Fossil microbial shark tooth decay documents in situ metabolism of enameloid proteins as nutrition source in deep water environments

Abstract

Alteration of organic remains during the transition from the bio- to lithosphere is afected strongly by biotic processes of microbes infuencing the potential of dead matter to become fossilized or vanish ultimately. If fossilized, bones, cartilage, and tooth dentine often display traces of bioerosion caused by destructive microbes. The causal agents, however, usually remain ambiguous. Here we present a new type of tissue alteration in fossil deep-sea shark teeth with in situ preservation of the responsible organisms embedded in a delicate flmy substance identifed as extrapolymeric matter. The invading microorganisms are arranged in nest- or chain-like patterns between fuorapatite bundles of the superfcial enameloid. Chemical analysis of the bacteriomorph structures indicates replacement by a phyllosilicate, which enabled in situ preservation. Our results imply that bacteria invaded the hypermineralized tissue for harvesting intra-crystalline bound organic matter, which provided nutrient supply in a nutrient depleted deep-marine environment they inhabited. We document here for the frst time in situ bacteria preservation in tooth enameloid, one of the hardest mineralized tissues developed by animals. This unambiguously verifes that microbes also colonize highly mineralized dental capping tissues with only minor organic content when nutrients are scarce as in deep-marine environments.

Full text



Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports
Fossil microbial shark tooth decay
documents in situ metabolism
of enameloid proteins as nutrition
source in deep water environments
Iris Feichtinger1*, Alexander Lukeneder1, Dan Topa2, Jürgen Kriwet3*, Eugen Libowitzky4 &
Frances Westall5
Alteration of organic remains during the transition from the bio- to lithosphere is aected strongly by
biotic processes of microbes inuencing the potential of dead matter to become fossilized or vanish
ultimately. If fossilized, bones, cartilage, and tooth dentine often display traces of bioerosion caused
by destructive microbes. The causal agents, however, usually remain ambiguous. Here we present a
new type of tissue alteration in fossil deep-sea shark teeth with in situ preservation of the responsible
organisms embedded in a delicate lmy substance identied as extrapolymeric matter. The invading
microorganisms are arranged in nest- or chain-like patterns between uorapatite bundles of the
supercial enameloid. Chemical analysis of the bacteriomorph structures indicates replacement
by a phyllosilicate, which enabled in situ preservation. Our results imply that bacteria invaded the
hypermineralized tissue for harvesting intra-crystalline bound organic matter, which provided nutrient
supply in a nutrient depleted deep-marine environment they inhabited. We document here for the
rst time in situ bacteria preservation in tooth enameloid, one of the hardest mineralized tissues
developed by animals. This unambiguously veries that microbes also colonize highly mineralized
dental capping tissues with only minor organic content when nutrients are scarce as in deep-marine
environments.
Teeth and bones are oen the only evidence of ancient vertebrate life because of the mineralized nature of tissues.
ere are numerous possibilities for chemical alteration during the transition from the bio- to the lithosphere
of which bacterial catabolysis of these tissues and organic matter within the carcass is an important example1,2.
Preservation of so tissue body fossils (e.g. skin) requires specic conditions of abiotic (e.g. salinity, tempera-
ture, pH-value, oxygen) and biotic (e.g. bacteria or bioturbation in general) factors, as well as marginal uctua-
tions in the continuum of processes during sedimentation, fossilization, lithication and preservation through
geological time3–5. Paradoxically, microbial activity can be both preserving and destructive. One of the most
famous preserving eects of microbial activity is documented by fossils from the Eocene Messel oil shale, which
exhibit a unique preservation of so tissues. Closer examination of the "skin shadows" preserved in these fossils
reveals an accumulation of lithied bacterial colonies3, mirroring the original contour. However, the preserva-
tive eects of bacteria are, denitely, rare phenomena requiring specic conditions. Generally, microorganisms
are predominantly responsible for destructive processes, removing digestible so tissues of carcasses preceding
diagenesis. Additionally, some organisms, like the bone-eating worm Osedax, literally invade bones to obtain
nutrients6 when food supply is limited, as in bathyal marine settings.
Decay of bony material in modern deep-sea environments is dominated by anaerobic microbial decomposi-
tion of the large lipid reservoirs within bones7. Studies of decay processes in both modern and fossil deep-sea
environments such as, whale-falls, show that they represent important nutrition supplies for deep-sea organisms.
However, these previous studies focused only rarely on bacteria or archaea, which are at the base of the food
webs8. Nevertheless, limitations in nutrients are also to be expected in ancient deep marine environments with
OPEN
         

            
           
             *

Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
similar microbial alteration of skeletal material, as in comparable modern habitats. Although microbial altera-
tion of bones (bioerosion) is known from aquatic environments, equivalent alteration in teeth has only rarely
been reported. Documented bioerosion patterns of dental tissues include endolithic macro- and microborings
of fossil teeth and small tubules within dentine represented by traces and holes of endolithic bivalves, clionaid
sponges, serpulid worms, and routes of microbial intrusion9,10. Here we review the fossil record of bioerosion
and document for the rst time insitu bacteria within the highly mineralized and organic-poor tooth enameloid
of an extinct deep-water shark. is nding represents a hitherto unrecognized bioerosion type for teeth and
nutrition source in deep-sea environments.
Results
Morphology. Based on the extensive enameloid investigations of previous studies11, several teeth of the
extinct shark Cretacladoides noricum were studied in detail, of which only two out of 40 examined teeth display
internal insitu microbial alteration of the supercial enameloid (Fig. 1C–G). Traces of bioerosion on tooth
surfaces of various species of the same fauna are common despite the scarce evidence of internal alteration
(Fig.1H–J).
e enameloid of the teeth analyzed here is characterized by parallel to subparallel bundles of uorapatite
(Ca5(PO4)3F) and corresponds exactly to the pattern present in extant shark teeth12. Two of the teeth exhibit
aggregates of mineralized, regularly shaped, coccoidal to short rod-shaped structures ranging from 0.5 to 1µm
in length and 0.4–0.5µm in diameter. ey are organized in chains that are arranged parallel to each other
between the enameloid bundles (Fig.1C,E). e surfaces of the coccoids are slightly irregular and most appear
to be attached to each other and to the enameloid substrate by a delicate, lmy substance with a partially aky
appearance (Fig.1F–G). Delicate brils can still be observed between the coccoidal structures (arrows in Fig.1G).
