When homologous cusps display non-homologous wear facets: An occlusal reorganization ensures functional continuity during dental evolution of Murinae (Rodentia, Mammalia)

archives of oral biology 56 (2011) 194–204

available at www.sciencedirect.com

journal homepage: http://www.elsevier.com/locate/aob

When homologous cusps display non-homologous wear facets: An occlusal reorganization ensures functional continuity during dental evolution of Murinae (Rodentia, Mammalia) Vincent Lazzari a,*, Paul Tafforeau b, Jacques Michaux c a

Steinmann-Institut, Pala¨ontologie, Universita¨t Bonn, Nussallee 8, 53115 Bonn, Germany European Synchrotron Radiation Facility, 6 rue Jules Horowitz, BP220, 38043 Grenoble Cedex, France c Institut des Sciences de l’Evolution, CNRS UMR 5554, Universite´ de Montpellier 2, 34095 Montpellier Cedex 5, France b

article info

abstract

Article history:

Objective: This study was designed to interpret the differences between the occlusal rela-

Accepted 13 September 2010

tionships in the murine rodents and those in their Miocene ‘‘cricetodont’’ ancestors. It aimed at understanding the functional transformations that led to the emergence of the

Keywords:

peculiar chewing motion of the Murinae, associating forwardly directed masticatory move-

Murine rodents

ments to cusp interlocking, a trait unique amongst mammals.

Evolution

Methods: Microwear analyses and simulations of occlusion achieved with size-increased 3D

Functional morphology

printings of teeth crafted from 3D data obtained by X-ray synchrotron microtomography at

Molar crown

the European synchrotron radiation facility allow us to carefully interpret the occlusal

X-ray microtomography

relationships in Muroidea.

3D printing

Results: A rotation of the direction of the chewing movements occurred from ‘‘Cricetodon-

Occlusion

tinae’’ to Murinae. This rotation emerged without any cusp removal contrary to previous interpretations, by the way of an occlusal reorganization involving a loss of contacts between some cusps, offset by a contact with other cusps. This new organization was already present in the early and middle Miocene genus Potwarmus. Conclusion: Molar tooth evolution in Murinae was characterized by the preservation and the reshaping of the primitive muroid cusps, the acquisition of supplementary cusps, and changes in the contacts between the opposite cusps. During evolution, changes of cusp patterns in upper and lower molar teeth are coordinated in order to retain a functional occlusion. Because of this functional constraint, one cusp was supposed to more likely occlude with the same opposite cusps during evolution, and therefore homologous cusps would always carry homologous attrition facets. Evolution of Murinae proves that functional continuity can also be preserved through changes in occlusal relationships independently from cusp removal. # 2010 Elsevier Ltd. All rights reserved.

* Corresponding author. Present address: Institut de Pale´oprimatologie et Pale´ontologie humaine: Evolution et Pale´oenvironnements – IPHEP, CNRS UMR 6046, Universite´ de Poitiers, 86022 Poitiers Cedex, France. Tel.: +33 (0)5 49 36 63 23; fax: +33 (0)5 49 45 40 17. E-mail address: [email protected] (V. Lazzari). 0003–9969/$ – see front matter # 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.archoralbio.2010.09.011

archives of oral biology 56 (2011) 194–204

1.

Introduction

Molar teeth provide direct evidence of their functioning in the form of attrition facets which are areas of the tooth surface that are worn as a result of contact with opposing teeth during mastication. Because each attrition facet on an upper tooth has a corresponding facet on the occluding lower tooth, facets show how the teeth fit together.1,2 The necessary functional continuity during evolution allows the hypothesis that changes in cusp patterns in upper and lower molars are coordinated in mammals in order to retain the interrelations between cusps during occlusion.3 Indeed it is generally considered that homologous cusps carry homologous facets, as one cusp should always occlude with the same opposite cusps.3,4 Facets in placental mammals are then expected to be homologous if cusps are homologous, molar patterns of all therian mammals deriving from the tribosphenic one, evolved from a single origin in the group Tribosphenida.5 Homologous dental facets can indeed be recognized in such different taxa as primates, ungulates and rodents.3,6 If the search for homology of wear facets amongst rodents becomes a meaningless question when crown flattening evolves in a lineage as in Arvicolinae and Gerbillinae,7 the question becomes much more difficult when teeth remain cuspidate and even develop new functional cusps, as demonstrated by the attempt of Butler3,4 for interpreting the evolution of Murinae (old world rats and mice). In a tentative effort to apply the facet nomenclature used for the mammalian upper and lower molars to the molars of murine rodents, this author was forced to recognize that one of the cusps of the upper molar of Murinae, previously interpreted as a metacone, does not function as a metacone but rather as a mesostyle.3,4 At the same time this author suggested a sequence of hypothetical stages explaining the transformation of the upper molars of a cricetid ancestor into those of a murine rodent.4 However, the earliest fossils of Murinae had not provided any morphology that fits with Butler’s hypothesis. Murine rodents illustrate one case of radiation that associates an evolutionary success (more than 500 species and 100 genera, Musser and Carleton8) with a highly derived dental pattern. Amongst the Muroid rodents, the Murinae are characterized by the murine dental plan, which largely differs from the ancestral cricetine dental plan by morphological9–13 and functional aspects.3,4,7 Cricetine upper and lower molars display two longitudinal cusp rows, with a longitudinal crest in a central position (see Fig. 1 in13 Ref. 13). Murinae are characterized by three longitudinal rows because of the presence of supplementary lingual cusps, whilst lower molars retain two rows of cusps as in cricetine rodents. In the upper and lower murine molars, crests connect rows of cusps into transversal lophs which have a characteristic V shape (also known as chevrons: three chevrons can be thus observed on murine first upper molars). There is rarely any central longitudinal crest (see Fig. 1 in13 Ref. 13). Schaub14 suggested that except for the lingual cusps, the other cusps of the murine molar were homologous to the cusps of the cricetine molar. Relying on studies of abrasion scratches on wear facets which indicate the direction of the masticatory movements,15 several later studies emphasized some differences in the functioning of the occlusion in Muroidea.3,4,7,16 Muroidea with cricetine

