Janelle Vultaggio

Abstract

G-quadruplexes, found at G-rich sequences of both DNA and RNA, are formed by Hoogsteen base pairing and stabilized by a metal ion such as K+ or Na+ within its central axis. Researchers are able to find and study G-quadruplexes by electrophoresis, chromatography, mass spectrometry, NMR spectroscopy, X-ray crystallography, and various sequencing techniques.  G-quadruplexes are preferentially formed at the very 3’ end of telomeres; however, they can also be found in various places throughout both the DNA and RNA structure.  In fact, they are found in high abundance in both 5’ UTRs and at the 3’ end of genes.  This review argues that G-quadruplexes have two main functions.   First, they provide regulation through structure of the genome.  This is found at the 3’ end of telomeres where they are able to stabilize the end of a linear DNA molecule.  Second, G-quadruplexes are important in gene regulation. This is found in telomeres, at the ‘5 end of UTRs, and downstream the 3′-end of genes where G-quadruplexes are suggested to play a critical role in gene suppression.  It is due to these two functional qualities of G-quadruplexes that they can be used in cancer treatments and therapies.   

Telomeres, the linear chromosome ends, ensure genome stability.  In fact, cancers and other diseases such as dyskeratosis congenital are caused by the miss regulation of telomeric proteins and telomerase (1).  Shortened telomeres can and often does lead to chromosomal instability and is a hallmark of human cancers.  In some cases, shortened telomeres have even been used as a prognostic marker (2).   Due to the (TTAGGG/CCCTAA)n repeats at end of telomeric DNA, G-quadruplex are often found (3).  These G-quadruplexes are well known as having the ability to inhibit telomerase.  The stabilization of these G-quadruplexes has been widely accepted as a therapeutic strategy against cancer because of this very ability.  This review is going to argue that the therapeutic advantages of G-quadruplex stabilization may not solely be due to its ability to inhibit telomerase.  In fact, it may only be a side effect.  This review will explore the bizarre phenomenon of G-quadruplexes located both within telomeres and intrinsically within the DNA.  It will show that G-quadruplex functions are much more complicated and elaborate than simply inhibition of telomerase.  In Tang et al. paper, G-quadruplex preferentially forms at the very 3′ end of vertebrate telomeric DNA, the researchers mentioned that “G-quadruplex may influence more telomere function than simply inhibiting telomerase” (4).   The main goal of this review is expand even a bit further on this and investigate what these other telomeric functions are and whether these telomeric functions are related to other functions of G-quadruplexes found intrinsically within DNA and RNA.

Background: What are G-quadruplexes?

G-quadruplexes on a very basic level have two main characteristics.  First and foremost, they are found at G-rich sequences of both DNA and RNA and formed by Hoogsteen base pairing (5,6,7).  Hoogsteen base pairing was first discovered in 1963 as the hydrogen-bonded complex between 1-methylthymine and 9-methyladenine (8).  It is a variation of base-pairing in which the N7 position of the purine base is used as a hydrogen bond acceptor and the C6 amino group is used as a donor (7).  Hoogsteen base pairing requires a lot of energy which is why the second characteristic of G-quadruplexes is a metal ion such as K+ or Na+ in the central axis to stabilize the Hoogsteen base pairing (5).  These complexes termed G-tetrads are then stacked on top of each other and held together by loops of DNA.  The loops of DNA are mixed and various sequences and not themselves G-tetrads (6).  The final result is the DNA G-quadruplexes: a four stranded structure that has nucleobases on the inside and sugar phosphate backbones on the outside (7).

How do researchers study G-quadruplexes?

Researchers are able to find and study G-quadruplexes in a variety of different ways.   Historically electrophoresis, chromatography, and mass spectrometry have been used to identify and give insight to the molecular sizes of G-quadruplexes.  One of the earlier studies of G-quadruplexes was done by using gel-electrophoresis to identify what they called the “formation of parallel four-stranded complexes by guanine-rich motifs in DNA” (9).  In their study, Sen et al. synthesized oligomers that corresponded to IgG switch regions or G-quadruplex regions.  They found that if they left the oligomers in buffer with salt molecules the result was significantly lower electerophoretic mobility (9).

NMR spectroscopy and X-ray crystallography have also played an important role in identification of G-quadruplexes both in single and double stranded DNA.  Both NMR spectroscopy and X-ray crystallography allow for atomic-resolution imaging of G-quadruplex structures.  NMR spectroscopy allows for researchers to run studies of kinetics, dynamics, and molecular interaction simultaneously (10).

