Nothing but the Tooth 1
2 Damaris Butters
3 School of Earth Sciences, University of Bristol
4 db13741@my.bristol.ac.uk
5
6 The origin of the tooth has a bearing on vertebrate evolution. But its origin is shrouded by
7 controversial end member hypotheses.
8
9 It is generally agreed upon that the foundations of vertebrate success hinged upon the development
10 of teeth and jaws. A deeper knowledge of this evolution increases our understanding of how tissues
11 such as the kidneys, heart and liver formed. The original dogma of tooth evolution recognised
12 similarities between the shape and structure of shark teeth and their placoid scales or ‘dermal
13 denticles (Agassiz, 1833-45). This supposed homology gave rise to the ‘outside-in’ hypothesis,
14 that teeth evolved from placoid scales. However, in the late 1990s in light of new fossil evidence,
15 an opposing hypothesis ‘inside-out’ was posited by (Smith and Coates, 1997). This suggested the
16 independent evolution of teeth and teeth-like structures inside and outside of the mouth. Neither
17 hypothesis has been completely scientifically disproven, or can be considered mutually exclusive,
18 as thus far studies attempting to support their side of the debate have done so by attempting to
19 impugn the opposite.
20
21 In order to begin to address the enigma, which is the origin of the tooth, it is important to consider
22 the criteria by which a tooth is defined. Living jawed vertebrates such as Chondrichthyans and
23 Osteichthyans are used as a proxy for for dental evolution with teeth comprising dentine encased
24 within an enameloid layer, with a bone of attachment (Donoghue and Ru¨cklin, 2016). Whereas
25 jawless vertebrates such as Lampreys and Hagfish have keratinous tooth-like structures which are
26 not considered homologous to teeth (Murdock et al., 2014; Kresja et al., 1990). It seems that a
27 tooth replacement mechanism is also an important criterion for defining true teeth. Even with this
28 in mind, the distinction between teeth and teeth-like structures is often ambiguous.
29
30 An odontode is a common developmental unit comprising dentine with an encasing enameloid
31 layer (Ørvig, 1977). From this definition, teeth and tooth-like structures may be considered odon32
todes. These features may appear as external dermal denticles (in Chondrichthyans), teeth and
33 tooth-like structures within the oro-pharyngeal cavity. However, the capacity for replacement is
34 not founded in dermal or pharyngeal denticles, but is within oro-pharyngeal teeth (Johanson and
35 Smith, 2003). But how did these odontodes evolve?
36 ‘Inside-Out’
37 The foundations of this hypothesis conclude that teeth, as well as tooth-like structures, appear
38 after jaw evolution and may only form from interior germ layer ‘endodermal’ tissue. These dental
39 features are considered to have evolved independently both inside and outside of the mouth. The
40 inside-out hypothesis developed by (Smith and Coates, 1997), proposes that in gnathostomes,
41 pharyngeal tooth evolution from pharynx to jaw shows support from the morphology and
42 molecular composition of teleost fishes. Such pharyngeal tooth evolution is also comparable to
43 structured, oro-pharyngeal odontogenesis within mice. This conclusion falls in line with their
44 model of dermal denticles being precursory to teeth and having a similar pattern of organisation
45 (Stock and Wise, 2006). Although, a study comparing gene expressions in mouse oral teeth and
46 zebrafish pharyngeal teeth indicates that molecular markers are specific for that species only
47 (Jackman et al., 2004) and can not be considered evidence for inside-out.
48
49 Conodonts are bizarre, early jawless vertebrate fish which possessed tooth-like structures called
50 ‘elements’, composed of calcium phosphate. Their astonishing similarity to odontodes brought
51 about the notion that they were tooth homologues. These were taken by the inside-out community
52 to represent the evolution of teeth long before the rise of a gnathostome dermal skeleton. There53
fore, it may be that odontodes first formed inside the mouth and then migrated to the external
54 portion of the creature. A study on various conodont elements of different ages by (Murdock et al.,
55 2013) contests these findings on the basis that conodont tooth-like structures were the product of
56 a self-contained trial and that they resemble nothing more than a feature of convergent evolution.
57 Phylogenetic support for this analysis arises due to the fact that enamel-bearing elements within
58 later euconodonts formed from through aggregation of hard tissues from earlier paraconodont
59 tooth-like structures, which were incomparable to odontodes. Therefore, the last common ancestor
60 of odontode-bearing vertebrates and conodonts did not have mineralised elements and euconodont
61 tooth-like structures evolved convergently of teeth and odontodes seen in gnathostomes. In
62 addition, these elements have been found not to be true teeth anyway, which somewhat voids the
63 argument.
64
65 Studies on extinct, jawed vertebrates by (Johanson and Smith, 2003) called ‘placoderms’, provides
66 the grounding for further inside-out evidence. These are described as having dentine-based
67 tooth-like structures with replacement capacity. However, (Ru¨cklin et al., 2012) have concluded
68 that the composition of these structures is actually bone, rather than dentine. On this basis, they
69 do not fulfill the compositional criteria for the definition of a tooth and are not evidence of
70 odontode-derived teeth. Another theory in support of inside-out lies with the jawless thelodont
71 Loganiella – now extinct. These fish were found to have compound scales (tooth-whorl structures),
72 formed by aggregation of internal scales. Their pharyngeal compound scales show alignment,
73 similar to the replacement teeth in sharks. From this, (Smith and Coates, 1997) proposed that
74 these were evidence for the evolution of teeth prior to jaws and were a phylogenetic precursor to
75 replacement teeth. Although a reasonable assumption, this replacement patterning is not found in
76 any common ancestor with gnathostomes and is the result of convergent evolution – as seen earlier
77 in conodonts. Which means that mechanisms of tooth replacement cannot have occurred before
78 the evolution of true teeth.