Some denser associations of aggregated structures also occur in spaces between the highly mineralized enameloid
bundles (Fig.1D,F) and within dentinal canals (Fig.1A).
Chemical composition. e coccoids and rods have a clearly mineralized appearance and are associated
with two types of minerals: a compact mineral with a nely pitted surface that forms the body of the coccoids
and rods, and tiny aky minerals with a phyllosilicate appearance that are attached to the surfaces of many of the
coccoids/rods and also to the lmy material that forms their immediate substrate (Fig.1F–G).
EDS analysis of the highly mineralized enameloid bundle of one of the teeth at 15keV at the NHM Vienna
(Fig.2A) shows distinct peaks of C (carbon coating), O, F, P, Ca (from le to right) resulting in stoichiometric
oxide values of CaO (54.29 wt%), P2O5 (40.75 wt%), and F (4.96 wt%). Aer conversion to mol% and correc-
tion of oxygen valence by one uorine, an apatite formula Ca5.03P2.98O11.98F1.35 close to the ideal composition is
obtained. e chemical analysis of the coccoidal aggregates (Fig.2B) indicates signicant amounts of SiO2 (26.46
wt%), Al2O3 (9.57 wt%), FeO (5.45 wt%), MgO (5.01 wt%), and Na2O (0.75 wt%), in addition to CaO (27.56
wt%), P2O5 (23.60 wt%), and F (1.60 wt%).
e EDS analysis made at 15keV clearly penetrated into the matrix and shows a compositional mixture
of ~ 50% matrix and ~ 50% coccoidal structures, given their small size compared with the volume of excitation
(by electrons) of the X-ray radiation (Fig.2B). Using the Anderson-Hasler formula of X-ray range13, approxi-
mately 2µm penetration depth must be considered at an average X-ray energy of 2keV, a density of ~ 3g/cm3
and 15keV acceleration voltage. us, the results conrm the uorapatite composition of the matrix but the
Na, Mg, Al, Si and Fe peaks also indicate the presence of an aluminosilicate mineral (Fig.2B). e composition
presented in Table1 (Spectrum B) therefore represents the combination of the chemistry of the matrix (= ena-
meloid bundle) and the coccoidal aggregates (Fig.2B). e analyses made at 5keV, on the other hand, did not
penetrate as deeply into the matrix (~ 0.25µm using the Anderson-Hasler formula from above) and conrmed
the presence of an aluminosilicate containing Na, Mg and Fe (L peak) (Fig.2B′). ese analyses corroborate the
presence of a coating of phyllosilicate, probably a clay mineral close to smectite (montmorillonite) or saponite,
encasing the bacteriomorph objects. Using the above composition without Ca and P (from the matrix), the
formula (based on 4 Si) (Al1.72Mg1.11Fe0.68)[Si4O10] ∙ Na0.25 is obtained, which compares well with a clay mineral
between montmorillonite, ~ (Al1.67Mg0.33)[(OH)2|Si4O10] ∙ Na0.33(H2O)4, and saponite, Mg3[(OH)2|(Si,Al)4O10] ∙
(Ca,Na)x(H2O)y14. Note that the 5keV EDS analysis (Fig.2B′) documents large C and O peaks, which are caused
by a higher emission yield of light vs. heavier elements at low-energy (5keV) excitation conditions. However,
to test for the possible presence of reduced carbon entrapped within the mineralized structures of the coccoids,
a comparative EDS study of enameloid and coccoids at 15, 10, and 5keV was performed (see supplementary
information for spectra).
At 15keV conditions, analysing for C, O, F, P, Ca independently (oxygen not by stoichiometry), the enam-
eloid composition resulted in a formula of ~ Ca5.07P3.00O12.13F0.81 which is close to the ideal formula of apatite (see
above). Carbon concentrations in enameloid were between 10.08 and 11.92 wt% in three spots. For comparison,
two spots on the bacterial remnants resulted in 11.16 and 11.51 wt%. At 10keV carbon contents on enameloid
were scattered strongly between 6.54 and 16.90 wt%, whereas the coccoids gave 9.41 and 10.58 wt%. At 5keV
the carbon signal was strongly corrupted by the strong EDS zero peak and gave 31.47 wt% C in enameloid and
15.12 and 19.40 wt% in the bacterial structures. us, none of these measurements conrmed an excess of carbon
from relics of organic material in the coccoids.
Contamination with recent bacteria can be excluded here because of the chemical ngerprint of extant bac-
teria, in which the three elements C, N, and O constitute 80–90 wt%15.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
Figure1. Sketch of bacteriomorph structures within the enameloid and scanning electron micrographs of
fossilized bacteria of tooth NHMW 2017/0055/0028. (A) sketch of bacteriomorph bodies providing an overview
about the frequency and arrangement of the coccoids within the enameloid bundles (for a better visualization,
coccoids and enameloid bundles are not to scale). (B) holotype of Cretacladoides noricum (NHMW
2017/0055/0001) in prole view for demonstration of the section plane. (C–E) coccoid/rod-shaped bacteria
arranged in chains embedded parallel to crystallite bundles of tooth enameloid. (F) denser associations in nest-
like structures of fossilized bacteria. (G) close up of nest-like structures, white arrows indicate lmy substance.
