PTENP1-AS contributes to BRAF inhibitor resistance and is associated with adverse clinical outcome in stage III melanoma

Approximately 50% of human cutaneous melanomas carry activating mutations in the serine/threonine protein kinase BRAF. BRAF inhibitors (BRAFi) selectively target the oncogenic BRAFV600E/K and are effective in approximately 80% of patients carrying the mutation. However, resistance to BRAFi is common and emerges within a median time of 6-7 months of treatment and is prolonged to 11 months when combined with MEK inhibitors. Better characterization of the underlying molecular processes is therefore needed to further improve treatments. Inactivation of the tumor suppressor gene PTEN has been suggested to occur during melanomagenesis and drug resistance development. We recently demonstrated that transcription of PTEN is negatively regulated by an antisense RNA from the PTEN pseudogene (PTENP1-AS) and here set out to investigate the impact of this molecular pathway on the resistance to BRAFi and clinical outcome. We used a panel of BRAFi resistant A375 sublines, and observed increased levels of PTENP1-AS associated with reduced expression of PTEN. Furthermore, this loss of PTEN expression was correlated to increased recruitment of Enhancer of zeste homolog 2 (EZH2) and formation of the transcriptional repression mark H3K27me3 at the PTEN promoter in the resistant cells. We demonstrated that targeting of PTENP1-AS was able to re-activate the expression of PTEN and sensitize resistant melanoma cells to BRAFi. Finally, we showed that PTENP1-AS is a promising prognostic marker for clinical outcome in melanoma patients as high expression of PTENP1-AS in regional lymph node metastases from stage III melanoma patients correlated with poor survival.


ABSTRACT
Approximately 50% of human cutaneous melanomas carry activating mutations in the serine/threonine protein kinase BRAF. BRAF inhibitors (BRAFi) selectively target the oncogenic BRAF V600E/K and are effective in approximately 80% of patients carrying the mutation. However, resistance to BRAFi is common and emerges within a median time of 6-7 months of treatment and is prolonged to 11 months when combined with MEK inhibitors. Better characterization of the underlying molecular processes is therefore needed to further improve treatments. Inactivation of the tumor suppressor gene PTEN has been suggested to occur during melanomagenesis and drug resistance development. We recently demonstrated that transcription of PTEN is negatively regulated by an antisense RNA from the PTEN pseudogene (PTENP1-AS) and here set out to investigate the impact of this molecular pathway on the resistance to BRAFi and clinical outcome. We used a panel of BRAFi resistant A375 sublines, and observed increased levels of PTENP1-AS associated with reduced expression of PTEN.
Furthermore, this loss of PTEN expression was correlated to increased recruitment of Enhancer of zeste homolog 2 (EZH2) and formation of the transcriptional repression mark H3K27me3 at the PTEN promoter in the resistant cells. We demonstrated that targeting of PTENP1-AS was able to re-activate the expression of PTEN and sensitize resistant melanoma cells to BRAFi.
Finally, we showed that PTENP1-AS is a promising prognostic marker for clinical outcome in melanoma patients as high expression of PTENP1-AS in regional lymph node metastases from stage III melanoma patients correlated with poor survival.

