- Email: [email protected]
Bacterial origin of iron-rich microspheres in Miocene mammalian fossils
Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
Contents lists available at ScienceDirect
Palaeogeography, Palaeoclimatology, Palaeoecology journal homepage: www.elsevier.com/locate/palaeo
Bacterial origin of iron-rich microspheres in Miocene mammalian fossils M. Dolores Pesquero a,c,⁎, Luis Alcalá a, Lynne S. Bell b, Yolanda Fernández-Jalvo c a b c
Fundación Conjunto Paleontológico de Teruel-Dinópolis, Avda. Sagunto s/n, 44002 Teruel, Spain Centre for Forensic Research, School of Criminology, Simon Fraser University, Burnaby, BC V5A 1S6, Canada Museo Nacional de Ciencias Naturales (CSIC), C/ José Gutiérrez Abascal 2, 28006 Madrid, Spain
a r t i c l e
i n f o
Article history: Received 6 May 2014 Received in revised form 1 December 2014 Accepted 5 December 2014 Available online 11 December 2014 Keywords: Iron microbial mineralization Vertebrate Miocene site Spain Taphonomy Bacteria Bone diagenesis
a b s t r a c t In their taphonomic study of a Cretaceous dinosaur fossil from the Gobi desert (Mongolia), Kremer et al. (2012) noted that the histological sections of this fossil preserved within their core iron oxide microspheres containing carbonaceous matter. They interpreted the carbonaceous nature of these structures as organic matter and suggested a microbial origin (probably bacterial) for the structures. Microspheres, similar both in composition and shape, have been identified in fossils from Cerro de la Garita, a Miocene mammalian site in Teruel, Spain. In the latter case, compact bone was also attacked by terrestrially associated bacteria (microscopic focal destruction [MFD]) which were enriched in iron and gives support to the idea that bacteria acted as the biological agent for iron precipitation during soft tissue decomposition in the early stages of bone diagenesis. Subsequent diagenetic episodes of mineralization related to the environmental context differ between these two sites; calcite precipitation at the palaeo-lakeshore of Cerro de la Garita and calcite and gypsum in the Gobi desert study case of Kremer et al. (2012). If the microsphere is bacterial in origin, it may be a useful taphonomic indicator of terrestrial exposure within a transitional environment of land and water. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Bone bacterial alteration and associated bacterial damage may be identified easily under the scanning electron microscope (SEM) in backscattered electron mode (BSE) as hypermineralized (bright) zones that contain small pores and thin channels: 0.1 to 2.0 μm in diameter (Hackett, 1981; Bell, 1990; Bell et al., 1991, 1996; Turner-Walker et al., 2002; Jans, 2005). These hypermineralized zones are the result of amorphous bioapatite reprecipitation after combined mineral–collagen removal/alteration from solubilised compact bone by bacteria (Jackes, 1990). During whole body decay bacterial activity is caused mainly by indigenous gut flora overgrowth that disperses through the post mortem intact vascular network, the activity and intensity of which may be reduced by the death history of the animal and by secondary predation (Bell et al., 1996), and may be affected by other environmental factors (Fernández-Jalvo et al., 2010). Although microbial attack frequently occurs in the early phase of soft tissue decomposition, Hedges (2002) speculated that microbial degradation may not necessarily be immediate, especially for waterlogged specimens where microbial attack is less intense (Jans et al., 2004). Microorganisms have been considered to be an important factor in mineral deposition (e.g. Borsato et al., 2000; Sánchez-Moral et al., 2003a,b, 2004, 2006; Cacchio et al., 2004; Baskar et al., 2006; Tormo, ⁎ Corresponding author at: Fundación Conjunto Paleontológico de Teruel-Dinópolis, Avda. Sagunto s/n, 44002 Teruel, Spain. Fax: +34 978 617 638. E-mail address: [email protected] (M.D. Pesquero).
http://dx.doi.org/10.1016/j.palaeo.2014.12.006 0031-0182/© 2014 Elsevier B.V. All rights reserved.
2007), and this work has demonstrated clear chemical evidence of the link existing between the substrate and biologically-mediated mineralization. Most microbial processes are based on redox reactions that transform the Fe II ion into the Fe III ion under both oxic and anoxic conditions under acidic to neutral pH ranges. Iron oxidation by bacteria produces a great quantity of Fe III, and bacteria also become iron bioaccumulators (Kremer et al., 2012). Metabolism by bacteria makes them ideal agents for mineral nucleation and precipitation. Iron oxides, iron-bearing carbonates, sulphides, and phosphates are the most common bacterially-mediated minerals. These include: hematite, ferro-ferric oxyhydroxides, lepidocrocite, goethite, magnetite, siderite, mackinawite, greigite, pyrite and vivianite (see Kremer et al., 2012). After experimental incubation of spring waters containing bacteria, Kawano and Tomita (2001) compared results with a simulated abiotic system to evaluate the role of bacteria in mineral formation. Their results showed that oxidation rates from experimental incubation are greater than those of the simulated abiotic system, suggesting that formation of iron minerals is promoted by bacterial oxidation of Fe2. Furthermore, under a moderate range of temperatures microorganisms “… can use Fe(II) from iron sulphides or ferrous carbonate as electron donors and form Fe(III) oxides which are tightly associated with the cell wall of the bacteria” (Frankel and Bazylinski, 2003, page: 101). In their study of a range of sites spanning Medieval to Pleistocene periods, Turner-Walker and Jans (2008) observed pyritic framboids with a clear granular texture filling natural bone porosity as well as some clear cases of bacterial-induced porosity. Pyrite deposits could be precursors of pyrite framboids, and therefore, sulphate-reducing bacteria may be
28
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
responsible for pyrite synthesis (Turner-Walker and Jans, 2008). During microbial decay of protein (type I collagen being the predominant protein in bone), sulphur is released as H2S raising the pH and causing partial dissolution of bone apatite and releasing small quantities of phosphate ions (Pfretzschner, 2000) with a simultaneous potential for redox (Pfretzschner, 1998). The aims of this study were twofold. Firstly, we compared remineralization linked to typical bacterial bone attack (MFD) and then compared them with the Fe-rich microspheres, considered as bacterial in origin by Kremer et al. (2012). These microspheres have a very distinctive morphology and have also been identified in the fossil bones of the Cerro de la Garita site (7 Mya). In both study cases, microspheres have been located in histological structures filled with calcite. The second aim of this study was to compare the results obtained by Kremer et al. (2012) from a Late Cretaceous dinosaur bone from the Gobi desert with our results from the Cerro de la Garita Miocene mammalian lakeshore site indicative of early and late diagenesis related to the environment. The study is based on observations using different microscopic techniques, mineralogical analysis and chemical composition.
origin of which has been related to western Mediterranean rifting (Simón, 1984). Cerro de la Garita (CG) is a National Heritage site of special historic value due to its early discovery in the 1920s and finds of holotype fossils (Alcalá, 1992; Pesquero et al., 2013). The site is characterized by a shallow, highly alkaline, lakeshore palaeoenvironment. The fossiliferous level (Fig. 1) was deposited during a progressive transition through time between alluvial and shallow lake environments in the area (Alcalá et al., 1999). Pesquero et al. (2010) and Pesquero et al. (2013) provide a detailed description of the taphonomic context of the site. Their studies included histological analyses and showed bacterial attack (MFD), which had affected a few fossils to a minor degree (OHI 4, according to Hedges et al.'s, 1995 classification). Calcite fillings appeared in pores, cracks and histological bone cavities as developed crystals (geodes) or massive calcite filling. The taphonomic study of Cerro de la Garita (Pesquero et al., 2013) concluded that the site was formed in a predominantly calm, lacustrine environment, eventually interrupted by rising water levels and a slight laminar water. Iron oxides are common deposits in the Cerro de la Garita sediments and fossils.
2. The site
3. Materials and methods
Cerro de la Garita (Concud, Teruel) is a site located on the western edge of the Teruel Neogene Basin (Fig. 1) in the north-eastern Iberian Peninsula within the Iberian Range and extending from Perales de Alfambra in the North, to Ademuz (Valencia) in the South. The Teruel Basin is a small graben filled with Neogene continental sediments, the
Given the special status of Cerro de la Garita as a National Heritage site of special historic value, the number of fossils that could be sectioned and processed for histological sections was limited to fourteen (Pesquero et al., 2013), three of which were specifically investigated for iron oxide microspheres and are presented here. Cross sections of
Fig. 1. Geographic and geological location of the Cerro de la Garita Site (Concud, Teruel) in the Iberian Peninsula, and a sedimentological section of the site.