As sequencing the genome has become easier and less expensive, researchers have been able to use sequences along with algorithms to computationally predict where G-quadruplexes are present.  Quadruplex forming ‘G’-rich sequences (QGRS)-Conserves are an example of this computational analysis.   Frees et al. used QGRS-Conserve to identify G-quadurplexes in humans that have been conserved across several mammalian and non mammalian species (11).  In order to predict sites of G-quadruplex formation, there first had to be enough studies done to know what DNA sequences G-quadruplexes correlate with.  The fact that Frees et al. has been able to create the QGRS-Conserve to identify G-quadruplexes is a true testament to how much research has been done in the field thus far.

Where are G-quadruplexes found?

It is well understood that human chromosomes are capped at both ends with telomere DNA (TTAGGG/CCCTAA)n repeats.  As the chromosomes are replicated the 3’ end of telomeres shortens (3).   By using DMS footprinting and exonuclease hydrolysis, researchers were able to discover that G-quadruplexes preferentially form at this very 3’ end of telomeres as opposed to intrinsically in the DNA (4).

Theoretically, G-quadruplexes should be able to form anywhere along a G-rich strand of DNA.  In 2013, Lam et al. was able to use an engineered antibody to detect and map G-quadruplexes in genomic DNA from human breast adenocarcinoma cells.  The results showed that G-quadruplexes are stable and present throughout the human genome.  Lam et al. found G-quadruplexes upstream of several genes’ transcriptional start sites.  Conclusively, although G-quadruplexes are preferentially formed at the very 3’ end of telomeres, they can also be found in various places throughout both the DNA and RNA structure (12).

G-quadruplexes function within the telomere: Inhibition of telomerase activity

It was first discovered that G-quadruplexes could inhibit the activity of telomerase in 1991.  Zahler et al. investigated the ability of folded telomeric DNA to serve as primer for telomerase in vitro.  The G-quadruplexes in primers, termed by Zahler et al. G-quartets, inhibited the use of telomerase.  They concluded from their study that telomerase does not require folding of DNA primers and that the folding of telomere DNA into G-quadruplexes negatively regulates the elongation of telomeres (13).  In addition, this inhibition of telomerase by G-quadruplexes can be overcome to allow telomere extension.

With the knowledge that G-quadruplexes can and does inhibit telomerase activity, researchers have been able to use G-quadruplex intercalating drugs to treat cancer cells.  Telomestatin, for example, has been shown to inhibit telomerase activity leading to a reduction in telomere length and therefore an initiation of apoptosis in cancer cells (14).  Shammas et al. was able to show that this treatment worked in ARD, MM1S, and ARP myeloma cells by inducing apoptosis (14).

G-quadruplexes function within the telomere: Damage Toleration

G-quadruplexes may also play a critical role in maintaining telomere length by protecting the telomere from external damage.  Telomere sequences are extremely sensitive to oxidative and UV-induced DNA damage because it can alter the telomeric DNA folding pattern.  However, the damage from UV-induced thymine dimer is minimal and does not cause telomere shortening.  This is due to the fact that UV-induced thymine dimers do not stop G-quadruplex formation at the end of telomeres (15).  Even when researchers introduced thymine dimers and thymine glycol directly into the loop of a G-quadruplex, the telomeres only experienced a moderate decrease in thermal stability (15).  This is evidence that even under stress and DNA damage, G-quadruplexes are still able to form.

G-quadruplexes function: 5’ UTRs and gene expression regulation

The 5’ untranslated region (5’ UTR) is the region of RNA located directly upstream of the initiation codon.  What does this have to do with G-quadruplexes and telomeres? Similarly to the importance of G-quadruplexes in telomeres, G-quadruplexes are found at 5’ UTRs and are extremely important in translation and transcription.  In a study of G-quadruplexes located at 5’ UTRs, Huppert et al. found that there was an excess of G-quadruplexes.  This is consistent with the previous studies suggesting that there was a link between 5′-UTR RNA G-quadruplexes and translational control.  In this same study, the researchers also found G-quadruplexes at the 3′-end of genes.  This suggests involvement of G-quadruplexes in termination of gene transcription (16).