195

dental plan, display chewing movements with cusp interlocking and an important transversal component relatively similar to the primitive condition observed in rodents.3,13 Occlusion in these rodents is thus characterized by both a transversal and an orthal component. Nevertheless, some Muroidea with a dental pattern derived from the cricetine one display longitudinal chewing movements (postero-anterior movement or propalinal) that they acquired through the flattening of the crown surface during evolution.4,13,16 Muroidea with the murine dental plan illustrate a different case: they display propalinal chewing movements whilst they retain a complex cuspidate tooth crown implying intercuspation during occlusion.3,4,7,13,16 Occlusion in these rodents is thus characterized by both a longitudinal and an orthal component. The murine dental plan that characterizes the Murinae therefore constitutes a puzzling problem: during evolution the cuspidate crown was not only preserved but has also become more complex whilst the chewing movement changed from transversal to longitudinal.4,7,13 If a change in chewing direction can easily be understood once the crown topography is flattened,4,7 the same does not apply when molar teeth remain cuspidate. Butler4 stressed that apart from a reorganization of the cusp positions, the upper molar morphology of primitive Muroidea such as the ‘‘Cricetodontinae’’ was not compatible with a shift of the chewing direction from oblique towards propalinal one because functional continuity has to be maintained during evolution. According to his view, this would imply both continuous rotation of the trajectories of the lower molar cusps and conservation of the contacts between cusps of the occluding teeth. Butler4 proposed that a distal displacement of the starting points of the trajectories of the opposite cusps could have allowed such a shift and that only the reduction of the metacone cusp on upper molars would have made it possible (Fig. 1A). Butler4 was consequently driven to the conclusion that the so-called metacone cusp of the Murinae was in fact a style (mesostyle) emerging from the paracone (Fig. 1A), which replaced the metacone. Butler’s model focused mainly on functional considerations and largely ignores the fossil record. The transition from cricetine muroids to murine muroids is now documented by numerous extinct species of which the morphology and the phylogenetic relationships are pictured in Fig. 1B. The oldest know Murinae is the Middle Miocene Pakistani genus Antemus (Fig. 1B), which displays the murine dental plan except for the transversal chevrons which connect only two cusps instead of three.17–20 The first genus to display the complete murine dental plan is Progonomys, which emerges directly from Antemus in Pakistan.18,20,21 Immediately preceding Antemus in the Pakistani stratigraphy, Potwarmus (Fig. 1B) displays an intermediary dental plan between that of cricetine rodents and that of murine rodents: there is only one supplementary lingual cusp (the enterostyle) on the upper molars and longitudinal crests although discontinuous on both upper and lower molars.22–25 Potwarmus is considered as a stem Murinae, closely related to Antemus.19,21,26 According to molecular data, Murinae share a relatively close common ancestor with Gerbillinae in the family Muridae.27,28 Together with palaeontological data, this indicates that Potwarmus and the Murinae could share a close common ancestor with extinct Gerbillinae of the tribe Myocri-

196

archives of oral biology 56 (2011) 194–204

[()TD$FIG]

Fig. 1 – Butler’s4 hypothesis about the evolution of occlusion in Murinae (A) versus the fossil record (B). Left first upper molars are figured in A and B. The long dark grey arrow indicates the polarity of the transformation. Black arrows: lower molar cusps trajectory during occlusion according to Butler’s hypothesis. Light grey arrows indicate the preservation of metacone in extinct Murinae. Ms: mesostyle. Mt: Metacone.B is modified fromRef. 37.

cetodontini such as Myocricetodon (Fig. 1B).19,21,23,26,29 Those rodents displayed the cricetine dental plan but also an enterostyle of varying size on upper molars. Muridae are supposed to have evolved along with Cricetidae (sensu Musser and Carleton8) from the extinct paraphyletic subfamily ‘‘Cricetodontinae’’,19,23,30 which includes genera such as Megacricetodon and Democriceton, early and middle Miocene genera displaying no supplementary lingual cusp on the first upper molars (M1) (Fig. 1B). ‘‘Cricetodontinae’’ as a taxon has to be revised significantly, but it is beyond the scope of this study. In any case these genera constitute a good illustration of the dental morphology of the hypothetical common ancestor of Murinae and Gerbillinae. Previous wear facet analyses revealed that Democricetodon, Megacricetodon and Myocricetodon displayed oblique chewing movements.7 The direction of the chewing movements in Potwarmus is debated: Tong31 indicated longitudinal chewing movements from the observation of the microwear pattern, whilst Wessels32 indicated a ‘‘slightly oblique’’ chewing motion from the orientation of the wear facets in slightly worn specimens. Anyway, this genus appears to display chewing movements that are at least very close to those of the crown Murinae, whilst displaying an incomplete murine plan: it could thus constitute the key for the understanding of the emergence of their peculiar mastication. Palaeontological data related to the emergence of Murinae support neither the loss of the metacone nor the emergence of a mesostyle (Fig. 1B).

