Home > Lab-blog > Structural elucidation of dimeric DNA G-quadruplexes

Structural elucidation of dimeric DNA G-quadruplexes

Stacking of G-quadruplexes: NMR structure of a G-rich oligonucleotide with potential anti-HIV and anticancer activity

Ngoc Quang Do, Kah Wai Lim, Ming Hoon Teo, Brahim Heddi1 and Anh Tuan Phan

Nucleic Acids Research, 2011, Vol. 39, No. 21, 9448–9457, doi:10.1093/nar/gkr539

A synopsis by Marilyn García-Arriaga

In order to gain better understanding of the nature of this structures, Phan and colleagues reported the structural analysis of a dimeric OGQ with the sequence (GGGT)4 (T30695) in K+ solution. This dimer is composed of two identical propeller-type parallel-stranded OGQ subunits each containing three tetrads that are stacked via the 5’-5’ interface. NMR structural studies of the OGQ formed by T30695 and T40214 ((GGGC)4), in K+ solution, share similar 1D and 2D NOESY spectral features. Furthermore, preliminary CD studies show the positive band at 260 nm characteristic of a parallel-stranded OGQ, in contrast to a previous report. In order to perform a detailed structural analysis by NMR an accurate assignment of the signals is essential. This task is more challenging in systems of high symmetry, thus, to overcome this problem they prepared a T30695 analogue with a single guanine-to-inosine substitution, GIGT(GGGT)3 (). This modification greatly improves the NMR spectra of the assembly allowing the assignment of the signals without ambiguity. Not only does this derivative show the same structural characteristics in the 1D and 2D NOESY spectra, but it also show similar positive band in the CD spectra and pattern in the gel electrophoresis experiments. 15N-labeling of the guanine imino protons and other protons of J19 enabled establishing the correlations in the COSY, TOCSY HSQC and NOESY spectra. In contrast to what was previously reported, the moderate intensity of the intra-residue H8/H6-H10 NOEs suggest that all residues adopt anti glycosidic bond conformation. The combined evidence suggested that the resulting structure is a propeller-type parallel-stranded OGQ with a three-tetrad core and three double-chain-reversal loops.

Evidence of the 5’-end stacking to form the dimer was obtained from gel electrophoresis in which the migration rate of J19 was similar to that of 93del, an interlocked dimeric OGQ. In addition, the migration rate of J19 was slower than that of a monomeric propeller-type OGQ. Furthermore, solvent-exchange experiments reveal that the imino protons of guanines in the outer tetrads (5’-end) are protected from exchange with D2O. Also, additions of bases in the 5’-end of J19 disrupt the dimer formation. The solution structure of J19 was generated after distance-restrained molecular dynamics refinement in this structure the core of the quadruplex show a close packing across the interfaces of the tetrads. Also, the directionality of the hydrogen bonds was the same for each subunit and opposite between the two structures in the dimer. The thymine bases are projected outwards in a double-chain-reversal loop. In more details, the sugars from the two end-tetrads are contiguous to one another and the backbones of the two dimer subunits are aligned in a staggered mode, maximizing the overlap of the five and six membered rings on the interface. They were able to conclude this by the identification of NOEs correlations among the base and the sugar protons of the end tetrads.

Finally, in order to assess if those OGQs conserved anti-HIV activity, they performed a reverse ‘disintegration’ reaction assay using T30695, J19 and their derivatives. They concluded that the derivatives containing thymines at the 5’-end were less active, which can be attributed to the lost of their ability to form stacked dimeric structures.

Once again Phan and colleagues present an impressive amount of work, but most of all, an incredible level of analysis. At the experimental level, the work is well supported and all the proper control experiments were performed. In contrast, the narrative and presentation of the data was least successful, in my opinion, it lacks details in the arguments of some conclusions.