(H–J) Dierent teeth with examples of bioerosion on the tooth surface of the same sample. (H–I) Paratype of
Similiteroscyllium iniquus (NHMW 2017/0058/0005) with branching type of bioerosion. (J) Pycnodont tooth
with boreholes. Dc, Dentinal canal; PBE, Parallel Bundled Enameloid; TBE, Tangle Bundled Enameloid.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
Discussion
Interpretation of the aggregates of coccoid/rod-shaped structures. e regular size and mor-
phology of the coccoidal/rod-shaped structures in our study, as well as their specic distribution in linear chains
in the cavities between the parallel bundles of tooth enameloid, is suggestive of microorganisms, such as bacte-
ria. Indeed, they strongly resemble fossilized microorganisms associated with decaying macro-organisms else-
Figure2. Quantitative energy dispersive spectrometry (EDS) of tooth NHMW 2017/0055/0028. (A) EDS
analysis of uorapatite crystal of tooth enameloid with 15keV. (B) EDS analysis of fossilized bacteria with
15keV. (B′) EDS analysis of fossilized bacteria with 5keV.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
where, such as the phosphatised bacteria of the “skin shadow” of fossil vertebrates in the Enspel oilshale3,16. Both
extant and fossilized bacteria exhibit a size range between 0.5 and 4µm15–17, which coincides with the size of the
mineralized bodies interpreted as fossilized bacteria herein. Bacteria and cyanobacteria are the predominant
bioerosion-causing organisms, however, both causal agents dier signicantly in the type of traces they produce.
Cyanobacteria are known to infest shells of marine bivalves in suitable habitats within the photic zone18. ese
phototrophic organisms create branching tunnels but do not produce traces with localized demineralization or
cung of redeposited mineral, which is typical for terrestrial or freshwater bacteria17,19,20. us, the taphonomic
alteration of the bacteriomorph invaders of the herein described shark teeth diers signicantly from bacteria
occupying terrestrial or freshwater environments. Nevertheless, the distinct morphology and size of the fossil-
ized bodies point towards bacteriomorph microorganism but exclude cyanobacteria due to their phototrophic
lifestyle.
Given the presence of bacteria, the observed lmy substance linking the individual bacteriomorph structures
together, as well as the enameloid crystallites, are thus interpreted here as microbial extrapolymeric substances
(EPS). EPS is a common exudate of microbes, used for attachment to substrates and as a control of the external
physico-chemical conditions21,22.
As noted above, microbial degradation of organic substances is very common; what is not so common, how-
ever, is the physical preservation and mineralization of the degrading heterotrophs for which specic physico-
chemical conditions are necessary, specically, a micro-scale anaerobic environment (easily achieved through
microbial oxidation of an organic substrate). Indeed, only two out of 40 tooth specimens exhibit this phenomenon
in our study. e small number of aected teeth, however, indicates either very specic conditions under which
invasion of bacteria into the hypermineralized tooth enameloid was feasible or lack of potential for fossilization
of colonising microbes on or in other teeth due to diverging time intervals of tooth shedding, pH conditions,
or other inuencing factors.
It is not possible to establish the exact composition of the fossilized bacteria owing to their small size. Never-
theless, the morphology of the tiny aky minerals attached to the surfaces of many of the coccoids/rods forming
their immediate substrate is reminiscent of phyllosilicates replacing or formed on EPS-like lm, as is also sug-
gested by the elements typical for aluminosilicates in spectrum B (15keV) and B′ (5keV) of Fig.2. In addition
to Si and Al, the spectrum also documents the presence of Fe, Mg, and Na. If related to a phyllosilicate, this
would indicate clay minerals close to smectite or saponite (see above). Substracting the clay composition from
the spectra just leaves the coccoids/rods with compositions close to the uorapatite matrix, although this may
simply be a consequence of the excitation energy of the electron beam (15 and 5keV) resulting in penetration
through the very small fossilized structures into the background uorapatite of the enameloid. Keeping in mind
the fact that the samples were coated with carbon before SEM observation and EDS analysis, the carbon signal
Table 1. Results of chemical composition analyzed by EDS of a uorapatite bundle of the tooth enameloid
(Spectrum A) and a fossilized bacteria (Spectrum B) of the tooth NHMW 2017/0055/0028 with 15keV and
fossilized bacteria (Spectrum B′) with 5keV.
Formula mass% mol% Cation
Spectrum A 15keV
F 4.96 17.21 0
P2O540.75 18.94 0.96
CaO 54.29 63.85 1.61
Tot a l 100 100
Spectrum B 15keV
F 1.6 5.65 0
Na2O 0.74 0.8 0.04
MgO 5.01 8.36 0.18
Al2O39.57 6.31 0.28
SiO226.46 29.59 0.65
P2O523.6 11.17 0.49
CaO 27.56 33.02 0.73
FeO 5.45 5.1 0.11
Tot a l 100 100
Spectrum B′ 5keV
F 4.16 13.85 0
Na2O 1.37 1.4 0.35
MgO 5.63 8.84 1.09
Al2O317.74 11.01 2.71
SiO254.69 57.59 7.09
P2O516.4 7.31 1.8
Tot a l 100 100
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
in the EDS spectra is easily explained. Beyond that, it is unlikely that the microorganisms were replaced by a
carbonate because the C peak is too low.
Whatever the exact composition of the replacement minerals is, they are likely to have formed as the result of
microbially inuenced changes of the immediate environment leading to enrichment in certain elements with
consequent precipitation onto functional groups of the degrading microbial structures23. During decomposition
and diagenesis, the composition and concentrations of elements in the surrounding uids control the type and
composition of minerals replacing organic substrates. ese elements come either from seawater and/or from
elements released by the degradation of an organo-mineral substrate (mostly transition metals). Release of cell/
EPS-bound elements is also inuenced by the metabolic activity of microbes. For example, an increase in local
alkalinity due to heterotroph degradation (by sulphate reducers) of primary photosynthetic mats releases Ca2+
ions into the uid medium, which then combine with CO2 in seawater to form Ca carbonate24. In the case of
the shark teeth in a deep-water environment, the organic substrate would be provided by the organic matrix
(collagen and other proteins) of the teeth itself. If the replacing mineral was carbonate, the latter would have
been enriched in transition elements, such as Fe and Mn (the latter not present at any detectable levels here), or
Mg, resulting from the degraded substrate. In the case of phyllosilicate formation, Na-Fe3+ phyllosilicates have
been experimentally produced on microbial EPS25. Indeed, EPS plays an important role in the biosynthesis of
dierent types of clays26. Reactive sites on the surfaces of microbial cells act as loci for the nucleation of clay
minerals in the poorly crystalline state27. Subsequent diagenesis and aging transforms these poorly crystalline
materials into crystalline phases.