In hopes to expand upon the Huppert et al study which showed evidence that there was an excess of G-quadruplexes towards the 5′-ends of 5′-UTRs, Beaudoin et al. used in silico, in vitro and in cellulo experiments to show that G-quadruplexes are translational repressors.  While investigating 5024 different human 5’ UTRs, Beaudoin et al. found a statistically significant presence of G-quadruplexes in a broad variety of genes.  With all experiments done in this study taken into consideration, the researchers concluded that the G-quadruplexes are very important in 5′-UTR sequences.  Specifically they are important to 5′-UTR sequences because they are able to repress translation (17).

Gene suppression as described has been shown to be a fundamental characteristic of G-quadruplexes.  Another study that further demonstrated the translational repressive characteristic of G-quadruplexes was done in 2009.  Halder et al. introduced artificially designed G-quadruplexes onto a report gene.  Then they, using quantitative real-time PCR, showed that G-quadruplexes have an impact on gene expression.  They found a result that suggests a universal ‘translational-suppressor’ effect of not only G-quadruplexes but all non-canonical RNA structures (18).  Halder et al. research was exclusively done with RNA and not DNA and therefore did not show the effects of G-quadruplexes on transcription.

In a more recent study, the G-quadruplex structure located in the 5’UTRs has been characterized as having a long central loop 10-70 nt long.  Jodoin et al. suggested, from this study, that the definition of G-quadruplex for sequencing purposes is too conservative and must be expanded to include these long central loops (19).  This information was adapted, as discussed previously, in the QGRS-Conserves algorithms (11) and has been used in further studies.  Because this is a newer discovery, more research needs to be done in order to see the true impact of expanding the sequencing parameters of G-quadruplex to include the long central loop conformation.

Gene suppression or gene silencing is used to explain epigenetic gene expression regulation.  G-quadruplexes disrupts the structure of conventional B-DNA and this structural alteration leads to gene suppression.   In the context of cancer, gene suppression is extremely important and has been exploited by many researchers in their use of G-quadruplex stabilization as a cancer therapy.  The reason behind the therapeutic advantage of stabilizing G-quadruplexes is still unknown (20).  The therapeutic advantage of targeting G-quadruplexes might be due to the fact that they are gene suppressors intrinsically or because of the role they play at the telomeres of chromosomes.  It may also have therapeutic advantages for neither of these reasons or both of these reasons.

Current use of G-quadruplexes as cancer therapy

G-quadruplexes are currently being used in clinical trials as cancer therapy treatments.  This is possible for various reasons.  As discussed previously in this review, G-quadruplexes can be used in cancer treatments because they directly affect the activity of telomerase.   This is seen in Telomestatin which has been shown to inhibit telomerase activity leading to a reduction in telomere length and an induction of apoptosis in cancer cells (14).  To expand upon this, researchers are currently working out ways in which they can use the gene suppression functionality of G-quadruplexes at the promoter region of oncogenes to suppress the oncogenes and therefore stop or prevent cancer growth.

There are many difficulties that arise from using this type of treatment.  Because G-quadruplexes can be found in various locations throughout the genome, targeted drug development has difficulty targeting a specific G-quadruplex.  As more research is done however, the hope is that conformational differences given rise to by G-quadruplex secondary structures may help overcome these targeting drug development difficulties (21).  An example of these secondary structures that could become useful in the future is the long central loops found in 5’ UTR G-quadruplexes discussed previously.

Conclusion

Although there has been a lot of progress in the study of G-quadruplexes, there are still many unanswered questions.  Though the function of G-quadruplexes can be explained very broadly as being important for regulation through structure and gene expression regulation, the specific functions of various secondary G-quadruplex structures is still not fully understood.  Not all G-quadruplexes secondary structures are the same in fact they are all quite different and so it would be interesting to see what the difference is in function of the various secondary structures.  I would hypothesize that the structure of promoter regions for various genes may be functionally relevant.  As for using G-quadruplexes in cancer research I believe that there is a lot more that has to be done.  There are too many G-quadruplexes throughout the genome to not worry about unwanted genes being suppressed from targeting G-quadruplexes.  I do believe however that using G-quadruplexes to regulate genes and not just to regulate telomerase activity is in the foreseeable future.