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
fossils were analysed using two environmental and low vacuum scanning electron microscopes (FEI QUANTA 200 and INSPECT respectively) housed at the Museo Nacional de Ciencias Naturales-CSIC (MNCN). The scanning electron microscopes (SEM) were equipped with X-ray energy dispersive spectrometer (EDS) integrated analysis systems (Oxford Instruments Analytical—Inca). Analyses were performed in backscattered electron (BSE-SEM) mode. EDS analyses provided the elemental composition of specific areas up to a few microns in diameter, coupled with EDS mapping which records the distribution of chemical elements at selected sample areas. Microsamples were analysed using a confocal Thermo-Fisher DXR Raman spectrograph (Madison, WI, USA), coupled with an Olympus BX-RLA2 microscope with a CCD detector (1024 × 256 pixels), a
29
motorized XY stage and auto-focus lenses from the Olympus microscope UIS2 series (West Palm Beach, FL, USA) controlled using OMNIC 8.1 software. A wavelength 780 nm solid laser (maximum power = 20 mW) was selected as the excitation light with double Nd frequency: YVO4 DPSS gave the best results. The spectra were obtained using 50× and 100× confocal lenses with a pinhole or slit diameter of 50 μm and a grating of 900 lines/mm. These conditions and an excitation at 780 nm provide an average spectral resolution of 2–4 cm− 1 in the spectral range of 100–3500 cm−1. The sample point size was ~1.2 μm, which is consistent with the lens used. An integration time of 10–20 s with four–ten accumulations was sufficient to obtain suitable signal-tonoise ratios (S/N). The spectrograph was calibrated and aligned using pure polystyrene.
Fig. 2. EDS mapping of chemical composition at an MFD alteration (CG98-1832). (A) Scanning electron photomicrographs (backscattered mode, SEM-BSE) showing an MFD alteration, (B) EDS analyses of non-affected bone (spectrum 1) and different zones of the MFD (spectrum 2 and spectrum 3), (C) detail of the MFD SEM-BSE image and EDS mapping showing Fe in blue (D), P in red (E) and Ca in green (F). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
30
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
4. Results
small pores (Fig. 4D). In some calcite fillings we have also found pyrite mineralization next to iron microspheres (Fig. 5).
4.1. Bone histology and scanning electron microscopy analysis Under the SEM, histological observations of cut sections show that preservation of the histological structure is good in most of the analysed samples. Bacterial attack (MFD) has been observed on some fossils dispersed around histological traits of the cortical bone, and along diagenetic cracks. EDS mapping in Fig. 2 shows a bacterially-induced MFD with iron being the most abundant element of its interior (Fig. 2D) whilst outside the MFD, phosphorous (Fig. 2E) and calcium (Fig. 2F) correspond to the bioapatite mineral of unaltered bone. EDS spot analysis inside the bacterial MFD hypermineralized area (Fig. 3A, spectrum 2) yielded calcium phosphate (bioapatite). Inside the small pores and thin channels bored by bacteria (Fig. 3A, spectrum 3) there is a Fe-rich phase, with iron content increasing according to the intensity of bacterial attack (spectrum 4 has a more intense and higher Fe content than spectrum 3). Fig. 3B shows an MFD structure with values of the outer rim (spectrum 5) that are similar to those obtained in spectrum 2 (calcium phosphate, bioapatite). The inner annular structures that encircle pores previously bored by bacteria are enriched in iron (spectra 6 and 7). Fig. 3C shows microspheres containing homogeneous granules enveloped by a distinct iron-rich rim (spectrum 8). Histological features (Haversian system) filled with massive calcite deposits (Fig. 4A) or calcite crystals (geodes, Fig. 4B) may contain Fe-rich microspheres in their interior. Iron-rich microspheres found in Cerro de la Garita vary from 3 to 10 μm in diameter. They are similar to those described by Kremer et al. (2012), although the outer rim is wider and shows concentric rings in the latter case. Microspheres in both studies have an outer rim that envelopes a core of granular structures (Fig. 4C) having oval to rhombic shapes, with the inner individual granules varying from 0.5 to 1 μm in size in Cerro de la Garita specimens and from 0.5 to 1.5 μm in the Gobi dinosaur specimen. A similar morphological arrangement can be distinguished in MFD bacterial colonies with an outer hypermineralized rim of amorphous calcium phosphate containing
4.2. Raman analysis Analysis by Raman spectroscopy of some microspheres from calcite fillings of the fossil bone CG-2619 (Fig. 6) revealed all spheres to have the same mineral composition. The spectrum obtained from these microspheres (Fig. 6A) concurs with the goethite spectrum reference sample analysed using the same laser power as that used with the microspheres, and agrees also with data reported in the literature (Oh et al., 1998). Goethite is characterized by peaks located at 552, 397, 298 and 242 cm− 1, which appear throughout the area with microspheres. A carbonaceous matter was identified in the centre (core) of the microspheres. Spectrum analysis of the core (Fig. 6A) shows a prominent peak at 1607 cm−1, which is the marker peak for organic matter as indicated by Kremer et al. (2012). In addition to goethite, another mineral phase, calcite, was detected on the rim surrounding the microsphere core (Fig. 6C) displaying a characteristic band at 1086 cm−1. The latter may correspond to phases that are found around microspherical structures (calcite filling). 5. Discussion The EDS analyses on bacterial alterations (MFD) from the Cerro de la Garita samples have revealed them to be iron-enriched. In Fig. 2, chemical composition maps show iron deposits in the area affected by bacteria (MFD). We have observed that initially, the main components of the outer rim of the original MFD continue to be P and Ca i.e. bioapatite (Fig. 3B, spectrum 5), although pores created by bacteria contain some iron deposits (Fig. 3A). Fig. 3B provides a close-up view of an MFD at an intermediate stage of mineralization, showing annular structures in the MDF interior and may correspond to phases of iron precipitation around bacterial aggregates. According to Bazylinski and Frankel (2000), Konhauser et al. (2011), and Kremer et al. (2012), depending
Fig. 3. EDS analyses of different areas of bacterial attack (MFD). (A) SEM microphotograph of typical terrestrial bacterial attack (specimen CG98-1832), showing specific areas analysed (spectra 1–4), (B) SEM microphotograph of another MFD structure (specimen CG98-1832), in which holes bored by bacteria are enriched in iron and acquire annular structures, and specific areas analysed (spectra 5–7), (C) SEM microphotograph of microspheres (specimen CG98-2619) showing outer rim enriched in iron (spectrum 8), (D) element composition of specific areas analysed. Analysis 1 from around a Havers canal on a spot where there are no signs of bacterial alteration which shows a composition of calcium phosphate without Fe. Analyses 2 and 5 from the rim of high electron density, shows a small quantity of Fe, which rises in analyses 3, 4, 5 and 6. Analysis 3 is from the centre of an incipient alteration and 4, 6 and 7 are from a spot with more advanced alteration, so the percentage of Fe is clearly higher. Spectrum 8, from the rim of the microspheres, shows the highest percentage of Fe.
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
31
Fig. 4. Scanning electron photomicrographs (backscattered mode SEM-BSE). (A) Cerro de la Garita microspheres (bright circles) contained within Haversian canals filled by calcite, (B) Cerro de la Garita microspheres contained within Haversian canals filled by geodes (note the angular shapes at the interior of the Haversian system of distinct crystals sectioned and polished in this histological preparation and bright microspheres), (C) close-up view of iron-rich microspheres from Cerro de la Garita, (D) typical terrestrial bacterial colonies (microscopic focal destruction, MFD) on bone showing pores bored by bacteria and outer rim surrounding the destructive foci.
on the depositional microenvironment, iron-reducing bacteria can induce precipitation of iron minerals. At a more advanced stage of mineralization, these annular structures might grow gradually, forming mineralized units that have a consistent rhombic geometry within the interior of the iron microspheres (Fig. 6C), and are similar to those
described by Kremer et al. (2012). Raman microanalysis of Cerro de la Garita microspheres yielded three different components: goethite, the main iron-phase mineral, on the rim of the outer layer (Fig. 6A), organic matter in the interior of microspheres (Fig. 6A, spectrum 1), and traces of calcite on the outer rim around the core (Fig. 6B, spectrum 2).
Fig. 5. Scanning electron images (backscattered mode, SEM-BSE) showing pyrite mineralization in a calcite filling of specimen CG99-2619. (A) SEM microphotograph of pyrite mineralization near Fe-rich microspheres showing specific areas analysed (spectra 1–4), (B) close up-view of pyrite mineralization, (C) element composition of the specific areas analysed.
32
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
Fig. 6. SEM microphotograph and Raman spectra of microsphere analysed. (A) Raman spectrum 1 collected from the microsphere core, (B) Raman spectrum collected from the outer part of the microsphere (number 2 in SEM picture). (C) electron photomicrograph (backscattered mode, SEM-BSE) showing the microspheres analysed by Raman spectroscopy (CG-2619) (1 = spectrum 1, A, 2 = spectrum 2, B). Note the organized rhombic intersecting geometry of the individual mineralized units within the interior of the iron microspheres.