Enameloid invading bacteria. e presence of bacteria associated with teeth is, during the lifetime
of an animal, normal but they also contribute to microbial degradation of so-tissues aer death during
decomposition3–5. Normally, bacteria are only found on surfaces or within the pulp cavity, which is easily acces-
sible. However, they also gain access to tissues such as dentine through tooth surface lesions or due to previous
bioerosion of other organisms. It is thus not surprising to nd their fossilized remains associated with skeletal
structures such as shark teeth and even within dental tissues. However, the presence of bacteria within the hyper-
mineralized capping tissues of teeth, such as enamel or enameloid providing only very small amounts of severe
accessible organic matter as a possible nutrition source, has not been documented up to now.
e resistant hypermineralized outermost layer of enameloid of shark teeth consists of uorapatite. e
uorapatite crystals are embedded in an organic matrix of about 4.5 wt%, as documented for a tooth of a great
white shark (C. carcharias)28 indicating a variable content of collagen, proteins, and other organic structures
(e.g. tubular vesicles) in enameloid depending on species and tooth maturation29. It has been demonstrated
that the uorapatite crystallites are devoid of any organic matter, while the crystallite bundles are encased by an
organic matrix that generally has a smooth, sometimes also a brous appearance12. Additionally, the collagen
bres within the enameloid of an extant salmon shark (L. ditropis) are arranged in a regular pattern and the
bres cross each other30.
Collagen plays a special role in the composition of all structures needed for many eukaryotes, from plesio-
morphic sponges to vertebrates31, and is signicantly resistant to post-mortem decay19. Two types of collagen,
un-mineralized and mineralized, whose resistance to deterioration are vastly dierent20,32, occur in dierent
maturation states of teeth and bones. Apart from the rapidly degradable un-mineralized collagen type of primar-
ily fresh bone, specic conditions are required for the assimilation of mineralized collagen, which is the type
found in hypermineralized tooth capping tissue and mature bone. Here, the isolated collagen bres are stabilized
by tiny hydroxylapatite platelets responsible for avoiding direct enzymatic degradation. us, removal of the
densely packed mineral platelets is necessary for eective microbial cracking of the large collagen molecules19,32,33.
A comparative process of tooth decay is caused by the demineralization causing caries in human teeth. Caries-
causing bacteria that occur in plaque ferment carbohydrates (e.g. glucose and fructose) and generate an acidic
microenvironment, which has the ability to demineralise the enamel34. Only as a result of this process, bacteria
gain access to the now un-mineralized tissue for utilization the matrix proteins of enamel and dentine34,35. us,
any kind of tooth decay is directly linked to low pH values or acidic microenvironments. As noted above, it is in
such low pH environments that elements, such as Ca, can be released from an organic substrate and subsequently
re-precipitated as a mineral24.
Accordingly, the organic matrix between the uorapatite bundles inside the teeth would have provided a
nutrient supply for invading microbes. In bones, microbial degradation of collagen brils using collagenases
provides a high-energy yield36–38 and bacteria invade the bone through haversian canals in order to attain this
valuable nutrient supply39. While they subsequently follow the collagen brils, they are unable to cross the cement
lines of secondary osteons19. Similarly, microbes can invade shark teeth via the nutritive foramina in the root and
ascend apically using dentinal canals, resulting in a chain-like arrangement of bacteria. Moreover, bacteria can
migrate into the enameloid because the enameloid/dentine boundary is not an insuperable separating layer in
chondrichthyans, as is the enamel/dentine boundary in mammals, but is penetrated by dentinal tubes extending
into the enameloid. is would form possible pathways for bacteria from the dentine into the enameloid corre-
sponding to the orientation, arrangement and location of the fossilized bacteria between the enameloid bundles.
However, invasion of enameloid by means of external bioerosion of the tooth surfaces, which is observable
in numerous teeth of dierent species deriving from the same faunal assemblage (Fig.1H–J), enables an easily
accessible entrance directly through the external single crystallite layer of the enameloid. Considering all pos-
sible pathways of penetration, intrusion facilitated by surface lesions caused by bioerosion or even as symbiont
of macro-organisms presents another plausible scenario.
Numerous incredible symbiotic relationships are known in the biosphere, resulting in a benet of both
involved parties. An obvious example of a remarkable symbiosis of a comparative, marine habitat represents
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
the bone-eating worm Osedax, which hosts microbial symbionts to benet from their collagenolytic enzyme
activity40. e symbionts comprise primarily Oceanospirillales and Epsilonproteobacteria, which colonies the root
tissues of the small worms enabling the degradation of dierent types of collagen during intrusion of the worm
into bones of a whale fall40. Indirect evidence provided by trace fossils (i.e. boreholes) in fossil whale bones,
document that this highly specialized polychaete worm has a fossil record since the Oligocene (~ 30 million
years)6, but most likely already originated in the Cretaceous41.
Comparing the borehole diameter generated by recent Osedax species (e.g. O. rubiplumus) and ancient traces
measured on fossil whale bones a possible trend in borehole size increasing from the Oligocene (0.1–0.45mm)
to today (0.1–2.0mm) is recognized6. us, the slightly smaller (~ 0.08mm) boreholes in teeth of this fauna
(Fig.1J) could be considered to result from an ancestral representative of Osedax or a similar organism, which
distinctly possessed the ability to digest collagen and other proteins due to microbial symbionts40. Although this
remains rather hypothetical, the ability to digest collagen using collagenolytic enzyme activity in the organic-
poor, oligotrophic environment of the deep-sea, is of well-documented benet to some microbes such as, e.g.,
Oceanospirillales40.
It is unclear, which organism is responsible for the bioerosion patterns (boreholes) present in the fossil teeth,
such as a pycnodont sh tooth (Fig.1J) from this locality, as well as the exact process of demineralization of
the mineralized collagen that is necessary for digestion of the small amount of collagen. However, some recent
microorganisms, such as Streptococcus mutans, which is responsible for tooth decay in humans42, and Oceano-
spirillales or Epsilonproteobacteria (common symbionts of the bone-eating Osedax), possess the ability to digest
dierent types of collagen40. Despite the fact that these are only a few examples of microbial species, which are
able to obtain nourishment from mineralized collagen, they show that some specialists adapted to this food
source and also survived in extreme habitats.
