Background
Molecular therapy targeting the BRAF
V600E mutation and anti-CTLA-4 antibody-based immunotherapy have shown some benefit for patients with advanced melanoma. However, in light of the fact that advanced melanoma, rapidly and very aggressively, mounts resistance to BRAF small-molecule inhibitor treatment; anti-CTLA-4 immunotherapy has efficacy in less than 10-20% of melanoma patients; and advanced melanomas harbor neither a large number of gene fusions or chimeric transcripts[
1] nor with the exception of the BRAF
V600E mutation, a high rate of mutations[
2,
3], the pertinent challenge regarding this disease continues to be the identification of genes that are upregulated to high levels in advanced melanoma, and when inhibited in their function, cause primary and metastatic melanoma cells to undergo apoptosis.
In line with our long-term effort to identify genes that are upregulated with progression from early to advanced melanoma, we recently profiled archived, formalin-fixed paraffin-embedded tissue samples representing early as well as advanced melanoma on DASL BeadChip arrays. The data from this whole-genome expression analysis revealed that one of the identified genes, upregulated to substantial levels with progression from noninvasive melanoma in situ > to invasive radial growth phase melanoma > primary melanoma > metastatic melanoma is the gene DDX11, which has never before been associated with melanoma.
First isolated as the human homologue of the yeast CHL1 gene[
4,
5], DDX11 (alias ChlR1) is a member of the DEAD/DEAH box family of helicases, which comprises more than 40 members. Sharing sequence similarity with the FANCJ helicase and the DEAH-box helicases, XPD and RTEL[
6], DDX11 is essential for the cohesion of chromosome arms and centromeres and when depleted, mitotic failure occurs due to replicated chromosomes failing to segregate after prometaphase arrest[
7]. More recently, biallelic mutations in DDX11 have been identified as the cause of the Warsaw breakage syndrome cohesinopathy, which among other clinical manifestations is associated with abnormal skin pigmentation[
8].
The findings of our study, summarized herein, demonstrate that DDX11 is expressed at high levels in primary and metastatic melanomas, but not in melanocytes of normal skin, atypical nevi, or melanoma in situ, and that suppressing DDX11 expression in advanced melanomas leads to severe defects in chromosome segregation, and with potential relevance to therapeutic intervention, inhibition of melanoma cell proliferation and rapid melanoma cell death.
Discussion
The findings of the study presented herein provide evidence that the DEAD/H (Asp-Glu-Ala-Asp/His) box helicase, DDX11, is upregulated with progression from early to advanced melanoma, and that this gene plays a pivotal role in shielding advanced melanomas from chromosome segregation defects and apoptosis.
Overshadowed for many years by efforts to find efficacious immunotherapies for advanced melanoma, and more recently by the focus on molecular therapy to target BRAF-mutated melanomas, considerably less attention has been paid to genes that are not only upregulated to high levels in advanced melanoma, but also play a pertinent role in protecting VGP and MGP melanomas from apoptosis. In a recent study[
13], we documented by way of molecular targeting of the cell cycle regulator, CDK2, that VGP and MGP melanoma cells are highly vulnerable to interference with their progression through S phase. The data, summarized herein, provide not only further and strong support for this observation, but also demonstrate that inhibiting expression of a helicase such as DDX11, which has a vital function in sister chromatid cohesion[
7,
9,
11], is deleterious for melanoma cells.
Thus far, little is known regarding the involvement, function, and importance of helicases in the progression from early to advanced melanoma, and their role in locally advanced and/or stage IV melanoma. Melanoma differentiation antigen 5 (MDA5), which comprises a caspase activation and recruitment domain (CARD) and an RNA helicase domain with ATP-dependent RNA-unwinding activity, was first isolated from a melanoma cell line[
14]. However, induced by Interferon-β (IFN-β), MDA5 is not expressed in cells representing advanced melanoma unless the cells are treated with the cytokine. A second helicase, recently shown to be expressed in a subpopulation of melanoma cells, referred to as ABCB5+ malignant melanoma-initiating cells, is HAGE (alias Cancer/testis antigen 13 (CT13); DDX43)[
15]. HAGE was shown to promote proliferation and tumor growth of this subpopulation of melanoma cells, and to regulate AKT and ERK phosphorylation through NRAS[
15].
