Targeting of G-Quadruplex Harboring Pre-miRNA 92b by LNA

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Targeting of G-Quadruplex Harboring Pre-miRNA 92b by LNA Rescues PTEN Expression in NSCL Cancer Cells Gayan Mirihana Arachchilage, Prakash Kharel, Joshua Reid, and Soumitra Basu ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.7b00749 • Publication Date (Web): 12 Mar 2018 Downloaded from http://pubs.acs.org on March 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

Targeting of G-Quadruplex Harboring Pre-miRNA 92b by LNA Rescues PTEN Expression in NSCL Cancer Cells

Gayan Mirihana Arachchilage1,#, Prakash Kharel1, Joshua Reid1 and Soumitra Basu1,*

1

Department of Chemistry and Biochemistry, Kent State University, Kent, Ohio, 44242, USA

#

Current Address: Howard Hughes Medical Institute, Department of Molecular, Cellular and

Developmental Biology, Yale University, New Haven, Connecticut, 06511, USA

* Corresponding author Email : [email protected]

1 ACS Paragon Plus Environment

ACS Chemical Biology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Since the elevated levels of microRNAs (miRNAs) often cause various diseases, selective inhibition of miRNA maturation is an important therapeutic strategy. Commonly used anti-miRNA strategies are limited to targeting of mature miRNAs, as the upstream targeting of miRNA maturation with an oligonucleotide is challenging due to the presence of stable pre-miRNA stem-loop structure. Previously, we reported that about 16% of known human pre-miRNAs have the potential to adopt G-quadruplex (GQ) structures alternative to canonical stem-loops. Herein, we showed a rationally designed locked nucleic acid (LNA) binds specifically the GQ conformation of pre-miRNA 92b and inhibits pre-miRNA maturation. Further, we showed that the LNA treatment rescues PTEN expression in non-small cell lung cancer (NSCLC) cells which is suppressed by the elevated level of miRNA 92b. Treatment of LNA significantly decreases the IC50 of doxorubicin for NSCLC cells. This strategy can be developed as a novel anti-pre-miRNA therapeutic approach to target GQ harboring miRNAs. This can potentially be a more powerful approach than targeting of the mature miRNA, as it is an upstream targeting and can reduce both 3′ and the 5′ mature miRNA levels at once.


Page 2 of 23

Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


MicroRNAs are short RNA sequences that are capable of regulating human protein coding gene expressions.1, 2 Mature miRNAs downregulate multiple gene expressions mostly at the translational level.3 Biogenesis of microRNA is a tightly regulated process as elevated levels are often linked to various diseases including neurodegenerative diseases and cancer.4, 5 Myriad studies have shown that elevated levels of mature miRNA cause the downregulation of expression of important genes which can eventually lead to diseases.6, 7 Thus, approaches to modulate miRNA levels inside cells have gained increased attention as possible therapeutics.8, 9

Several strategies have been utilized to develop therapeutics to decrease the level of mature miRNAs, such as the introduction of expression vectors (miRNA sponges)10, small-molecule inhibitors11, 12, and anti-sense oligonucleotides13. Among those strategies, considerably more attention has been paid to anti-sense miRNAs (anti-miRs) as they bind with high complementarity to miRNA-induced silencing complex (miRISCs) and block their binding to endogenous mRNA targets.14 In contrast to the strategies that block mature miRNA binding, a few studies have attempted to block the biogenesis of miRNA as an upstream targeting event by using small molecules, 15 oligonucleotides or peptide nucleic acids (PNA).16, 17 However, almost all oligonucleotide-based premiRNA targeting approaches have focused on targeting the canonical stem-loop conformation.18 It is believed that all pre-miRNAs adopt a stem-loop conformation but their lengths and thermal stabilities can vary.19 A majority of the nucleotides within premiRNA stem-loops are already base paired and thus are not available for new base



pairings, except at the loop region. A complementary oligonucleotide strand may invade and bind to the stem-loop structure only if it can compete with the stem-loop base pairing.16 However, such a strategy may not be very efficient if the stem region is long with a high degree of thermodynamic stability.