Cyanobacteria also produce collagenases to tunnel into marine shells18, but their phototrophic lifestyle pre-
cludes them as candidate organisms for degradation of shark teeth in a deep-sea setting. Additionally, cyanobacte-
ria normally form branching tunnels rather than creating chain-like arranged globular structures. Consequently,
considering dierent modes of life and especially environmental limitations, we argue for a heterotroph lifestyle
of the herein described bacteria.
The fossil record of bioerosion. Bioerosion results in the loss of information and thus plays a crucial
role aecting the structure and consequently the potential for preservation of hard tissues like bones and teeth.
Nevertheless, microbial alteration of skeletal structures provides much evidence about taphonomic conditions.
Bioerosion caused by microbes is a phenomenon that is known since at least the nineteenth century when the
anatomist and histologist Rudolf Albert von Kölliker described meandering tunnels in e.g. a fossil gastropod
(Aporrhais pespelecani) and in a Cretaceous sh scale (Beryx ornatus)43. However, subsequent studies of the
Viennese Pathologist Carl Wedl, focusing on tunnel-like structures in human teeth, had received much more
publicity44. Kölliker attributed these tunnels to a fungal attack, while Wedl is not specic about the causal agents
and described them either as parasitic plants, microscopic parasites, or as fungi43,44. Further investigations by
Wedl additionally demonstrated the occurrence of these microstructures in a horse bone, in teeth of fossil car-
tilaginous shes (Hemipristis and Myliobates), and one bony sh (Pycnodus)44. However, the interpreted fungi
do not penetrate the enamel, being limited to the spongy bone and tooth cementum9,44,45. Following investiga-
tions on bone (e.g. Nothosaurus, Plesiosaurus, and Ichthyosaurus) and cartilage (e.g. Squatina, Galeocerdo, and
Carcharias) in various groups, Roux46 described the bone-penetrating fungus, Mycelites ossifragus. Subsequent
investigations led Bernhauser47 to conclude that the structures described by Wedl44 and Roux46 belong to an
inchnogenus rather than a distinct fungal species.
Hackett48 was the rst to identify bacteria as causal agents for dierent structures described in teeth and bones,
thus introducing a Wedl-type (apical expansion of meandering, bifurcating tunnels) and three non-Wedl-types
(linear longitudinal, lamellate, and budded foci) of structure. Of these four types of bioerosion structures, only
the Wedl-type is supposed to originate via fungal colonization and the other three types are interpreted to be
the product of bacterial activities.
Since the fundamental contributions of Wedl44, Roux46, and Hackett48, numerous studies have dealt with
histological microstructures of exhumed teeth and bones of ancient humans39,45,49–51, marine and terrestrial
vertebrates6,8 and marine invertebrates52,53. e focus of some studies was on causal agents46,48, others, conversely,
examined the inuence of dierent environmental settings and the subsequent impact of the alteration of the
tissue by specic micro-organisms45,54,55. However, the identity of the microorganisms responsible for bioerosion
patterns in teeth and bones usually remains ambiguous54.
Fossil bacteria are well known throughout the rock record and are the oldest known preserved traces of
life, the latter occurring as silicied remains56–58. In rare cases, fossilized insitu colonies are documented, e.g.,
in various Eocene vertebrate fossils with so tissue preservation from the Messel pit16, in an Early Cretaceous
pterosaur head crest from Brazil59, and in the dentine of a historic human tooth45. Another example of so tissue
preservation replicated by microbial biolms is the conservation of muscle bres in a Jurassic horseshoe crab60.
However, insitu bacteria invading tooth enameloid, one of the hardest and most highly mineralized biogenic
tissues developed by an animal and lacking signicant cavities or lacunae in contrast to dentine, have not been
reported up to now.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
Conclusion
We identied fossilized microorganisms inhabiting the hypermineralized outermost layer of enameloid of teeth in
the extinct, Early Cretaceous shark, Cretacladoides noricum. e 0.5–1µm-sized organisms, associated in linear,
chain-like colonies occur in the enameloid of the shark teeth between the parallel bundles of the uorapatite
crystallites.
A delicate lmy substance coating and linking the fossilized bacteriomorph bodies was replaced by a Fe–Mg
rich phyllosilicate, most probably a clay mineral close to smectite (montmorillonite) and saponite. However,
limitations of the analytical conditions preclude denitive identication of the mineral phase replacing the
microfossils.
e microorganisms were likely fossilized during decay and diagenesis of the shark teeth since microbial
invasion of bones and teeth (reservoirs of proteins) is a known behaviour of heterotrophs in the oligotrophic
deep-sea environment. is observation is particularly signicant because it is the rst time that fossilized
microorganisms have been observed in the highly mineralized enameloid of shark teeth. Moreover, it shows
that microbes obviously not only colonize less mineralized skeletal structures such as bones or dentine, but also
target the scarce organic matter (collagen and other proteins) in highly mineralized tissues such as enameloid
if the boundary is permeable or the surface is damaged. Even though the organic content in the enameloid is
rather low, it seemingly nevertheless provides an additional, high-energy source of nutrient matter in an other-
wise nutrient-poor environment.
Materials and methods
Material. e bacteria-bearing teeth forming the focus of this study belong to the extinct, Early Cretaceous
shark species, Cretacladoides noricum11 and were found in association with the published Early Cretaceous deep-
water chondrichthyan assemblage reported in Fuchs etal.61 and Feichtinger etal.11,62.
Methods. e teeth were extracted from the limestone matrix using 12% acetic acid (W. Neuber`s Enkel,
Vienna). e residual sediment was screen washed using dierent mesh sizes (500, 250, 125, and 63µm) and
dried by 60°C for 24h. e extraction process was made on two, separate samples in order to exclude contami-
nation. e teeth were separated from the residual sediment using a ne preparation needle and were mounted
in resin (Körapox 439, Kömmerling) in longitudinal direction, subsequently wet-ground with siliciumcarbid
Figure3. (A) locality map of the Klausrieglerbach 1 section (KB1-A) in the Northern Calcareous Alps of
Upper Austria with the indicated fossil locality (white star). (B) KB1-A outcrop with the older red Steinmühl
Formation (le) and the grey Schrambach Formation (right). (C) lithologic and stratigraphic column of the
KB1-A section with indicated shark teeth layer (black star). (Da) and (Db) shark teeth on naturally dissolved
rock surface. (Ea) and (Eb) thin sections of the shark teeth bearing bed. (Ea) bioclastic wackestone, mud
supported, with crinoid fragments, ammonites, ostracods, bivalves, and foraminifera. (Eb) bioclastic wackestone
to packstone, partly mud or grain supported, with crinoid fragments, ammonites, ostracods, bivalves, and
foraminifera (note two fragments of shark scale or teeth in le lower area).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
1000 (~ 4µm grain size) and the resulting surface (as indicated in Fig.1B) polished with Micropolish II Alumina
0.3µm Powder (Buehler, U.S.A.), and nally etched with 10% HCl (W. Neuber`s Enkel, Vienna) for ve seconds.
