Grantee Research Project Results
McGill University, Montreal, Quebec, Canada
We have identified two very different modes of access, one in bacteria and one in mammalian cells. Both are dependent upon the physical chemistry of the particle.
In bacteria. Bacteria do not endocytose; thus, they are unlikely to be able to take up particles several nanometers in size unless there is a particular mechanism permitting these particles to penetrate. We have identified two features critical for uptake in bacteria. First, the QD must be conjugated to a biomolecule to which the bacterium has receptors, and the conjugation must not interfere with the recognition site of that molecule. Second, the QD must be readily photooxidized, and light exposure must occur during incubation with the bacteria. Certain biomolecules, such as the purines adenine and guanine, the neurotransmitter dopamine, and the amino acids tyrosine and tryptophan make the QDs more effective at this light-induced cell penetration. Penetration is through membrane defects in the cell surface; however, treated bacteria do not show membrane compromise or associated cell death.
In mammalian cells. The presence of specific receptors to the QD conjugate is required for uptake. Light exposure is not required, consistent with endocytic mechanisms of uptake. QDs can be visualized in cellular endosomes, and endosomes can be ruptured by exposure to UV light.
In all cells. Surprisingly, even cells that endocytose show preferential uptake of certain colors (sizes) of quantum dots. This is correlated with the ability of the different sizes to transfer electrons or holes to their conjugates.
Reactivity. With membranes, hydrophobic QDs, capped with pyridine, show integration into biomimetic lipid membranes and can be delivered (with the aid of a surfactant) to the membranes of living cells. Surprisingly, the application of large voltages (120 mV) and exposure to high-intensity UV light (Hg lamp, 60 min) are not sufficient to produce any measurable leakage events. However, hydrophilic QDs, capped with mercaptoacetic or mercaptoproprionic acids, create fast leakage events suggesting penetration through lipid membranes. Exposure to large voltages appears to be necessary to begin the process; after leakage events have begin, smaller voltages (20 mV) are sufficient to evoke additional events. Events are nonquantal and do not require light exposure.
With DNA. DNA damage by QDs appears to be related to Cd release after photodegradation of particles. Severe effects on bacterial growth were seen whenever bare core CdSe QDs were assimilated, especially in Gram positive strains. We investigated growth using measurements of cell density (optical density at 600 nm), by inspection under epifluorescence microscopy (to investigate cell morphology, spore formation, and QD-related fluorescence), by plating, and by quantitative analysis for hydroxyguanine. When QDs were added to Gram positive Bacillus cultures in early log phase (OD 600 <0.2), optical density decreased, indicating cell loss, and endospores could be visualized (n >10). When incubation was begun at mid-log phase, OD continued to increase, but more slowly than in controls: percent of control at 2 h, 60 2%; at 6 h, 39 3% (n = 3). However, the cells were unable to form colonies (n = 5; dilutions ranged from 1:5 to 1:1 for cultures of OD600 between 0.3 and 0.5). Extraction of genomic DNA and assays for oxidative DNA damage showed an average of 7.75 0.006 abasic sites per 105 base pairs in Bacillus subtilis cells exposed to yellow QD-adenine (n = 6). B. subtilis exposed to cell-wall targeted (nonassimilated) QDs under room light showed 2.33 0.005 abasic sites per 105 base pairs (n = 3). With CdSe/ZnS core-shell QDs, less than 1 abasic site was seen per 105 base pairs. However, much less QD assimilation was observed in this case: 2-fold weaker fluorescence using particles with a 15-fold greater quantum yield. Hence, it is possible that Cd release assists particle uptake.
Stability. The QD conjugates we have used have all linked a biomolecule to the QD surface cap via an amide bond. Proteases in cellular lysosomes break these bonds, liberating the QD from its conjugate. Small conjugates can thus diffuse out of the lysosomes and into the cell. Using QD-dopamine, this process may be directly observed as oxidized dopamine shows blue fluorescence.
Prolonged exposure to light and oxidizing conditions do not appear to lead to significant release of Cd from CdSe/ZnS core-shell particles. Ion chromatography/ mass spectrometry (ICP-MS) revealed less than 1 ppm Cd in aqueous solutions of core-shell dopamine conjugates after 4 hours of UV exposure. The addition of antioxidants to these solutions (beta-mercaptoethanol or ascorbic acid) reduced the Cd release to less than 20 ppb.
Conclusions and future work. The photophysics of QDs, particularly their size, redox potential (which depends not only on band gap, but on the presence of surface states) has a great influence on the interactions of these particles with biological cells and structures. Electron transfer processes should be sought as causes of selective uptake of specific sizes of QDs. Common biomolecules can also sensitize QDs, making them more photoreactive and able to damage cell membranes with or without Cd release. DNA damage, however, has so far only been seen when Cd is released in significant quantities.