79
80 A final point on the inside-out theory concerns one of its foundation theories – that teeth may
81 only form from pharyngeal endodermal tissue. These vertebrate fish have 3 germ layers of tissue:
82 ectoderm, mesoderm and endoderm. (Soukup et al., 2008) find from germ layer tracking, that teeth
83 don’t have to develop from endoderm, but may also form from ectodermal or mixed epithelial germ
84 layers. Therefore, there becomes less of a distinction between the development of dermal denticles
85 and teeth, so it is incorrect to base the inside-out theory on this disproven idea.
86 ‘Outside-in’
87 Comparisons between shark teeth and dermal denticles in the 1800s developed the framework
88 for the outside-in hypothesis, which has recently come into favour again. This theory suggests
89 that teeth evolved when dermal denticles of the exterior germ layer ‘ectoderm’ migrated into the
90 oro-pharyngeal cavity and fostered a tooth-like function. The notion that dermal denticles have
91 an extensive, deep phylogeny supports the foundation of this hypothesis – that dermal denticles
92 evolved before teeth (Donoghue and Sansom, 2002). But what is the evidence for this? The lack
93 of an obvious transition fossil showing gradation between dermal denticles and teeth somewhat
94 limits the progress of this argument.
95
96 In response to this missing link, evidence was presented by (Blais et al., 2011), bringing specimens
97 of Canadian Ischnacanthus gracilis into the mix. They found that oral scales become more tooth98
like with proximity to the margins of the organism’s mouth and take on a tooth-whorl form with
99 a bone of attachment and pulp cavity. These then appear to grade into head scales exteriorly from
100 the mouth. Similar morphology between modified cheek scales and head scales implies that the
101 two have undergone the same constructive processes. If taken to be a true transitional fossil, these
102 features may truly represent an intermediate stage between dermal denticle and tooth. Despite the
103 plausability of these observations, they have come under scrutiny, for reasons that the dentition of
104 these acanthodians is actually disparate to their external dermal ornimentation and the lineage of
105 the fish has not been fully resolved.
106 In opposition to the inside-out notion that teeth emerged from assimilation of oro-pharyngeal den107
ticles independently in seperate lineages, a ‘modified outside-in’ has been suggested by Fraser et
108 al., 2010 that internal odontodes of extinct jawless gnathostomes were a recurring feature in nasal,
109 oral and pharyngeal areas. This is thought to represent migration of dermal denticles into these
110 orifices by means of an odontogenic continuum between external and internal epithelia. Likely re111
sulting from a combination of neural crest cells and mesenchyme, as opposed to epithelial tissues,
112 to initiate the process of tooth development. Support for this external-internal cell mixing arises
113 because pharyngeal teeth only occur in fish with gill slits.
114 ‘Inside and out’
115 Recently, an ‘inside and out’ theory proposed by (Fraser et al., 2010) has attempted to com116
bat the shoehorning of evidence into one camp or another. This provokes a mixed end-member
117 hypothesis which generates both endoderm-derived pharyngeal teeth and ectoderm-derived oral
118 teeth. Aforementioned germ layer mixing would be responsible for enabling this combination.
119 Further molecular-scale evidence for inside and out comes from experimental studies on dogfish
120 by (Debiais-Thibaud et al., 2011). They note that equivalent stages of odontogenesis lead to the
121 development of both oral and dermal denticles. Therefore, molecular as well as morphological sim122
ilarities between them indicate serial homology between odontodes which cannot discount either
123 one hypothesis.
124 Discussion and conclusion
125 And so, we arrive at a paradox. Outside-in appears to be favoured, but is established on compara126
tively little evidence. Whereas the inside-out hypothesis is grounded upon a plethora of evidence,
127 but this is in most cases unfounded. Comprehensive phylogenetic and comparative odontogenetic
128 studies have been made, so where do we go from here? An important future hope would be to
129 find a transitional fossil demonstrating the crossover between dermal denticles and oral teeth; in
130 light of no such finding so far, does it imply evidence of absence or absence of evidence? The
131 inside-out hypothesis has recently come under much scrutiny from the scientific community who
132 are in favour of outside-in and any furthering of support for the latter appears to derive solely from
133 ‘successful’ attempts to disprove the former. The invalidation of the foundation hypotheses for
134 inside-out, including derivation capability of teeth from ectoderm, endoderm and mixed epithelia
135 and attributing evidence of teeth before dermal denticles as features of covergent evolution, has
136 inflicted the most damage of the two hypotheses. But that is not to say that it can be entirely
137 discounted, in fact neither of the two need be truly mutually exclusive as shown by the suggestion
138 of an inside-and-out hypothesis.
139
140 It seems that comparatively little thought has been put into the requirements of the organism itself
141 – which are dependent on environmental factors such as food supply, feeding habits and predation.
142 For example, external denticles may serve for abrasion protection, defense and reducing hydro143
dynamic drag. Likewise, teeth could serve to reduce oro-pharyngeal abrasion and filter or grind
144 food. These various functions raise important questions. Did teeth develop from dermal denticles
145 due to evolutionary progression towards predation? Can a tooth be defined as such if there is no
146 jaw mechanism by which to use them? Would defining the purpose of odontodes in an individual
147 species give clues as to which hypothesis is most fitting? Is it possible to progress from this sci148
entific stalemate by shifting the focus of these hypotheses away from mutual exclusion towards a
149 combination of odontode function and odontogenesis?