Advertisements
  1. Jean
    February 1, 2012 at 22:22

    In this article by the Phan Group, the structure of a G-Rich DNA sequence was elucidated via NMR analysis. The structure was found to form a GQ consisting of 3 tetrads. This structure then stacks into a dimer. The first half of the article concerns the elucidation of the structure of the GQ and the second half the evaluation of the stacking interactions found in the dimer; including several experiments to confirm that the sequence in fact forms a dimer. These results are novel, since this particular structure, as well as the fact that it formed a dimer was not confirmed.

    The authors spent a lot of time confirming their findings. In order to appropriately characterize the structure, they replaced a base (guanosine for inosine), which resulted in loss of symmetry and simplified the NMR spectra. They confirmed the similarity of this structure to the original via CD and gel electrophoresis. Analysis of the stacking interactions reveal that they stack 5’-5’ with partial π-π interactions between the guanine bases. They could also reduce the dimerization by adding non G residues.

    In general, I like the narrative of the article I think it is fine the way it is. I could not find any major spelling or grammatical mistakes. I didn’t like the placement of the figures but that is understandable. The conclusion seemed a bit short (3 sentences) but the article does not suffer from that.

    Marilyn’s synopsis is really good as always. However, there were a few grammar mistakes such as this instead of these. The mistakes, however, are not unbearable and in no way affect the quality of her synopsis. I like the picture; the message gets across (I know structure determination is a lot of work). It is a good job in general.

  2. AJS
    February 2, 2012 at 00:01

    This week’s paper concerns a thorough structure determination, primarily by detailed NMR analysis but incorporating other methods, of a quadruplex-forming synthetic oligonucleotide known as J19. J19 is very similar to another synthetic oligonucleotide, T30695 (only one base difference); T30695 is of particular interest thanks to its anti-HIV activity and anticancer activity, and its structure has been investigated by other groups, and caused some controversy. The authors show that in solution (K+ buffer) J19 adopts a propeller-type G-quadruplex conformation, with these quadruplex subunits forming dimers, both stacking at the face containing the 5’ terminus; furthermore, they show ample evidence to infer that the structure of T30695 incorporates these features.

    Despite being someone who does not spend a lot of time reading structure papers, and despite the level of detail, I found the paper readable and well-argued; this is somewhat at odds with Marilyn, and I’ll be interested to see what she feels it lacks in detail. If I must make a criticism I would say that it is rather a shame that, with the oligonucleotides in hand, more detailed analysis wasn’t made in the ‘Control of stacking between G-quadruplexes’ section, or at least some NMR work. Given our interests it would have made very exciting reading; perhaps they’re working on that as a follow-up.

    Marilyn’s synopsis gives at detailed account of what was done, and generally quite clearly; the picture is fine.

  3. Mariana
    February 2, 2012 at 02:18

    This article is presenting the detail and revised elucidation of the Oligo GQ formed by the sequence T30695 (GGGT)4 with K+ in solution. They determined that this sequence forms a dimer between identical propeller-type parallel-stranded GQs containing each 3 tetrads stacked with a 5’-5’ interface. Previously the NMR elucidation of this OligoGQ by other group presented some contradictory information base on NMR signal overlapping. This cause a misunderstanding assigning the structure as an atiparallel GQ even though the CD showed the band corresponding to a parallel GQ. Phan group take advantage of the characterization and behavior of similar sequences and MDS studies and performed a detailed study by corrupting the general symmetry of the dimer by incorporating an ionosine for the 2nd guanine in the sequence (GIGT(GGGT)3), which they called J19. Cleverly enough this modification gave rise to the elucidation of the dimer be performing several 1D and 2D NMR experiments with better signal, as well as gel-electrophoresis studies to confirm the dimer and CD for the strand orientation. They also performed molecular dynamic calculations using the NMR structural data for the construction of their model. At the end they showed some inhibition experiments, although they mentioned them but it gave the impression that they were not confortable with the results.
    I have to say that structural characterization articles are not in my list of “articles I want to read”, but when I have the opportunity to do so, it is nice to appreciate the amount of details that NMR can give you about a supramolecular structure. Not only that, this article in particular demonstrated how important it is to take into consideration various techniques to confirm your structure, since by simply making a wrong assignment you can ended with the wrong structure. I guess that for me due to the amount of NMR experiments done (the fact that they have NMR of 600 and 700 Mhz), they were missing DOSY experiments. I know the gel electrophoresis gave them basically the information they were looking for, but they could run diffusion NMR to make the article even more fancy.
    The narrative is clear and easy to follow I think they make it attractive to people who does not like to get drawn for many structural details, so I guess form this part I have a discrepancy with Marilyn. The pictures in the article are not my favorite but god enough to show and confirm their hypothesis.
    The synopsis I think is complete and with minor errors including Marilyn’s opinion. Better than anyone in the lab she knows the amount of analysis that this type of work involve. I like the picture because you can get the idea of article by just reading the tittle and looking at the picture.