However, we cannot discard the possibility that this result could have been influenced by the calcite filling of the Haversian canals. Kremer et al. (2012) also found goethite around the core of microspheres, together with hematite and organic matter within the spheres that they suggest have a biological (bacterial) origin. The bacterial appearance of this type of change is different in morphology to the typical bacterial change associated with terrestrial environments (Fig. 3A & D). Whilst the microspheres distribute similarly in bone to that of terrestrial, they do not share the same internal morphology. In fact, what is striking about the microsphere is that it has a highly organized and constrained rhombic structure, which looks much like crystal growth within a void (Fig. 6C). Certainly a different bio-chemical mechanism is at play than that seen in terrestrially associated bacterial change in bone. Ferris (1993), Fortin and Beveridge (2000) and Southam (2000) consider that goethite, as with other metal oxides, is formed mainly via passive cell surface-mediated mineralization occurring on biological surfaces e.g. bacterial cell walls. Some bacteria such as Thiobacillus ferrooxidans promote Fe2+ oxidation to gain chemical energy that aids metabolism, which results in the formation of Fe-oxide and/or hydroxide minerals on their cell surfaces (Ehrlich, 1990). The metabolic activity of the organisms secretes metabolic products that react with ions or compounds in the environment resulting in the subsequent deposition of mineral particles (Frankel and Bazylinski, 2003). Field observations by Frankel and Bazylinski (2003) suggest that occurrences of these iron minerals are acstrongly dependent on solution chemistry, specifically pH and SO2− 4 tivity. Goethite usually occurs between neutral pH 7 and pH 3. Kremer et al. (2012) point out that the presence of Fe-minerals in paleontological materials does not provide unequivocal evidence for
biological mediation, and that biologically mediated Fe-minerals are usually deposited on or within organic matrices. In the particular case of iron microspheres described by Kremer et al. (2012), we have found similar microspheres at Cerro de la Garita (Fig. 3), also restricted to the Haversian system. Haemoglobin decay would also release free iron ions (Pfretzschner, 2000, 2001a,b), and Bell et al. (1996) and Bell (2011) point out that immediately after death, in a temperate environment, indigenous gut bacteria overgrow into the intact post mortem vasculature and from there to all parts of the body including organs, muscle, bone and teeth. This change is used as a forensic marker known as “marbling” and is used to estimating the post mortem interval i.e. how much time has passed since death. Marbling occurs within a 3– 5 day period and more quickly when temperatures are elevated. It is this early spread of bacteria, via the post mortem vasculature, that drives the visible soft tissue decay and will contribute to bacterial change to the microstructure of bone and teeth (Bell et al., 1991). Therefore, the potential iron source of haemoglobin and the dispersal vehicle through the post mortem vascular network may explain the apparent restriction of these microspheres to the interior of the Haversian canals, or put another way, the main bacterial activity would be expected to occur in the interior of the Haversian system canals where the internal post mortem bone blood supply resides. In addition, the Cerro de la Garita site also yielded iron deposits associated to bacterial MFDs on compact bone (Figs. 2 and 3) at initial (Fig. 3A) and transitional stages when the more regular annular shape (Fig. 3B) of iron deposits is formed where bacterial pores had once been. Our results, therefore, appear to reinforce the indication of bacteria involvement as precursors to iron oxide microspheres. At the Gobi and Cerro de la Garita sites, both of
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
which are associated with aquatic environments (the terrestrial-related bacterial change has never been documented in freshwater or marine environments), histological canals are filled with calcite, possibly a result of calcareous fluids and crystallization indicating that in both cases mineralization occurred during late diagenesis. For the purpose of clarity, given authors define early and late diagenesis differently, we include soft tissue decomposition as part of early diagenesis, and that early diagenesis ends at the point of skeletonization and/or its incorporation into a sediment. This means that late diagenesis begins post skeletonization and is post depositional, extending from near present into deep geological time. Environmental contexts play an important part in mineral formation and can moderate, promote or fully inhibit bacterial attack to bone. In this respect, the Gobi Desert site (Mongolia) was in a fluvial system, perhaps a river delta (as with the Okavango delta today, Kremer et al., 2012). Dinosaur skeletons were found preserved in an anatomically articulated state suggesting rapid burial and the absence of secondary predation. Cerro de la Garita was on a palaeo-lakeshore in a relatively calm calcareous aquatic environment with an intermittent laminar flow in a non-turbulent hydraulic system where scavengers were present (Pesquero et al., 2013). At both sites, microorganisms may have played an important role in Fe precipitation as oxides during the early stages of diagenesis, although differences, evident in subsequent episodes of mineralization, are considered related to environmental factors. In the case of the alkaline palaeo-lakeshore of Cerro de la Garita, calcite-rich fluids would have filled the Haversian canal network with early geodes or calcite. In the fluvial environment of the Gobi desert, cementation of calcite and gypsum took place during later mineralization episodes (Kremer et al., 2012). The Cerro de la Garita fossil samples also showed pyrite deposits located between some Fe microspheres. Sulphur was noted in most spectra analysed by EDS (Fig. 3D), although always in small amounts, which may be linked to bacterial activity and decay processes. Whilst Kremer et al. (2012) did not observe any pyrite mineralization in the Gobi desert bones, they could not exclude the presence of pyrite as a primary mineral phase during an early diagenetic stage that could be transformed through oxidation into hematite as a secondary mineral (present in the outer rings of microspheres) at a later diagenetic stage. At Cerro de la Garita site, pyrite framboids are present but scarce, always forming near microspheres. However, in general, environmental conditions appear to have inhibited sulphur deposition and, consequently, framboid formation and pyrite oxidation. In this regard, Pfretzschner (2000) observed that extensive pyritization is rare or absent in freshwater deposits because, although iron is fairly abundant, the sulphur content is low and the redox potential is usually not low enough for pyritization. Under these conditions, pyrite formation should be localized next to the source of sulphur, which is usually decaying organic matter, especially proteins (Canfield and Raiswell, 1991). It is possible that the environmental conditions in which the Cerro de la Garita bones fossilized inhibited the formation of large amounts of pyrite, instead favouring iron mineralization. Finally, microstructural changes associated with marine and freshwater bodies are seen as endolithic tunnelling (Bell et al., 1996; Bell and Elkerton, 2008; Pesquero et al., 2010), which looks very different and is peripheral in its distribution. Given that the fossil bones described here, and elsewhere, were recovered from transitional environments between land and water, some thought needs to be given to the presence of microspheres restricted to the Haversian system and what taphonomic information can be derived from them. One question of course is, are these truly bacterial in origin? Their location and size and their distribution in the Haversian canal suggest that they may well be, but it is curious that they do not invade the bone microstructure as terrestrial bacteria do, and most importantly, bacterial change of this type has not been seen in other water-derived deposits (Bell et al., 1996; Bell and Elkerton, 2008). Perhaps a working explanation might be that the animal was decomposing in a context that was terrestrial for periods
33
of time and flooded or tidal at others. You could then expect to see bacterial change associated with terrestrial decomposition, located in what would otherwise be described as a water-worked environment.
6. Conclusions Analysis of the Cerro de la Garita fossils shows iron deposits in the interior of putative bacterial colonies (MFD) displaying transitional morphological stages characteristic of MFD pores to more highly constrained mineralized rhombic units, both rich in iron. This rhombic organization has not been seen in bacterially altered bone from purely terrestrial contexts. Nor does it share the morphology associated with marine and lacustrine tunnelling of bone. The most advanced stage of the microsphere is goethite-rich microspheres that contain organic matter in the internal core in Haversian canals. Chemical composition, morphology and location of microspheres in Cerro de la Garita are very similar to those described by Kremer et al. (2012) for the Gobi Desert specimens. Transitional MFDs and microspheres with organic material in the centre of the sphere observed in Cerro de la Garita, give support to a microbial origin for iron microspheres as proposed by Kremer et al. (2012). In the early stages of formation, microorganisms may alter the pH of their immediate environment causing nucleation and precipitation of Fe oxides on the bacterial cellular wall. These structures would be the precursors of the microsphere iron-rich deposits (with occasional pyrite). Later, calcite precipitates filling many of these cavities in the bone indicating a post-depositional environment characterized by a shallow phreatic level and surrounding fluids oversaturated in calcium carbonate. If these microspheres are truly bacterial in origin, then they may be useful as a taphonomic indicator of terrestrial exposure in a transitional environment between land and water.
Acknowledgements Excavations at Cerro de la Garita, fossil preparation, and later palaeontological research, were made possible through the support of the Spanish Ministry of Science and Innovation (projects CGL200766231/CGL2010-19825) and a Juan de la Cierva contract to M.D.P. Ref. JCI-2007-132-565). We are especially grateful to Norah Moloney for comments on the text, and to the Electron Microscopy Laboratory of the Museo Nacional de Ciencias Naturales-CSIC (MNCN). Thanks are also owed to Francisco Esteban, owner of the Cerro de la Garita site, for his assistance in this project.
