Accepted Manuscript

Directionality of noncoding human RNAs: how to avoid artifacts

Sivan Tzadok, Yarden Caspin, Yafit Hachmo, Dan Canaani Iris Dotan

PII: S0003-2697(13)00159-0

DOI: http://dx.doi.org/10.1016/j.ab.2013.03.031

Reference: YABIO 11305

To appear in: Analytical Biochemistry

Received Date: 17 January 2013

Revised Date: 13 March 2013

Accepted Date: 20 March 2013

Please cite this article as: S. Tzadok, Y. Caspin, Y. Hachmo, D.C.I. Dotan, Directionality of noncoding human

RNAs: how to avoid artifacts, Analytical Biochemistry (2013), doi: http://dx.doi.org/10.1016/j.ab.2013.03.031

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

��

�

Directionality of noncoding human RNAs: how to avoid artifacts

Sivan Tzadoka, Yarden Caspina, Yafit Hachmoa, Dan Canaania Iris Dotana*

a Department of Biochemistry and Molecular biology, Faculty of Life Sciences, Tel Aviv

University, Ramat Aviv, Tel Aviv 69978, Israel.

*To whom correspondence should be addressed. Tel: 972-3-6424270; Fax: 972-3-6408985;

Email:�IrisD@tauex.tau.ac.il�

Short title: Noncoding RNAs directionality: avoiding artifacts

Subject category: DNA recombinant techniques and nucleic acids

Abstract

��

�

Inactivation of tumor suppressor and metastasis suppressor genes via epigenetic silencing is a

frequent event in human cancers. Recent work has shown new mechanisms of epigenetic

silencing, based upon the occurrence of long noncoding promoter-spanning antisense and/or

sense RNAs (lncRNAs), which constitute part of chromatin silencing complexes. Using RT-PCR

we have started to scan Triple Negative and Her2-overexpressing breast cancer cell lines, for

directional/bidirectional transcription through promoters of tumor suppressor and metastasis

suppressor genes, known to be epigenetically silenced in vivo. Surprisingly, we found that RT-

PCR amplified products were obtained at high frequency in the absence of exogenous primers.

These amplified products resulted from RT priming, via transcripts originating from promoter or

upstream spanning regions. Consequently, this priming overruled directionality determination

and led to false detection-identification of such lncRNAs. We show that this prevalent “no

primer” artifact, can be eliminated by periodate treatment of the RNA preparations, performing

RT reactions at highly elevated temperatures, or combination of both. These experimental

improvements enabled determination of the presence and directionality of individual promoter-

spanning long noncoding RNAs, with certainty. Examples for the BRMS1 metastasis

suppressor gene, as well as RAR-�2 and CST6 human tumor suppressor genes, in breast

carcinoma cell lines, are presented.

Keywords: RT-PCR; Epigenetic gene silencing; Long noncoding RNAs

Introductory statement

��

�

Epigenetic silencing of tumor suppressor gene promoters is a common observation in cancer

[reviewed in 1]. Over the past few years genome-wide promoter analysis, aided by

pharmacological activation, has uncovered tens of known and putative tumor suppressor genes,

that are epigenetically silenced in human cancers [2-4 and references therein]. Likewise,

reduced transcription, rather than mutation, can explain a large part of the loss of metastasis

suppressor gene expression observed in different tumor types, including breast carcinoma [5-7].

Recent large-scale RNA sequencing and RNA polymerase II mapping studies have shown, that

actually thousands of mammalian genes are transcribed in both directions upstream and

downstream of the known RNA polymerase II start sites [8-16]. Moreover, in several reports of

bidirectional transcription through human RNA polymerase II promoters, where the occurrence

of small sense promoter RNAs [11, 17, 18] as well as natural non-coding short or long antisense

RNAs were manifested [19-22], promoter silencing was demonstrated to be accompanied by

heterochromatin formation requiring Ago-1, Ago-2, DNMT3a, and HDAC-1. DNMT1, as well as

DNMT3a, are required thereafter for maintenance of silencing. In the latter case, it is thought

that the non-coding promoter antisense RNA recognizes the promoter region, perhaps through

