Information processing and communication in neuronal circuits is enabled by networks of dynamic inter-areal coupling of neuronal oscillations. These networks exhibit frequency-specific spatial and topological patterns and rely on hubs that are crucial for information routing. Oscillations are generated by fast synaptic neurotransmission among pyramidal cells and interneurons, and are subject to neuromodulation on slower timescales in a frequency- and region-specific manner. Efferent connections, receptor densities, and neurotransmitter reuptake regulation of neuromodulatory systems are highly heterogeneous across the cortical mantle and influence oscillations locally. Yet, the fundamental question of how this variability shapes and constrains inter-areal connectivity of neuronal oscillations remains unanswered. Here, we analyzed source-reconstructed human magnetoencephalography (MEG) data to assess how large-scale phase-synchrony (PS) and amplitude-correlation (AC) networks are shaped by local neurotransmitter systems. Specifically, we estimated node centrality, which indexes ‘hub-ness’ of brain regions, in PS and AC networks and estimated its covariation across 200 brain regions with 19 neurotransmitter receptor and transporter density maps. We found that network centrality covaried positively in delta to alpha, and gamma frequency bands (PS) or delta and gamma (AC) with GABA, NMDA, muscarinic, dopaminergic, and most serotonergic (5HT) receptor densities. In contrast, for frequencies in between, covariance of these receptors with network centrality was zero or negative. Negative covariance was also observed for most frequency bands with histaminergic, opioid, nicotinergic, cannabinoid, and 5HT1b receptor densities. These results establish how local microarchitecture influences large-scale connectivity networks of neuronal oscillations in the human brain in frequency- and spatially specific patterns.