Butler’s interpretation is consequently not supported by the presently known fossil record and the important functional point that he raised remains to date unresolved. The present paper aims at revisiting the facet homologies of Muroidea proposed by Butler3,4 in order to provide an interpretation of the evolution of the murine molar pattern coherent with the functional continuity principle, the microwear pattern and the morphological stages provided by the fossil record.

2.

Material and methods

2.1.

Material

We intended to compare the occlusal relationships between cricetine muroids and murine muroids. We sampled lightly worn isolated first upper and lower molars of several extinct muroid taxa, Democricetodon, Cricetodon, Megacricetodon and Progonomys, from the palaeontological collections of the University Montpellier 2 and Potwarmus, from the Collection of Geology of the University Lyon I. The first three represent the putative ancestors of the Muridae, displaying a simple cricetine dental plan, whilst Progonomys is the most ancient muroid rodent displaying a complete murine dental plan. Potwarmus is a stem Murinae, displaying an intermediary morphology and a nearly longitudinal chewing motion. We also observed complete tooth rows from recent skulls of

197

archives of oral biology 56 (2011) 194–204

Table 1 – Material: for each taxon, subfamily, collection, geographical origin (with locality) and relative geological age are indicated. COUM: collection of Osteology of the University of Montpellier 2. CPUM: collection of Palaeontology of the University of Montpellier 2. CGUL: collection of geology of the University Lyon 1. Taxon Cricetodon albanensis Democricetodon franconicus Megacricetodon aunayi Mesocricetus auratus Mus musculus Rattus rattus Peromyscus yucatanicus Potwarmus thailandicus Progonomys cathalai Progonomys clauzoni

Subfamily

Collection

Locality

‘‘Cricetodontinae’’ ‘‘Cricetodontinae’’ ‘‘Cricetodontinae’’ Cricetinae Murinae Murinae Neotomyinae Stem Murinae Murinae Murinae

CPUM CPUM CPUM COUM COUM COUM COUM CGUL CPUM CGUL

La Grive M (France) (France) Blanquate`re-1 (France) France France France Mexico Thailand Montredon (France) Dionay (France)

Murinae (Mus, Rattus) and from taxa retaining the cricetine dental plan amongst another family of Muroidea, the Cricetidae (Peromyscus, Mesocricetus) from the osteological collections of University Montpellier 2. Analysed species are presented in Table 1.

2.2.

Microwear analysis

We crafted high resolution resin casts of fossil teeth of Murinae (Progonomys clauzoni, Progonomys cathalai, Potwarmus thailandicus) and ‘‘Cricetodontinae’’ (Cricetodon albanensis, Democricetodon franconicus, Megacricetodon tautavelensis and Megacricetodon aunayi). Teeth elements were first cleaned with alcohol and acetone in order to remove dirt and glue from the occlusal surface. Then casts of the teeth were made using polyvinylsiloxane (President Microsystem 6015, Coltene) and transparent epoxy resin (In EpoxR, ADAM Montparnasse) which was heated at 30 8C during 12 h following the protocol developed by Merceron et al.33 Dental facets on the casts were digitized using a stereomicroscope ZEISS1 SV11 M2B with the 115 objective and transmitted-light at the Centre Re´gional d’Imagerie Cellulaire of the Montpellier RIO Imaging facility.

2.3.

X-ray synchrotron microtomography and 3D printing

To optimize our observations of the intercuspal contact relationships, we digitized first molars of Democricetodon franconicus and Progonomys cathalai and tooth rows of Mus musculus and Peromyscus yucatanicus using X-ray synchrotron microtomography at the European Synchrotron Radiation Facility (ESRF, Grenoble, France).13,34 These experiments were carried out on the beamlines ID19 and BM5 with voxel sizes of 2.8 mm (isolated teeth) and 5.06 mm (tooth rows) using moderate propagation phase contrast and a monochromatic X-ray beam at energy of 25 keV. 3D processing of teeth and tooth rows was performed with the software VGStudiomax 1.2 (Volume Graphics, Heildelberg, Germany).13 Nevertheless, such 3D processing, even if it can give good approximations of occlusion pattern, does not provide real physical contacts between teeth. We thus used the scans of teeth of Progonomys cathalai and Democricetodon franconicus to print in 3D enlarged plastic models (Fig. 2A). 3D volumes were oriented, binarized and cut at the cervix level with VG-studiomax following the protocol described in Lazzari et al.13 The surface area was then extracted as STL files. 50 size-increased 3D printings in

Relative age Middle Miocene Early Miocene Early Miocene Extant Extant Extant Extant Early/middle Miocene Late Miocene Late Miocene

ABSplus plastic were then made from these files using a Dimension Elite Printer (Stratasy Inc, Eden Prairie, USA) with printing resolution of 170 mm. To simulate the contacts between the cusps of these opposite molars, some watercolour was applied on the observed wear facets of the printings of the upper molars. During the simulated occlusion (Fig. 2B) the paint left coloured marks on the corresponding facets of the lower molar (Fig. 2C). This process was repeated with watercolour on the printings of the lower molars. This new method allows a precise detection and localization of the contacts between cusps in micro-mammals.

