The key is to construct a vector $\alpha$ as @Michael Hoppe did.:
If $R$ is on a sphere. Then think of the vector $\alpha$ as exactly the vector that points from the centre of sphere to the curve $R$. So $\alpha$ is different from the original curve $R$ by just by a translation, hence they have the same tangent, curvature and torsion.
We can always express $\alpha$ as a linear combination (as suggested by Shifrin) of:
$$\alpha(s) = \lambda(s)T(s) + \mu(s)N(s)+\nu(s)B(s)$$
for some functions $\lambda$, $\mu$, $\nu$
(Note that $\alpha$ itself is not necessarily arc length parametrised.)
Then since the $R$ is on the sphere, and that $\alpha$ is from the centre to the sphere, we require that $\alpha$ is tangent to its tangent vector (notice that if $\alpha$ does not point from the centre to the sphere, it will not be tangent to its tangent vector $T$), that is,
$$<\alpha, T>=0$$
By differentiating above, we will find the expression for $\alpha$ as shown in @Michael Hoppe's post:
$$\alpha=-\frac{1}{\kappa}N-\left(\frac{1}{\kappa}\right)'\frac{1}{\tau}B,$$
$\alpha$ must be of constant length. This implies hat the relationship in the problem holds.