Conformational fluctuations in the native state ensemble enhance the complexity in designing de novo protein sequences that may fold correctly into a desired target structure. In this work, the results of a self-consistent mean field theory are applied to a cubic lattice model of proteins and real nonhomologous proteins to assess the designability of folded, misfolded, and unfolded conformations. This theory, for the first time, accounts for the properties of misfolded sequences in terms of a generalized foldability criterion and characterizes the topography of the sequence energy landscape in terms of folded, misfolded, and unfolded ensemble of conformations. For a given foldability criterion, the folded, misfolded, and unfolded conformations may be distinctly classified by tuning the energy variance of the native state ensemble. This implies a promising route to de novo protein design and provides useful insights into understanding the impact of conformational similarity/diversity on the folding-misfolding-unfolding transition.