3.

Results

3.1. Occlusal contacts and chewing movements in Muroidea with cricetine plan In tribosphenic and derived tribosphenic mammals the hypoconid makes contact with metacone, paracone and protocone (Fig. 3A), and the protocone makes contact with metaconid, entoconid and hypoconid because of the oblique movement from the labio-distal side to the linguo-proximal side of the lower molar during chewing.3 We first checked this assertion in cricetine muroids, which display first molars with a derived tribosphenic morphology, due to the adjunction of supplementary anterior cusps: the anterocones/ids. Wear facets were identified on casts of upper and lower first molars of Democricetodon franconicus, Megacricetodon tautavelensis and Peromyscus yucatanicus. Each facet on first lower molar (m1) fits one to one with one facet of M1 according to the occlusal pattern hypothesized above for tribosphenic mammals (Fig. 4), which was tested with 3D prints of the first molars of Democricetodon franconicus. Our observations of wear patterns and occlusions in cricetine muroids support previous observations.3 The observed facets can be distributed amongst three groups. (i) Facets formed by contact between the labial cusps of opposing teeth (light grey facets 1,2,3,4,5,20 in Fig. 4); (ii) facets formed between the lingual cusps of opposing teeth (light grey facets 1,9,10,11,12,90 in Fig. 4); (iii) Facets formed when the protoconid and hypoconid respectively meet the lingual side of the anterocone and the protocone (dark facets 6 and 7 in Fig. 4). Facet 8 is formed by the protoconid of m2 (working with the posterolophid of m1) and the hypocone of M1 (Fig. 4). Facets of

198

archives of oral biology 56 (2011) 194–204

[()TD$FIG]

Fig. 2 – Occlusion pattern of first molars in Progonomys cathalai (murine dental plan) revealed by 3D printings. (A) 3D printings of left M1 (on the left) and of left m1 (on the right) of Progonomys cathalai. The scale is in centimetres. (B) Occlusal contacts between labial cusp row of first lower molar and central cusp row of first upper molar simulated thanks to 3D printings. Dark arrow indicates the direction of chewing movement. (C) Occlusal contacts between upper and lower first molar enlightened by watercolour applied on 3D printings. Paint was applied on upper molar cusps, and then coloured the opposite cusps during the occlusion simulated in (B) Blue: M1 anterior cusp row. Green: M1 central cusp row. Orange: M1 posterior cusp row. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

the two first groups are produced during phase I or buccal phase defined for tribosphenic mammals.35,36 Facets 6–8 are formed during phase II. The occlusal relations in cricetine dental plan however display several differences with those found in other tribosphenic mammals, notably in primates. First paracone and metacone in rodents have lost their primitive shearing function and are low and blunt3: thus phase I is not a shearing phase in Muroidea. Moreover, contrary to primates both phases are aligned in the horizontal plan in Muroidea.3,16

3.2. Occlusal contacts and chewing movements in murine plan: reinterpretation Occlusion in Murinae is characterized by a longitudinal and an orthal component. These rodents were suggested to display

cusp occlusal relationships similar to cricetine muroids, with for example the hypoconid making contact with the metacone, the paracone and the protocone during occlusion.3 Having identified facets on upper and lower molars of tooth casts of upper and lower first molars of Progonomys clauzoni, Progonomys cathalai and Rattus rattus, we observed that if the hypoconid occludes with the protocone, the protoconid occludes with the lingual anterocone and therefore no contacts are possible for anteroconids since Muroidea do not display any premolar. However, the anteroconids display wear facets indicating that they do occlude with cusps of the upper molar. The occlusal relations of Murine rodents were therefore worked out by fitting 3D printings of teeth together using watercolour (Fig. 2B and C). Fitting printings in a way that anteroconids and anterocones occlude, we showed that the hypoconid does not occlude with the protocone but

archives of oral biology 56 (2011) 194–204

[()TD$FIG]

199

Fig. 3 – Comparison of occlusion of first molars between ‘‘Cricetodontinae’’ (A) and Murinae (B). Trajectories of hypoconid, protoconid and labial anteroconid along first upper molar gutters are presented, as well as their contact with first upper molar cusps, in both subfamilies.

instead with the hypocone (Fig. 2B and C). The occlusal relationships in Murinae can be summed up as follows: in Murinae the primitive cusps of the anterior cusp row of M1 occlude with the anterior cusp row of m1, whilst the neoformed lingual cusp of the anterior cusp row of M1 occludes with central cusp row of m1 (blue paint on metaconid in Fig. 2C); the primitive cusps of the central cusp row of M1 occlude with the central cusp row of m1 whilst the neo-formed lingual cusp of the central cusp row of M1 occludes with the posterior row of m1 (green paint on entoconid in Fig. 2C); the posterior cusp row of M1 occludes with the posterior cusp row of m1 (orange paint in Fig. 2C). Such observations were at the same time corroborated by observations of complete tooth rows from skulls of Rattus rattus37 and Mus musculus. Differences of occlusal relationships between ‘‘Cricetodontinae’’ and Murinae are summarized in Fig. 3. In Murinae, the hypoconid