RNA-DNA hybridization, followed by recruitment of chromatin silencing proteins to the promoter

site [reviewed in 20-21]. These findings allude to the potential participation of natural non-

coding promoter spanning antisense RNAs in silencing of human gene promoters, including

those belonging to tumor suppressor genes, in cancerous tissues. Specifically, ectopic

expression of either miRNA-373 directed against the E-cadherin tumor suppressor gene

promoter antisense RNA, siRNAs against the promoter antisense RNA of the p15 tumor

suppressor, the E-Cadherin, the p21 tumor suppressor, VEGF, progesterone receptor, and AIR,

all activate their corresponding genes [23-29]. These results indicate the potential of promoter-

directed siRNAs, shRNAs or phosphorothioate oligonucleotides (ODNs) [17, 30], to activate

gene expression of tumor suppressor genes, silenced in cis by promoter sequence-bidirectional

transcription. Such mode of action of noncoding antisense RNAs can be one of several potential

��

�

mechanisms of lncRNA, that are not mutually exclusive [31]. The long range aim of this work is

to uncover lncRNA mediated regulation of breast carcinoma tumor suppressor and metastasis

suppressor genes. Upon initiation of this project, we have encountered artifacts leading to

overestimation of the presence or false determination of directionality of such lncRNAs. How to

avoid such artifacts is the subject matter of this report. In this context, we also report the

occurrence of long noncoding antisense RNAs that span the human promoters of the

metastasis suppressor gene BRMS1 and the tumor suppressor genes RAR-�2 and CST6.

Materials and Methods

Cell growth

MDA-MB-231 [32] was a kind gift from Prof. J. Price, MD Anderson. MDA-MB-435 and Hs578T

'Triple Negative' (TN) breast carcinoma cell lines, as well as the Her2-overexpressing SKBR3

breast carcinoma cell line, were purchased from ATCC. All four breast carcinoma cell lines were

routinely cultured at 37°C in 5% CO2 in full Dulbecco's modified Eagle’s medium [DMEM],

supplemented with 10% Fetal Bovine Serum (FBS, Sigma-Aldrich, Israel), 4 mM L-glutamine,

and antibiotics (20 units/ml of penicillin and 20 �g/ml streptomycin). All tissue culture treatments

were performed under mycoplasma free conditions, with routine check for mycoplasma infection

using EZ-PCR mycoplasma test kit (Biological Industries, Israel) according to the supplier

instructions.

Periodate treatment

In order to reduce endogenous 3'-OH RNA ends, from serving as primers for polymerization by

the RT enzyme, total RNA was subjected to a beta-elimination reaction in a total volume of 80�l

containing 30mM sodium periodate and 300mM sodium acetate buffer [33]. Incubation with the

periodate was carried out in the dark for 1 hr at room temperature, followed by addition of 20�l

��

�

of 500mM dextrose. After 10 min. incubation with the dextrose quencher, RNA was directly

precipitated in ethanol.

RNA extraction, reverse transcription (RT), and PCR

Total RNA was extracted by the EZ-RNA kit (Biological Industries, Israel), DNase treated [RQ

DNase, Promega], and periodate treated, where required. Nuclear RNA was isolated by

perforating the cells with Hank’s Balanced Salt Solution (HBSS ) containing Digitonin and PMSF

(composition per liter: 8.0g NaCl, 0.4g KCl, 0.14g CaCl2, 0.1g MgCl2.6H2O, 0.1g MgSO4•7H2O,

0.06g Na2HPO4.2H2O, 0.06g KH2PO4, 1.0g Glucose, 0.35g NaHCO3, 1g Digitonin, and

PMSF to a final concentration of 1mM).���After adding a minimum of 20 cell-pellet volumes, cells

were incubated on ice for 10 min. Cell ghosts were then collected by a low speed spin and the

pellet was further subjected to RNA extraction, with the EZ-RNA kit (Biological Industries,

Israel). �

RT was carried out using RevertAidTM RNase H Minus enzyme (Fermentas) at 42°C, or

RevertAidTM Premium enzyme (Fermentas), at higher temperatures. RT reactions were

performed using gene specific primers enabling the determination of directionality. cDNA was

PCR-amplified using Taq DNA Polymerase master mix (Lambda, NEB). PCR conditions were

as follows: initial incubation at 94°C for 2 min, 25-40 cycles of 94°C for 45 sec, 56-60°C for 45

sec, and 72°C for 30 sec. A common terminal incubation step at 72°C for 7 min was applied for

all amplifications.