In a next step, the samples were rinsed for 2min in distilled H2O, cleaned with an ultrasonic bath for 5min, and
dried by 60 degrees for 2h. Tooth NHMW 2017/0055/0028 was then examined with a FEI Quanta 3D FEG at the
Department of Lithospheric Research at the University of Vienna, without any coating with an excitation energy
of the electron beam of 5kV. Quantitative energy dispersive spectrometry (EDS) analysis of the enameloid bun-
dle and aggregates of tooth NHMW 2017/0055/0028 were obtained by a JEOL “Hyperprobe” JXA 8530-F eld-
emission electron microprobe (FE-EPMA) in combination with an online JEOL quantitative ZAF-correction
program at the Central Research Laboratories of the Natural History Museum Vienna (NHMW). For the EDS
analyses, the sample was coated with a 10nm carbon lm. An accelerating voltage of 15 and 5keV, a beam cur-
rent of 5 nA, and fully focused electron beam (with an estimated beam diameter of ~ 70–80nm) were used. e
Count Rate was 1055.00 CPS. e comparative EDS studies of enameloid and coccoids were performed with a
FEI Inspect-S scanning electron microscope with an EDAX Apollo XV SDD EDS detector at 15, 10 and 5keV
acceleration voltage. Spectra were acquired for 30–90s to obtain a good signal to noise ratio, and intensities
were corrected with the ZAF algorithm. e teeth (NHMW 2020/0042/0001 and NHMW 2017/0055/0028)
are housed in the Geological—Palaeontological Department of the Natural History Museum Vienna, Austria
(NHMW). e brightness and contrast of the images were adjusted using Adobe Photoshop Elements 8.0. Ink.
Geological setting. e teeth described here derived from the KB1-A section that consists of bioclastic
wacke- to packstones of the so-called Steinmühl Formation occurring in the northern tectonic units of the
Northern Calcareous Alps in Upper Austria (Fig.3A–C). e exact position of the KB1-A section was deter-
mined by global positioning system (E 14°21′10″, N 47°54′32″) and is dated as upper Berriasian to lower
Valanginian63. e teeth-yielding rock comprises abundant remains of crinoids, aptychi, bivalves, foraminifera,
ostracods, radiolaria, and signicant calpionellids (Fig.3D,E). Ammonites, belemnites and brachiopods domi-
nate the macrofossil content, with the extraordinarily frequent teeth of the extinct shark genera Cretacladoides,
Natarapax, Altusmirus, Fornicatus, and Similiteroscyllium comprising ve percent of the rock volume. Addition-
ally deeper water bivalves and pelagic foraminifera (planktonic favusellids) hint to open marine conditions and
deeper depositional environments for the condensed shark tooth layer. Condensation took place in deep marine
areas by deep-water currents and winnowing of sediment, leading to condensation and enrichment of bioclastic
material. e assumed deep-water environment is also mirrored by the presence of the microfossil group of
calpionellids typical for pelagic to hemipelagic sedimentation. e facies and fossil assemblage from macro- and
microfossils observed in thin sections is also characteristic for deep-water pelagic deposits and basinal settings
from 200 to 1000m in the Tethyan Lower Cretaceous63,64.
Received: 29 April 2020; Accepted: 26 October 2020
References
1. Waldron, T. e relative survival of the human skeleton: Implications for palaeopathology. In Death, Decay and Reconstruction:
Approaches to Archaeology and Forensic Science (eds Boddington, A. et al.) 55–64 (Manchester University Press, Manchester, 1987).
2. Carpenter, K. Experimental investigation of the role of bacteria in bone fossilization. N. Jb. Geol. Paläont. Mh. 2, 83–95 (2005).
3. Wuttke, M. “Weichteil-Erhaltung” durch lithizierte Mikroorganismen bei mitteleozänen Vertebraten aus den Ölschiefern der
“Grube Messel” bei Darmstadt. Lethaea 64, 509–527 (1983).
4. Gaines, R. R., Kennedy, M. J. & Droser, M. L. A new hypothesis for organic preservation of Burgess Shale taxa in the middle
Cambrian Wheeler Formation, House Range, Utah. Palaeogeogr. Palaeoclimatol. Palaeoecol. 22, 193–205 (2005).
5. McNamara, M. E. et al. High-delity organic preservation of bone marrow in ca. 10 Ma amphibians. Geology 34, 641–644 (2006).
6. Kiel, S., Goedert, J. L., Kahla, W.-A. & Rouse, G. W. Fossil traces of the bone-eating worm Osedax in early Oligocene whale bones.
PNAS 107, 8656–8659 (2010).
7. Deming, J., Reysenbach, A. L., Macko, S. A. & Smith, C. R. e microbial diversity at a whalefall on the seaoor: Bone-colonizing
mats and animal-associated symbionts. Microsc. Res. Techniq. 37, 162–170 (1997).
8. Shapiro, R. S. & Spangler, E. Bacterial fossil record in whale-fa lls: petrographic evidence of microbial sulfate reduction. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 274, 196–203 (2009).
9. Underwood, C., Mitchell, S. F. & Veltkamp, C. J. Microborings in mid-Cretaceous sh teeth. Proc. Yorkshire Geol. Soc. 52, 269–274
(1999).
10. Maisch, H. M., Becker, M. A. & Chamberlain, J. A. Jr. Macroborings in Otodus megalodon and Otodus chubutensis shark teeth from
the submerged shelf of Onslow Bay, North Carolina, USA: implications for processes of lag deposit formation. Ichnos 27, 122–141
(2019).