BACKGROUND
Although cutaneous malignant melanoma is a molecularly diverse disease, approximately 50% carry activating mutations in the serine/threonine protein kinase BRAF. The majority of BRAF mutations are represented by a valine (V) to glutamic acid (E) or lysine (K) substitution at position 600 (BRAF V600E/K ) 1 . These missense mutations occur in the BRAF kinase domain, which results in constitutively active BRAF and subsequently activated MAPK signaling.
Recent drug development efforts using a targeted approach have successfully led to the development of small inhibitory molecules (BRAFi), which efficiently and specifically target the oncogenic BRAF V600E/K . Vemurafenib represents one such inhibitor and 80% of patients with advanced melanoma carrying a BRAF V600E/K mutation initially respond well to this treatment and show tumor regression 2 . However, the development of resistance to BRAFi is a major issue and emerges within a median time of 6-7 months of treatment which can be prolonged to 11 months when combined with MEK inhibitors. The 5-year overall survival rate is only 34% for the combination treatment 3 . In recent years there has been a growing understanding of the underlying molecular mechanisms involved in acquired BRAFi resistance 4,5 . Activation of the PI3K/AKT pathway due to loss of PTEN has been shown to be one of the mechanisms contributing to BRAFi resistance 6 . Thus, dissecting the mechanisms of PTEN loss is vital in order to design novel approaches for treatment, which may circumvent or delay the onset of resistance.
Approximately 14,000 pseudogenes have been identified in the human genome and hundreds of those shown to be transcribed into non-coding RNAs (ncRNAs) 7 . Transcribed pseudogenes have been suggested to be capable of acting as regulators of gene expression 8-10 , for instance through mechanisms such as microRNA (miRNA) sponging and trans acting antisense RNA (asRNA) regulation. One of these, the PTEN pseudogene, PTENP1 (also known as PTENpg1, PTENΨ), has been shown to be transcribed into both sense and asRNA transcripts. While the sense transcript (PTENP1-S) functions as a positive regulator of PTEN by acting as a miRNA sponge for PTEN related miRNAs 10 , the asRNA transcript (PTENP1-AS) functions as a negative regulator of PTEN expression through inducing epigenetic alterations at the PTEN promoter 8 . PTEN is a tumor suppressor gene which is frequently inactivated across a wide range of cancers such as breast cancer, prostate cancer and melanoma (reviewed in 11 ). Loss of PTEN expression has been reported to play a role not only in the development of resistance to BRAFi 6 , but also in the process of melanoma metastasis 12 . Moreover, mutated BRAF itself is not sufficient to induce melanoma, other molecular events such as inactivation of PTEN 12,13 , are required for melanoma initiation.
In this study, our aim was to study the involvement of PTENP1-AS in melanoma tumors and in resistance to the BRAFi vemurafenib. We observe that the expression of PTENP1-AS is induced in BRAFi resistant cell lines, which coincides with transcriptional suppression of PTEN. We further show that targeting of the PTENP1-AS transcript sensitizes resistant melanoma cells to vemurafenib. Mechanistically, we conclude that up-regulation of the transcription factor C/EBPβ initiates transcription of PTENP1-AS and that PTEN is suppressed through the recruitment of EZH2 to the PTEN promoter with subsequent formation of H3K27me3. Finally, we also demonstrate that expression of PTENP1-AS is increased in tumor samples from stage III melanoma patients with poor survival.

Inverse correlation of PTEN and PTENP1-AS in BRAFi resistant A375 sublines
Previous studies suggested downregulation of PTEN to be an important mechanism for the development of resistance to BRAFi 6,12 . Based on these reports, we hypothesized that the  (Figures 1B, S1B). In summary, these initial observations motivated us to further elucidate the underlying molecular events mediating suppression of PTEN.
We next sought to determine the involvement of PTENP1 encoded lncRNAs in this process.
To this end, we measured the levels of sense and antisense transcripts and found PTENP1-AS levels to be significantly upregulated in all three resistant sublines (Figures 1C, S1C). Notably, we also observed a modest, although non-significant, induction of PTENP1-S (Figures 1D,   S1D). This is, however, in contrast to the previously reported miRNA-sponge function of PTENP1-S, since such a mechanism would result in increased expression of PTEN due to the release of miRNA-mediated suppression of PTEN. We therefore considered PTENP1-AS to hold the main regulatory function under these conditions. Previously, we have reported multiple isoforms of PTENP1-AS with distinct regulatory functions 8 . We selected the A375PR1 subline for further characterization of isoforms using a total of four different primer sets. A concurrent induction was observed for all primer sets ( Figure 1E). Notably, the level of unspliced form of PTENP1-AS was found to be induced in the resistant cells, indicating that induction of PTENP1-AS takes place at the transcriptional level.

PTENP1-AS suppresses PTEN in vemurafenib resistant cells through chromatin remodelling
To investigate the molecular interplay between PTENP1-AS and PTEN in greater detail, we next designed a gapmer antisense oligonucleotide (ASO) (Figures 2A-B) as well as an siRNA targeting the PTENP1-AS transcript (Figures 2C-D). Both approaches decreased the amounts of PTENP1-AS with more than 60% in A375 as well as in A375PR1 cells. Intriguingly, knockdown of PTENP1-AS only induced the expression of PTEN in the A375PR1 cells, while PTEN remained unaffected in the vemurafenib sensitive A375 cells (Figures 2A-D), suggesting that PTENP1-AS is only involved in suppression of PTEN in the resistant cells.
To further investigate the function of PTENP1-AS, we next asked if the resistant A375PR1 subline had acquired epigenetic changes at the PTEN promoter. We first assessed DNA methylation using the McrBc enzyme that cleaves methylated DNA at Pu m CG sequence elements. Elevated DNA methylation was demonstrated by increased enzymatic cleavage at the PTEN promoter in the resistant A375PR1 subline compared to parental A375 cells ( Figure   3A). Since PTENP1-AS has previously been shown to recruit the Enhancer of zeste homolog 2 (EZH2) and DNA methyltransferase 3a (DNMT3A) to the PTEN promoter 8 , we next Taken together, these data suggest that suppression of PTEN is mediated by PTENP1-AS through the recruitment of chromatin remodeling factors utilizing a mechanism that may be exclusively active in the vemurafenib resistant cells.