Primers (genomic coordinates are derived from NCBI36/hg18 assembly)

BRMS1 RT gene specific primers:

RT-S1: F-TTGGGAGGCTCAGATGAGAT chr11:65869814-65869833

RT-S2: F- GGGTAAGAAATGTCGCTCCA chr11:65870009-65870028

��

�

RT-AS1: R- GCTCGACTAACCGGAGAGG chr11:65868968-65868986

BRMS1 PCR primers:

F1- AAGCTTCGTACAAATGTGAAGTATT chr11:65869767-65869791

R1- CTACCCACCCAAGATTGTTAAT chr11:65869704-65869725

F2- AAGCACCGATAGGCTCTGC chr11:65869139-65869157

R2- TGGAGCCTCTGGCCTCAC chr11:65869028-65869045

F3- TGGGATCGATTCCTAGCTTT chr11:65869938-65869957

R3- GGTAACCGAGGCTCAGAGAG chr11:65869878-65869897

RAR- �2 RT gene specific primers:

RT-S1: F- AAGGCGCACAGAGGAATTTA chr3:25444336-25444355

RT-AS1: R-TGCTAACTCTGCTAAGCCAAA chr3:25479032-25479052

RT-AS2: R- TGTCTTGCTACCAATGCAGTTT chr3:25445461-25445482

RAR- �2 PCR primers:

F1- CTCCTCCCCTGCTCATTTTA chr3:25444499-25444518

R1-CCGGCGTTTTCTTTCCTATT chr3:25444551-25444570

F2- TTTGAACTAGCCCTTCCCTAGA chr3:25478882-25478903

R2- TGCCATGGAGTACCTGTAACAT chr3:25479003-25479024

��

�

CST6 RT gene specific primers:

RT-S1: F- TCCAGCACCAGACCTCTTCT chr11:65535604-65535623

RT-AS1: R- AAGCCAGGAAGGCAGGTAGT chr11:65536533-65536552

CST6 PCR primers:

F1- GCTGCGGTTGGTAGTTCATT chr11:65535764-65535783

R1- GACAGTCCCCAGCAACAAGT chr11:65535820-65535839

R2- CCCTCCAGGACACACCTG chr11:65535862-65535879

Results and Discussion

In an effort to study the in vivo regulation of tumor suppressor and metastasis suppressor genes

in breast carcinomas, we have tried to identify directional/bidirectional transcription through

promoters of such genes, known to be epigenetically silenced in vivo [1-4]. Identification of

Sense (S) and Anti-Sense (AS) transcripts to promoter regions was performed by RT-PCR,

where the RT specific primer determines directionality. In our initial studies, total cellular RNA

was used. However, in cases where products were questionable, or, the identity of amplified

products was unclear, nuclear fractions of RNA were used as templates. As might be expected,

the use of nuclear-enriched RNA often resulted in better detection of low abundance transcripts.

For the latter purpose, we have devised a protocol based on mild fractionation of cultured cells,

using digitonin, rather than the classical NP-40 detergent treatment. Digitonin generates

complexes with cholesterol, upon which the plasma membrane becomes perforated, rather than

solubilized. Cell ghosts remaining after this treatment are then subjected to RNA isolation.

Yields and quality of nuclear RNA prepared by our protocol (outlined in M & M) were

�

�

substantially greater than those obtained by commercial kits such as Ambion's PARISTM kit

(data not shown).

As a first test case, we chose the human BRMS1 (HGNC: 17262) metastasis suppressor gene.

This gene was shown to be epigenetically silenced, coupled to its promoter methylation and loss

of expression, in breast cancer [34, 35]. Bearing in mind the claims for the large variety of

directional/bidirectional transcripts spanning the human/mouse gene promoters [8-16], we

decided to include a "no primer" RT control reaction. This control is rarely used. Figure 1A

shows the relative positions on the human BRMS1 gene promoter/first exon of the primers

used herein for the RT and PCR reactions. Figure 2A displays the occurrence of human

BRMS1 metastatic suppressor gene transcripts, spanning its promoter. Lanes 2 & 3 from the

right, show, that under standard RT-PCR conditions (RT performed at 42ºC) there are both

sense and antisense promoter-spanning RNA transcripts. However, the inclusion of a "no

primer" control reaction (4th lane from the right) indicates, that there is endogenous priming by

RNA, leading to first strand cDNA synthesis followed by PCR amplification; priming by

nondigested endogenous DNA is unlikely because the –RT control is clean (Fig. 2A & 2C).

Thus, the determination of putative promoter spanning noncoding RNA of the S or AS type by

this standard RT-PCR protocol has turned out inconclusive. In cases where the promoter RNA

transcripts snap back at their 3’ ends, or hybridize to endogenous short complementary

RNA/DNA fragment, this priming can lead to the false determination of RNA transcripts and their

corresponding directionality. One type of obvious known solution which has been already

outlined entails exhaustive DNase treatment of the RNA preparation in order to exclude any

endogenous DNA priming of the RNA during reverse transcription. Moving on to false RT

priming by endogenous RNAs, we initially attempted abolishing the false priming, driven by

such naturally occurring short RNAs through the usage of RT enzymes possessing high optimal

catalytic temperatures, up to 60ºC (Fermentas RevertAid Premium reverse transcriptase). As

�

�

shown in Figure 2B, one can abolish this particular false “endogenous priming”, by raising the

RT temperature to 52.4ºC.