References

1. Lu, Weisi et al. “Telomeres – Structure, Function, and Regulation.”Experimental cell research 319.2 (2013): 133–141. PMC. Web.
2. Heaphy, Christopher M. et al. “Prostate Cancer Cell Telomere Length Variability and Stromal Cell Telomere Length as Prognostic Markers for Metastasis and Death.” Cancer discovery 3.10 (2013): 1130–1141. PMC. Web.
3. Phan, Anh Tuân, and Jean-Louis Mergny. “Human Telomeric DNA: G-Quadruplex, I-Motif and Watson–Crick Double Helix.” Nucleic Acids Research30.21 (2002): 4618–4625. PubMed.
4. Tang, Jun et al. “G-Quadruplex Preferentially Forms at the Very 3′ End of Vertebrate Telomeric DNA.” Nucleic Acids Research 36.4 (2008): 1200–1208.PMC. Web.
5. Cáceres, Carmen et al. “A Thymine Tetrad in d(TGGGGT) Quadruplexes Stabilized with Tl+/Na+ Ions.” Nucleic Acids Research 32.3 (2004): 1097–1102. PMC. Web.
6. Brázda, Václav et al. “DNA and RNA Quadruplex-Binding Proteins.” Ed. Genxi Li. International Journal of Molecular Sciences 15.10 (2014): 17493–17517.PMC. Web.
7. DePamphilis, Melvin L., and Stephen D. Bell. “Quadruplex DNA.”Genome Duplication. London: Garland Science, 2011. N. pag. Print.
8. Hoogsteen K (1963). “The crystal and molecular structure of a hydrogen-bonded complex between 1-methylthymine and 9-methyladenine”. Acta Crystallographica 16: 907–916.
9. Sen D, Gilbert W. “Formation of parallel four-stranded complexes by guanine-rich motifs in DNA and its implications for meiosis.” Nature. 1988 Jul 28;334(6180):364-6. PubMed
10. Adrian M, Heddi B, Phan AT. “NMR spectroscopy of G-quadruplexes.” Methods. 2012 May;57(1):11-24. doi: 10.1016/j.ymeth.2012.05.003. Epub 2012 May 24. Review. PubMed
11. Frees, Scott et al. “QGRS-Conserve: A Computational Method for Discovering Evolutionarily Conserved G-Quadruplex Motifs.” Human Genomics 8.1 (2014): 8. PMC. Web.
12. Lam, Enid Yi Ni et al. “G-Quadruplex Structures Are Stable and Detectable in Human Genomic DNA.” Nature communications 4 (2013): 1796. PMC. Web.
13. Zahler AM, Williamson JR, Cech TR, Prescott DM. “Inhibition of telomerase by G-quartet DNA structures.” Nature. 1991;350:718–720.  PubMed
14. Shammas MA, Shmookler Reis RJ, Li C, Koley H, Hurley LH, Anderson KC, Munshi NC. “Telomerase inhibition and cell growth arrest after telomestatin treatment in multiple myeloma.” Clin Cancer Res. 2004 Jan 15;10(2):770-6. PubMed
15. Wolna AH, Fleming AM, Burrows CJ. “Single-molecule analysis of thymine dimer-containing G-quadruplexes formed from the human telomere sequence.” Biochemistry. 2014 Dec 9;53(48):7484-93. doi: 10.1021/bi501072m. Epub 2014 Nov 26. PubMed PMID: 25407781; PubMed
16. Huppert, Julian Leon et al. “G-Quadruplexes: The Beginning and End of UTRs.”Nucleic Acids Research 36.19 (2008): 6260–6268. PMC. Web.
17. Beaudoin, Jean-Denis, and Jean-Pierre Perreault. “5′-UTR G-Quadruplex Structures Acting as Translational Repressors.” Nucleic Acids Research 38.20 (2010): 7022–7036. PMC. Web.
18. Halder, Kangkan, Markus Wieland, and Jörg S. Hartig. “Predictable Suppression of Gene Expression by 5′-UTR-Based RNA Quadruplexes.”Nucleic Acids Research 37.20 (2009): 6811–6817. PMC. Web.
19. Jodoin, Rachel et al. “The Folding of 5′-UTR Human G-Quadruplexes Possessing a Long Central Loop.” RNA 20.7 (2014): 1129–1141. PMC. Web.
20. Hirashima, Kyotaro, and Hiroyuki Seimiya. “Telomeric Repeat-Containing RNA/G-Quadruplex-Forming Sequences Cause Genome-Wide Alteration of Gene Expression in Human Cancer Cells in Vivo.” Nucleic Acids Research43.4 (2015): 2022–2032. PMC. Web.
21. Yang D, Okamoto K. “Structural insights into G-quadruplexes: towards new anticancer drugs.” Future Med Chem. 2010 Apr;2(4):619-46. Review. PubMed PMID: 20563318; PubMed