References Alcalá, L., 1992. Macromamíferos neógenos de la fosa de Alfambra-Teruel. (Ph D Thesis). Univ. Complutense, Madrid, Spain. Alcalá, L., Martín Escorza, C., Pesquero, M.D., 1999. Implicaciones tafonómicas de la orientación de huesos de mamíferos en el Neógeno continental de Concud (Teruel). Rev. Soc. Geol. Esp. 12, 501–511. Baskar, S., Baskar, R., Mauclaire, L., McKenzie, J.A., 2006. Microbially induced calcite precipitation in culture experiments: possible origin for stalactites in Sahastradhara caves, Dehradun India. Curr. Sci. India 90, 58–64. Bazylinski, D.A., Frankel, R.B., 2000. Biologically controlled mineralization of magnetic iron minerals by magnetotactic bacteria. In: Lovley, D.R. (Ed.), Environmental Microbe–Metal Interactions. ASM Press, NW Washington, pp. 109–143. Bell, L.S., 1990. Palaeopathology and diagenesis: an SEM evaluation of structural changes using backscattered electron imaging. J. Archaeol. Sci. 17, 85–102. Bell, L.S., 2011. Histotaphonomy. In: Crowder, C., Stout, S. (Eds.), Bone Histology: An Anthropological Perspective. CRC Press, New York, pp. 241–251. Bell, L.S., Elkerton, A., 2008. Human remains recovered from a sixteenth century mass fatality: unique marine taphonomy in human skeletal material from the medieval warship the Mary Rose. Int. J. Osteoarchaeol. 18, 523–535. Bell, L.S., Boyde, A., Jones, S.J., 1991. Diagenetic alteration to teeth in situ illustrated by backscattered electron imaging. Scanning 13, 173–183. Bell, L.S., Skinner, M.F., Jones, S.J., 1996. The speed of post-mortem change to the human skeleton and its taphonomic significance. Forensic Sci. Int. 82, 129–140. Borsato, A., Frisia, S., Jones, B., Van Der Borg, K., 2000. Calcite moonmilk: crystal morphology and environment of formation in caves in the Italian Alps. J. Sediment. Res. 70, 1179–1190.
34
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
Cacchio, P., Contento, R., Ercole, C., Capuccio, G., Preite-Martínez, M., Lepidi, A., 2004. Involvement of microorganisms in the formation of carbonate speleothems in the Cervo Cave (L'Aquila, Italy). Geomicrobiol J. 21, 497–509. Canfield, D., Raiswell, K., 1991. Pyrite formation and fossil preservation. In: Allison, P.A., Briggs, D.E. (Eds.), Taphonomy. Plenum Press, New York, pp. 337–387. Ehrlich, H.L., 1990. Microbiology. second ed. Marcel Dekker, New York. Fernández-Jalvo, Y., Andrews, P., Pesquero, D., Smith, C., Marín-Monfort, D., Sánchez, B., Geigl, E.M., Alonso, A., 2010. Early bone diagenesis in temperate environments. Part I: surface features and histology. Palaeogeogr. Palaeoclimatol. Palaeoecol. 288, 62–81. Ferris, F.G., 1993. Microbial biomineralization in natural environments. Earth Sci. (Chikyu Kagaku) 47, 233–250. Fortin, D., Beveridge, T.J., 2000. Mechanistic routes to biomineral surface development. In: biomineralization. In: Bäuerlein, E. (Ed.), Biology to Biotechnology and Medical Application. Wiley-VCH, Weinheim, Germany, pp. 7–24. Frankel, R.B., Bazylinski, D.A., 2003. Biologically induced mineralization by bacteria. Rev. Mineral. Geochem. 54, 95–114. Hackett, C.J., 1981. Microscopical focal destruction (tunnels) in exhumed human bones. Med. Sci. Law 21, 243–265. Hedges, R.E.M., 2002. Bone diagenesis: an overview of processes. Archaeometry 44, 319–328. Hedges, R.E., Millard, A.P., Pike, A.W.G., 1995. Measurements and relationships of diagenetic alteration of bone from three archaeological sites. J. Archaeol. Sci. 22, 201–210. Jackes, M., 1990. Diagenetic change in prehistoric Portuguese human bone (4000 to 8000 BP). Symposium on Bone Chemistry, Canadian Association for Physical Anthropology, 18th Annual Meeting. Jans, M.M.E., 2005. Histological characterisation of diagenetic alteration of archaeological bone. Geoarchaeological and Bioarchaeological Studies 4. Institute for Geo and Bioarchaeology, Vrije Universiteit, Amsterdam. Jans, M.M.E., Nielsen-Marshb, C.M., Smithc, C.I., Collinsd, M.J., Karsa, H., 2004. Characterisation of microbial attack on archaeological bone. J. Archaeol. Sci. 31, 87–95. Kawano, M., Tomita, K., 2001. Microbial biomineralization in weathered volcanic ash deposit and formation of biogenic minerals by experimental incubation. Am. Mineral. 86, 400–410. Konhauser, K.O., Kapple, A., Roden, E.E., 2011. Iron in microbial metabolisms. Elements 7, 89–93. Kremer, B., Owocki, K., Królikowska, A., Wrzosek, B., Kazmierczak, J., 2012. Mineral microbial structures in a bone of the Late Cretaceous dinosaur Saurolophus angustirostris from the Gobi Desert, Mongolia — a Raman spectroscopy study. Palaeogeogr. Palaeoclimatol. Palaeoecol. 358–360, 51–61. Oh, S.J., Cook, D.C., Townsend, H.E., 1998. Characterization of iron oxides commonly formed as corrosion products on steel. Hyperfine Interact. 112, 59–65.
Pesquero, M.D., Ascaso, C., Fernández-Jalvo, Y., Alcalá, L., 2010. A new taphonomic bioerosion in a Miocene lakeshore environment. Palaeogeogr. Palaeoclimatol. Palaeoecol. 295, 192–198. Pesquero, M.D., Alcalá, L., Fernández-Jalvo, Y., 2013. Taphonomy of the reference Miocene vertebrate mammal site of Cerro de la Garita, Spain. Lethaia 46, 378–398. Pfretzschner, H.-U., 1998. Frühdiagenetische Prozesse bei der Fossilisation von Knochen. Neues Jahrb. Geol. Palaontol. Abh. 210, 369–397. Pfretzschner, H.-U., 2000. Micro-cracks and fossilization of Haversian bone. Neues Jahrb. Geol. Palaontol. Abh. 216, 413–432. Pfretzschner, H.-U., 2001a. Pyrite in fossil bone. Neues Jahrb. Geol. Palaontol. Abh. 220 (1), 1–23. Pfretzschner, H.-U., 2001b. Iron oxides in fossil bone. Neues Jahrb. Geol. Palaontol. Abh. 220 (3), 417–429. Sánchez-Moral, S., Bedoya, J., Luque, L., Cañaberas, J.C., Jurado, V., Laiz, L., Saiz-Jiménez, C., 2003a. Biomineralization of different crystalline phases by bacteria isolated from catacombs. In: Saiz-Jiménez, C. (Ed.), Molecular Biology and Cultural Heritage. Balkema, Lisse, pp. 179–185. Sánchez-Moral, S., Cañaberas, J.C., Laiz, L., Saiz-Jiménez, C., Bedoya, J., Luque, L., 2003b. Biomediated precipitation of calcium carbonate metastable phases in hypogean environments. Geomicrobiol J. 20, 491–500. Sánchez-Moral, S., Luque, L., Cañaberas, J.C., Laiz, L., Jurado, V., Hermosin, B., Saiz-Jiménez, C., 2004. Bioinduced barium precipitation in St. Callixtus and Domitilla catacombs. Ann. Microbiol. 54, 1–12. Sánchez-Moral, S., González, J.M., Cañaberas, J.C., Cuezva, S., Lario, J., Cardell, C., Elez, J., Luque, L., Saiz-Jiménez, C., 2006. Procesos de precipitación mineral bioinducidos en sistemas kársticos subterráneos: breve revisión y nuevas tendencias. Estud. Geol. 62, 43–52. Simón, J.L., 1984. Comprensión y distensión alpinas en la cadena Ibérica oriental. Inst. Estudios Turolenses (269 pp.). Southam, G., 2000. Bacterial surface-mediated mineral formation. In: Lovley, D.R. (Ed.), Environmental Microbe–Mineral Interactions. ASM Press, Washington, DC, pp. 257–276. Tormo, L., 2007. Interacción bacteria-mineral en substratos de cuevas volcánicas. Unpublished Trabajo Tutelado de Iniciación a la Investigación, Facultad de Ciencias, Universidad Autónoma de Madrid, Spain, 34 pp. Turner-Walker, G., Jans, M.M.E., 2008. Reconstructing taphonomic histories using histological analysis. Palaeogeogr. Palaeoclimatol. Palaeoecol. 266, 207–235. Turner-Walker, G., Nielsen-Marsh, C.M., Syversen, U., Kars, H., Collins, M.J., 2002. Submicron spongiform porosity is the major ultra-structural alteration occurring in archaeological bone. Int. J. Osteoarchaeol. 12, 407–414.