occludes with the metacone and the hypocone (Fig. 3B), whilst it occludes with the metacone, the paracone and the protocone in ‘‘Cricetodontinae’’ (Fig. 3A). In Murinae, the protoconid occludes with the paracone and the protocone (Fig. 3B), whilst it occludes with the paracone, the labial anterocone and the lingual anterocone in ‘‘Cricetodontinae’’ (Fig. 3A). The labial cusps of m1 in Murinae do not slide in the same gutters of M1 as the labial cusps of m1 in ‘‘Cricetodontinae’’ (Fig. 3). M1 crown gutters in Murinae are longitudinal, whereas they are oblique in ‘‘Cricetodontinae’’.13 We emphasize here that occlusion considerably differs between ‘‘Cricetodontinae’’ and Murinae, both subfamilies displaying facets that are not present in the other. The wear facets of Murinae are presented in Fig. 5. Facets related to phase II observed in ‘‘Cricetodontinae’’ such as Democricetodon (dark grey areas in Fig. 4) have disappeared in

200

archives of oral biology 56 (2011) 194–204

[()TD$FIG]

Fig. 4 – Wear facets of first molars in Democricetodon franconicus (cricetine dental plan). (A) Epoxy resin cast of left first upper molar. (B) Epoxy resin cast of the right first lower molar. (C) Map of the wear facets of the first upper molar. (D) Map of the wear facets of the first lower molar. Light grey areas in (B) and (D) correspond to wear facets produced during phase I. Dark grey areas in (B) and (D) correspond to wear facets produced during phase II. Grey arrow indicates the direction of chewing movement.

Murinae, because these rodents do not display any contacts between hypoconid and protocone, nor between protoconid and anterocone. Light grey areas in Fig. 5 represent homologous facets between ‘‘Cricetodontinae’’ and Murinae, which have been preserved during evolution. Murinae exhibit new facets (dark grey areas in Fig. 5) that are not present in

[()TD$FIG]

‘‘Cricetodontinae’’, because of two factors: (i) new contacts between the hypoconid and the hypocone (facet 6 in Fig. 5); the protoconid and the protocone (facet 5 in Fig. 5); the labial anteroconid and the labial anterocone (facet 4 in Fig. 5); and (ii) new contacts due to the emergence of new cusps in both upper and lower molars (facets 10,11,12,13 in Fig. 5).

Fig. 5 – Wear facets of first molars in Progonomys clauzoni (murine dental plan). (A) Epoxy resin cast of the first lower molar. (B) Epoxy resin cast of first upper molar. (C) Map of the wear facets of the first upper molar. (D) Map of the wear facets of the first lower molar. Light grey areas in (C) and (D) correspond to wear facets present in cricetine dental plan. Dark grey areas in (C) and (D) correspond to new wear facets relative to murine dental plan. Grey arrow indicates the direction of chewing movement.

archives of oral biology 56 (2011) 194–204

[()TD$FIG]

201

Fig. 6 – Wear facets of first molars in Potwarmus thailandicus (intermediary dental plan). (A) Epoxy resin cast of the first lower molar. (B) Epoxy resin cast of first upper molar. (C) Map of the wear facets of the first upper molar. (D) Map of the wear facets of the first lower molar. Light grey areas in (C) and (D) correspond to wear facets present in cricetine dental plan. Dark grey areas in (C) and (D) correspond to new wear facets relative to murine dental plan. Grey arrow indicates the direction of chewing movement. The wear facets of Potwarmus thailandicus are homologous to the wear facets of Progonomys clauzoni and receive the same number as those in Fig. 5.

3.3. Occlusal contacts and chewing movements in the stem Murinae Potwarmus The occlusal pattern of Potwarmus is a very important issue as this genus is one of oldest attested stem Murinae.19,21,26,32 Different studies already indicated that this taxon had a slightly oblique32 or even nearly longitudinal31 direction of chewing, intermediary between the conditions observed in Murinae and Muroidea with cricetine plan. From these observations, one question can be addressed: was the wear pattern of Potwarmus more similar to Muroidea with cricetine plan or to Murinae? Wear facets were identified on casts of upper and lower first molars of Potwarmus thailandicus (Fig. 6). They appear to be slightly obliquely oriented in lower molars (Fig. 6D) as already observed by Wessels.32 In slightly worn specimens, the wear facets also appear slightly vertical, which indicates an important orthal component in mastication (Fig. 6D). The wear pattern of Potwarmus clearly appears to be more similar to Murinae than to Muroidea with cricetine plan: the facets related to phase II observed in the latter (dark grey areas in Fig. 4) are also absent in Potwarmus. Light grey areas in Fig. 6 represent homologous facets between Muroidea with cricetine plan and Potwarmus which have been preserved during evolution and which can be also observed in Murinae (Fig. 5). Like crown Murinae, Potwarmus exhibit new facets (dark grey areas in Figs. 5 and 6) that are not present in ‘‘Cricetodontinae’’ and Myocricetodontini. Most of them are due to new contacts between the hypoconid and the hypocone (facet 6 in Fig. 6); the protoconid and the protocone (facet 5 in Fig. 6); the labial anterolophid and the labial anterocone (facet 4 in Fig. 6). Only one facet represents the new contact due to

the emergence of a new cusp, presently the enterostyle, in upper molars (facets 11 in Fig. 6). This new facet is very small, due to the relatively small size of the enterostyle in Potwarmus compared to the larger enterostyle of Progonomys. Potwarmus thus displays an incomplete murine wear pattern.