C/EBPβ is a transcriptional regulator of PTENP1-AS
To explore the molecular events underlying transcriptional induction of PTENP1-AS in the BRAFi resistant sublines, we took advantage of FANTOM5 CAGE data and ChIP-seq data from the UCSC Genome Browser, and identified enrichment of the transcription factor C/EBPβ upstream of the PTENP1-AS transcriptional start site (TSS) (Figure 5A). C/EBPs can homodimerize or heterodimerize and bind to the consensus DNA sequence RTTGCGYAAY 15 ,

PTENP1-AS is a clinically relevant target and a prognostic marker in melanoma
We next explored if the resistant A375PR1 cells could be sensitized to BRAFi by targeting the PTENP1-AS transcript. Notably, gapmer ASO-induced knockdown of PTENP1-AS resulted in increased induction of apoptosis to 10 μM of vemurafenib in the A375PR1 cells ( Figure 6A).
We finally set out to investigate the relevance of PTENP1-AS for clinical outcome in melanoma. The expression of PTENP1-AS was first determined in lymph node metastases from a cohort of 20 stage III melanoma patients prior to the treatment. The samples were initially chosen to include two groups of patients, with either long (≥ 60 months) or short (≤13 months) overall survival 16 , and the expression of PTENP1-AS was determined by qRTPCR. Notably, 9/10 (90%) patients having tumors with high expression of PTENP1-AS (median cut-off) had poor survival, while low expression of PTENP1-AS seemed to be indicative of prolonged survival, although not reaching significance (p=0.06) ( Figure 6B). This motivated us to evaluate the expression of PTENP1-AS in an independent second cohort of stage III melanoma samples. Lymph node metastases from 29 stage III melanoma patients were analyzed for the expression of PTENP1-AS. The patient samples from the second cohort were divided in high (n=12) or low (n=17) expression of PTENP1-AS, as determined by qRTPCR, and a survival plot was generated ( Figure 6C). A significant difference in overall survival was observed between patients with high and low expression of PTENP1-AS. In summary, the expression of PTENP1-AS appears to be a promising prognostic marker for clinical outcome, where increased expression of PTENP1-AS correlates with poor overall survival.

DISCUSSION
The data presented in this study suggest a role for the PTENP1-AS in the development of Notably, this study also shows that the suppression of PTEN in BRAFi resistant cells may be reversible through targeting of EZH2, DNMT3A as well as PTENP1-AS. The, individual knockdown of EZH2 or DNMT3A alone did not re-activate the expression of PTEN to the same extent as co-knockdown of these proteins (Figures 3B-D). We believe this is due to incomplete knockdown of DNMT3A and EZH2 where some active protein complexes may still be present under these circumstances (Figure S2). An interaction between EZH2 and DNMT3a have been demonstrated 21 and the combination of EZH2 and DNMT3A targeting may in this case be more efficient in reducing the protein complexes more completely. Importantly, the induction of PTEN is more prominent in the A375PR1 cells than in the parental cells, indicative of a therapeutic window for re-activation of PTEN in BRAFi resistant melanoma cells. In support to this, we show that manipulation of the PTENP1-AS pathway re-sensitized the A375PR1 resistant cells to treatment with vemurafenib ( Figure 6A).
We finally evaluated the expression of PTENP1-AS in regional lymph node metastases of stage III melanoma and found that high expression correlated with poor survival (Figure 6C). This indicates that PTENP1-AS is involved in melanoma tumor progression and is not only important during the development of resistance to BRAFi. Also, development of a particular type of resistance to vemurafenib may depend on the pre-existing factors/pathways in the tumors. High expression of PTENP1-AS could for example indicate that the tumor is less likely to respond to treatment. Moreover, loss of PTEN expression has previously been linked to metastasis 12 . Therefore, one may envision PTENP1-AS being involved in such inactivation and consequently enhance initiation of distant metastasis. Additional studies will be required in order to better understand the functional role and importance of PTENP1-AS in melanoma progression and drug resistance by taking patient samples pre-and post-treatment with BRAFi.
In summary, we have identified PTENP1-AS to be involved both in melanoma tumor progression and in development of resistance to vemurafenib. The data presented here show that PTENP1-AS is not only a promising target for re-activation of PTEN, but also a possible prognostic marker for clinical outcome in stage III melanoma.

