7 Retinoic Acid Signaling and the Zebrafish Dentition …
187
It was also determined that global RA application was not necessary to induce
supernumerary teeth, as RA-coated beads inserted into the zebrafish pharyngeal
region during this same time window produced ectopic tooth formation surrounding
the bead (Seritrakul et al. 2012). Interestingly, there was a corresponding disruption
in nearby cartilage formation that correlated with supernumerary tooth formation.
Given that both the odontoblasts of teeth and chondrocytes of the viscerocranial
skeleton are derived from CNC cells, this correlation was particularly notable. Additionally, the pharyngeal region was abnormal in embryos exposed to exogenous RA
at this stage and the embryos also had severe pleiotropic effects, as might have been
expected given the numerous other roles RA is known to play in development. Thus,
the supernumerary tooth formation seen may have been the result of regional changes
and/or changes to cell fate specification.
To assess whether increasing RA signaling later in development would also have
an effect on zebrafish tooth formation, Gibert et al. (2015) exposed zebrafish embryos
after 24 hpf to exogenous RA and monitored tooth number in zebrafish larvae at
12 days post fertilization (dpf). In wild type zebrafish by this stage, five teeth are
present on each side of the midline, which is the maximum number of teeth formed
on the ventral part of the 5th ceratobranchial arch at any age in zebrafish. However,
after this relatively late RA exposure, a sixth tooth was formed in this region (Gibert
et al. 2015).
Then, more refined and focused, RA-coated bead experiments were carried out.
This was achieved by taking advantage of a loss-of-function mutation in one of the
enzymes that locally degrades retinoic acid, cyp26b1 (Spoorendonk et al. 2008).
Unlike RA, cyp26b1 does not diffuse outside of the cell where it is synthesized
and therefore, loss of CYP26b1 function in the cyp26b1
−/− mutants leads to localized excess of RA signaling. Consequently, there are few pleiotropic effects from
the mutation, and heterozygous individuals are able to grow relatively normally to
adulthood. Computed tomography of the dentition from adult heterozygous mutants
revealed the presence of a fully formed sixth tooth on the ventral part of the fifth ceratobranchial arch in more than 50% of individuals (Compare Figs. 7.3B and 7.4C),
confirming that altering RA signaling in a localized fashion is sufficient to change
zebrafish tooth number (Gibert et al. 2015). This result has particularly interesting
evolutionary implications, as it points to fine-tuning RA signaling as a possible mechanism by which the dentition of vertebrate species can be gradually changed over
time.
Models of RA Action in Tooth Development and Evolution
The specific effects that have been observed after RA inhibition and/or overexpression during zebrafish tooth development provide material for further consideration of RA’s control of tooth formation in general. Regarding the loss-of-function
experiments, it is likely that very early RA inhibition is causing regional problems
in the embryos, severe enough that essential cell types are missing and that teeth
187
It was also determined that global RA application was not necessary to induce
supernumerary teeth, as RA-coated beads inserted into the zebrafish pharyngeal
region during this same time window produced ectopic tooth formation surrounding
the bead (Seritrakul et al. 2012). Interestingly, there was a corresponding disruption
in nearby cartilage formation that correlated with supernumerary tooth formation.
Given that both the odontoblasts of teeth and chondrocytes of the viscerocranial
skeleton are derived from CNC cells, this correlation was particularly notable. Additionally, the pharyngeal region was abnormal in embryos exposed to exogenous RA
at this stage and the embryos also had severe pleiotropic effects, as might have been
expected given the numerous other roles RA is known to play in development. Thus,
the supernumerary tooth formation seen may have been the result of regional changes
and/or changes to cell fate specification.
To assess whether increasing RA signaling later in development would also have
an effect on zebrafish tooth formation, Gibert et al. (2015) exposed zebrafish embryos
after 24 hpf to exogenous RA and monitored tooth number in zebrafish larvae at
12 days post fertilization (dpf). In wild type zebrafish by this stage, five teeth are
present on each side of the midline, which is the maximum number of teeth formed
on the ventral part of the 5th ceratobranchial arch at any age in zebrafish. However,
after this relatively late RA exposure, a sixth tooth was formed in this region (Gibert
et al. 2015).
Then, more refined and focused, RA-coated bead experiments were carried out.
This was achieved by taking advantage of a loss-of-function mutation in one of the
enzymes that locally degrades retinoic acid, cyp26b1 (Spoorendonk et al. 2008).
Unlike RA, cyp26b1 does not diffuse outside of the cell where it is synthesized
and therefore, loss of CYP26b1 function in the cyp26b1
−/− mutants leads to localized excess of RA signaling. Consequently, there are few pleiotropic effects from
the mutation, and heterozygous individuals are able to grow relatively normally to
adulthood. Computed tomography of the dentition from adult heterozygous mutants
revealed the presence of a fully formed sixth tooth on the ventral part of the fifth ceratobranchial arch in more than 50% of individuals (Compare Figs. 7.3B and 7.4C),
confirming that altering RA signaling in a localized fashion is sufficient to change
zebrafish tooth number (Gibert et al. 2015). This result has particularly interesting
evolutionary implications, as it points to fine-tuning RA signaling as a possible mechanism by which the dentition of vertebrate species can be gradually changed over
time.
Models of RA Action in Tooth Development and Evolution
The specific effects that have been observed after RA inhibition and/or overexpression during zebrafish tooth development provide material for further consideration of RA’s control of tooth formation in general. Regarding the loss-of-function
experiments, it is likely that very early RA inhibition is causing regional problems
in the embryos, severe enough that essential cell types are missing and that teeth