4.

Discussion

4.1.

Mastication in Murinae

Since facets related to phase II observed in ‘‘Cricetodontinae’’ have disappeared in Murinae, all facets in this latter subfamily are thus produced during a single phase characterized by a longitudinal and an orthal component. Such a reduction to one masticatory phase was already observed during the emergence of the Arvicolinae and Gerbillinae and was associated to crown flattening and loss of intercuspation7: in these cases, the orthal component of the occlusion was lost and both phases are fused into one. The Murinae however are very peculiar because intercuspation has been preserved during the origination of a new cusp organization. New facets evidently result from the occurrence of new cusps on the lingual side of the upper molar, but also because cusps of the upper molars occlude differently with the cusps of the lower molars. As stated previously, phase I does not functionally differ a lot from phase II in cricetine Muroidea because paracone and metacone in rodents have lost their primitive shearing function and are low and blunt.3 Therefore, in Murinae the loss of facets related to phase II of ‘‘Cricetodontinae’’ might not have been a major functional disadvantage

202

archives of oral biology 56 (2011) 194–204

[()TD$FIG]

Fig. 7 – New interpretation on the evolution of occlusion in Murinae. Black arrows indicate the trajectory of hypoconid (1) and entoconid (2) along first upper molar gutters during occlusion. Black dots indicate the initiation points of the course of these cusps during occlusion.Modified from Ref. 37.

since phase I and II can be considered as fused into a single phase in these rodents.

4.2.

Transition from the cricetine plan to the murine plan

Wear facets enable the interpretation of the functional meaning of morphological evolution known amongst rodents, in this context Murine rodents, illustrating a trend towards a propalinal chewing movement. The present study demonstrated that the change is much deeper than previously noted by Butler.3,4 The increase in width of the upper molars consecutive to the emergence of supplementary lingual cusps, together with the change in chewing direction, forces us to consider a new position for the interlocking teeth at the start of the masticatory movement. Contrary to Butler’s interpretation4 the starting point of the trajectory of the hypoconid did not only take a more distal position, but a more linguo-distal position (Fig. 7). Such a condition allows both the preservation of metacone and the rotation of the chewing movement from an oblique condition to a propalinal one. This condition must have appeared before the emergence of the murine dental plan, in the stem Murinae Potwarmus: even if this genus did not display a strictly longitudinal chewing motion,32 it displayed a wear pattern nearly similar to crown Murinae (Fig. 6). Fig. 7 summarizes the morpho-functional transition from cricetine plan to murine plan. In Fig. 7, Democricetodon illustrates the primitive morphological and functional conditions in Muroid rodents, the cricetine dental plan being associated to a hypoconid initiating its oblique trajectory on the labial side of M1. The primitive Gerbillinae Myocricetodon displays a more derived morphological condition, with the emergence of a small enterostyle, but retains the primitive functional condition (Fig. 7). The emergence of a small enterostyle could have induced the less oblique direction of chewing observed in this

genus4,7 (Fig. 7). The stem Murinae Potwarmus displays both derived masticatory and morphological conditions, with a nearly propalinal chewing motion and a larger supplementary lingual cusp.22 This genus retains a well-developed metacone, and such a functional condition could only arise with a linguodistal displacement of the starting point of the trajectory of the hypoconid (Fig. 7). A more important orthal component might have also been involved especially in this taxon (shorter arrows in Fig. 7), facilitating the rotation of the chewing movements. A similar condition is also seen in Progonomys, which is the first Murinae displaying the complete morphology of the murine dental plan with two lingual supplementary cusps on the upper M1 and complete transverse chevrons (Fig. 7). Progonomys also displays a well-developed enterostyle which excludes other chewing movements than strictly longitudinal ones32 (Fig. 7). The present conclusion is also in complete agreement with previous analyses that have revealed a change in individual cusp shape associated to the emergence of the murine dental plan.7,13 In Muroidea with a cuspidate molar, the planes of symmetry of the main cusps are roughly parallel to the direction of chewing. Moreover, these changes in orientation and chewing direction occur at the same time, which is linked to a rotation of the gutters delimited by the cusps.7,13 The reorganization of the individual molar cusp shape together with the shift in the starting position of the opposite cusps and the emergence of supplementary lingual cusps on upper molars made the emergence of the masticatory condition of Murinae possible without losing the functional continuity.

5.

Conclusion

3D printings of molar teeth of micro-mammals proved here their utility in functional analyses, until software taking