  4. Ana Victoria
    February 2, 2012 at 03:20

    Phan and coworkers report the structure of a dimeric GQ containing three tetrads. The structure was determined by NMR structural analysis. CD studies confirmed the presence of a parallel stranded OGQ by the appearance of the positive band at 260 nm. They also made a derivative (it had a single guanine to inosine substitution to loose the symmetry) to facilitate the assignment of the signals in the NMR. Gel electrophoresis was used to provide evidence of the 5′-end stacking that contributes to the dimer formation. Other studies such as the reverse disintegration reaction assay was performed to see wether the OGQs had anti-HIV activity. The conclusion was that the derivatives with the thymines at the 5′-end were less active probably because of the loss of the dimeric structure.
    The narrative of the article was easy to follow, it had fluidity and understanding it was not difficult. Marilyn’s synopsis was very clear, she did a good job. The picture helps to get the general idea of the article.

  5. Nicole Zoé
    February 2, 2012 at 04:12

    Phan and colleagues in this full article present the structure determination and characterization of a G-rich oligonucloetide that forms a dimeric structure, which stacks in a K+ solution. Also they present studies done regarding the activity of the dimeric structure with -anti-HIV and anticancer studies. The NMR and DC studies performed to the compound, which are essential and I think the most significant part of the article, help elucidate and confirm the structure and a parallel-stranded OGQ, with the help of a change in symmetry of the molecules by replacing the guanine fot an inosine, so all the protons could be correlated and identified in the spectrum, using COSY, TOCSY HSQC and NOESY spectra.

    The other part of the article deals with the studies regarding the ant-HIV and anticancer studies performed with the OGQs. The results regarding this studies showed that derivatives containing thymines at the 5′-end position were less successful, apparently due to the lack or not favorable formation of stacked dimeric structures.

    I have to say that although the article is long, it contains and presents a lot of studies, which indicate the effort and work performed and that it payed off. I found the narrative to be easy to follow, maybe a few more details in some parts, but an overall good work. Marilyn’s synopsis, very good and helpul, as always, straight to the point but without leaving anything behind and focusing on the essential details. Regarding the picture, I liked it a lot, very nice and captures the essence of the article.

