equivalent amounts of materials were tested. We suggested that these two proteins,
and probably many more proteins of the prisms and nacre matrices, belong to the
class of low affinity–high capacity calcium-binding proteins (Maurer et al. 1996;
Marin and Luquet 2007). The binding of calcium ions would be performed via the
negatively charged side chains of aspartate and glutamate residues, and maybe via
some of their posttranslational modifications, many of which may exhibit anionic
properties: glycosylations, in particular those involving sulfated sugars, and phosphorylation of serine and threonine residues. Although the full sequences of
caspartin and of calprismin are not known, it is very likely that these proteins do
not exhibit canonical calcium-binding domains, such as EF-hands (Kretsinger
1976), which are characteristic of the class of low capacity–high affinity calciumbinding protein. For a shell protein, we believe that binding reversibly numerous
calcium ions is probably more compatible with a crystal synthesis process than
sequestering strongly few calcium ions. We do not exclude the possibility that
bound calcium ions also stabilize shell proteins in vivo and help them to maintain a
precise conformation.
The in vitro inhibition property is another point in case. In our hands, we always
observed that the acetic acid-soluble matrices of prisms and nacre of P. nobilis had
good inhibitory effect. However, we observed that the inhibitory effect of the
prisms matrix (and of caspartin and calprismin) was much higher than that of
nacre matrix, for similar tested concentrations (Marin 1992). This effect seems to
be correlated with the “acidity” of both matrices, the prisms matrix being much
more “acidic” than the nacre matrix. The higher acidity of proteins associated to
calcitic prisms is supported by two observations: first, we have shown in a previous
review paper (Marin et al. 2008) that molluscan shell proteins associated to calcite
(foliated or prismatic) and the ones associated to aragonite (mainly nacre) exhibit
different distributions, when plotted on a diagram showing their theoretical calculated pI vs. their molecular weight. The first ones have a bimodal distribution, either
very acidic, either very basic, while the second ones are characterized by weakly
acidic to weakly basic pIs. As the very basic proteins of the prisms may be
associated to the insoluble framework, the soluble prism matrix proteins are
necessarily extremely acidic. Second, anionic posttranslational modifications,
such as sulfated sugars, are present in high quantity in the prisms of P. nobilis but
are almost absent from the nacre matrix proteins. In the case of P. nobilis, these two
combined factors explain the difference of inhibitory effect of the prisms and the
nacre matrices.
At last, proteins of the shell matrix of P. nobilis interfere with the precipitation of
calcium carbonate in vitro. This effect is dose dependent, up to a certain threshold
above which no crystals are formed due to inhibitory effects (Marin et al. 2005).
Among the most common crystal figures, one can see the formation of polycrystalline aggregates of calcite and the formation of apparent “single” crystals, which
exhibit a layering and truncated corners. Although these effects can be observed
with several shell proteins in general, we notice that they are particularly drastic
when acidic proteins such as caspartin and calprismin are employed.
380
F. Marin et al.
and probably many more proteins of the prisms and nacre matrices, belong to the
class of low affinity–high capacity calcium-binding proteins (Maurer et al. 1996;
Marin and Luquet 2007). The binding of calcium ions would be performed via the
negatively charged side chains of aspartate and glutamate residues, and maybe via
some of their posttranslational modifications, many of which may exhibit anionic
properties: glycosylations, in particular those involving sulfated sugars, and phosphorylation of serine and threonine residues. Although the full sequences of
caspartin and of calprismin are not known, it is very likely that these proteins do
not exhibit canonical calcium-binding domains, such as EF-hands (Kretsinger
1976), which are characteristic of the class of low capacity–high affinity calciumbinding protein. For a shell protein, we believe that binding reversibly numerous
calcium ions is probably more compatible with a crystal synthesis process than
sequestering strongly few calcium ions. We do not exclude the possibility that
bound calcium ions also stabilize shell proteins in vivo and help them to maintain a
precise conformation.
The in vitro inhibition property is another point in case. In our hands, we always
observed that the acetic acid-soluble matrices of prisms and nacre of P. nobilis had
good inhibitory effect. However, we observed that the inhibitory effect of the
prisms matrix (and of caspartin and calprismin) was much higher than that of
nacre matrix, for similar tested concentrations (Marin 1992). This effect seems to
be correlated with the “acidity” of both matrices, the prisms matrix being much
more “acidic” than the nacre matrix. The higher acidity of proteins associated to
calcitic prisms is supported by two observations: first, we have shown in a previous
review paper (Marin et al. 2008) that molluscan shell proteins associated to calcite
(foliated or prismatic) and the ones associated to aragonite (mainly nacre) exhibit
different distributions, when plotted on a diagram showing their theoretical calculated pI vs. their molecular weight. The first ones have a bimodal distribution, either
very acidic, either very basic, while the second ones are characterized by weakly
acidic to weakly basic pIs. As the very basic proteins of the prisms may be
associated to the insoluble framework, the soluble prism matrix proteins are
necessarily extremely acidic. Second, anionic posttranslational modifications,
such as sulfated sugars, are present in high quantity in the prisms of P. nobilis but
are almost absent from the nacre matrix proteins. In the case of P. nobilis, these two
combined factors explain the difference of inhibitory effect of the prisms and the
nacre matrices.
At last, proteins of the shell matrix of P. nobilis interfere with the precipitation of
calcium carbonate in vitro. This effect is dose dependent, up to a certain threshold
above which no crystals are formed due to inhibitory effects (Marin et al. 2005).
Among the most common crystal figures, one can see the formation of polycrystalline aggregates of calcite and the formation of apparent “single” crystals, which
exhibit a layering and truncated corners. Although these effects can be observed
with several shell proteins in general, we notice that they are particularly drastic
when acidic proteins such as caspartin and calprismin are employed.
380
F. Marin et al.