Geobiology (2013), 11, 139–153
DOI: 10.1111/gbi.12022
Nanometer-scale characterization of exceptionally preserved bacterial fossils in Paleocene phosphorites from Ouled Abdoun (Morocco) E,1 B. BOUYA4 AND J . C O S M I D I S , 1 , 2 K . B E N Z E R A R A , 1 E . G H E E R B R A N T , 3 I . E S T EV M. AMAGHZAZ4 1
Institut de Mine´ralogie et de Physique des Milieux Condense´s, Universite´ Pierre et Marie Curie, CNRS, UMR 7590, Campus Jussieu, F-75005, Paris, France 2 Equipe Ge obiosphe re Actuelle et Primitive, IPGP, Sorbonne Paris Cite , Universite´ Paris Diderot, CNRS, UMR 7154, F-75005, Paris, France 3 Centre de Recherches sur la Pale obiodiversite et les Pale oenvironnements, Muse´um National d’Histoire Naturelle, CNRS, UMR 7207, CP38, F-75005, Paris, France 4 Office Che rifien des Phosphates, Centre minier de Khouribga (Geological Survey), Khouribga, Morocco
ABSTRACT Micrometer-sized spherical and rod-shaped forms have been reported in many phosphorites and often interpreted as microbes fossilized by apatite, based on their morphologic resemblance with modern bacteria inferred by scanning electron microscopy (SEM) observations. This interpretation supports models involving bacteria in the formation of phosphorites. Here, we studied a phosphatic coprolite of Paleocene age originating from the Ouled Abdoun phosphate basin (Morocco) down to the nanometer-scale using focused ion beam milling, transmission electron microscopy (TEM), and scanning transmission x-ray microscopy (STXM) coupled with x-ray absorption near-edge structure spectroscopy (XANES). The coprolite, exclusively composed of francolite (a carbonate-fluroapatite), is formed by the accumulation of spherical objects, delimited by a thin envelope, and whose apparent diameters are between 0.5 and 3 lm. The envelope of the spheres is composed of a continuous crown dense to electrons, which measures 20–40 nm in thickness. It is surrounded by two thinner layers that are more porous and transparent to electrons and enriched in organic carbon. The observed spherical objects are very similar with bacteria encrusting in hydroxyapatite as observed in laboratory experiments. We suggest that they are Gram-negative bacteria fossilized by francolite, the precipitation of which started within the periplasm of the cells. We discuss the role of bacteria in the fossilization mechanism and propose that they could have played an active role in the formation of francolite. This study shows that ancient phosphorites can contain fossil biological subcellular structures as fine as a bacterial periplasm. Moreover, we demonstrate that while morphological information provided by SEM analyses is valuable, the use of additional nanoscale analyses is a powerful approach to help inferring the biogenicity of biomorphs found in phosphorites. A more systematic use of this approach could considerably improve our knowledge and understanding of the microfossils present in the geological record. Received 19 June 2012; accepted 21 November 2012 Corresponding author: J. Cosmidis. Tel.: +33 1 44 27 75 41; fax: +33 1 44 27 37 85; e-mail:
[email protected]
INTRODUCTION Marine phosphorites are phosphate-rich sedimentary formations that result from the reworking and concentration of calcium phosphate minerals. A number of studies have shown that microbial processes are implicated in the
© 2012 Blackwell Publishing Ltd
precipitation of phosphatic minerals and contribute to the formation of phosphorites (e.g., Reimers et al., 1990; Krajewski et al., 1994; F€ ollmi, 1996; Soudry, 2000; Schulz & Schulz, 2005). According to these studies, bacteria act by transitorily concentrating and/or releasing phosphorus issued from the degradation of organic P-rich compounds.