Previously, several groups have analyzed bidirectional transcription residing in coding regions

of genes. Those groups have suggested the use of “no-primer control”, but did not come up with

satisfactory experimental solutions to this problem [36-39]. Specifically, they have suggested the

use of RT at 50ºC in order to prevent RNA self priming. As proven below performing RT at

elevated temperature in order to prevent the RNA from snapping back upon itself obstructing

self priming, does not always solve the problem. Therefore, in another line of experiments, we

tried to abolish the phenomenon of endogenous priming, by treating the RNA preparation with

periodate. This treatment (see M & M section) initiates the beta-elimination reaction of 2' and 3'

free hydroxyl residues on the 3' ends of RNA. The reduction of endogenous 3' RNA ends

available for polymerization by the RT enzyme, renders the reaction more likely to occur by the

exogenously added DNA primer. As shown in Figure 2C, this periodate treatment enabled us to

nullify the false “no primer” product and to identify BRMS1 promoter spanning RNAs, of both the

sense (S) and antisense (AS) polarity. In the MDA-MB-435 breast carcinoma cell line, the

antisense RNA has a contiguous segment of at least 526 bp (devoid of long ORF) which starts

555 bp upstream of the major BRMS1 transcription start site (TSS). In contrast, the sense RNA

is scattered throughout at least 2 kb of the BRMS1 promoter. It may therefore constitute bona

fide noncoding RNA transcript. Alternatively, because the RT primer is located downstream of

the TSS in exon 1, the sense transcript may represent BRMS1 mRNA/s initiated upstream of

the canonical TSS. The location of the RT primer downstream of the BRMS1 TSS, precludes

the noncoding sense RNA from being part of the B3GNT1 gene (HGNC: 15685), as the latter

mRNA 3’ end maps to 271 bp upstream of BRMS1 TSS.

After obviating artifacts resulting in false priming of RT reactions, we set to probe for promoter

spanning lncRNA, in two tumor suppressor genes, known to be epigenetically silenced in the

triple negative (TN) or Her2-overexpressing breast carcinomas [40]. The first tumor suppressor

���

�

gene attempted, was RAR-�2 (HGNC 9865, whose promoter/first two exons and relevant

primers used herein are shown schematically in Fig. 1B), in the context of the MDA-MB-231

human breast carcinoma cells [41].

Indeed, as shown in Fig. 3A (lanes 3 & 5 from the left), under standard RT-PCR conditions,

these "no primer" reactions resulted in the endogenous priming of a presumably desired RNA,

leading to first strand cDNA synthesis and PCR amplification. Fig. 3B exemplifies the effect of

RT temperature on the false endogenous priming of template RNA in the RAR-�2 gene coding

region. When the RT reaction is performed at temperatures as high as 49.3ºC, false

endogenous priming occurred. Upon shifting to 52.3ºC, this artifact disappears. Fig. 3C, shows

that periodate treatment of RNAs derived from the triple negative breast carcinoma Hs578T

(panel 1) and MDA-MB-231 cell lines (panel 2) eliminated the product present in the "no primer"

control, thus enabling the identification of RAR-�2 promoter upstream Sense and Anti Sense

transcripts, in an unequivocal manner. The case exemplified in Fig. 3D with nuclear RNA

derived from the Her2-positive breast carcinoma cell line SKBR3, illustrates a need to combine

periodate treatment with RT reactions performed at no less than 59ºC.

These results presented herein and others (data not shown) reveal the occurrence of RAR-�2

promoter- spanning RNAs of both sense and antisense polarity. The identified antisense RNA

spans at least 383 bp and starts 40 bp upstream of the major RAR-�2 transcription start site

(TSS) in the MDA-MB-231/Hs578T/SKBR3 breast carcinoma cell lines. The sense promoter-

spanning RNA is present in the Hs578T and SKBR3 cell lines, but not in the MDA-MB-231 cell

line. The sense RNA spans a contiguous region of at least several hundred nucleotides

upstream of RAR-�2 TSS localized within intronic sequences of RAR-�5 and/or RAR-�201

prespliced mRNAs.