archives of oral biology 56 (2011) 194–204

into account physical contacts becomes available. The murine dental pattern illustrates one of the many changes that affected the evolution of the basal tribosphenic molar plan, during which a change in the direction of mastication is made in a group that not only retains cuspidate molars but whose molars acquired supernumerary cusps. Our results reconcile paleontological data with functional interpretation based on wear facet analyses: functional continuity was maintained by the way of a reorganization of the occlusal cusp relationships, by the way of a cusp reshaping,13 and by the way of new cusp acquisition32 instead of the loss of a primitive cusp. New facets differentiated whilst some other facets already present in the ancestor disappeared or were maintained. The muroid rodent wear pattern demonstrates that homologous cusps in mammals can display non homologous wear facets. Functional continuity during evolution can be maintained without retaining the interrelations of the cusps during occlusion, through coordinated changes in cusp patterns and occlusal patterns in upper and lower molars: evolution of the occlusal pattern is consequently more complex than previously appreciated in placental mammals. The rearrangement observed in Murinae implies a shift in the position of the occluding teeth at the very beginning of the course of the chewing movement. It corresponds to a lingual shift of the lower molar in order that its longitudinal gutter fits with the median longitudinal cusp row of the upper molar. Further paleontological discoveries may show if this shift is explained by changes of relative position or orientation of the dental rows. This functional transformation also needs to be studied by the way of tooth topographic analyses that will allow the quantification of the associated morphological changes,7,13 especially in the case of the stem Murinae and Gerbillinae (the different species of the genera Potwarmus and Myocricetodon), to decipher the timing and the relative order of the involved modifications. Fossil data may reveal which character, between direction of chewing, cusp morphology and enterostyle increase was the first to change. The emergence of a longitudinal direction of chewing in the context of a cusp interlocking masticatory apparatus could have been promoted in relation with some functional advantages and adaptive constraints. A propalinal movement would allow a symmetrical and simultaneous functioning of both jaws during chewing.38 The loss of wear facets of phase II could also have occurred in the context of a modification of the diet as suggested in a convergent cuspy diatomyid rodent, Marymus dalanae,39 which has independently acquired three cusp rows on upper molars as well as longitudinal chewing motion. Anyway, Potwarmus, the first muroid genus displaying the wear pattern of the Murinae, had only 10 wear facets on first upper molars (Fig. 6), whereas Muroidea with cricetine plan had 14 facets (Fig. 4) and Murinae 13 (Fig. 5). Further studies will have to interpret whether this important loss of attrition in Potwarmus was related to a change of diet. Similar studies should also concern the homoplastic emergence of the murine dental plan in other muroid sub-families (Cricetomyinae, Dendromurinae and Deomyinae)27 to understand if they followed parallel or convergent evolutionary pathways during their evolution, and if the same adaptive constraints were involved.

203

Acknowledgements We thank J. Baruchel and ID 19 and BM5 staff of the European Synchrotron Radiation Facility of Grenoble for 3D data acquisition, 3D printings, and funding; Nicole Lautredou and Montpellier Rio Imaging staff for microwear imaging; Abel Prieur and Pierre Mein of the University Lyon I for loan of fossil material; Wighart von Koenigswald of Steinmann Institut of University Bonn for comments on this paper, L. Foley-Ducrocq for English revision; two anonymous reviewers for their astute and useful comments. VL is a research fellow of Alexander von Humboldt Foundation. This work was partially funded by financial assistance from the Agence Nationale de la Recherche ANR-09-BLAN-0238. Funding: Vincent Lazzari was a research fellow of the Alexander von Humboldt Foundation. This work was partially funded by financial assistance from the Agence Nationale de la Recherche ANR-09-BLAN-0238. 3D experiments and casts were funded by the European Synchrotron Radiation Facility (Grenoble, France). Competing interests: None declared. Ethical approval: None.

references

1. Butler PM. Molar wear facets of tertiary north American primates. In: Zingeser MR, editor. Craniofacial biology of primates, vol. 3, Symp 4th internat Congr Primatol1973:1–27. 2. Greaves WS. The inference of jaw motion from tooth wear facets. J Paleontol 1973;47:1000–1. 3. Butler PM. Functional aspects of the evolution of rodent molars. Palaeovertebrata; Me´m Jubil R Lavocat 1980:249–62. 4. Butler PM. Homology of cusps and crests, and their bearing on assessments of rodent phylogeny. In: Luckett WP, Hartenberger J-L, editors. Evolutionary relationships among rodents. NewYork: Plenum; 1985. p. 381–401. 5. McKenna MC. Toward a phylogenetic classification of the Mammalia. In: Luckett WP, Szalay FS, editors. Phylogeny of the primates. NewYork: Plenum; 1975. 6. Butler PM. Relationships between upper and lower molar patterns. In: Internat colloq on the evolution of mammals. Brussels, Belgium: Kon Vlaamse Acad Wetens Lett sch Kun Belgie; 1961. p. 117–26. 7. Lazzari V, Charles C, Tafforeau P, Vianey-Liaud M, Aguilar J-P, Jaeger J-J, et al. Mozaic convergence of rodent dentition. PLoS One 2008;3:1–13. e3607. 8. Musser GG, Carleton MD. Superfamily Muroidea. In: Wilson DE, Reeder DM, editors. African and Indo-European Muridae; evolutionary trends. Baltimore: John Hopkins University; 2005. p. 894–1531. ¨ ber das System der Nagethiere: Eine 9. Tullberg T. U Phylogenetische Studie. Nova Acta Regiae Societatis Scientarum Upsaliensis 1899;3:1–514. 10. Simpson GG. The principles of classification and a classification of mammals. Am Museum National Hist Bull 1945;85:1–350. 11. Stehlin HG, Schaub S. Die Trigonodontie der Simplicidentaten. Neger Schweizerische Pala¨ontologische Abhandlungena 1951;67:1–385. 12. Schaub S, Simplicidentata. In: Piveteau J, editor. Traite´ de Pale´ontologie. Paris, France: Masson; 1958. 13. Lazzari V, Tafforeau P, Aguilar J-P, Michaux J. Topographic maps applied to comparative molar morphology: the case of

204

14. 15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

25. 26.