RNA extraction and cDNA
RNA was extracted using the RNA NucleoSpin II kit (Macherey-Nagel) and treated with DNase (Ambion Turbo DNA-free, Life Technologies). DNase treated RNA (~500 ng) was used for the generation of cDNAs using MuMLV (Life Technologies) and a mixture of oligo(dT)15 with nanomers.

Semi-qRTPCR
PCR was performed by using the KAPA2G FAST mix (Kapa Biosystems) according to the manufacturer's recommendations and by using the corresponding oligos in supplementary

ChIP of EZH2, H3K27me3 and C/EBPβ
ChIP assays were performed as previously described 8 . Briefly, the ChIP assay Kit (Upstate/Millipore) was used by crosslinking the cells in 1% formaldehyde for 10 minutes, quenched in 0.125M Glycine for 5 minutes and lysed according to the manufacturer's recommendations. The samples/nuclei were sonicated with a Bioruptor Sonicator (Diagenode) at 30sec ON, 30sec OFF (setting = high) for a total of 18 cycles. The water was replaced with ice-cold water after every sixth cycle. The samples were diluted 1:10 in ChIP dilution buffer, pre-cleared and incubated over night with the appropriate antibody. Salmon sperm DNA/Protein A-agarose (Upstate/Millipore) was used to pull down the antibody. The DNA was eluted in elution buffer (1% SDS, 100 mM NaHCO3), followed by reverse cross-linking at 65°C over night. The samples were RNase-A and protease-K treated and finally eluted using the Qiagen PCR purification kit (Qiagen). The following antibodies were used (4 μg/sample): H3K27me3 (Upstate/Millipore, cat. no. 17-622), EZH2 (Upstate/Millipore, cat. no. 07-689) and C/EBPβ (Santa Cruz Biotechnology cat. No sc-150).

Assessment of methylated DNA
The enzyme McrBc (New England Biolabs) was used to cleave methylated DNA. Briefly, 200 ng DNA was digested with McrBc at 37°C overnight and heat inactivated the next day at 65° for 1 h. Samples were run on a qPCR and standardized to uncut input. Delta CT values were converted to fold-change values and the ratio between A375 PR1/A375 calculated.

PI-annexin V staining
The cells were harvested, washed twice in PBS and resuspended in 100 μL annexin V incubation buffer (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 5 mM CaCl2) containing 1% annexin V FLOUS (Roche Molecular Biochemicals) and 500 μg/μl PI stain. The samples were incubated for 15min at room temperature followed by adding 400 μL of ice-cold annexin V incubation buffer and subsequently analyzed on a cytometry machine.

Patient samples
Samples from stage III melanoma patients with first regional lymph node metastases that had not received any systemic treatment were collected. The RNA was extracted using the Qiagen RNeasy mini kit (Qiagen). The RNA was DNase treated on column according to the manufacturer's recommendations. The expression of PTENP1-AS was analyzed using qRTPCR and standardized to beta-actin (dCt) by using the corresponding oligos in Supplementary Table 1. ddCt was calculated by using the mean Ct value of all samples. High expression of PTENP1-AS ( Figure 7B) was defined as Ct < 30.

Statistics
Two tailed Student's T-test was used to determine statistical significance. Error bars represent the standard error of the mean. Statistical significance of the melanoma patient samples in figure 7B is evaluated using log-rank test. Throughout the paper, * P < 0.05; ** P < 0.01; *** P < 0.005.
wrote the manuscript with the support from A.A, S.E.B, J.H and K.P.T. D.G and P.J supervised the project.

COMPETING INTEREST
The authors declare no competing interests.

Ethical approval and consent to participate
This study has been approved by the Research Ethics Committee of Karolinska Institutet (Dnr: 2006/1373-31/3) and the Regional Ethics Committee in Lund (Dnr 2013/101). All involved individuals have provided informed consent to be included in the study.     qRTPCR presenting the expression of PTENP1-AS in a set of 20 first regional lymph node metastases from stage III melanoma patients. Based on clinical follow up data, the patients were categorized as long-or short-term survivors, >60 months or <13 months. (C) The expression of PTENP1-AS was determined in an independent set of 29 stage III melanoma patients with first regional lymph node metastases. The patients were divided according to high (n=12) or low (n=17) expression of PTENP1-AS and overall survival was analyzed in a Kaplan-Meier plot. The p-value represents a log-rank test.