The other tumor suppressor gene which we have tested was CST6 (HGNC: 9865), also

suggested of being regulated by noncoding RNA/s, due to epigenetic silencing in breast

���

�

carcinomas [42, 43, 35]. Figure 1C shows the relative positions on the human CST6 gene

promoter/first exon of the primers used herein for the RT and PCR reactions. Fig. 4A indicates

that for unraveling bona fide CST6 promoter overlapping transcripts, it is essential to treat RNA

preparations, from the Hs578T cell line, with periodate (while RT was carried out at 50ºC). In

contrast, when probing periodate-treated RNA derived from the SKBR3 cell line, performing the

RT at 50ºC still gives rise to endogenous priming (Fig. 4B), and one has to go to 55 ºC in order

to identify genuine CST6 upstream transcripts.

The latter results and others to be reported elsewhere, reveal the occurrence of CST6 promoter

spanning RNAs of both sense and antisense polarity. The antisense RNA is at least 276 bp

long, initiating 159 bp upstream of the major CST6 transcription start site (TSS) of the Hs578T

and SKBR3 breast carcinoma cell lines. The sense RNA is at least 276 bp long, starting

upstream of the CST6 known TSS. As noted for the two other genes, at this point, we cannot

rule out the possibility, that the origin of the determined promoter spanning sense RNA of the

CST6 gene, originates from a minor CST6 mRNA variant, whose transcript initiation site is

located upstream of the known TSS.

In this report, we demonstrated that endogenous priming in RT reactions directed at individual

promoter-spanning or intergenic RNAs is rather prevalent. We suggest here a combined novel

protocol to obviate such artifacts originating from reverse transcriptase (RT) reactions, primed

by endogenous RNA, rather than exogenous primers. The method proof of concept is given

here by the periodate treatment experiments, as well as by the loss of these false transcripts,

when priming individual specific RT reactions at high temperatures, and when necessary

combining the two (Figs. 3D & 4B). The abundance of the human lncRNA transcripts (14,880

according to GENCODE v7, ref. 44), which could eventually even significantly increase [45],

make the primer-independent RT polymerization artifact (endogenous priming) particularly

important. Our improved protocols and controlled experiments, have allowed the unequivocal

���

�

identification of long noncoding antisense RNAs of the human BRMS1 metastasis suppressor

gene, the RAR-�2 and the CST6 human tumor suppressor gene promoters of breast carcinoma

derived cell lines. Importantly, none of the Antisense noncoding RNAs identified in our studies is

listed in any of GENCODE lncRNA databases, up to version 14.

Noteworthy is the presence of a particular noncoding RNA in a cell line dependent manner; the

lineage specificity of lncRNAs has been recognized before [46]. The potential regulatory role of

these noncoding RNAs is currently under study in our laboratory.

Acknowledgments

We thank Prof. J. Price for a biological reagent, and Prof. G. Kaufmann for his

thoughtful suggestion in treating RNA preparations with periodate.

This work was supported by a grant of The David Orgler Fund for Cancer

Research to D.C. Sivan Tzadok was supported by The TAU Rector fellowship.

Yafit Hachmo was supported by The van Baytes cancer research fellowship.

References

���

�

[1] P.A. Jones, S.B. Baylin, The epigenetics of cancer, Cell 128 (2007) 683-692.

[2] S.B. Baylin, P.A. Jones, A decade of exploring the cancer epigenome-biological and

translational implications, Nature Reviews Cancer 11 (2011) 726-734.

[3] M.O. Hoque, M.S. Kim, K.L. Ostrow, J. Liu, G.B. Wisman, H.L. Park, M.L. Poeta, C

Jeronimo, R Henrique, A Lendvai, et al. Genome-wide promoter analysis uncovers portions

of the cancer methylome, Cancer Res. 68 (2008) 2661-2670.

[4] K.L. Ostrow, H.L. Park, M.O. Hoque, M.S. Kim, J. Liu, P. Argani, W. Westra, W. Van

Criekinge, D. Sidransky, Pharmacological unmasking of epigenetically silenced genes in

breast cancer Clin Cancer Res. 15 (2009) 1184-1191.

[5] K.S. Vaidya, D.R. Welch, Metastasis suppressors and their roles in breast carcinoma, J.

Mammary Gland Biol. Neoplasia 12 (2007) 175-190.

[6] S.C. Smith, D. Theodorescu, Learning therapeutic lessons from metastasis suppressor

proteins, Nat. Reviews Cancer 9 (2009) 253-264.