Contents lists available at ScienceDirect
Palaeogeography, Palaeoclimatology, Palaeoecology journal homepage: www.elsevier.com/locate/palaeo
Bacterial origin of iron-rich microspheres in Miocene mammalian fossils M. Dolores Pesquero a,c,⁎, Luis Alcalá a, Lynne S. Bell b, Yolanda Fernández-Jalvo c a b c
Fundación Conjunto Paleontológico de Teruel-Dinópolis, Avda. Sagunto s/n, 44002 Teruel, Spain Centre for Forensic Research, School of Criminology, Simon Fraser University, Burnaby, BC V5A 1S6, Canada Museo Nacional de Ciencias Naturales (CSIC), C/ José Gutiérrez Abascal 2, 28006 Madrid, Spain
a r t i c l e
i n f o
Article history: Received 6 May 2014 Received in revised form 1 December 2014 Accepted 5 December 2014 Available online 11 December 2014 Keywords: Iron microbial mineralization Vertebrate Miocene site Spain Taphonomy Bacteria Bone diagenesis
a b s t r a c t In their taphonomic study of a Cretaceous dinosaur fossil from the Gobi desert (Mongolia), Kremer et al. (2012) noted that the histological sections of this fossil preserved within their core iron oxide microspheres containing carbonaceous matter. They interpreted the carbonaceous nature of these structures as organic matter and suggested a microbial origin (probably bacterial) for the structures. Microspheres, similar both in composition and shape, have been identified in fossils from Cerro de la Garita, a Miocene mammalian site in Teruel, Spain. In the latter case, compact bone was also attacked by terrestrially associated bacteria (microscopic focal destruction [MFD]) which were enriched in iron and gives support to the idea that bacteria acted as the biological agent for iron precipitation during soft tissue decomposition in the early stages of bone diagenesis. Subsequent diagenetic episodes of mineralization related to the environmental context differ between these two sites; calcite precipitation at the palaeo-lakeshore of Cerro de la Garita and calcite and gypsum in the Gobi desert study case of Kremer et al. (2012). If the microsphere is bacterial in origin, it may be a useful taphonomic indicator of terrestrial exposure within a transitional environment of land and water. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Bone bacterial alteration and associated bacterial damage may be identified easily under the scanning electron microscope (SEM) in backscattered electron mode (BSE) as hypermineralized (bright) zones that contain small pores and thin channels: 0.1 to 2.0 μm in diameter (Hackett, 1981; Bell, 1990; Bell et al., 1991, 1996; Turner-Walker et al., 2002; Jans, 2005). These hypermineralized zones are the result of amorphous bioapatite reprecipitation after combined mineral–collagen removal/alteration from solubilised compact bone by bacteria (Jackes, 1990). During whole body decay bacterial activity is caused mainly by indigenous gut flora overgrowth that disperses through the post mortem intact vascular network, the activity and intensity of which may be reduced by the death history of the animal and by secondary predation (Bell et al., 1996), and may be affected by other environmental factors (Fernández-Jalvo et al., 2010). Although microbial attack frequently occurs in the early phase of soft tissue decomposition, Hedges (2002) speculated that microbial degradation may not necessarily be immediate, especially for waterlogged specimens where microbial attack is less intense (Jans et al., 2004). Microorganisms have been considered to be an important factor in mineral deposition (e.g. Borsato et al., 2000; Sánchez-Moral et al., 2003a,b, 2004, 2006; Cacchio et al., 2004; Baskar et al., 2006; Tormo, ⁎ Corresponding author at: Fundación Conjunto Paleontológico de Teruel-Dinópolis, Avda. Sagunto s/n, 44002 Teruel, Spain. Fax: +34 978 617 638. E-mail address: [email protected] (M.D. Pesquero).
http://dx.doi.org/10.1016/j.palaeo.2014.12.006 0031-0182/© 2014 Elsevier B.V. All rights reserved.
2007), and this work has demonstrated clear chemical evidence of the link existing between the substrate and biologically-mediated mineralization. Most microbial processes are based on redox reactions that transform the Fe II ion into the Fe III ion under both oxic and anoxic conditions under acidic to neutral pH ranges. Iron oxidation by bacteria produces a great quantity of Fe III, and bacteria also become iron bioaccumulators (Kremer et al., 2012). Metabolism by bacteria makes them ideal agents for mineral nucleation and precipitation. Iron oxides, iron-bearing carbonates, sulphides, and phosphates are the most common bacterially-mediated minerals. These include: hematite, ferro-ferric oxyhydroxides, lepidocrocite, goethite, magnetite, siderite, mackinawite, greigite, pyrite and vivianite (see Kremer et al., 2012). After experimental incubation of spring waters containing bacteria, Kawano and Tomita (2001) compared results with a simulated abiotic system to evaluate the role of bacteria in mineral formation. Their results showed that oxidation rates from experimental incubation are greater than those of the simulated abiotic system, suggesting that formation of iron minerals is promoted by bacterial oxidation of Fe2. Furthermore, under a moderate range of temperatures microorganisms “… can use Fe(II) from iron sulphides or ferrous carbonate as electron donors and form Fe(III) oxides which are tightly associated with the cell wall of the bacteria” (Frankel and Bazylinski, 2003, page: 101). In their study of a range of sites spanning Medieval to Pleistocene periods, Turner-Walker and Jans (2008) observed pyritic framboids with a clear granular texture filling natural bone porosity as well as some clear cases of bacterial-induced porosity. Pyrite deposits could be precursors of pyrite framboids, and therefore, sulphate-reducing bacteria may be
28
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
responsible for pyrite synthesis (Turner-Walker and Jans, 2008). During microbial decay of protein (type I collagen being the predominant protein in bone), sulphur is released as H2S raising the pH and causing partial dissolution of bone apatite and releasing small quantities of phosphate ions (Pfretzschner, 2000) with a simultaneous potential for redox (Pfretzschner, 1998). The aims of this study were twofold. Firstly, we compared remineralization linked to typical bacterial bone attack (MFD) and then compared them with the Fe-rich microspheres, considered as bacterial in origin by Kremer et al. (2012). These microspheres have a very distinctive morphology and have also been identified in the fossil bones of the Cerro de la Garita site (7 Mya). In both study cases, microspheres have been located in histological structures filled with calcite. The second aim of this study was to compare the results obtained by Kremer et al. (2012) from a Late Cretaceous dinosaur bone from the Gobi desert with our results from the Cerro de la Garita Miocene mammalian lakeshore site indicative of early and late diagenesis related to the environment. The study is based on observations using different microscopic techniques, mineralogical analysis and chemical composition.
origin of which has been related to western Mediterranean rifting (Simón, 1984). Cerro de la Garita (CG) is a National Heritage site of special historic value due to its early discovery in the 1920s and finds of holotype fossils (Alcalá, 1992; Pesquero et al., 2013). The site is characterized by a shallow, highly alkaline, lakeshore palaeoenvironment. The fossiliferous level (Fig. 1) was deposited during a progressive transition through time between alluvial and shallow lake environments in the area (Alcalá et al., 1999). Pesquero et al. (2010) and Pesquero et al. (2013) provide a detailed description of the taphonomic context of the site. Their studies included histological analyses and showed bacterial attack (MFD), which had affected a few fossils to a minor degree (OHI 4, according to Hedges et al.'s, 1995 classification). Calcite fillings appeared in pores, cracks and histological bone cavities as developed crystals (geodes) or massive calcite filling. The taphonomic study of Cerro de la Garita (Pesquero et al., 2013) concluded that the site was formed in a predominantly calm, lacustrine environment, eventually interrupted by rising water levels and a slight laminar water. Iron oxides are common deposits in the Cerro de la Garita sediments and fossils.
2. The site
3. Materials and methods
Cerro de la Garita (Concud, Teruel) is a site located on the western edge of the Teruel Neogene Basin (Fig. 1) in the north-eastern Iberian Peninsula within the Iberian Range and extending from Perales de Alfambra in the North, to Ademuz (Valencia) in the South. The Teruel Basin is a small graben filled with Neogene continental sediments, the
Given the special status of Cerro de la Garita as a National Heritage site of special historic value, the number of fossils that could be sectioned and processed for histological sections was limited to fourteen (Pesquero et al., 2013), three of which were specifically investigated for iron oxide microspheres and are presented here. Cross sections of
Fig. 1. Geographic and geological location of the Cerro de la Garita Site (Concud, Teruel) in the Iberian Peninsula, and a sedimentological section of the site.
