (MOR 2598). We focused first on the demineralized, fibrous extracellular matrix that, in living organisms, is comprised predominately of the protein collagen, and demonstrated collagen sequence [3], [4], [38], [39]. Because we have already shown the persistence of these structures morphologically in multiple fossils, we chose to concentrate analytical efforts on the two dinosaurs already partially characterized and for which handling history and depositional setting were known. While our previous analyses were directed toward characterizing the fibrous demineralized matrix, we have not similarly analyzed the other components. Now, we conduct similar studies on the osteocyte-like microstructures (this document) and structures morphologically consistent with blood vessels [44].

We isolated transparent cell-like microstructures with dendritic processes, some containing internal contents, from bone of T. rex (MOR 1125) and B. canadensis (MOR 2598), and hypothesized that if these were cells persisting into geological time, they would demonstrate not only preservation of morphology and location, but also chemical fingerprints consistent with extant vertebrate osteocytes. We reasoned that structural proteins, including actin filaments [45], [46] and tubulin [47] would have greater preservation potential than other, more labile compounds. However, because actin and tubulin are ubiquitous in vertebrate cells, and because proteins similar in function to actin have been found in microbes [48], [49], we also tested for the presence of an osteocyte specific protein, phosphate-regulating endopeptidase homologous to those found on X-chromosomes in mammals (PHEX) [12], [50]. To identify PHEX in these dinosaur cells, we used the monoclonal antibody OB7.3, which is specific for avian osteocytes and is reported to not bind avian osteoblasts or non-avian osteocytes [12], [50]. Because phylogenetic analyses support the hypothesis that non-avian dinosaurs are most closely related to birds among extant taxa [51], [52], [53], [54], [55], we reasoned that this relationship may extend to the molecular level. Here, we present morphological, microscopic, and chemical evidence that these are indeed altered remnants of original cells.

Section snippets

Experimental procedures

An in-depth discussion of the methods used to generate these data is included in the accompanying appendix. Briefly, MOR 1125 (T. rex) was recovered from the base of the Hell Creek Formation in Eastern Montana. The sediments were well-sorted sandstones, interpreted to be estuarine. Growth curve analyses indicate a chronological age of 18–20 years at death [36]. MOR 2598 (B. canadensis) is represented by an articulated hindlimb, recovered from Judith River sediments in Northern Montana, and

Microscopy

We refer hereafter to the cell-like microstructures as ‘cells’ or ‘osteocytes’ for convenience and readability. The ‘osteocytes’ were isolated and concentrated as described (Materials and methods, Appendix A Supplementary data), then imaged by transmitted light microscopy (LM, Fig. 1A–F), scanning electron microscopy (SEM, Fig. 1G–L), transmission electron microscopy (TEM, Fig. 2 and Supplemental Fig. S1) and fluorescence microscopy (FM, Fig. 3, Fig. 4 and Supplemental Fig. S5). Fig. 1 shows

Suggested alternatives to endogenous source are not supported

The persistence of original cells, retaining morphology, transparency and flexibility comparable to those in living tissues, in fossils dating to the Mesozoic (~ 250–65 MYA) is highly controversial. It has been proposed, as mentioned above, that the ‘vessels’ and ‘cells’ arise as a result of biofilm infiltration [43]; but no data exist to support this hypothesis. In fact, localized antibody binding to these tissues that show different patterns of binding, using antibodies to proteins microbes do

Conclusion

The potential of molecular paleontology to address questions of evolution and physiology can be realized through the identification and recovery of informative molecules. Here, we give unequivocal evidence for the localization of specific antibodies, including the monoclonal antibody to avian PHEX, OB 7.3, to osteocytes recovered from two different dinosaurs from which protein sequences have previously been reported [3], [4], [39], [62]. These data not only support the hypothesis that the

Note added in proof

A recent paper by Allentoft et al. (2012) hypothesizes a half-life for DNA of ~ 521 years in an optimal depositional environment, suggesting that DNA should be degraded to single bases by a little under 7 million years, even though they also state that “considerable sample-to-sample variance in DNA preservation could not be accounted for by geologic age”. Their half-life estimate was based upon extrapolations of data taken from > 150 relatively recent Holocene bones (less than 10,000 years old).

Acknowledgments

We thank C. Seimens and M. Helfrich for providing samples of OB 7.3 for use in this study; J. Horner, B. Harmon and the Museum of the Rockies paleontology crews for the provision of these two dinosaur samples; LAELOM facility at NCSU Veterinary Medical School; N. Equall, R. Avci and S. Brumfield for use of facilities at ICAL and Plant Pathology, Montana State University. The A. mississipiensis specimen was provided by Bryan Stewart, NCMNS, and the E. coli strain used as negative control was

References (91)

  • P.H. Tartaix et al.

    In vitro effects of dentin matrix protein-1 on hydroxyapatite formation provide insights into in vivo functions

    J Biol Chem

    (2004)
  • B.S. Noble et al.

    Osteocyte function, osteocyte death andbone fracture resistance

    Mol Cell Endocrinol

    (2000)
  • P. Semal et al.

    Collagen extraction from recent and fossil bones: quantitative and qualitative aspects

    J Arch Sci

    (1995)
  • C.L. Hawkins et al.

    Generation and propagation of radical reactions on proteins

    Biochim Biophys Acta

    (2001)
  • R.V. Lloyd et al.

    The origin of the hydroxyl radical oxygen in the Fenton reaction

    Free Radic Biol Med

    (1997)
  • M. Spanner et al.

    The iron-binding protein ferritin is expressed in cells of the osteoblastic lineage in vitro and in vivo

    Bone

    (1995)
  • M.H. Schweitzer et al.

