Parkinson's disease (PD) is associated with widespread structural and functional brain abnormalities, yet whether morphology-constrained functional organization is disrupted in a frequency-specific manner remains unclear. To address this gap, we derived frequency-resolved features of morphology-constrained functional organization in PD by integrating morphological networks from T1-weighted magnetic resonance imaging with resting-state functional signals in the graph spectral domain. Low- and high-frequency eigenmodes and the relative decoupling index (RDI) were used to identify abnormal patterns of morphology-function coupling in PD. We further evaluated the cognitive relevance, biological significance, and disease-related discriminative information of these features using meta-analytic decoding, behavioral prediction, gene expression analysis, and exploratory patient-control classification. Our results revealed that low-frequency eigenmode abnormalities in PD were predominantly localized to the default mode network, dorsal and ventral attention networks, limbic network, and subcortical network. In contrast, high-frequency eigenmode and RDI abnormalities were primarily observed in the limbic network and subcortical network. In addition, low-frequency alterations were linked to social cognition and anxiety functions, while high-frequency changes were associated with motor and memory functions. Spatial patterns of abnormal regions showed preliminary spatial correspondence with gene expression profiles enriched for ABC transporter activity and fatty acid metabolism. Classification analyses indicated that frequency-resolved features contained disease-related discriminative information, with selected features preferentially involving motor, attentional and limbic systems. Together, these findings indicate that morphology-function coupling is disrupted in PD in a frequency-resolved manner and suggest that frequency-resolved morphology-constrained functional features may provide candidate MRI-derived features for further disease characterization.