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
fossils were analysed using two environmental and low vacuum scanning electron microscopes (FEI QUANTA 200 and INSPECT respectively) housed at the Museo Nacional de Ciencias Naturales-CSIC (MNCN). The scanning electron microscopes (SEM) were equipped with X-ray energy dispersive spectrometer (EDS) integrated analysis systems (Oxford Instruments Analytical—Inca). Analyses were performed in backscattered electron (BSE-SEM) mode. EDS analyses provided the elemental composition of specific areas up to a few microns in diameter, coupled with EDS mapping which records the distribution of chemical elements at selected sample areas. Microsamples were analysed using a confocal Thermo-Fisher DXR Raman spectrograph (Madison, WI, USA), coupled with an Olympus BX-RLA2 microscope with a CCD detector (1024 × 256 pixels), a
29
motorized XY stage and auto-focus lenses from the Olympus microscope UIS2 series (West Palm Beach, FL, USA) controlled using OMNIC 8.1 software. A wavelength 780 nm solid laser (maximum power = 20 mW) was selected as the excitation light with double Nd frequency: YVO4 DPSS gave the best results. The spectra were obtained using 50× and 100× confocal lenses with a pinhole or slit diameter of 50 μm and a grating of 900 lines/mm. These conditions and an excitation at 780 nm provide an average spectral resolution of 2–4 cm− 1 in the spectral range of 100–3500 cm−1. The sample point size was ~1.2 μm, which is consistent with the lens used. An integration time of 10–20 s with four–ten accumulations was sufficient to obtain suitable signal-tonoise ratios (S/N). The spectrograph was calibrated and aligned using pure polystyrene.
Fig. 2. EDS mapping of chemical composition at an MFD alteration (CG98-1832). (A) Scanning electron photomicrographs (backscattered mode, SEM-BSE) showing an MFD alteration, (B) EDS analyses of non-affected bone (spectrum 1) and different zones of the MFD (spectrum 2 and spectrum 3), (C) detail of the MFD SEM-BSE image and EDS mapping showing Fe in blue (D), P in red (E) and Ca in green (F). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
30
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
4. Results
small pores (Fig. 4D). In some calcite fillings we have also found pyrite mineralization next to iron microspheres (Fig. 5).
4.1. Bone histology and scanning electron microscopy analysis Under the SEM, histological observations of cut sections show that preservation of the histological structure is good in most of the analysed samples. Bacterial attack (MFD) has been observed on some fossils dispersed around histological traits of the cortical bone, and along diagenetic cracks. EDS mapping in Fig. 2 shows a bacterially-induced MFD with iron being the most abundant element of its interior (Fig. 2D) whilst outside the MFD, phosphorous (Fig. 2E) and calcium (Fig. 2F) correspond to the bioapatite mineral of unaltered bone. EDS spot analysis inside the bacterial MFD hypermineralized area (Fig. 3A, spectrum 2) yielded calcium phosphate (bioapatite). Inside the small pores and thin channels bored by bacteria (Fig. 3A, spectrum 3) there is a Fe-rich phase, with iron content increasing according to the intensity of bacterial attack (spectrum 4 has a more intense and higher Fe content than spectrum 3). Fig. 3B shows an MFD structure with values of the outer rim (spectrum 5) that are similar to those obtained in spectrum 2 (calcium phosphate, bioapatite). The inner annular structures that encircle pores previously bored by bacteria are enriched in iron (spectra 6 and 7). Fig. 3C shows microspheres containing homogeneous granules enveloped by a distinct iron-rich rim (spectrum 8). Histological features (Haversian system) filled with massive calcite deposits (Fig. 4A) or calcite crystals (geodes, Fig. 4B) may contain Fe-rich microspheres in their interior. Iron-rich microspheres found in Cerro de la Garita vary from 3 to 10 μm in diameter. They are similar to those described by Kremer et al. (2012), although the outer rim is wider and shows concentric rings in the latter case. Microspheres in both studies have an outer rim that envelopes a core of granular structures (Fig. 4C) having oval to rhombic shapes, with the inner individual granules varying from 0.5 to 1 μm in size in Cerro de la Garita specimens and from 0.5 to 1.5 μm in the Gobi dinosaur specimen. A similar morphological arrangement can be distinguished in MFD bacterial colonies with an outer hypermineralized rim of amorphous calcium phosphate containing
4.2. Raman analysis Analysis by Raman spectroscopy of some microspheres from calcite fillings of the fossil bone CG-2619 (Fig. 6) revealed all spheres to have the same mineral composition. The spectrum obtained from these microspheres (Fig. 6A) concurs with the goethite spectrum reference sample analysed using the same laser power as that used with the microspheres, and agrees also with data reported in the literature (Oh et al., 1998). Goethite is characterized by peaks located at 552, 397, 298 and 242 cm− 1, which appear throughout the area with microspheres. A carbonaceous matter was identified in the centre (core) of the microspheres. Spectrum analysis of the core (Fig. 6A) shows a prominent peak at 1607 cm−1, which is the marker peak for organic matter as indicated by Kremer et al. (2012). In addition to goethite, another mineral phase, calcite, was detected on the rim surrounding the microsphere core (Fig. 6C) displaying a characteristic band at 1086 cm−1. The latter may correspond to phases that are found around microspherical structures (calcite filling). 5. Discussion The EDS analyses on bacterial alterations (MFD) from the Cerro de la Garita samples have revealed them to be iron-enriched. In Fig. 2, chemical composition maps show iron deposits in the area affected by bacteria (MFD). We have observed that initially, the main components of the outer rim of the original MFD continue to be P and Ca i.e. bioapatite (Fig. 3B, spectrum 5), although pores created by bacteria contain some iron deposits (Fig. 3A). Fig. 3B provides a close-up view of an MFD at an intermediate stage of mineralization, showing annular structures in the MDF interior and may correspond to phases of iron precipitation around bacterial aggregates. According to Bazylinski and Frankel (2000), Konhauser et al. (2011), and Kremer et al. (2012), depending
Fig. 3. EDS analyses of different areas of bacterial attack (MFD). (A) SEM microphotograph of typical terrestrial bacterial attack (specimen CG98-1832), showing specific areas analysed (spectra 1–4), (B) SEM microphotograph of another MFD structure (specimen CG98-1832), in which holes bored by bacteria are enriched in iron and acquire annular structures, and specific areas analysed (spectra 5–7), (C) SEM microphotograph of microspheres (specimen CG98-2619) showing outer rim enriched in iron (spectrum 8), (D) element composition of specific areas analysed. Analysis 1 from around a Havers canal on a spot where there are no signs of bacterial alteration which shows a composition of calcium phosphate without Fe. Analyses 2 and 5 from the rim of high electron density, shows a small quantity of Fe, which rises in analyses 3, 4, 5 and 6. Analysis 3 is from the centre of an incipient alteration and 4, 6 and 7 are from a spot with more advanced alteration, so the percentage of Fe is clearly higher. Spectrum 8, from the rim of the microspheres, shows the highest percentage of Fe.
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
31
Fig. 4. Scanning electron photomicrographs (backscattered mode SEM-BSE). (A) Cerro de la Garita microspheres (bright circles) contained within Haversian canals filled by calcite, (B) Cerro de la Garita microspheres contained within Haversian canals filled by geodes (note the angular shapes at the interior of the Haversian system of distinct crystals sectioned and polished in this histological preparation and bright microspheres), (C) close-up view of iron-rich microspheres from Cerro de la Garita, (D) typical terrestrial bacterial colonies (microscopic focal destruction, MFD) on bone showing pores bored by bacteria and outer rim surrounding the destructive foci.
on the depositional microenvironment, iron-reducing bacteria can induce precipitation of iron minerals. At a more advanced stage of mineralization, these annular structures might grow gradually, forming mineralized units that have a consistent rhombic geometry within the interior of the iron microspheres (Fig. 6C), and are similar to those
described by Kremer et al. (2012). Raman microanalysis of Cerro de la Garita microspheres yielded three different components: goethite, the main iron-phase mineral, on the rim of the outer layer (Fig. 6A), organic matter in the interior of microspheres (Fig. 6A, spectrum 1), and traces of calcite on the outer rim around the core (Fig. 6B, spectrum 2).
Fig. 5. Scanning electron images (backscattered mode, SEM-BSE) showing pyrite mineralization in a calcite filling of specimen CG99-2619. (A) SEM microphotograph of pyrite mineralization near Fe-rich microspheres showing specific areas analysed (spectra 1–4), (B) close up-view of pyrite mineralization, (C) element composition of the specific areas analysed.
32
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
Fig. 6. SEM microphotograph and Raman spectra of microsphere analysed. (A) Raman spectrum 1 collected from the microsphere core, (B) Raman spectrum collected from the outer part of the microsphere (number 2 in SEM picture). (C) electron photomicrograph (backscattered mode, SEM-BSE) showing the microspheres analysed by Raman spectroscopy (CG-2619) (1 = spectrum 1, A, 2 = spectrum 2, B). Note the organized rhombic intersecting geometry of the individual mineralized units within the interior of the iron microspheres.
