September 5, 2018
by Kil Ho Lee
QDs are inorganic nanoparticles that are coated with surface capping agent known as ligands. Depending on the type of ligand, QDs can be soluble either in organic solvent (i.e. chloroform), or in aqueous solvent (i.e. water). For biological applications, QDs need to be soluble in an aqueous environment. Therefore, the ligands must be suitable to render QDs stable in aqueous solvent to enable their use as imaging agents for STORM.
Previously, we demonstrated successful ligand exchange procedures that replace organic ligands (i.e. octadecylamine) with aqueous ligands (3-MPA), yielding QDs stable in aqueous solution. Thus, in these experiments, we attempted to further modify those aqueous QDs by coating with single-stranded DNA (ssDNA). As discussed in my previous blog post, the “DNA embedding” procedure requires the deposition of additional layer of material onto the surface of the QDs. Specifically, CdSe/ZnS QDs are subjected to the deposition of additional ZnS layer during the DNA embedding step. This is accomplished by introducing Zn ions, additional 3-MPA, and thiol (-SH)-modified ssDNA into a solution containing aqueous QDs while increasing the solution temperature to 90 °C. As a reminder, 3-MPA includes a thiol functional group (-SH). Hence, both 3-MPA and ssDNA serve as S sources for forming ZnS layers.
In these experiments, we tested different ratios of Zn ions, 3-MPA, and ssDNA and evaluated their effect on QD stability following the DNA embedding procedure. We also tested other variables, including pH and purification procedures. We examined the final products based on the degree of aggregation, fluorescence intensity, as well as the red shift of the fluorescence emission wavelength. Aggregation indicates weakly stabilize QD surface; therefore, we expected to observe QD aggregation when the DNA embedding was not successfully accomplished. Fluorescence intensity can also indicate the stability of QD surface because weakly stabilize QD surface often results in the reduced fluorescence intensity. Lastly, the red-shift of the emission wavelength occurs when QDs are coated with thicker ZnS layer. Therefore, we expected to observe the red-shift when the deposition of additional ZnS layer and/or the DNA embedding procedure was successfully accomplished.
Ligand Exchange Experiments
Figure 1. 3-MPA-aqueous QDs. (Left) Image under a handheld UV light. (B) absorbance (red) and fluorescence (green) spectra. Results are consistent with the desired green fluorescence emission.
We used green QDs with an emission wavelength of 560 nm. Using the same procedure as in the previous report, we produced 3-MPA coated aqueous CdSe/ZnS QDs. Again, the procedure was reproducible and produced stable QDs in water. The absorbance and fluorescence emission spectra (Figure 1) were consistent with those reported in previous results.
DNA Embedding Experiments
Using these aqueous QDs, we attempted the “DNA embedding” procedure. Table 1 summarizes groups of sample conditions tested in this study. Group 1-Sample1 (G1-1) used the ratio of 3-MPA, Zn+, and DNA (as well as the purification method) consistent with literature reports and served as the control.
Table 1. Summary of reagents, purification setting, and post-purification QD stability
Using G1-1 as a standard, Group 1 and 2 were varied to test the effect of increasing Zn+ ion or DNA concentration on the properties of the resultant product. We would expect increasing Zn availability results in thicker ZnS layer, whereas increasing DNA results in higher DNA embedding efficiency.
In these experiments and consistent with the literature, purification was conducted using a centrifugal filter device. Purification is necessary to remove any remaining, unreacted reagents in the solution. All samples in Groups 1 and 2 appeared to aggregate during the initial heating step. Following purification, QDs were no longer dispersible in water, and they settled to the bottom of the test tubes (Figure 2).
Figure 2. Results for Group 1 and Group2: Aggregated QDs were clearly visible with excitation by a handheld UV lamp after DNA embedding and purification procedures.
Because some aggregation was observed during the initial embedding procedure, we suspected that the either the ratio or the duration of heating needed to be optimized. Nonetheless, the solution also appeared to include dispersed QDs. We attempted to retrieve the dispersed QDs through purification. However, the QDs formed larger aggregates and were no longer dispersible in water. Based on this observation, we suspected that QDs in Group 1 and Group 2 experienced some aggregation during the embedding procedure, resulting in theormation of larger QDs with weakly stabilized surfaces. The source of irreversible aggregation of the dispersed QDs after purification is still not clear. This may have resulted from the relatively rapid precipitation of the solution, containing large aggregates and dispersed particles together.
Next, using G1-1 as the standard, we introduced Group 3 to test the effect of a lower centrifuge speed during the purification step. If the rapid centrifugal filtration was the cause of irreversible aggregation, we expected that lowering the centrifuge speed would reduce the degree of aggregaton. This variable was tested by comparing G3-1 to G1-1 directly, for which conditions were identical except for centrifuge speed during purification.
Through the remainder of group 3, G3-1 was used as a standard, and the other samples (G3-2 to G3-4) were varied to test the effect of increasing DNA availability on resultant particle stability. Reducing the centrifuge speed by 3000 rpm, resulted in QDs that were well dispersed in water (Figure 3). However, the QD emission wavelength red-shifted from a green to yellow color for all QDs in group 3.
Figure 3. Results for Group 3: (Top) Well-dispersed QDs were seen in water showing. (Bottom) the fluorescence spectrum indicated red-shifted fluorescence emission.
The literature on which we based our process suggests that the formation of an additional ZnS layer with embedded ssDNA results in the observed red-shift. In fact, one of the examples in the literature shows a fluorescence shift of ~ 13 nm toward the red end of the spectrum after coating CdSe QD with a thick CdS shell. 
In the literature we are following, the pH of the solution was often adjusted to pH 12. However, in our earlier attempts, adjusting the H to 12 often resulted in QD aggregation, even before mixtures were subjected to heating. Therefore, we evaluated initial groups at a pH 10, which did not result in aggregation in our prior efforts. Thus for our final experiments, using G3-1 as a standard, we varied the pH of the aqueous phase to pH 12 in sample Group 4.
Figure 4. Results for Group 4: (top) Well-dispersed QDs were seen in water, and (Bottom) the emission spectra indicated only a slight red-shifted in fluorescence emission.
First, adjusting the pH to 12 did not cause QD to aggregate. Under this condition, the heating step did not cause the slight aggregation that appeared in Group 1 – 3. We suspect that our new, well-optimized ligand exchange procedure provided better QD stability against higher pH. Compared to samples in Group 3, the purified QDs in group 4 showed only a slight red-shift in their fluorescence emission spectra (Figure 4), and they were highly soluble in water.
Conclusion and Future Directions
Both Group 3 and Group 4 showed high solubility and red-shifted emission spectra, indicating successful ZnS deposition and possible DNA embedding. Comparing Group 3 and Group 4, QDs processed at pH 10 seemed to form thicker ZnS layers than those processed at pH 12. We aim to examine the embedding efficiency of QDs in Group 3 and Group 4 next. We are currently working on binding QDs to DNA origami structures presenting the complementary ssDNA sequences. This will allow us to confirm the presence of DNA on the QD surface. Once we confirm QD-DNA origami binding, we will attempt to quench QD fluorescence via FRET with gold nanoparticles presenting complementary ssDNA.
 Deng, Zhengtao, et al. "Robust DNA-functionalized core/shell quantum dots with fluorescent emission spanning from UV–vis to near-IR and compatible with DNA-directed self-assembly." Journal of the American Chemical Society 134.42 (2012): 17424-17427.