archives of oral biology 56 (2011) 194–204

murine and cricetine dental plans (Rodentia, Muroidea). Paleobiology 2008;34:46–64. Schaub S. Tertia¨re Quarta¨re Murinae. Abhandlungen der Schweizerischen Pala¨ontologischen Gesellschaft 1938;61:1–38. Butler PM. The milk-molars of Perissodactyla with remarks on molar occlusion. Proc Zool Soc Lond 1952;121:777–817. Charles C, Jaeger J-J, Michaux J, Viriot L. Dental microwear in relation to changes in the direction of mastication during the evolution of Myodonta (Rodentia, Mammalia). Naturwissenschaften 2007;94:71–5. Jacobs LL. A new genus of murid from the Miocene of Pakistan and comments on the origin of the Muridae. Paleobios 1977;25:1–11. Jacobs LL. Fossils rodents (Rhizomyidae & Muridae) from Neogene Siwalik deposits, Pakistan. Museum Northern Arizona Press Bull Ser 1978;52:1–103. Flynn LJ, Jacobs LL, Lindsay EH. Problems in muroid phylogeny: relationship to other rodents and origin of major groups. In: Luckett WP, Hartenberger J-L, editors. Evolutionary relationships among rodents. NewYork: Plenum; 1985. p. 589–616. Jacobs LL, Flynn LJ, Downs WR. Neogene rodents of southern Asia, in Papers on fossil rodents, in honor of Albert Elmer Wood. Nat Hist Mus Los Angeles Co Sci Ser 1989;33:157–77. Jacobs LL, Flynn LJ. Of mice. . .again: the Siwalik rodent record murine distribution molecular clocks. In: Liberman DE, Smith RJ, Kelley J, editors. Interpreting the past: essays on human, primate, and Mammal evolution in honor of David Pilbeam. Boston: Brill Academic Publishers; 2005. p. 63–80. Wessels W, Bruijn Hd, Hussain T, Leinders JJM. Fossil rodents from the Chinji Formation, Banda Daud Shah, Kohat, Pakistan. Koninklijke Nederlandse Akademie van Wetenschappen Proc Ser B 1982;85:337–64. Jaeger J-J, Tong H, Buffetaut E, Ingavat R. The first fossils rodents from the Miocene of Northern Thailand and their bearing on the problem of the origin of the Muridae. Rev Pale´obiol 1985;4:1–7. Lindsay EH. Cricetid rodents from Siwalik deposits near Chinji village. Part I: Megacricetodontinae, Myocricetodontinae and Dendromurinae. Paleovertebrata 1988;18:95–154. Jacobs LL, Downs WR. The evolution of murine rodents in Asia. Nat Sci Mus Monogr 1994;8:149–56. Tong H, Jaeger J-J. Muroid rodents from the Middle Miocene locality of Fort Ternan (Kenya) and their contribution to the

27.

28.

29.

30.

31.

32.

33.

34.

35. 36. 37.

38. 39.

phylogeny of Muroids. Palaeontographica Abteilung A Pala¨ozoologie-Stratigraphie 1993;229:51–73. Jansa SA, Weksler M. Phylogeny of muroid rodents: relationships within and among major lineages as determined by IRBP gene sequences. Mol Phylogenet Evol 2004;31:256–76. Steppan SJ, Adkins RM, Anderson J. Phylogeny and divergence-date estimates of rapid radiations in muroid rodents based on multiple nuclear genes. Syst Biol 2004;53:533–53. Jaeger J-J, Tong H, Denys C. Age de la divergence Mus-rattus: comparaison des donne´es pale´ontologiques et mole´culaires. Comptes Rendus Acade´mie des Sciences Paris 1986;302:917–22. Mein P, Freudenthal M. Une nouvelle classification des Cricetidae (Mammalia, Rodentia) du Tertiaire de l’Europe. Scripta geologica 1971;2:1–37. Tong H. Origine et e´volution des Gerbillidae (Mammalia, Rodentia) en Afrique du Nord. Me´moires de la Socie´te´ Ge´ologique de France 1989;155:1–120. Wessels W. Miocene rodent evolution and migration Muroidea from Pakistan, Turkey and Northern Africa. Geologica Ultraiectina 2009;387:1–290. Merceron G, Blondel C, Bonis Ld. A new method of dental microwear analysis: application to extant primates and Ouranopithecus macedoniensis (late Miocene of Greece). PALAIOS 2005;20:551–61. Tafforeau P, Boistel R, Boller E, Bravin A, Brunet M, Chaimanee Y, et al. Applications of X-ray synchrotron microtomography for non-destructive 3D studies of paleontological specimens. Appl Phy A 2006;83:195–202. Mills JRE. Ideal dental occlusion in the primates. Dent Practit 1965;6:47–61. Kay RF, Hiiemae KM. Jaw movement and tooth use in recent and fossil primates. Am J Phys Anthropol 1974;40:227–56. Lazzari V, Origin and evolution of the murine dental pattern in Muroidea (Rodentia, Mammalia): contribution of palaeontology, functional morphology and development in the study of an evolutionary innovation. PhD Thesis, Montpellier; 2006. Offermans M, Vree Fd. Mastication in springhare: acineradiographic study. J Morphol 1990;205:353–67. Flynn LJ. Origin and evolution of the Diatomyidae with clues to paleoecology from the fossil record. Bull Carnegie Mus Nat Hist 2007;39:173–81.