The novel and important findings regarding the herein described pivotal role of DDX11 in advanced melanoma is that following inhibition of DDX11 expression, the cells not only exhibited a significantly higher number of chromosomes with partially closed as well as open/separated arms, but also that compared with the control, the average length of their chromosomes was shorter. To date, little if anything is known about how VGP and MGP melanoma cells guard against DNA damage, control and maintain their genome stability, and related to these survival processes, retain telomere length. In the context of a study published a few years ago[
16], a hypothesis was put forward but not tested that DDX11 might be involved in telomere length determination. Recently, however, it has been documented that loss of DDX11 leads to changes in telomeric chromatin formation[
17], and that DDX11 interacts with the flap structure-specific endonuclease 1 (FEN-1) gene[
9], which has a vital role in telomere stability. Thus, it is possible that like DDX39, which when overexpressed leads to progressive telomere elongation and to telomere shortening when depleted[
18], DDX11 has an important function in maintaining telomere length and stability in a malignancy such as advanced melanoma. The second and even more pertinent finding described herein is that inhibition of DDX11 expression leads to rapid and massive melanoma cell apoptosis. In the context of mouse mutant studies, it has been shown that loss of Ddx11 causes apoptosis[
11,
19], but this is the first study which shows that inhibiting DDX11 expression in a malignancy that is refractory to virtually all apoptosis-inducing agents/therapies, leads to rapid and massive programmed cell death.
The biochemical functions of DDX11 have been established[
20], but hitherto, a DDX11-specific small-molecule inhibitor is not available that would make it possible to systemically treat human melanoma xenografts and establish in vivo, therapeutic efficacy of blocking the function of DDX11. Given the possibility that like in the case of the Werner syndrome helicase[
21], a human DDX11-specific small-molecule inhibitor will be isolated in the near future, it will be of importance to determine whether systemic therapy with such an inhibitor, alone or in combination with an inhibitor that blocks the function of FGFR1, which together with bFGF (FGF2) is a key regulator of melanoma proliferation, will have efficacy for advanced melanomas, and in particular for melanomas that are BRAF wild-type, which are the most aggressive type of advanced melanoma. Another interesting and important aspect that begs exploration is whether there is a link between DDX11 and pigmentation. Genetic disorders such as Bloom syndrome, Fanconi anemia, and the recently identified cohesinopathy, Warsaw breakage syndrome[
8], are associated with abnormal skin pigmentation. Thus, it is a possibility that there is link between helicases such as DDX11 and the microphthalmia-associated transcription factor (MITF), which is a master regulator of melanocyte development.
In addition to the findings of our study presented herein, novel and compelling evidence that some members of the family of DEAD box helicases have pertinent functions in certain types of cancer has recently also been obtained in two other cases – one being breast cancer and the other medulloblastoma, which like melanoma is a neural crest-derived malignancy. In the case of breast cancer, the DEAD box helicase, DDX5, was found to be amplified and often co-amplified along with ERBB2, and breast cancer cell lines with amplification of the DDX5 locus were considerably more sensitive to its knockdown than breast cancer cell lines lacking this amplification[
22]. In the case of WNT-subgroup medulloblastomas that unlike the other three medulloblastoma subtypes have a good long-term prognosis, exome sequencing revealed somatic missense mutations in the gene DDX3X[
23‐
25], which is a paralogue of the DEAD box helicase, DDX3.
Conclusions
The novel finding presented herein documents that DDX11, a member of the DEAD-box family of helicases, is expressed at high levels in primary and metastatic melanoma, but not in melanocytes of normal skin. Furthermore, our data demonstrate that interfering with the expression of DDX11 has severe consequences for melanoma cells. In particular, we document that inhibiting DDX11 expression causes substantial chromosome segregation defects and telomere shortening, major inhibition of melanoma cell proliferation, and rapid and massive melanoma cell apoptosis. Taken together, these data suggest that molecular targeting of DDX11 could be a new avenue and powerful approach to treat advanced melanoma, which is refractory to chemotherapy and radiation therapy.
Methods
Melanoma cell lines and tissues
VGP (WM983-A) and MGP (WM983-B, WM852, WM1158) human melanoma cell lines were propagated in vitro as previously described[
26]. Standard immunohistochemistry analysis of deidentified, post-diagnosis excess cryopreserved 20 human tissue samples (deemed exempted (4e) by the University of Pittsburgh IRB), representing normal skin, atypical nevus, melanoma in situ, VGP melanoma, MGP melanoma, and melanoma-infiltrated lymph nodes was performed with an anti-human DDX11 mouse monoclonal antibody (Sigma-Aldrich, St. Louis, MO, USA). The chromogen used in the immunohistochemistry analysis was Vulcan Fast Red, and all DDX11 antibody-probed tissue sections were counterstained with hematoxylin.
Immunoblot and immunofluorescence analysis
Protein lysates (25 μg/sample), separated on sodium dodecyl sulfate-polyacrylamide gels (SDS-PAGE) and transferred onto nylon membrane, were probed with antibody to human DDX11 (Sigma-Aldrich) or α-tubulin (Cell Signaling Technology, Inc., Danvers, MA, USA), followed by incubation with a horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology) and Luminol reagent (Millipore, Billerica, MA, USA).
For immunofluorescence analysis, melanoma cells were fixed with 2% paraformaldehyde (2 min at room temperature) and thereafter, with methanol (30 min at −20°C). The fixed cells were subsequently blocked with goat serum, probed with primary antibody to human DDX11 (Sigma-Aldrich) followed by an Alexa Fluor 555-conjugated secondary antibody (Invitrogen, Carlsbad, CA, USA), counterstained with fluorescent 40-6-diamidino-2-phenylindole (DAPI) (Invitrogen), and imaged with an inverted, epifluorescent TE2000 Nikon microscope and a charge-coupled device (CCD) camera (Roper Scientific, Inc./Photometrics, Tucson, AZ, USA).