    Soft tissue and cellular preservation in vertebrate skeletal elements from the Cretaceous to the present

    Proc R Soc B Biol Sci

    (2007)
  • M.H. Schweitzer et al.

    Soft-tissue vessels and cellular preservation in Tyrannosaurus rex

    Science

    (2005)
  • J.M. Asara et al.

    Interpreting sequences from mastodon and Tyrannosaurus rex

    Science

    (2007)
  • J.M. Asara et al.

    Protein sequences from mastodon (Mammut americanum) and dinosaur (Tyrannosaurus rex) revealed by mass spectrometry

    Science

    (2007)
  • C.L. Organ et al.

    Molecular phylogenetics of mastodon and Tyrannosaurus rex

    Science

    (2008)
  • J.D. San Antonio et al.

    Dinosaur peptides suggest mechanisms of protein survival

    PLoS One

    (2011)
  • E.M. Aarden et al.

    Function of osteocytes in bone

    J Cell Biochem

    (1994)
  • M.L.K. Tate et al.

    The osteocyte

    Int J Biochem Cell Biol

    (2004)
  • S. Weiner et al.

    The material bone: structure–function relations

    Annu Rev Mater Sci

    (1998)
  • I. Westbroek et al.

    Osteocyte-specific monoclonal antibody MAb OB7.3 is directed against PHEX protein

    J Bone Miner Res

    (2002)
  • B.K. Hall

    Origin and evolution of vertebrate skeletal tissues

  • M. Laurin et al.

    A microanatomical and histological study of the paired fin skeleton of the Devonian Sarcopterygian Eusthenopteron foordi

    J Paleontol

    (2007)
  • J. Klein-Nulend et al.

    Osteocyte and bone structure

    Curr Osteoporos Rep

    (2003)
  • R. Pawlicki

    Metabolic pathways of the fossil dinosaur bones. Part V. Morphological differentiation of osteocyte lacunae and bone canaliculi and their significance in the system of extracellular communication

    Folia Histochem Cytochem

    (1985)
  • C. Palumbo et al.

    Osteocyte dendrogenesis in static and dynamic bone formation: an ultrastructural study

    Anat Rec A

    (2004)
  • F. Shapiro

    Variable conformation of GAP junctions linking bone cells: a transmission electron microscopic study of linear, stacked linear, curvilinear, oval and annular junctions

    Calcif Tissue Int

    (1997)
  • D. Guo et al.

    Advancing our understanding of osteocyte cell biology

    Ther Adv Musculoskelet Dis

    (2009)
  • L.F. Bonewald

    Establishment and characterization of an osteocyte-like cell line, MLO-Y4

    J Bone Miner Metab

    (1999)
  • L.F. Bonewald

    Osteocytes as multifunctional cells

    J Musculoskelet Neuronal Interact

    (2006)
  • Y. Wang et al.

    A model for the role of integrins in flow-induced mechanotransduction in osteocytes

    Proc Natl Acad Sci

    (2007)
  • Y. Han et al.

    Mechanotransduction and strain amplificationm in osteocyte cell processes

    PNAS

    (2004)
  • L.-D. You et al.

    Ultrastructure of the osteocyte process and its pericellular matrix

    Anat Rec A

    (2004)
  • S.C. Cowin et al.

    Candidates for the mechanosensory system in bone

    J Biomech Eng

    (1991)
  • S.C. Cowin

    Mechanosensation and fluid transport in living bone

    JMNI

    (2002)
  • D.H. Enlow et al.

    A comparative histological study of fossil and recent bone tissues. Part I

    Tex J Sci

    (1956)
  • D.H. Enlow et al.

    A comparative histological study of fossil and recent bone tissues. Part II

    Tex J Sci

    (1957)
  • D.H. Enlow et al.

    A comparative histological study of fossil and recent bone tissues. Part III

    Tex J Sci

    (1958)
  • A.J. De Ricqles

    Tissue structures of dinosaur bone: functional significance and possible relation to dinosaur physiology

  • R. Pawlicki et al.

    Cells, collagen fibrils and vessels in dinosaur bone

    Nature

    (1966)
  • Cited by (69)

    • DNA staining in fossil cells beyond the Quaternary: Reassessment of the evidence and prospects for an improved understanding of DNA preservation in deep time

      2021, Earth-Science Reviews
      Citation Excerpt :

      Anti-DNA antibodies (called autoantibodies) are often the result of an autoimmune response (for example in the autoimmune disease Lupus) where antibodies recognize their own DNA as an antigen. The specific anti-DNA antibody used in Schweitzer et al., 2013 was raised in a mouse (Millipore MAB030) but it supposedly recognizes any DNA backbone (not just mouse DNA) and it is clearly stated that it is ‘not based on sequence’. The main results of this study showed dinosaur cells (both of T. rex and the hadrosaur) that reacted positively with the three assays: the antibodies raised against DNA showed binding inside the cells, and positive PI (Fig. 1E) and DAPI staining was seen as single circular intracellular areas (see Fig. 5 in Schweitzer et al., 2013 for all results).

    • Preservation frequency of tissue-like structures in vertebrate remains from the upper Campanian of Alberta: Dinosaur Park Formation

      2020, Cretaceous Research
      Citation Excerpt :

      McGowan (1983) theorized that ironstone surrounding and within fossil bone likely derived from the iron within blood. Schweitzer et al. (2013) gave an explanation to how this may occur, suggesting that the iron released from haemoglobin, myoglobin, and cytochromes bind to proteins and lipids, leading to their crosslinking. Investigations into the preservation of collagen and osteocytes recovered from Lufengosaurus (Early Jurassic, 190–197 Ma) proposed that hematite cementation may play a role in preventing the degradation of soft tissue within bone (Lee et al., 2017).

    View all citing articles on Scopus
    View full text