MicroRNA coding genes are transcribed into primary microRNAs (pri-miRNAs) by enzymatic action of RNA polymerase II or III enzymes.1 Pre-miRNAs are originated inside the nucleus after the cleavage of pri-miRNAs by the enzyme complex Drosha.1 These pre-miRNAs adopt a stem-loop structure and get exported into the cytoplasm for further processing. It is well known that the structural features of pre-miRNA stem-loops are recognized by the enzyme Dicer in the cytoplasm and cleaved internally to produce mature miRNAs.20 Recently, we discovered that approximately 16% of known human pre-miRNAs have the potential to form G-quadruplex (GQ) structures alternative to canonical stem-loop structure.20 Using pre-miRNA 92b, a clinically important miRNA, we demonstrated that the GQ formation modulates formation of the canonical stem-loop structure and the equilibrium between the stem-loop and the GQ structure regulates Dicer-mediated maturation of pre-miRNA both in vitro and in vivo. Subsequently, Pandey et al also showed that the GQ formation controls mature miRNA production inside cells.21 These findings provided a platform for developing new strategies to target pre-miRNAs in order to inhibit their maturation.

Several studies have used PNA, RNA and DNA oligonucleotides to target GQ structures and induce or inhibit their formation.22-24 In this study, we developed a novel approach to inhibit maturation of GQ harboring miRNAs using a rationally designed 4 ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


oligonucleotide. In our previous study, we showed that GQ formation of pre-miRNA 92b disrupts the stem-loop and leaves an ~ 60 nt long single stranded segment near the 3′ end. This single stranded region was utilized to design an oligonucleotide which can specifically bind the GQ containing pre-miRNA 92b. To enhance the thermodynamic stability and the sustainability inside the cells we used a Locked Nucleic Acid (LNA) to target the GQ conformation. A number of studies have shown that LNAs can be used more effectively as anti-sense miRNAs than DNA, RNA or other common modifications due to the high thermodynamic stability in terms of complementary base pairing.16, 25-27 Using pre-miRNA 92b, which has a very long stem, resulting in a thermodynamically stable stem-loop, we showed that a rationally designed LNA can be used to inhibit the maturation of pre-miRNA 92b and rescues the PTEN tumor suppressor gene expression in human non-small cell lung cancer (NSCLC) cells 28 a strategy that can potentially be developed further to treat NSCLC. Since, approximately 16% of premiRNAs are predicted to fold into GQ structures, this approach to inhibit miRNA biogenesis will likely have a broad impact on the development of anti-miRNA therapeutics.

First, we designed three different DNA oligonucleotide sequences (DNA-ON 1 through 3) to bind different sites in the single stranded region of the pre-miRNA 92b GQ conformation. In order to investigate the best binding site that can be used to change the equilibrium more towards the GQ conformation, a gel mobility shift assay was performed with pre-miRNA 92b in the presence of various concentrations of DNA-ONs. DNA-ON3 which binds to the last 16 nucleotides was observed to change the equilibrium better than other DNA-ONs (Supplementary Figure S1). Then we designed a locked nucleic acid oligonucleotide (LNA-ON) 5 ACS Paragon Plus Environment


sequence analogue to bind to the last 16 nt at the 3′ end (Supplementary Table S1). Since the last 16 nt at the 3′ end encompass the complementary region of the first four G-stretches (Figure 1A), we presumed that by providing an LNA-ON which can competitively bind to this region, will release the first four G-stretches from being engaged in the stem-loop and facilitate the GQ formation. Our previous studies have shown that both the GQ and stem-loop

conformations are thermodynamically very stable. It has been well studied that an LNA:RNA duplex is more stable than DNA:RNA or RNA:RNA duplexes due to its modified chemical structure (Figure 1B).26 Since the stem-loop is made up of RNA:RNA base pairings, we hypothesized that LNA oligonucleotide can stabilize the GQ conformation by preventing the sequence from folding back to the stem-loop form.