[7] C.S. Cropp, R. Lidereau, A. Leone, D. Liscia, A.P. Cappa, G. Campbell, E. Barker, V. Le

Doussal, P.S. Steeg, R. Callahan, NME1 protein expression and loss of heterozygosity

mutations in primary human breast tumors, J. Natl. Cancer Inst. 86 (1994) 1167-1169.

[8] P. Bertone, V. Stolc, T.E. Royce, J.S. Rozowsky, A.E. Urban, X. Zhu, J.L. Rinn, W.

Tongprasit, M. Samanta, S Weissman, et al. Global identification of human transcribed

sequences with genome tiling arrays, Science 306 (2004) 2242-2246.

[9] P. Carninci, T. Kasukawa, S. Katayama, J. Gough, M.C. Frith, N. Maeda, R. Oyama, T.

Ravasi, B. Lenhard, C. Wells, et al. The transcriptional landscape of the mammalian

genome, Science 309 (2005) 1559-1563.

[10] P. Kapranov, J. Cheng, S. Dike, D.A. Nix, R. Duttagupta, A.T. Willingham, P.F. Stadler, J.

Hertel, J. Hackermüller, I.L. Hofacker, et al. RNA maps reveal new RNA classes and a

possible function for pervasive transcription, Science 316 (2007) 1484-1488.

���

�

[11] L.J. Core, J.J. Waterfall, J.T. Lis, Nascent RNA sequencing reveals widespread pausing

and divergent initiation at human promoters, Science 322 (2008) 1845-1848.

[12] P. Preker, J. Nielsen, S. Kammler, S. Lykke-Andersen, M.S. Christensen, C.K.

Mapendano, M.H. Schierup, T.H. Jensen, RNA exosome depletion reveals transcription

upstream of active human promoters, Science 322 (2008) 1851-1854.

[13] A.C. Seila, J.M. Calabrese, S.S. Levine, G.W. Yeo, P.B. Rahl, R.A. Flynn, R.A. Young, P.A.

Sharp, Divergent transcription from active promoters, Science 322 (2008) 1849-1851.

[14] Y. He, B. Vogelstein, V.E. Velculescu, N Papadopoulos, K.W. Kinzler The antisense

transcriptomes of human cells, Science 322 (2008) 1855-1857.

[15] M. Guttman, I. Amit, M. Garber, C. French, M.F. Lin, D. Feldser, M. Huarte, O. Zuk, B.W.

Carey, J.P. Cassady, et al. Chromatin signature reveals over a thousand highly conserved

large non-coding RNAs in mammals, Nature 458 (2009) 223-227.

[16] AM Khalil, M. Guttman, M. Huarte, M. Garber, A. Raj, M.D. Rivea, K. Thomas, A. Presser,

B.E. Bernstein, A.van Oudenaarden, et al. Many human large intergenic noncoding RNAs

associate with chromatin-modifying complexes and affect gene expression, Proc. Natl.

Acad. Sci. USA 106 (2009) 11667-11672.

[17] J. Han, D. Kim, K.V. Morris, Promoter-associated RNA is required for RNA-directed

transcriptional gene silencing in human cells. Proc Natl Acad Sci USA 104: (2007) 12422-

12427.

[18] P.G. Hawkins, S. Santoso, C. Adams, V. Anest, K.V. Morris, Promoter targeted small

RNAs induce long-term transcriptional gene silencing in human cells, NAR 37 (2009) 2984-

2995.

[19] K.V. Morris, S.W. Chan, S.E. Jacobsen, D.J. Looney, Small interfering RNA-induced

transcriptional silencing in human cells, Science 305 (2004) 1289-1292.

[20] K.V. Morris, Long antisense non-coding RNAs function to direct epigenetic complexes that

regulate transcription in human cells, Epigenetics 4 (2009) 296-301.

���

�

[21] KV Morris, Non-coding RNAs, epigenetic memory and the passage of information to

progeny, RNA Biol 6 (2009) 242-247.

[22] S. Knowling, K.V. Morris, Epigenetic regulation of gene expression in human cells by

noncoding RNAs, Prog. Mol. Biol. Transl. Sci. 102 (2011) 1-10.

[23] W. Yu, D. Gius, P. Onyango, K. Muldoon-Jacobs, J. Karp, A.P. Feinberg, H. Cui, Epigenetic

silencing of tumour suppressor gene p15 by its antisense RNA, Nature 451 (2008) 202-206.

[24] R.F. Place, L.C. Li, D. Pookot, E.J. Noonan, R. Dahiya, MicroRNA-373 induces expression

of genes with complementary promoter sequences, Proc Natl Acad Sci USA 105 (2008)

1608-1613.

