Molecular Adaptation of Microtubules and Microtubule Motors from Antarctic Fish
145
Thus, we have suggested that a decrease in the acidic residue content of the
solvent-exposed carboxy-terminal tails of the a- and j3-tubulin chains might
contribute to microtubule formation at cold temperatures by reducing
electrostatic repulsion between dimers as they approach one another [4,15].
However, the neural Ncn~1 polypeptide [11] and six a-tubulin isotypes from
N coriiceps (H.W. Detrich and S. K. Parker, unpublished results) do not,
relative to vertebrate tubulin isotypes, contain reduced numbers of glutamyl
and/or aspartyl residues in the primary sequences of their carboxyl termini.
Furthermore, the critical concentrations of brain tubulins from an Antarctic
fish and a mammal decrease similarly in response to carboxyl-group
neutralization largely, but not entirely, restricted to the tubulin C termini
[16]. However, carboxy-terminal charge reduction might still play an
adaptive role should posttranslational polyglutamylation [17] of the C
termini of Antarctic fish tubulins be reduced relative to mammalian tubulins
[11]. Indeed, preliminary analyses of the polyglutamyl content of the a and
~ isotypes of brain tubulins from Antarctic fish (A. Frankfurter and H. W.
Detrich, unpublished results) are consistent with this hypothesis.
Multiple Structural Contributions to Microtubule Cold Adaptation
The results reviewed in this section allow us to draw several interesting, if
not yet definitive, conclusions regarding the structural bases of the cold
adaptation of microtubule assembly in Antarctic fish. First, the hypothesis of
hydrophobic remodeling of the interdimer interfaces [3,9] remains
provisional. Although most of the unique residue replacements observed in
Ncn~1 tubulin (Table 2) are found in the amino-terminal domain that
participates in longitudinal interdimer contacts, no clear physicochemical
pattern emerges from consideration of the unique residue substitutions and
insertions of this single brain isotype. A slight tendency towards increased
polarity (e.g., Ser for Ala at position 18, Ser for Ala/Gly at 283, Tyr insertion
at 442) is offset by reciprocal changes (Gly for Ser at 126,278). Phe 2 °O, Ile 267 ,
and Ala 333 increase, maintain, and decrease, respectively, the hydrophobicity
of the Ncn~1 chain relative to other vertebrate ~ isotypes. Validation or
rejection of the hydrophobicity-enhancement hypothesis must await further
structural analyses of other a- and ~-tubulin isotypes from Antarctic fish.
Second, increased flexibility of domains involved in dimer-dimer contact
very likely contributes to the favorable polymerization energetics of
Antarctic fish tubulins. The tubulin dimer apparently undergoes a guanine
nucleotide-dependent conformational change upon incorporation into or loss
from a microtubule end [18-20]. The energy cost associated with this
conformational change would be expected to decrease concomitant with a
145
Thus, we have suggested that a decrease in the acidic residue content of the
solvent-exposed carboxy-terminal tails of the a- and j3-tubulin chains might
contribute to microtubule formation at cold temperatures by reducing
electrostatic repulsion between dimers as they approach one another [4,15].
However, the neural Ncn~1 polypeptide [11] and six a-tubulin isotypes from
N coriiceps (H.W. Detrich and S. K. Parker, unpublished results) do not,
relative to vertebrate tubulin isotypes, contain reduced numbers of glutamyl
and/or aspartyl residues in the primary sequences of their carboxyl termini.
Furthermore, the critical concentrations of brain tubulins from an Antarctic
fish and a mammal decrease similarly in response to carboxyl-group
neutralization largely, but not entirely, restricted to the tubulin C termini
[16]. However, carboxy-terminal charge reduction might still play an
adaptive role should posttranslational polyglutamylation [17] of the C
termini of Antarctic fish tubulins be reduced relative to mammalian tubulins
[11]. Indeed, preliminary analyses of the polyglutamyl content of the a and
~ isotypes of brain tubulins from Antarctic fish (A. Frankfurter and H. W.
Detrich, unpublished results) are consistent with this hypothesis.
Multiple Structural Contributions to Microtubule Cold Adaptation
The results reviewed in this section allow us to draw several interesting, if
not yet definitive, conclusions regarding the structural bases of the cold
adaptation of microtubule assembly in Antarctic fish. First, the hypothesis of
hydrophobic remodeling of the interdimer interfaces [3,9] remains
provisional. Although most of the unique residue replacements observed in
Ncn~1 tubulin (Table 2) are found in the amino-terminal domain that
participates in longitudinal interdimer contacts, no clear physicochemical
pattern emerges from consideration of the unique residue substitutions and
insertions of this single brain isotype. A slight tendency towards increased
polarity (e.g., Ser for Ala at position 18, Ser for Ala/Gly at 283, Tyr insertion
at 442) is offset by reciprocal changes (Gly for Ser at 126,278). Phe 2 °O, Ile 267 ,
and Ala 333 increase, maintain, and decrease, respectively, the hydrophobicity
of the Ncn~1 chain relative to other vertebrate ~ isotypes. Validation or
rejection of the hydrophobicity-enhancement hypothesis must await further
structural analyses of other a- and ~-tubulin isotypes from Antarctic fish.
Second, increased flexibility of domains involved in dimer-dimer contact
very likely contributes to the favorable polymerization energetics of
Antarctic fish tubulins. The tubulin dimer apparently undergoes a guanine
nucleotide-dependent conformational change upon incorporation into or loss
from a microtubule end [18-20]. The energy cost associated with this
conformational change would be expected to decrease concomitant with a