139
140
J. COSMIDIS et al.
Traces of microbial activity have been suggested in many ancient phosphorites and modern phosphogenic environments (e.g., Schulz & Schulz, 2005; Arning et al., 2009; Berndmeyer et al., 2012), and experimental studies have been conducted to gain further insight into the molecular mechanisms involved in the microbial precipitation of calcium phosphate minerals (e.g., Lucas & Prev^ ot, 1984; Hirschler et al., 1990; Benzerara et al., 2004a; Goldhammer et al., 2010). Phosphorites are also known as preferential sites for finescale fossilization and exceptional preservation of biologic structures. The most emblematic example is provided by the ~570-million-year-old fossils of the Ediacaran Doushantuo phosphorite formation in China (Xiao et al., 1998; Chen et al., 2009). These fossils were lately reinterpreted as encysting protists, and cellular details as fine as the nucleus may have been preserved (Huldtgren et al., 2011). Taphonomic experiments with modern embryos (Raff et al., 2008) suggest that these fossils may have formed by the replacement of the cellular structures by the bacterial biofilms and subsequent biomineralization of the bacteria. The report of fossil traces of micro-organisms in phosphorites is an old matter. In 1936, Cayeux observed spherules measuring 0.5–2.5 lm in diameter and surrounded by a thick envelope in phosphorites. He interpreted them as bacteria (Cayeux, 1936). Based on systematic observations, he suggested that they might be present in phosphorites throughout the geological record. With the development of scanning electron microscopy (SEM), numerous observations of purported fossil micro-organisms have been reported in phosphorites since then (e.g., Soudry, 2000; Zanin & Zamirailova, 2011 and references therein). The identification of bacterial structures in phosphorites has systematically been based on the morphologic resemblance between micrometer-sized spherical, rod-shaped, or filamentous apatite particles found in phosphorites and modern microbial cocci, bacilli, or filaments (e.g., O’Brien et al., 1981; Lamboy, 1994; Toporski, 2002; Lundberg & McFarlane, 2011). However, the identification of microbial fossils by SEM alone is not straightforward. Indeed, precipitation experiments under interstitial water physico-chemical conditions as well as thermodynamical calculations have shown that apatite particles grow to sizes in the order of 0.1–10 lm during burial in modern phosphatic sediments (VanCappellen & Berner, 1991). SEM observations of experimentally synthetized apatite minerals have moreover shown that abiotic reactions can produce textural features similar to those formed by microbially mediated precipitation (Blake et al., 1998). Calcium phosphate particles resembling bacteria in size and shape at the SEM scale (biomorphs) can therefore be obtained by abiotic processes. In contrast, some structural details reminiscent of a wall or an envelope have sometimes been observed by SEM in putative bacterial par-
ticles (Breheret, 1991; Soudry, 1992; Alvaro & Clausen, 2010). However, SEM provides limited insight into the exact nature of these envelopes and can thus not ascertain their biological origin. As a result, Baturin and Titov (2006) have reinterpreted globular and elongated particles measuring several hundreds of nanometers up to a few micrometers in size and observed in phosphorites from the Namibian shelf as non-microbial in origin. This drastic reinterpretation was based on the observation of a wide range of variations in sizes, the existence of mutual intergrowths and their irregular distribution in the sediments, all of which are features supposedly characteristic of minerals instead of bacteria. Baturin and Titov (2006) moreover proposed that ‘the overwhelming part of the bacteriomorphic particles in recent phosphorites refer to mineral rather than to biogenic formations’. Despite great progress achieved in the understanding of phosphorite genesis by laboratory experiments, the presence of micrometer-sized forms interpreted as fossil bacteria in phosphorites has often been used as a major evidence of the implication of microbes in their formation (Mullins & Rasch, 1985; Soudry & Lewy, 1988; Lamboy, 1994; Purnachandra Rao et al., 2000). The correct interpretation of these forms is therefore crucial and calls for additional clues than morphology only. Furthermore, biogenic structures maybe altered during diagenesis and metamorphism, which make them even more difficult to distinguish from abiotic biomorphs (Brasier et al., 2002; Javaux & Benzerara, 2009; Schiffbauer et al., 2012a). It has been shown, however, that the combination of several high-resolution techniques can help in identifying remnants of biological structures in ancient and/or metamorphosed rocks (e.g., Lepot et al., 2011; Galvez et al., 2012). Here, we present new observations performed on a phosphatic coprolite of Paleocene age containing numerous purported bacteriomorphs. In addition to SEM, we use microscopy techniques, including focused ion beam (FIB) milling, transmission electron microscopy (TEM), and scanning transmission x-ray microscopy (STXM), that provide a higher spatial resolution and complementary information on the mineralogical and chemical composition of the samples at the nanoscale.