  6. Diana S.
    February 2, 2012 at 11:54

    This paper presents the story of the elucidation of a GQ NMR structure. It’s particularly interesting because of the relatively small number of dimeric GQ structures published to date. Substitution of a single guanine for inosine helps elucidate the structure by disrupting the symmetry that results in overlap of a lot of signals.
    Things I found interesting from the paper: the mixing time is actually a lot lower than the ~600 ms that Marilyn and I’ve used. (I think María’s used more similar mixing times). I know this depends on the size of the molecule, but I’d think that this dimeric oligomer GQ would be if anything, larger that on of our quadruplexes. Also, they seem to use different mixing times in different figures, which is something I’ve never done. Maybe using several NOESY spectra with different mixing times would give us more information.
    Figure 8 is supposed to show HIV IN activity. If I understand correctly, the bottom band means activity and that’s really very little intense in the lanes corresponding to the GQ they’re studying. I guess that from a biochemist’s perspective, it’s enough…
    I’m curious as to the observation that the NOESY they obtained is completely different from the one obtained by Jing. There has to be a reason for this change, maybe in sample preparation? They don’t mention they’re efforts to corroborate this, which I would have liked.
    The paper is a very easy, nice read from a potentially dense topic. The synopsis is good and gives credit to Marilyn’s experience (which is far more than mine). The figure I’m assuming alludes to the amount of printed and analyzed spectra needed for such a job. Nice.

  7. Luis Negron
    February 2, 2012 at 17:43

    Another chapter in the quest for G-rich oligonucleotides that forms GQs but many structural details to understand these systems functions remains unclear. More than anti-HIV or anticancer activities, control of function of these systems need more tools to understand GQs. In this case, Phan work reveals strategies to solve the problem of ambiguity in some reported structures like T30695 (or T30923), with the sequence of (GGGT)4, and T40214, with the sequence of (GGGC)4. There is contrast in reposted structures using CD and HNMR. According with the presented structural data, the importance in this work is the control in the stacking of two monomeric subunits by sequence modifications. By using 600 and 700 MHz NMR Bruker spectrometers and A single guanine-to-inosine substitution (GIGT(GGGT)3), they can improve the NMR spectra of T30695.

    With the help of molecular simulation systems and powerful 2D NMR experiments they provide a potential tool to analyze stacking of GQ that will help to understand more complex assemblies of other G-rich sequences. For me, more than the data provided, is interesting how one substitution conserve the structure of a dimeric GQ. What is the limit of this recognition motif or dimeric GQ when substitutions that preserve structure are applied? More over, how this structural scaffolds are affected at lower concentrations than the ones reported in terms of maintaining the more stable folding that will be have a potential function in the inhibition of cancer or anti-HIV.

    I like to read these article, my favorite part are the NOESY readings to solve the structure. It will be interesting how the stability of fluorescent label derivatives is affected with these substitutions and then using these tools of structural determination, test the potential applications in medicinal chemistry. We have many tools to study these systems, but those that study GQs in solution and have the ability to monitor in terms of concentration and time are the ones that can provide the answer to many unsolved questions in these area.

    About MGA synopsis, was clear enough to help me to understand this complex topic. She focuses in the dimeric GQ structure with the two identical propeller-type parallel-stranded GQ subunits and how this structure was elucidated. Nice picture, very creative to describe the presented work.

  8. Maxier A. S.
    February 2, 2012 at 19:42

    Anh Tuân Phan presents a full structural analisis via NMR studies and other correlating them with MD structures. On their interpretarion of the NOE of T30695 and T40214 there was a syn interaction of glycosidic conformation due to the lack of cross-peak of H8-H1’. Later for the T30695 sequence they modified the second G to an I so it simplifies the NMR spectra yet it breaks symmetry. But it does not seem to make major difference for the overall expected spectra. They saw 11 sharp G imino protons correlated to 3 G-tetrads. Later via NOESY this pattern indicates the same correlation with J90. With CD spectra they confirm a parallel-stranded GQ and it was all confirmed via electrophoresis experiments to J90, T30695 and T40214. Later stacking studies via NOE showed that J19 forms a dimeric GQ of propeller-type and parallel-stranded with 3 tetrads if a 5’ to 5’ interface.

    Personally I enjoyed reading this paper, the correlation of experimental data between different sequences shows a good understanding of OGQs. In the case of the HIV-1 IN inhibition I wish they would have explored more of this area of maybe with the discussion like they did with the rest of the experiments. What it may concern me is that the guanosine to inosine substitution, I understand that they did it to “simplify” the NMR yet, wouldn’t that affect the structure now that the eliminated a guanine that would have possible interacted in the GQ to an inosine that it would be expelled from it,?