11. Feichtinger, I., Engelbrecht, A., Lukeneder, A. & Kriwet, J. New chondrichthyans characterised by cladodont-like tooth morpholo-
gies from the Early Cretaceous of Austria, with remarks on the microstructural diversity of enameloid. Hist. Biol. 32, 823–836
(2020).
12. Enax, J., Janus, A. M., Raabe, D., Epple, M. & Fabritius, H.-O. Ultrastructural organization and micromechanical properties of
shark tooth enameloid. Acta Biomater. 10, 3959–3968 (2014).
13. Goldstein, J. et al. Scanning Electron Microscopy and X-Ray Microanalysis 3rd edn. (Springer, Berlin, 2003).
14. Strunz, H. & Nickel, E. H. Strunz Mineralogical Tables 9th edn. (Schweizerbart, Berlin, 2001).
15. Kahn, M. S. I., Oh, S.-W. & Kim, Y.-J. Power of scanning electron microscopy and energy dispersive X-ray analysis in rapid micro-
bial detection and identication at the single cell level. Sci. Rep. 10(2368), 1. https ://doi.org/10.1038/s4159 8-020-59448 -8 (2020).
16. Liebig, K., Westall, F. & Schmitz, M. A study of fossil microstructures from the Eocene Messel Formation using transmission
electron microscopy. N. Jb. Geol. Paläontol. Mh. 4, 218–231 (1996).
17. Pesquero, M.-D., Ascaso, C., Alcalá, L. & Fernándes-Jalvo, Y. A new taphonomic bioerosion in a Miocene lakeshore environment.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 295, 192–198 (2010).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
18. Kaehler, S. Incidence and distribution of phototrophic shell-degrading endoliths of the brown mussel Perna perna. Mar. Biol. 135,
505–514 (1999).
19. Turner-Walker, G. e chemical and microbial degradation of bones and teeth. In Advances in Human Palaeopatholog y (eds Pinhasi,
R. & Mays, S.) 3–29 (Wiley, New York, 2008).
20. Turner-Walker, G. Light at the end of the tunnels? e origins of microbial bioerosion in mineralised collagen. Palaeogeogr. Pal-
aeoclimatol. Palaeoecol. 529, 24–38 (2019).
21. Westall, F. et al. Polymeric substances and biolms as biomarkers in terrestrial materials: Implications for extraterrestrial samples.
J. Geophys. Res. Planets. 105(E10), 24511–24527 (2000).
22. Costa, O. Y. A., Raaijmakers, J. M. & Kuramae, E. E. Microbial extracellular polymeric substances: Ecological function and impact
on soil aggregation. Front. Microbiol. https ://doi.org/10.3389/fmicb .2018.01636 (2018).
23. Robin, N. et al. Diagenesis of bacterial colonies. Minerals 5(3), 488–506 (2015).
24. Dupraz, C. et al. Processes of carbonate precipitation in modern microbial mats. Earth Sci. Rev. 96, 141–162 (2009).
25. Ueshima, M. & Tazaki, K. Possible role of microbial polysaccharides in Nontronite formation. Clays Clay Miner. 49, 292–299
(2001).
26. Dong, H. Clay-microbe interaction and implication for environmental mitigation. Elements 8, 113–118 (2012).
27. Konhauser, K. O. Introduction to Geomicrobiology 1–425 (Blackwell Publishing, Oxford, 2006).
28. Lübke, A. et al. Dental lessons from past to present: ultrastructure and composition of teeth from plesiosaurs, dinosaurs, extinct
and recent sharks. RSC Adv. 5, 61612–61622 (2015).
29. Sasagawa, I. Mineralization patterns in Elasmobranch Fish. Microsc. Res. Technol. 59, 396–407 (2002).
30. Kiso, T. M. Organic components in enameloid of extant and fossil shark teeth. Trans. Proc. Palaeont. Soc. Jpn. 179, 169–174 (1995).
31. Duarte, A. S., Correia, A. & Esteves, A. C. Bacterial collagenases—a review. Crit. Rev. Microbiol. 42, 106–126 (2014).
32. Nielsen-Marsh, C. M. et al. e chemical degradation of bone. In Human Osteology in Archaeology and Forensic Science (eds Cox,
M. & Mays, S.) 439–454 (Cambridge University Press, London, 2000).
33. Gernaey, A. M., Waite, E. R., Collins, M. J., Craig, O. E. & Sokol, R. J. Survival and interpretation of archaeological proteins. In
Handbook of Archaeological Science (eds Brothwell, D. R. & Pollard, A. M.) 323–329 (Wiley, New York, 2001).
34. Colby, S. M. & Russell, R. R. B. Sugar metabolism by mutans streptococci. J. Appl. Microbiol. 83, 80–88 (1997).
35. Klont, B. & Ten Gate, J. M. Susceptibility of the collagenous matrix from bovine incisor roots to proteolysis aer invivo lesion
formation. Caries Res. 25, 46–51 (1991).
36. Ba lzer, A. et al. Invitro decomposition of bone collagen by soil bacteria: the implications for stable isotope analysis in Archaeom-
etry. Arachaeometry 39, 415–429 (1997).
37. Grupe, G. & Turban-Just, S. Amino acid composition of degraded matrix collagen from archaeological human bone. Anthropol.
Anz. 56, 213–226 (1998).
38. Tuross, N. Alterations in fossil collagen. Archaeometry 44, 427–434 (2002).
39. Jans, M. M. E., Nielsen-Marsh, C. M., Smith, C. I., Collins, M. J. & Kars, H. Characterisation of microbial attack on archaeological
bone. J. Archaeol. Sci. 31, 87–95 (2004).
40. Goredi, S. K., Johnson, S. B. & Vrijenhoek, C. Genetic diversity and potential function of microbial symbionts associated with
newly discovered species of Osedax Polychaete Worms. Appl. Environ. Microbiol. 73, 2314–2323 (2007).
41. Vrijenhoek, R. C., Johnson, S. B. & Rouse, G. W. A remarkable diversity of bone-eating worms (Osedax; Siboglinidae; Annelida).