However, we cannot discard the possibility that this result could have been influenced by the calcite filling of the Haversian canals. Kremer et al. (2012) also found goethite around the core of microspheres, together with hematite and organic matter within the spheres that they suggest have a biological (bacterial) origin. The bacterial appearance of this type of change is different in morphology to the typical bacterial change associated with terrestrial environments (Fig. 3A & D). Whilst the microspheres distribute similarly in bone to that of terrestrial, they do not share the same internal morphology. In fact, what is striking about the microsphere is that it has a highly organized and constrained rhombic structure, which looks much like crystal growth within a void (Fig. 6C). Certainly a different bio-chemical mechanism is at play than that seen in terrestrially associated bacterial change in bone. Ferris (1993), Fortin and Beveridge (2000) and Southam (2000) consider that goethite, as with other metal oxides, is formed mainly via passive cell surface-mediated mineralization occurring on biological surfaces e.g. bacterial cell walls. Some bacteria such as Thiobacillus ferrooxidans promote Fe2+ oxidation to gain chemical energy that aids metabolism, which results in the formation of Fe-oxide and/or hydroxide minerals on their cell surfaces (Ehrlich, 1990). The metabolic activity of the organisms secretes metabolic products that react with ions or compounds in the environment resulting in the subsequent deposition of mineral particles (Frankel and Bazylinski, 2003). Field observations by Frankel and Bazylinski (2003) suggest that occurrences of these iron minerals are acstrongly dependent on solution chemistry, specifically pH and SO2− 4 tivity. Goethite usually occurs between neutral pH 7 and pH 3. Kremer et al. (2012) point out that the presence of Fe-minerals in paleontological materials does not provide unequivocal evidence for
biological mediation, and that biologically mediated Fe-minerals are usually deposited on or within organic matrices. In the particular case of iron microspheres described by Kremer et al. (2012), we have found similar microspheres at Cerro de la Garita (Fig. 3), also restricted to the Haversian system. Haemoglobin decay would also release free iron ions (Pfretzschner, 2000, 2001a,b), and Bell et al. (1996) and Bell (2011) point out that immediately after death, in a temperate environment, indigenous gut bacteria overgrow into the intact post mortem vasculature and from there to all parts of the body including organs, muscle, bone and teeth. This change is used as a forensic marker known as “marbling” and is used to estimating the post mortem interval i.e. how much time has passed since death. Marbling occurs within a 3– 5 day period and more quickly when temperatures are elevated. It is this early spread of bacteria, via the post mortem vasculature, that drives the visible soft tissue decay and will contribute to bacterial change to the microstructure of bone and teeth (Bell et al., 1991). Therefore, the potential iron source of haemoglobin and the dispersal vehicle through the post mortem vascular network may explain the apparent restriction of these microspheres to the interior of the Haversian canals, or put another way, the main bacterial activity would be expected to occur in the interior of the Haversian system canals where the internal post mortem bone blood supply resides. In addition, the Cerro de la Garita site also yielded iron deposits associated to bacterial MFDs on compact bone (Figs. 2 and 3) at initial (Fig. 3A) and transitional stages when the more regular annular shape (Fig. 3B) of iron deposits is formed where bacterial pores had once been. Our results, therefore, appear to reinforce the indication of bacteria involvement as precursors to iron oxide microspheres. At the Gobi and Cerro de la Garita sites, both of
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
which are associated with aquatic environments (the terrestrial-related bacterial change has never been documented in freshwater or marine environments), histological canals are filled with calcite, possibly a result of calcareous fluids and crystallization indicating that in both cases mineralization occurred during late diagenesis. For the purpose of clarity, given authors define early and late diagenesis differently, we include soft tissue decomposition as part of early diagenesis, and that early diagenesis ends at the point of skeletonization and/or its incorporation into a sediment. This means that late diagenesis begins post skeletonization and is post depositional, extending from near present into deep geological time. Environmental contexts play an important part in mineral formation and can moderate, promote or fully inhibit bacterial attack to bone. In this respect, the Gobi Desert site (Mongolia) was in a fluvial system, perhaps a river delta (as with the Okavango delta today, Kremer et al., 2012). Dinosaur skeletons were found preserved in an anatomically articulated state suggesting rapid burial and the absence of secondary predation. Cerro de la Garita was on a palaeo-lakeshore in a relatively calm calcareous aquatic environment with an intermittent laminar flow in a non-turbulent hydraulic system where scavengers were present (Pesquero et al., 2013). At both sites, microorganisms may have played an important role in Fe precipitation as oxides during the early stages of diagenesis, although differences, evident in subsequent episodes of mineralization, are considered related to environmental factors. In the case of the alkaline palaeo-lakeshore of Cerro de la Garita, calcite-rich fluids would have filled the Haversian canal network with early geodes or calcite. In the fluvial environment of the Gobi desert, cementation of calcite and gypsum took place during later mineralization episodes (Kremer et al., 2012). The Cerro de la Garita fossil samples also showed pyrite deposits located between some Fe microspheres. Sulphur was noted in most spectra analysed by EDS (Fig. 3D), although always in small amounts, which may be linked to bacterial activity and decay processes. Whilst Kremer et al. (2012) did not observe any pyrite mineralization in the Gobi desert bones, they could not exclude the presence of pyrite as a primary mineral phase during an early diagenetic stage that could be transformed through oxidation into hematite as a secondary mineral (present in the outer rings of microspheres) at a later diagenetic stage. At Cerro de la Garita site, pyrite framboids are present but scarce, always forming near microspheres. However, in general, environmental conditions appear to have inhibited sulphur deposition and, consequently, framboid formation and pyrite oxidation. In this regard, Pfretzschner (2000) observed that extensive pyritization is rare or absent in freshwater deposits because, although iron is fairly abundant, the sulphur content is low and the redox potential is usually not low enough for pyritization. Under these conditions, pyrite formation should be localized next to the source of sulphur, which is usually decaying organic matter, especially proteins (Canfield and Raiswell, 1991). It is possible that the environmental conditions in which the Cerro de la Garita bones fossilized inhibited the formation of large amounts of pyrite, instead favouring iron mineralization. Finally, microstructural changes associated with marine and freshwater bodies are seen as endolithic tunnelling (Bell et al., 1996; Bell and Elkerton, 2008; Pesquero et al., 2010), which looks very different and is peripheral in its distribution. Given that the fossil bones described here, and elsewhere, were recovered from transitional environments between land and water, some thought needs to be given to the presence of microspheres restricted to the Haversian system and what taphonomic information can be derived from them. One question of course is, are these truly bacterial in origin? Their location and size and their distribution in the Haversian canal suggest that they may well be, but it is curious that they do not invade the bone microstructure as terrestrial bacteria do, and most importantly, bacterial change of this type has not been seen in other water-derived deposits (Bell et al., 1996; Bell and Elkerton, 2008). Perhaps a working explanation might be that the animal was decomposing in a context that was terrestrial for periods
33
of time and flooded or tidal at others. You could then expect to see bacterial change associated with terrestrial decomposition, located in what would otherwise be described as a water-worked environment.
6. Conclusions Analysis of the Cerro de la Garita fossils shows iron deposits in the interior of putative bacterial colonies (MFD) displaying transitional morphological stages characteristic of MFD pores to more highly constrained mineralized rhombic units, both rich in iron. This rhombic organization has not been seen in bacterially altered bone from purely terrestrial contexts. Nor does it share the morphology associated with marine and lacustrine tunnelling of bone. The most advanced stage of the microsphere is goethite-rich microspheres that contain organic matter in the internal core in Haversian canals. Chemical composition, morphology and location of microspheres in Cerro de la Garita are very similar to those described by Kremer et al. (2012) for the Gobi Desert specimens. Transitional MFDs and microspheres with organic material in the centre of the sphere observed in Cerro de la Garita, give support to a microbial origin for iron microspheres as proposed by Kremer et al. (2012). In the early stages of formation, microorganisms may alter the pH of their immediate environment causing nucleation and precipitation of Fe oxides on the bacterial cellular wall. These structures would be the precursors of the microsphere iron-rich deposits (with occasional pyrite). Later, calcite precipitates filling many of these cavities in the bone indicating a post-depositional environment characterized by a shallow phreatic level and surrounding fluids oversaturated in calcium carbonate. If these microspheres are truly bacterial in origin, then they may be useful as a taphonomic indicator of terrestrial exposure in a transitional environment between land and water.
Acknowledgements Excavations at Cerro de la Garita, fossil preparation, and later palaeontological research, were made possible through the support of the Spanish Ministry of Science and Innovation (projects CGL200766231/CGL2010-19825) and a Juan de la Cierva contract to M.D.P. Ref. JCI-2007-132-565). We are especially grateful to Norah Moloney for comments on the text, and to the Electron Microscopy Laboratory of the Museo Nacional de Ciencias Naturales-CSIC (MNCN). Thanks are also owed to Francisco Esteban, owner of the Cerro de la Garita site, for his assistance in this project.