    About Marilyn synopsis, she always does a great job at those. Giving an explanation of the paper without having to enter in specifics. The picture reminds me more of Marilyn’s current research state rather than the paper, yet it works perfectly.

  9. PGMS
    February 2, 2012 at 20:17

    This article presents the efforts of Phan and colleagues in reporting the structural analysis of a dimeric OGQ formed by the sequence T30695 (GGGT)4 in K+ solution through NMR. This dimer is proposed to be composed of two identical propeller- type-parallel strands each containing three tetrads stacked via 5’-5’ interface.

    The idea of preparing an analogue for T30695 with a single guanine-to-inosine substitution at the 2nd guanine position, GIGT(GGGT)3 is appreciable as it greatly improved the NMR spectra, allowing the assignment of signals, which would rather be difficult to elucidate structure with systems of high symmetry. To their efforts, they could prove the same structural characteristics in 1D and 2D NOESY spectra, similar positive band in the CD spectra and stacking pattern in gel electrophoresis.

    The 5’-end stacking to form a dimer was easily evidenced through the comparison of migration rate of an interlocked dimeric OGQ by gel electrophoresis. Finally, the conserved anti-HIV activity by the OGQs was assessed by reverse disintegration reaction assay, wherein they could conclude that derivatives containing thymines at 5’ end were less active, which was a known fact their inability to form stacked dimeric structures.
    This article is simple, results are meticulous and easily understandable and well presented.

  10. LRR
    February 2, 2012 at 20:37

    In this article Anh Tuan Phan and coworkers studied the structural details of the G-quadruplexes formed by three different G-rich oligos in a K+ enrich solution using gel electrophoresis, circular dichroism, molecular dynamics, and a plethora of NMR techniques. The (GGGT)4 and (GGGC)4 are two of the many known sequences with anticancer and anti-HIV activity. To help with the NMR characterization they developed a new sequence by mutating (GGGT)4 to (GIGT)(GGGT)3. Inosine breaks the symmetry of the original sequence and as a consequence more information about the GQ secondary structure was obtained by NMR. Although, I wasn’t able to see the 11 peaks from the resonance in figure 1c, I only see 9 peaks. To study the mode of dimerization new sequences having thymine at the 3’ and 5’ ends were constructed and information gathered based on their electrophoretic mobility.
    Qualitative analysis for HIV testing show no notable difference between parent and mutated strand and this positive test was enough to hypothesized that this G-rich oligo has potential as anti cancer agent. This portion of the title was what attracts me the most of this paper and this is only a hypothesis based on a qualitative test. Where the disintegration assays part of the qualitative biological screening?
    Even though this is not the type of article I will chose to read I did like the narrative used and the order in which the results were discussed. In my opinion it’s a very well constructed paper starting from the chosen title.
    Marilyn synopsis was good, it’s clear and summarized the main results. The cartoon looks very nice, I don’t know if what she means is that this is how the desk of some one who works in the elucidation of GQ structure using NMR analysis looks like.

  11. Loruhama
    February 2, 2012 at 21:41

    In this article, Phan and coworkers present the structure determination of a Guanosine rich DNA sequences by doing NMR studies. They explain that their structures consist of three tetrads that adopt a dimeric GQ structure by the stacking of two propeller-type parallel-stranded GQ subunits at the 5’-end position. CD studies were also used, to elucidate the presence of parallel or antiparallel species. These studies confirmed the presence of a parallel stranded OGC because of the appereance of a positive peak at 260nm. They also used Gel electrophoresis to elucidate the 5’ end stacking. For further characterization processes, they replaced guanosine for inosine, which basically simplified their spectra.
    For me this article had a very engaging narrative with a lot of information. I think that they did all the proper control experiments, and they explain their data very well. As for Marylin’s synopsis, she did a good job, summarizing all the important parts. She had a few grammatical errors here and there, but the overall synopsis was great. The picture was nice and creative.

