The volatility of a codon is defined as the probability that a random point mutation in the codon generates a nonsynonymous change. To study the impact of genomic volatility under constrained evolution we recoded the structural region (117 codons in the P1 region) of the genome of the human enterovirus Coxsackievirus B3 (CVB3). Thus, we genetically engineered two mutants with different genomic volatility according to a mathematical framework. Hence, one mutant bears the most volatile synonymous codons of Serine and Leucine (MoreV). On the contrary, the second mutant was designed to use only the lowest volatile codons of these two amino acids (LessV). In this study, we performed plaque-to-plaque passages in tissue culture of both mutants and the Wild Type virus (WT). This approach was carried out during ten passages using three lineages for each virus. During the evolution, we saw that infective particles recovered from plaques were decreasing in titers, as expected. Moreover, we evaluated the impact of the mutations fixed in passages three, six, and ten by measuring the relative fitness. We then sequenced all these passages using Nanopore and we related every genotype to the relative fitness and a plaque size phenotype. Our results suggest that the WT and LessV viral populations lost their fitness due to the action of the Muller's ratchet but the MoreV population had the most chaotic behavior (in virus titer and plaque phenotype) after the subsequent bottlenecks imposed.