Chem 8157

Hi Melissa,

I thought your talk was excellent, but I had a couple of questions that we didn't have time for:
1. How well characterized was the size of the cadmium nanoparticles used to bind the DNA strands?
2. How efficient is the reaction that binds the nanoparticles to the strands?

As a direct followup to these two questions - is this a quantitative method? It seems that if you could know exactly how much cadmium was contained in a nanoparticle bound to a strand then you could get an accurate idea of how much DNA annealed to the target strands on the gold surface.

I also liked her talk. There are a couple of questions for Melissa and the rest of the class:

1. Methodology - Why choosing CdS? Should other materials be more suitable for dissolution and chronoamperometric detection?

2. Data interpretation - I disagree with a main conclusion by the authors: "The method is selective". They use Figure 5 to support their claim. However, I believe that the anodic current magnitude depends on the copies of DNA on the SAM and bound to the CdS particles. Can somebody elaborate?

3. The electron microscopy images are bad.

I like Melissa's talk. But I have some questions about the paper too.

First, I think one reason for the author to choose CdS is that CdS nanoparticle preparation is very mature. It is easy to control shape and size.

I doubt about the model the author use. Does CdS and DNA strand connect as a 1:1 ratio? For me, I think one CdS could connect more DNA strand and it is not easy to control every nanoparticle to connect the same number DNA strand. If one nanoparticle has several DNA strands on it. Could it hybridize with more target DNA at the same time or could it cause a hindrance to other nanoparticle? I don't know. So I doubt it could be a quantitative method.

Like Jing said, CdS nanoparticles are attractive to use because it is easy to synthesize and control the size. The electron/hole pairs of the semiconductor nanoparticle also make it a useful material to use with respect to other materials. Another reason why the authors may have chosen this material is because it has been commonly used in past DNA sensor experiments. Unfortunately, the authors of the paper did minimal characterization of the nanoparticle, only taking poor TEM images and determining the diameter to be approximately 20nm. Ideally, dynamic light scattering, zeta-potential, and better UV Vis measurements (for determination of the area of nanoparticle absorbance) should have been taken. In regards to covalently linking the complementary oligonucleotide to the CdS particle, the authors modified the surface with mercaptoacetic acid (MAA). While I agree that multiple strands would fit on the nanoparticle, if the number of MAA molecules on the surface of the nanoparticle were limited, you could potentially limit the oligonucleotide strands on the nanoparticle. I tried accessing the paper that the authors based their synthesis, but I was unable to access it so my thoughts on MAA binding are not supported by the authors work.

The authors didn’t specify the efficiency of the DNA hybridization, but I can imagine that the rate of binding is related to the monolayer packing density, the concentration of the DNA sensor, and the ability of the oligonucleotide probe to enter the target DNA monolayer. The packing density could be determined from SEM images of the surface and would have been a nice edition to the paper instead of simply saying the monolayer was “densely packed.” If the monolayer is too dense, this may inhibit binding because the oligonucleotide probe may not be able to efficiently enter the layer to bind and therefore, would affect the signal response.

Finally, I think the term selectivity is used loosely here but I still think it is accurate to say that it is more selective to DNA of interest. I agree with Edgar that signal will depend on the amount of DNA on the SAM and bound to the CdS, but if the samples of different oligonucleotides were prepped in the same manner, I would imagine that the amount of DNA in the SAM and on the nanoparticle would be similar. Also, in figure 5, there is a 10 fold concentration difference in the amount of the different oligonucleotide sequences on the electrode. That is, the concentration of noncomplementary sequences is 10 times more than that of the complementary SAM and still yields a much larger signal.

I liked the presentation. The question I have is why did they use a chronoamperometric method rather than potentiometry with an ion-selective electrode? For an interesting example of the use of potentiometry in a DNA array, see:

R. Thurer, et al. Anal. Chem, 2007, 79, 5107-5110

The method depicted in this paper also does not involve modification of the analyte DNA.

Also, I agree with the other posts that CdS was used because they are well studied. CdS nanoparticles also provide an inherent amplification of the DNA based upon the size of nanoparticle used.

For the selectivity question, the authors use three base pair mismatched DNA to demonstrate the high selectivity. I am just wondering whether this detection technique still works if only one base pair mismatched DNA is used.