DDX11 siRNA conjugation and transfection
Five μg of a custom-synthesized siRNA (Thermo Fisher Scientific Dharmacon (Lafayette, CO, USA) based upon the human DDX11 exon 3-specific sequence 5’ CCU GUG UCU GUC UUC UUC CUG CGA A 3’[
10], which as we determined via a BLAST search, does not align with any other sequence, was conjugated to the fluorochrome Cy5 via a Label IT siRNA Tracker Intracellular Localization Cy5 kit (Mirus Bio LLC, Madison, Wl, USA). Melanoma cells, transfected for 24 hr with 5 nM of the Cy5-conjugated DDX11 siRNA, were fixed with 4% paraformaldehyde, counterstained with fluorescent DAPI (Invitrogen), and imaged. Dual color images were processed with MetaMorph software 7.7.5.0 (Molecular Devices, LLC, Sunnyvale, CA, USA).
Melanoma cells were transfected with the single DDX11 siRNA not conjugated or a control ON-TARGETplus non-targeting siRNA pool using Lipofectamine 2000 (Invitrogen) as the siRNA delivery vehicle. Phase-contrast images of the transfected melanoma cells were acquired with an inverted, epifluorescent TE2000 Nikon microscope and a CCD camera (Roper Scientific, Inc./Photometrics).
qPCR analysis
Total RNA was isolated from DDX11 and control siRNA transfected melanoma cells with Trizol reagent (Sigma-Aldrich) and in each case, cDNA was transcribed with qscript cDNA Supermix (Quanta BioSciences, Inc., Gaithersburg, MD, USA) from 1 μg of RNA. qPCR reactions were performed with a pair of qPCR primers (life technologies/Applied Biosystems, Foster City, CA, USA) spanning human DDX11 exon boundary 20–22 that generated an 88 bp amplicon. A set of human 18s RNA primers (life technologies/Applied Biosystems) served as the internal control. Using PerfeCTa FastMix II, ROX (Quanta BioSciences, Inc.), 40 qPCR cycles (3 min at 95°C, 15 seconds at 95°C, 1 min at 95°C) were carried out via a StepOnePlus Real-Time PCR system (life technologies/Applied Biosystems). The qPCR data were analyzed using the 2-ΔΔCT method.
Chromosome staining and analysis
Following siRNA transfection, melanoma cells were treated for 90 min with 100 ng/ml of colchicine, followed by addition of hypotonic buffer (0.075 M KCl) for 15 min at 37°C and subsequent fixation with Carnoy's solution. The chromosome spreads were stained with fluorescent DAPI or Giemsa. Images of the fluorescent DAPI-stained chromosome spreads were captured with an inverted, epifluorescent TE2000 Nikon microscope and a CCD camera (Roper Scientific, Inc./Photometrics). Images of the Giemsa-stained chromosome spreads were acquired with a Hamamatsu NanoZoomer 2.0-HT (Hamamatsu Corporation USA, Bridgewater, NJ, USA). The number of chromosomes with closed, partially closed, or open/separated arms was determined for 40 Giemsa-stained chromosome spreads. The ‘Linear Measure’ tool in the NDP.view software (Hamamatsu Corporation) was used to determine the length of three randomly selected chromosomes in each of 50 Giemsa-stained chromosome spreads.
Proliferation and apoptosis analysis
Melanoma cell proliferation was determined by counting cells with a hemocytometer. At each time point, duplicate samples were analyzed for DDX11 as well as control siRNA-transfected cells, and cells that had received only Lipofectamine 2000 (Invitrogen).
Whole-cell lysates (30 μg/sample) of melanoma cells transfected with DDX11 or control siRNA were separated by SDS-PAGE, transferred onto nylon membrane, and probed with antibody to c-PARP (Cell Signaling Technology, Inc.) or α-tubulin (Cell Signaling Technology, Inc.), followed by incubation with a horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology) and Luminol reagent (Millipore).
For immunofluorescence-based detection of apoptosis, cytospin preparations of DDX11 as well as control siRNA-transfected melanoma cells were fixed, permeablized, labeled with the In Situ Cell Death Detection Kit, TMR red (Roche Applied Science, Indianapolis, IN, USA), counterstained with fluorescent DAPI, and imaged.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CB performed the studies that involved siRNA transfection, qPCR, preparation and analysis of chromosome spreads, and proliferation and apoptosis assays. CB also helped to draft the manuscript. XW carried out the immunohistochemistry study, and the immunoblot and optical imaging analysis. DB conceived of the study, participated in its design and coordination, and wrote the manuscript. All authors read and approved the final manuscript.