MATERIALS AND METHODS Sample The phosphatic coprolite originates from the Ouled Abdoun phosphate basin (Morocco), which has been studied thoroughly because it contains very rich marine vertebrate fauna, and notably mammals such as the earliest known proboscideans Phosphatherium escuilliei (Gheerbrant et al., 1996, 1998) and Eritherium azzouzorum (Gheerbrant, 2009). The phosphate series of Ouled Abdoun were
© 2012 Blackwell Publishing Ltd
Fossil bacteria in Paleocene phosphorites
141
Geochemical analyses
Fig. 1 Photograph of two phosphatic coprolites originating from the Ouled Abdoun phosphate basin. These two coprolites were found in the same bed (local bed IIa) and present similar macroscopic features as the specimen described in this study. Scale bar is 10 cm 9 2 cm.
deposited in an epicontinental sea open on to the Atlantic toward the west, whose depth has varied during the Cretaceous and Cenozoic (Cappetta, 1981). More information about the paleogeography of the Ouled Abdoun basin can be found in Gheerbrant et al. (2003). The coprolite was collected from the phosphorite local Bed IIa and is comprised in the lower Bone Bed horizon yielding Eritherium azzouzorum, of possible Selandian age (59–60 Ma; Gheerbrant, 2009). The sedimentary matrix of Bed IIa is composed of gray and yellow to brown sandy ‘pseudo-oolitic’ phosphorites (phospharenites). The stratigraphy and lithology of the Ouled Abdoun phosphate series are fully described in Gheerbrant et al. (2003). The coprolite specimen is roughly cylindrical with rounded ends and measures ~2.3 cm in diameter and ~5 cm in length. It presents a yellowish-white color and a very smooth and homogeneous texture and shows no striation. It has been inferred that the animal producing this coprolite was a small crocodilian or a turtle (France de Lapparent de Broin, MNHN, pers. comm.). Two similar phosphatic coprolites, originating from Bed IIa, are shown in Fig. 1.
X-ray diffraction The bulk mineralogical composition of the coprolite was determined by x-ray diffraction (XRD). About 1 g of the coprolite was crushed in an agate mortar in pure ethanol, and the powder was deposited on an aluminum sample holder. XRD measurements were performed on a Panalytical X’Pert Pro MPD diffractometer equipped with a copper anode (Cu Ka). Data were recorded at 40 kV and 40 mA in the continuous-scan mode between 3 and 120° (in 2h) with a step of 0.017° and a total counting time of around 6 h. XRD data were analyzed using the PANalytical X’Pert Highscore software for background subtraction, peak finding, and matching with XRD patterns of reference compounds from the International Crystal Structure Database (ICSD, Fachinformationszentrum Karlsruhe, Germany; US Institute of Standards and Technology, USA).
© 2012 Blackwell Publishing Ltd
Concentrations of major and trace elements, total organic carbon (TOC), and loss on ignition (LOI) measurements were performed by the Service d’Analyse des Roches et Mineraux (SARM, Centre de Recherches Petrographiques et Geochimiques, Nancy, France). Elemental analyses were performed by alkali fusion of rock samples (LiBO2), followed by concentration measurements using ICP-AES and ICP-MS for major and trace elements, respectively. The uncertainty of measurements was