BMC Biol. https ://doi.org/10.1186/1741-7007-7-74 (2009).
42. Loesche, W. J. Microbiology of dental decay and periodontal disease. In Medical Microbiology 4th edn (ed. Baron, S.) 1–30 (Uni-
versity of Texas Medical Branch, Galveston, 1996).
43. Kölliker, A. Ueber das ausgebreitete Vorkommen von panzlichen Parasiten in den Hartgebilden niederer iere. Zeitschr. wiss.
Zool. 10, 215–232 (1859–1860).
44. Wedl, C. Über einen im Zahnbein und Knochen keimenden Pilz. Sitzungsber. Kaiserl. Akad. Wiss. 50, 171–193 (1864).
45. Bell, L. S., Boyd, A. & Jones, S. J. Diagenetic alteration to teeth in situ illustrated by backscattered electron imaging. Scanning 13,
173–183 (1991).
46. Roux, W. Über eine Knochen lebende Gruppe von Fadenpilzen (Mycelites ossifragus). Z. wiss. Zool. 45, 227–254 (1887).
47. Bernhauser, A. Über Mycelites ossifragus Roux Aureten und Formen im Tertiär des Wiener Beckens. Sitzungsber. Kaiserl. Akad.
Wiss Math.-Naturwiss. Cl. 162, 119–127 (1953).
48. Hackett, C. J. Microscopical focal destruction (tunnels) in exhumed human bones. Med. Sci. Law. 21, 243–265 (1981).
49. Trueman, C. N. & Martill, D. M. e long-term preservation of bone: the role of bioerosion. Archaeometry 44, 371–382 (2002).
50. Farlow, J. O. & Argast, A. Preservation of fossil bone from the Pipe Creek sinkhole (Late Neogene, Grant County, Indiana, USA).
J. Palaeontol. Soc. Korea 22, 51–75 (2006).
51. Bell, L. S. & Elkerton, A. Unique marine taphonomy in human skeletal material recovered from the Medieval warship Mary Rose.
Int. J. Osteoarchaeol. 18, 523–535 (2007).
52. Glaub, I., Vogel, K. & Gektidis, M. e role of modern and fossil cyanobacterial borings in bioerosion and bathymetry. Ichnos 8,
185–195 (2001).
53. Vogel, K. & Marincovich, L. Palaeobathymetric implications of microborings in Tertiary strata of Alaska, USA. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 206, 1–20 (2004).
54. Turner-Walker, G. Early bioerosion in skeletal tissues: persistence through deep time. N. Jb. Geol. Paläontol. Abh. 265, 165–183
(2012).
55. Turner-Walker, G. & Jans, M. M. E. Reconstructing taphonomic histories using histological analysis. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 266, 207–235 (2008).
56. Awramik, S. M., Schopf, J. W. & Walter, M. R. Filamentous fossil bacteria from the Archean of Western Australia. Precambrian
Res. 20, 357–374 (1983).
57. Westall, F. Silicied bacteria and associated biolm from the deep-sea sedimentary environment. Kaupia 4, 29–43 (1994).
58. Westall, F., Hickman-Lewis, K. & Cavalazzi, B. Biosignatures in deep time. In Biosignatures for Astrobiology (eds Cavalazzi, B. &
Westall, F.) 146–164 (Springer, New York, 2018).
59. Pinheiro, F. L., Horn, B. L. D., Schultz, C. L., De Andrade, F. A. F. G. & Sucerquia, P. A. Fossilized bacteria in a Cretaceous pterosaur
headcrest. Letheia 45, 495–499 (2012).
60. Briggs, D. E. G., Wade, R., Schultz, J. W. & Schweigert, G. Mineralization of so-part anatomy and invading microbes in the
horseshoe crab Mesolimulus from the Upper Jurassic Lagerstätte of Nusplingen, Germany. Proc. R. Soc. B 272, 627–632 (2005).
61. Fuchs, I., Engelbrecht, A., Lukeneder, A. & Kriwet, J. New Early Cretaceous sharks (Chondrichthyes, Elasmobranchii) from deep-
water deposits of Austria. Cret. Res. 84, 245–257 (2018).
62. Feichtinger, I., Lukeneder, A. & Guinot, G. A Lower Cretaceous chondrichthyan dermal denticle assemblage and its bearing on
placoid scale diversity and histology. Cret. Res. https ://doi.org/10.1016/j.cretr es.2020.10444 4 (2020).
63. Lukeneder, A. & Reháková, D. Lower Cretaceous section of the Ternberg Nappe (Northern Calcareous Alps, Upper Austria):
Facies–changes, biostratigraphy and paleoecology. Geol. Carpath. 55, 227–237 (2004).
64. Lukeneder, A. & Reháková, D. Chronostratigraphic signicance of an early Valanginian (Cretaceous) calpionellid association
(Hochkogel section, Upper Austria, Northern Calcareous Alps). Geol. Quart. 51, 27–38 (2007).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:20979 | 
www.nature.com/scientificreports/
Acknowledgements
Christian Baal and Gerlinde Habler (both University of Vienna) are thanked for SEM assistance and Goran Batic
(Department for Mineralogy and Petrography, NHM Vienna) is warmly thanked for preparing the samples. Neu-
meister Birgid (Ravensburg, Germany) is acknowledged for providing information of dierent morphologies of
bacteria. We additionally thank Lutz Nasdala (University of Vienna) for helpful discussions and two anonymous
reviewers, whose comments improved the manuscript. Open access funding provided by University of Vienna.
Author contributions
I.F. designed the study, prepared the samples and draed the manuscript. D.T. conducted the EDS analysis
analysis at the NHM Vienna, E.L. did the comparative EDS study at the University of Vienna. A.L provided
the material and reviewed dras of the manuscript. I.F and A.L produced the gures. F.W. participated in the
interpretation of the study, I.F and F.W. wrote the manuscript, and J.K reviewed dras of the manuscript. All
authors gave nal approval for publication.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary information is available for this paper at https ://doi.org/10.1038/s4159 8-020-77964 -5.
Correspondence and requests for materials should be addressed to I.F.orJ.K.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access is article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.
© e Author(s) 2020
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com