Sub-nanoscale gold clusters are promising materials due to their unique molecular structures, aurophilicity, and photoluminescence properties, including phosphorescence. Improving their photoluminescence is an active area of research, with strategies including alloying, supramolecular networking, surface hardening, and ligand modification. The octahedral hexagold(I) cluster with a hyper-coordinated carbon center (CAu6) is a classical model, but its luminescence is often limited. While phosphine ligands have been used, N-heterocyclic carbene (NHC) ligands offer advantages due to their strong electron-donating properties and ability to enhance stability. Previous work demonstrated that NHC ligands can alter the photochemical properties of CAu6 clusters. This study aims to synthesize and characterize heteronuclear Au(I)-Ag(I) clusters using bidentate NHC ligands linked to pyridyl ligands, investigate their photochemical properties, and explore their cellular uptake and organelle-selective translocation pathways. The goal is to develop highly phosphorescent clusters for organelle-selective phosphorescence imaging and dynamic analysis of molecular distribution in cells.

Significant advancements have been made in enhancing the photoluminescence of gold clusters through various methods, including alloying metal kernels, supramolecular networking, surface hardening, and chemical modification of capping ligands. The CAu6 cluster, initially synthesized with phosphine ligands, serves as a key model. However, its emission is limited to the solid state. The use of pyridyl-phosphine bidentate ligands has enabled the creation of heteronuclear metal clusters exhibiting strong red phosphorescence in solution. More recently, NHC ligands have emerged as promising alternatives for Au cluster stabilization. Their strong electron-donating capabilities enhance the stability of metal surfaces, nanoparticles, and nanorods. Studies have shown that different NHC ligand modifications significantly impact the photochemical properties of CAu6 clusters, shifting emission wavelengths and influencing luminescence.

The researchers synthesized a series of bidentate ligands comprising an NHC ligand linked to a pyridyl ligand (1a-d). These ligands were used to synthesize CAu6 complexes [(C)(AuI-L)](BF4)2 (2a-d) where L = 1a-d. The heterometallic CAu6AgI2 clusters [(C)(AuI-L)Ag2](BF4)4 (3a-d) were then synthesized from 2a-d using AgBF4. The synthesis involved several steps, including reactions with K2CO3, NaBF4, KOH, and AgBF4. The characterization employed various techniques, including single-crystal X-ray diffraction (ScXRD) to determine the solid-state molecular structures, NMR spectroscopy to confirm solution-phase structures, and mass spectrometry to identify the species present. UV-vis absorption and phosphorescence spectroscopy were used to characterize the optical properties, and phosphorescence quantum yields and lifetimes were determined. Theoretical calculations using time-dependent density functional theory (TD-DFT) were conducted to analyze the absorption and phosphorescence properties, including the role of spin-orbit coupling. For the cellular studies, HeLa cells were incubated with the synthesized clusters (3a, 3b, and 4) and imaged using confocal microscopy to assess cellular uptake and organelle localization. Co-localization studies with organelle markers and time-lapse imaging were used to track the translocation pathways. Finally, endocytosis inhibition studies using wortmannin, sucrose, and genistein were performed to investigate the cellular uptake mechanism.

The synthesized CAu6AgI2 clusters (3a-d) exhibited strong yellow luminescence in the solid state and solution. Clusters 3a and 3b, protected by benzimidazolylidene ligands, displayed exceptionally high phosphorescence quantum yields (0.88 and 0.86 in CH2Cl2, respectively), significantly higher than those of clusters 3c and 3d with imidazolylidene ligands (0.14 and 0.01). The phosphorescence lifetimes of 3a and 3b were also substantially longer (1.85 and 1.66 µs) compared to 3c and 3d (0.32 and 0.16 µs). TD-DFT calculations revealed that the carbene ligands accelerate radiative decay by influencing spin-orbit coupling, and the benzimidazolylidene ligands effectively suppress non-radiative relaxation pathways. Cellular studies demonstrated that the NHC ligand-protected clusters (3a and 3b) accumulated in specific organelles, specifically the endoplasmic reticulum (ER), contrasting with the phosphine-protected cluster (4) which showed non-selective cytosolic distribution. Time-lapse imaging revealed rapid cellular uptake and ER accumulation within 10 minutes, with subsequent nuclear accumulation at longer incubation times. Furthermore, endocytosis inhibition studies suggested that caveolin-dependent endocytosis was the primary uptake pathway for NHC-protected clusters, while phosphine-protected clusters were taken up via multiple endocytic pathways.

The findings demonstrate that NHC ligands significantly impact the photochemical properties and cellular behavior of CAu6AgI2 clusters. The remarkably high phosphorescence quantum yields and lifetimes of 3a and 3b are attributed to the synergistic effect of the bidentate NHC ligands and AgI ions, which enhance radiative decay and suppress non-radiative processes. The organelle-selective translocation observed for the NHC-protected clusters is a significant departure from the non-selective distribution of the phosphine-protected cluster, highlighting the importance of ligand design in controlling intracellular behavior. The caveolin-dependent endocytosis pathway for NHC-protected clusters suggests a potential for targeted drug delivery.

This study provides a rational design and synthesis strategy for highly phosphorescent Au(I)-Ag(I) clusters with tunable photochemical properties and cellular translocation. The use of NHC ligands, particularly benzimidazolylidene, significantly enhances phosphorescence quantum yields and lifetimes. The demonstration of organelle-selective targeting to the ER opens possibilities for developing advanced bioimaging probes and targeted drug delivery systems. Future research could explore other NHC ligands to achieve even greater control over the emission properties and subcellular localization of these clusters.

The study primarily focused on HeLa cells. Further investigation is needed to determine the generality of the organelle-selective translocation in other cell types. The mechanistic details of the interaction between the clusters and the ER membrane require further exploration. The long-term effects of cluster accumulation in cells also warrant investigation.