[25] K.V. Morris, S. Santoso, A.M. Turner, C. Pastori, P.G. Hawkins, Bidirectional transcription

directs both transcriptional gene activation and suppression in human cells, PloS Genet. 4

(2008) e1000258

[26] L.C. Li, S.T. Okino, H. Zhao, D. Pookot, R.F. Place, S. Urakami, H. Enokida, R. Dahiya,

Small dsRNAs induce transcriptional activation in human cells, Proc Natl Acad Sci USA 103

(2006) 17337-17342.

[27] B.A. Janowski, S.T. Younger, D.B. Hardy, R. Ram, K.E. Huffman, D.R. Corey Activating

gene expression in mammalian cells with promoter-targeted duplex RNAs, Nat Chem Biol 3

(2007) 166-173.

[28] J.L. Run, M. Kertesz, J.K. Wang, S.L. Squazzo, X. Xu, S.A. Brugmann, Goodnough L.H.,

J.A. Helms, P.J. Farnham, E. Segal, H.Y. Chang, Functional demarcation of active and

silent chromatin domains in human HOX loci by noncoding RNAs, Cell 129 (2007) 1311-

1323.

���

�

[29] T. Nagano, J.A. Mitchel, L.A. Sanz, F.M. Pauler, A.C. Ferguson-Smith, R. Feil, P. Fraser,

The Air noncoding RNA Epigenetically silences transcription by targeting G9a to chromatin,

Science 322 (2008) 1717-1720.

[30] T. Ideue, K. Hino, S. Kitao, T. Yokoi, T. Hirose, Efficient oligonucleotide-mediated

degradation of nuclear noncoding RNAs in mammalian cultured cells, RNA 15 (2009) 1578-

1587.

[31] K.C. Wang, H.Y. Chang, Molecular mechanisms of long noncoding RNAs, Molecular Cell

43 (2011) 904-914.

[32] R. Calleau, M. Olive’, Q.V. Cruciger, Long-term human breast carcinoma cell lines of

metastatic origin: preliminary characterization. In Vitro 14 (1978) 911-915.

[33] H. Neu, L. Heppel, Nucleotide sequence analysis of polyribonucleotides by means of

periodate oxidation followed by cleavage with an amine, J. Biol. Chem. 239 (1964) 2927-

2934.

[34] B.J. Metge , A.R. Frost , J.A. King , D.L. Dyess , D.R. Welch , R.S. Samant , L.A. Shevde,�

Epigenetic silencing contributes to the loss of BRMS1 expression in breast cancer, Clin.

Exp. Metastasis 25 (2008) 753-763.

[35] M. Chimonidou , A. Strati , A. Tzitzira , G. Sotiropoulou , N. Malamos , V. Georgoulias ,

E.S. Lianidou, DNA methylation of tumor suppressor and metastasis suppressor genes in

circulating tumor cells, Clin Chem. 57 (2011) 1169-1177.

[36] A. Stahlberg, J. Hakansson, X. Xian, H. Semb, M. Kubista, Properties of the reverse

transcription reaction in mRNA quantification, Clin. Che. 50 (2004) 509-515.

[37] F. Haddad, A.X. Qin, J.M. Giger, H. Guo, K.M. Baldwin, Potential pitfalls in the accuracy of

analysis of natural sense-antisense RNA pairs by reverse transcription-PCR. BMC

Biotechnology 7 (2007) 21.

���

�

[38] M.F. Adrover, M.J. Munoz, M.V. Baez, J. Thomas, A.R. Kornblihtt, A.L. Epstein, D.A.

Jerusalinsky, Characterization of specific cDNA background synthesis introduced by

reverse transcription in RT-PCR assays, Biochimie 92 (2010) 1839-1846.

[39] E.C.H Ho, M.E. Donaldson, B.J. Saville, Detection of antisense RNA transcripts by strand-

specific RT-PCR, Methods Mol. Biol. 630 (2010) 125-138.

[40] A. Goldhirsch , J.N. Ingle , R.D. Gelber , A.S. Coates , B. Thürlimann , H.J. Senn; Panel

members Thresholds for therapies: highlights of the St Gallen International Expert

Consensus on the primary therapy of early breast cancer, Ann Oncol. 20 (2009) 1319-1329.

[41] S.M. Sirchia, M. Ren, R. Pili, E. Sironi, G. Somenzi, R. Ghidoni, S. Toma, G. Nicolo, N.

Sacchi, Endogenous reactivation of the RARbeta2 tumor suppressor gene epigenetically

silenced in breast cancer, Cancer Res. 62 (2002) 2455-2461.

