I think your question is interesting, though the use of "expression" is misleading in context of your rabbit example.
Examples of environmentally influenced gene-expression in humans?
That is, a phenotype which requires BOTH a certain genetic background and particular environmental influences in order to be expressed?
In the example you give, the color-changing phenotype is not due to changes in gene expression but rather a conformational change in a tyrosinase variant at low temperatures that leads to decoupling of an inhibitor from the enzyme. This is not to say that there aren't any changes in expression with changes in temperature -- only that the phenomenon you describe is ultimately a change in enzyme activity and not the induction of a specific allele. See Kidson and Fabian, J Exp Zool (1981):
It is proposed that at the normal body temperature of 37 degree C, tyrosinase from Himalayan skin is strongly bound to an inhibitor. At lower body temperatures, the affinity of the enzyme for the inhibitor decreases, thus allowing the synthesis of melanin to increase. This change in affinity of the enzyme for the inhibitor could be regulated by temperature-induced conformational changes in either the enzyme or the inhibitor or both.
Note that this study was conducted in "Himalayan" mice, not rabbits, though the authors do reference an earlier rabbit study
These results are in support of Danneel's visual observations ('41) that "ferment" activity of Himalayan rabbit skin is absent at temperatures above 25 degree C.
the full text of which can be found, in German, on Google Books.
Concerning a human example of allele-environment interactions, look no further! The same temperature-sensitive tyrosinase mutation has been reported in people with oculocutaneous albinism (OCA), albeit with a slightly different phenotype. From the abstract:
In this paper, we report a new type of OCA that results from a tyrosinase allele producing a temperature-sensitive enzyme. The proband had white hair in the warmer areas (scalp and axilla) and progressively darker hair in the cooler areas (extremities) of her body. Melanocyte and melanosome architecture were normal. Quantitative hairbulb tyrosinase (dopa oxidase) assay demonstrated a loss of activity above 35-37 degrees C. Plasma pheomelanin and urine eumelanin intermediates were reduced and correlated with hair melanin content. This is the first temperature-sensitive tyrosinase mutation to be reported in humans and is analogous to the Siamese mutation in the cat and the Himalayan mutation in the mouse.