  12. Y.Melendez
    February 2, 2012 at 23:24

    In this article, they present structural studies of G-quadruplexes that consist of T30695 and derivative sequences via NMR spectra. An improvement was observed on the spectra with a guanosine-to-inosine substitution. They also give us possible structures at the stacking interface and the conditions that are affecting the stacking. The sequence formed was shown to adopt a dimeric G-quadruplex form and the stacking interactions were shown to be impaired by the extension of non-G residues from the terminals.

    I liked this article, I guess it shows a different side of G-quadruplexes that I wasn’t necessarily familiar with, since usually we kind of center on applications for G-quadruplexes and other supramolecular devices, rather than structure analysis. However, it wasn’t at all boring and the narrative was very easy to flow with. One thing I was very impressed with was the images; they complemented the article very well, and some of them were pretty gorgeous (specifically, figure #5). As for the studies, they pretty much pulled all the stops, I mean, the article is decked with different experiments that support their claims and their work. However, I did find their conclusion lacking, I didn’t feel like they went anywhere with it.

    As for Marilyn’s Synopsis, I found it very complete and complementary to the article. It was a very good summary of the key points in the paper. I thought the picture looked really cool too ☺

  13. Maria
    February 3, 2012 at 00:08

    The challenges behind atomic-resolution characterization of GQ-DNA/RNA structures are proportional to the combined complexity of the strands orientations, types of loops and syn/anti conformation of the nucleosides in a GQ structure. In this particular article Anh Tuan Phan and his colleagues reported on the NMR-based characterization of the GQ structure formed by T30695 and its derivative sequences in K+ solutions to clarify the current ambiguity in their structures. Herein the data reported suggest that these sequences have the tendency to form GQ dimers stacked through the 5’-end of two propeller-type parallel-stranded GQ with three G-tetrads and three double-chain-reversal loops. But why so much work dedicated to understand the structure of these sequences? Actually, these particular sequences have shown anticoagulant, anti-HIV and anticancer activity. Interestingly, when testing the anti-HIV IN activity of T30695, J19, and its derivative sequences, the authors observed decreased activity for sequences having Thymines at the 5’-end which might inhibit dimer formation. These results suggest the likely importance of dimer formation for an efficient anti-HIV activity.
    For the structural characterization, the authors used common techniques such as 1D- and 2D-NMR experiments, circular dichroism, gel electrophoresis and molecular modeling (e.g. molecular dynamics). However for the NMR characterization their advantage was the addition of inosine at a strategic position to disrupt the symmetry of the GQs improving the spectral resolution. Once they achieve this, they were able to extract cherished information from through-bond and through-space correlations complimented with strategic site-specific 15N-labeling and solvent exchange experiments. I consider that, experimentally, the authors carried out the appropriate control experiments, especially considering the effects of having additional bases at the 5’-end. In addition I compliment the authors for the subtle and respectful way in which they compare their results to previous data that might not be in agreement with their observed results.
    This was a very nice article to read considering that even though the authors are thorough in the discussion of their results the narrative didn’t seem overwhelming. The article was actually easy to read and with the introduction and discussion about the HIV-1 IN disintegration inhibition assays it became simple to understand the relevance of their results. How can I improve the manuscript?? I guess with the simple incorporation of Figure S4 (from the SI) into the actual manuscript the narrative could have been complimented even more and the article will seem much more appealing.
    I love Marilyn’s figure! Oh Marilyn, It reminds me of the long days we had just looking at many NMR spectra and playing with the molecular models in MacroModel jajaja! Her representation is so accurate highlighting the overwhelming volume of work that lies behind having “pretty” but accurate models like the one she included in the center of her figure. Regarding her synopsis, it was appealing, I think it has good details and her personal insights were included specially in the last paragraph.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: