Gas-phase techniques are promising for studying intrinsically disordered peptides and proteins, as they can be implemented in a conformer-selective way. One of the issues that prevent them from finding a broader use is the question of to which extent the structure of a biological molecule is preserved when transferring it from solution to the gas phase via electrospray. We focus on cis/trans isomerization of prolyl-peptide bonds, which is known to have high barrier on the potential energy surface. Upon electrospraying, solution-like structures can find themselves kinetically trapped behind these barriers, as we explicitly show for the case of bradykinin 1-5 fragment. Using nuclear magnetic resonance, we determine that it adopts mostly all-trans conformation in solution, while three distinct conformational families are observed in the ion mobility experiment in the gas phase. Cryogenic ion spectroscopy combined with first-principles simulations allows us to identify the major conformers and show that the solution structural preferences of the prolyl-peptide bonds are preserved in the kinetically trapped conformational family, while the lowest-energy gas-phase structures have one of the prolines in the cis conformation. We show how collisional cross-sections and infrared spectra are used to guide and verify the calculations, giving rise to a new type of symbiosis between theory and experiment. This is especially valuable for the kinetically trapped structures, as in this case the energy criterion cannot be decisive. As a next step, we characterize the full nonapeptide bradykinin in the +3 charge state. Using field-asymmetric ion mobility spectrometry combined with cryogenic ion spectroscopy, we demonstrate the presence of three major conformational families in the gas phase, one of which is kinetically trapped. This result is in agreement with the drift-tube ion mobility data published previously, and we propose a correspondence between the conformational families separated by field-asymmetric and drift-tube ion mobility spectrometry. We obtain conformer-specific infrared spectra of the major conformers and assign the vibrational bands using 15N isotopic labeling. Substituting carbon atoms in the phenyl ring with their 13C isotope allows us to separate two types of structures according to their fragmentation pattern upon electronic excitation. This method can serve as a complementary way to introduce conformer-selectivity when studying large molecules. Finally, we assess the result of substituting one or two prolines in the bradykinin sequence with alanine. This method has been used previously to assign the peaks in the drift-time distributions to certain conformations of the prolyl-peptide bonds. We show that the mutants do reproduce a part of the conformers of the original peptide, but also form additional structures due to the higher flexibility of the alanine backbone compared to proline. We conclude that the proline-to-non-proline substitutions are helpful to assign the structures, but have to be used in conjunction with spectroscopic techniques, which allow detailed comparison of the structures of the mutant and the native peptide. In general, this work is one of the first steps towards a database of atomic-resolution peptide structures in the gas phase and a better understanding of the capabilities and limits of spectroscopy and ion mobility in the gas phase applied to intrinsically disordered peptides.