References Alcalá, L., 1992. Macromamíferos neógenos de la fosa de Alfambra-Teruel. (Ph D Thesis). Univ. Complutense, Madrid, Spain. Alcalá, L., Martín Escorza, C., Pesquero, M.D., 1999. Implicaciones tafonómicas de la orientación de huesos de mamíferos en el Neógeno continental de Concud (Teruel). Rev. Soc. Geol. Esp. 12, 501–511. Baskar, S., Baskar, R., Mauclaire, L., McKenzie, J.A., 2006. Microbially induced calcite precipitation in culture experiments: possible origin for stalactites in Sahastradhara caves, Dehradun India. Curr. Sci. India 90, 58–64. Bazylinski, D.A., Frankel, R.B., 2000. Biologically controlled mineralization of magnetic iron minerals by magnetotactic bacteria. In: Lovley, D.R. (Ed.), Environmental Microbe–Metal Interactions. ASM Press, NW Washington, pp. 109–143. Bell, L.S., 1990. Palaeopathology and diagenesis: an SEM evaluation of structural changes using backscattered electron imaging. J. Archaeol. Sci. 17, 85–102. Bell, L.S., 2011. Histotaphonomy. In: Crowder, C., Stout, S. (Eds.), Bone Histology: An Anthropological Perspective. CRC Press, New York, pp. 241–251. Bell, L.S., Elkerton, A., 2008. Human remains recovered from a sixteenth century mass fatality: unique marine taphonomy in human skeletal material from the medieval warship the Mary Rose. Int. J. Osteoarchaeol. 18, 523–535. Bell, L.S., Boyde, A., Jones, S.J., 1991. Diagenetic alteration to teeth in situ illustrated by backscattered electron imaging. Scanning 13, 173–183. Bell, L.S., Skinner, M.F., Jones, S.J., 1996. The speed of post-mortem change to the human skeleton and its taphonomic significance. Forensic Sci. Int. 82, 129–140. Borsato, A., Frisia, S., Jones, B., Van Der Borg, K., 2000. Calcite moonmilk: crystal morphology and environment of formation in caves in the Italian Alps. J. Sediment. Res. 70, 1179–1190.
34
M.D. Pesquero et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 420 (2015) 27–34
Cacchio, P., Contento, R., Ercole, C., Capuccio, G., Preite-Martínez, M., Lepidi, A., 2004. Involvement of microorganisms in the formation of carbonate speleothems in the Cervo Cave (L'Aquila, Italy). Geomicrobiol J. 21, 497–509. Canfield, D., Raiswell, K., 1991. Pyrite formation and fossil preservation. In: Allison, P.A., Briggs, D.E. (Eds.), Taphonomy. Plenum Press, New York, pp. 337–387. Ehrlich, H.L., 1990. Microbiology. second ed. Marcel Dekker, New York. Fernández-Jalvo, Y., Andrews, P., Pesquero, D., Smith, C., Marín-Monfort, D., Sánchez, B., Geigl, E.M., Alonso, A., 2010. Early bone diagenesis in temperate environments. Part I: surface features and histology. Palaeogeogr. Palaeoclimatol. Palaeoecol. 288, 62–81. Ferris, F.G., 1993. Microbial biomineralization in natural environments. Earth Sci. (Chikyu Kagaku) 47, 233–250. Fortin, D., Beveridge, T.J., 2000. Mechanistic routes to biomineral surface development. In: biomineralization. In: Bäuerlein, E. (Ed.), Biology to Biotechnology and Medical Application. Wiley-VCH, Weinheim, Germany, pp. 7–24. Frankel, R.B., Bazylinski, D.A., 2003. Biologically induced mineralization by bacteria. Rev. Mineral. Geochem. 54, 95–114. Hackett, C.J., 1981. Microscopical focal destruction (tunnels) in exhumed human bones. Med. Sci. Law 21, 243–265. Hedges, R.E.M., 2002. Bone diagenesis: an overview of processes. Archaeometry 44, 319–328. Hedges, R.E., Millard, A.P., Pike, A.W.G., 1995. Measurements and relationships of diagenetic alteration of bone from three archaeological sites. J. Archaeol. Sci. 22, 201–210. Jackes, M., 1990. Diagenetic change in prehistoric Portuguese human bone (4000 to 8000 BP). Symposium on Bone Chemistry, Canadian Association for Physical Anthropology, 18th Annual Meeting. Jans, M.M.E., 2005. Histological characterisation of diagenetic alteration of archaeological bone. Geoarchaeological and Bioarchaeological Studies 4. Institute for Geo and Bioarchaeology, Vrije Universiteit, Amsterdam. Jans, M.M.E., Nielsen-Marshb, C.M., Smithc, C.I., Collinsd, M.J., Karsa, H., 2004. Characterisation of microbial attack on archaeological bone. J. Archaeol. Sci. 31, 87–95. Kawano, M., Tomita, K., 2001. Microbial biomineralization in weathered volcanic ash deposit and formation of biogenic minerals by experimental incubation. Am. Mineral. 86, 400–410. Konhauser, K.O., Kapple, A., Roden, E.E., 2011. Iron in microbial metabolisms. Elements 7, 89–93. Kremer, B., Owocki, K., Królikowska, A., Wrzosek, B., Kazmierczak, J., 2012. Mineral microbial structures in a bone of the Late Cretaceous dinosaur Saurolophus angustirostris from the Gobi Desert, Mongolia — a Raman spectroscopy study. Palaeogeogr. Palaeoclimatol. Palaeoecol. 358–360, 51–61. Oh, S.J., Cook, D.C., Townsend, H.E., 1998. Characterization of iron oxides commonly formed as corrosion products on steel. Hyperfine Interact. 112, 59–65.
Pesquero, M.D., Ascaso, C., Fernández-Jalvo, Y., Alcalá, L., 2010. A new taphonomic bioerosion in a Miocene lakeshore environment. Palaeogeogr. Palaeoclimatol. Palaeoecol. 295, 192–198. Pesquero, M.D., Alcalá, L., Fernández-Jalvo, Y., 2013. Taphonomy of the reference Miocene vertebrate mammal site of Cerro de la Garita, Spain. Lethaia 46, 378–398. Pfretzschner, H.-U., 1998. Frühdiagenetische Prozesse bei der Fossilisation von Knochen. Neues Jahrb. Geol. Palaontol. Abh. 210, 369–397. Pfretzschner, H.-U., 2000. Micro-cracks and fossilization of Haversian bone. Neues Jahrb. Geol. Palaontol. Abh. 216, 413–432. Pfretzschner, H.-U., 2001a. Pyrite in fossil bone. Neues Jahrb. Geol. Palaontol. Abh. 220 (1), 1–23. Pfretzschner, H.-U., 2001b. Iron oxides in fossil bone. Neues Jahrb. Geol. Palaontol. Abh. 220 (3), 417–429. Sánchez-Moral, S., Bedoya, J., Luque, L., Cañaberas, J.C., Jurado, V., Laiz, L., Saiz-Jiménez, C., 2003a. Biomineralization of different crystalline phases by bacteria isolated from catacombs. In: Saiz-Jiménez, C. (Ed.), Molecular Biology and Cultural Heritage. Balkema, Lisse, pp. 179–185. Sánchez-Moral, S., Cañaberas, J.C., Laiz, L., Saiz-Jiménez, C., Bedoya, J., Luque, L., 2003b. Biomediated precipitation of calcium carbonate metastable phases in hypogean environments. Geomicrobiol J. 20, 491–500. Sánchez-Moral, S., Luque, L., Cañaberas, J.C., Laiz, L., Jurado, V., Hermosin, B., Saiz-Jiménez, C., 2004. Bioinduced barium precipitation in St. Callixtus and Domitilla catacombs. Ann. Microbiol. 54, 1–12. Sánchez-Moral, S., González, J.M., Cañaberas, J.C., Cuezva, S., Lario, J., Cardell, C., Elez, J., Luque, L., Saiz-Jiménez, C., 2006. Procesos de precipitación mineral bioinducidos en sistemas kársticos subterráneos: breve revisión y nuevas tendencias. Estud. Geol. 62, 43–52. Simón, J.L., 1984. Comprensión y distensión alpinas en la cadena Ibérica oriental. Inst. Estudios Turolenses (269 pp.). Southam, G., 2000. Bacterial surface-mediated mineral formation. In: Lovley, D.R. (Ed.), Environmental Microbe–Mineral Interactions. ASM Press, Washington, DC, pp. 257–276. Tormo, L., 2007. Interacción bacteria-mineral en substratos de cuevas volcánicas. Unpublished Trabajo Tutelado de Iniciación a la Investigación, Facultad de Ciencias, Universidad Autónoma de Madrid, Spain, 34 pp. Turner-Walker, G., Jans, M.M.E., 2008. Reconstructing taphonomic histories using histological analysis. Palaeogeogr. Palaeoclimatol. Palaeoecol. 266, 207–235. Turner-Walker, G., Nielsen-Marsh, C.M., Syversen, U., Kars, H., Collins, M.J., 2002. Submicron spongiform porosity is the major ultra-structural alteration occurring in archaeological bone. Int. J. Osteoarchaeol. 12, 407–414.