To validate the binding site of LNA-ON in pre-miRNA 92b, RNase T1 footprinting was performed (Supplementary Figure S2). The binding site of LNA-ON at the 3′ end of premiRNA 92b was protected from RNase T1 cleavage in the presence of LNA-ON indicating its specific binding to the complementary site. Interestingly, LNA-ON remained bound to pre-miRNA 92b even under denaturing conditions which suggests the strong interactions between LNA and RNA. Next, a gel mobility shift assay was performed in the presence of LNA-ON to assess the effect of LNA-ON on the equilibrium. Physiologically relevant K+ concentration (100 mM) was used to facilitate the attainment of equilibrium condition (Figure 1C). In the absence of LNA-ON treatment, both the GQ and the stem-loop structures were observed, similar to what we had reported in our previous study.20 The LNA-ON treatment shows the appearance of an additional slowly migrating band above the band corresponding to the GQ, indicating the LNA-ON bound GQ species which should be heavier in molecular weight and slower


Page 6 of 23

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in migration than the GQ conformation. However, no additional band was observed in the absence of K+ ions even at the highest concentration of LNA-ON (100 nM), indicating that LNA-ON does not bind to the stem-loop directly. Thus, the gel mobility shift assay shows that LNA-ON specifically binds to the GQ containing form of the premiRNA 92b but not to the stem-loop structure.

Densitometric analysis shows that the stem-loop population is approximately 2.5 fold higher than the GQ population before the LNA-ON treatment (Figure 1D). In the presence of LNA-ON, the stem-loop population decreases while the GQ population (GQ and LNA-ON bound GQ species) increases in a concentration-dependent manner. At 100 nM LNA-ON, the stem-loop population was approximately 2.5 fold lower than the GQ population suggesting that the LNA-ON binding can reverse the equilibrium and significantly lower the stem-loop population. It should be noted that the DNA version of the LNA-ON (DNA-ON3) failed to achieve such a significant change in equilibrium even at a much higher concentration (5 µM). Interestingly, the band corresponding to the GQ conformation shifted up at higher concentrations of LNA-ON. We speculate this might be due to the formation of intermolecular GQ between the G-rich LNA-ON and premiRNA 92b, but these additional species require further investigations to arrive at a more definitive conclusion. Nevertheless, only the stem-loop population facilitates Dicermediated maturation of pre-miRNA 92b. Collectively, these results strongly indicate that LNA-ON treatment can significantly decrease the levels of the stem-loop population by shifting the equilibrium more towards the GQ conformation.



This capability of LNA to prevent stem-loop structure formation shows the advantage of LNA over the DNA or RNA for anti-miRNA oligonucleotide therapeutics. A miRNA, like miRNA 92b, which has a long and thermodynamically very stable stem-loop cannot simply be perturbed by a complimentary sequence. For example, our gel mobility shift data show no binding of LNA-ON, even at the highest concentration when only the stem-loop conformation of the target is available (Figure 1C). It should be noted that the absence of K+ ions precludes GQ formation and therefore favor the canonical stem-loop conformation. On the other hand, GQ formation provides a platform for the binding of LNA-ON by providing a single stranded target region. Since, our previous bioinformatics studies showed that approximately 16% of human pre-miRNAs have the potential to form GQ structures, targeting these GQ conformations with LNA can have a broader impact on the therapeutic development for various diseases where GQ harboring miRNAs are upregulated.

Since the gel mobility shift assay suggested that the LNA-ON treatment can decrease the levels of stem-loop population, we next investigated whether the LNA-ON treatment can decrease the maturation of miRNA 92b in A549 cells, a human non-small cell lung cancer (NSCLC) cell line. It has been reported that miR-92b is overexpressed in human NSCLC cells.28 Since only the stem-loop conformation undergoes Dicer-mediated maturation, decrease in the stem-loop population should result in a decrease in the maturation of miR-92b. After transfection of A549 cells (a NSCLC cell line) with 100 nM of LNA-ON, the levels of endogenous mature miR-92b were quantified by RT-qPCR. The same concentration of a scramble LNA sequence was transfected separately and


Page 8 of 23

Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


served as a control. After normalizing to U6 snRNA, mature miR-92b levels were measured to be significantly (~6 fold, P

Targeting of G-Quadruplex Harboring Pre-miRNA 92b by LNA

Recommend Documents