[42] L. Ai, W.-J. Kim, T.-Y. Kim, C.R. Fields, N.A. Massoll, K.D. Robertson, K.D. Brown

Epigenetic silencing of the tumor suppressor cystatin M occurs during breast cancer

progression, Cancer Res. 66 (2006) 7899-7909.

[43] A.G. Rivenbark, W.D. Jones, W.B. Coleman, DNA methylation-dependent silencing of

CST6 in human breast cancer cell lines, Laboratory Invest. 86 (2006) 1233-1242.

[44] T. Derrien, R. Johnson, G. Bussotti, A. Tanzer, S. Djebali, H. Tilgner, G. Guernec, D.

Martin, A. Merkel, D.G. Knowles, et al. The GENCODE v7 catalog of human long noncoding

RNAs: Analysis of their gene structure, evolution, and expression, Genome Res. 22 (2012)

1775-1789.

[45] J. Harrow, A. Frankish, J.M. Gonzalez, E. Tapanari, M. Diekhans, F. Kokocinski, B.L. Aken,

D. Barrell, A.Z. Zadissa, S. Searle, et al. GENCODE: The reference human genome

annotation for The ENCODE Project, Genome Res. 22 (2012) 1760-1774.

[46] M.S. Kowalczyk, D.R. Higgs, T.R. Gingeras, Molecular biology: RNA discrimination, Nature

482 (2012) 310-311.

��

�

Figure legends

Fig.1. Schematic illustration of the investigated gene promoter region, the positions of the

primers used for the RT first strand formation and the PCR primers location. The former primers

and their polarity are indicated as RT-S1, RT-AS1 etc. The second type of primers and the

direction of PCR synthesis are labelled as PCR-F1, PCR-R1 etc. (A) The human BRMS1 gene.

(B) The human RAR-�2 gene isoform. (C) The human CST6 gene.

Fig.2. Sorting out of BRMS1 promoter-spanning RNAs from false endogenous priming. Nuclear

RNA from MDA-MB-435 human breast carcinoma cells was subjected to RT at the indicated

temperatures using either no exogenous primer or primer complementary to the Sense or Anti

Sense strand of the BRMS1 metastasis suppressor gene promoter. First strand RT reactions

were taken for PCR with a pair of defined primers located at the promoter region. Genomic DNA

served as the PCR positive control. Negative controls included: no template, no RT, no primer.

(A) Endogenous priming of RT. RT was performed with both BRMS1 primer RT-S1 or RT-AS1

at 42ºC, and the PCR with primers F1 & R1. (B) The effect of RT reaction temperature on

endogenous priming. RT reactions were carried out with the RT-AS1 primer at different

temperatures without adding an exogenous primer, while the PCR was made with primers F2 &

R2. (C) Ablation of endogenous priming following periodate treatment; Recovery of S and AS

noncoding RNAs. Nuclear RNA from MDA-MB-435 cells was subjected to periodate treatment

as outlined in M & M. RT was performed at 47ºC with primers RT-S2 & RT-AS1, while the

PCRs carried out with primers F3 & R3. The S and AS RNAs derived RT-PCR products are of

80 bp sizes.

Fig.3. Sorting out of RAR-�2 promoter-spanning RNAs from false endogenous priming. Total

RNA from MDA-MB-231 human breast carcinoma cells was subjected to RT at the indicated

temperatures using either no exogenous primer or primer complementary to the Sense or Anti

��

�

Sense strand of the RAR-�2 metastasis suppressor gene promoter. First strand RT reactions

were taken for PCR with a pair of defined primers located at the promoter region. Genomic DNA

served as the PCR positive control. Negative controls included: no template, no RT, no primer.

(A) Endogenous priming of RT. RT was performed at 42ºC with the RT-S1 primer, while the

PCR employed primers F1 & R1. The expected RAR-�2 PCR positive control product has a

size of 72bp. (B) The effect of RT reaction temperature on endogenous priming. RT reactions

were carried out with primer RT-AS1 at different temperatures without adding an exogenous

primer leading in some (in the presence of coding region spanning primers F2 & R2) to the

production of a 143 bp RT-PCR product. (C) Ablation of endogenous priming following periodate

treatment; Recovering bona fide noncoding RNA/s. Total RNAs from Hs578T (Panel 1) and

MDA-MB-231 (Panel 2) breast carcinoma cells were subjected to DNaseI and periodate

treatment as outlined in M & M. The RT reaction was performed with either primer RT-S1 or RT-

AS2 at 47ºC. The expected RAR-�2 PCR product (with primers F1 & R1) has a size of 72 bp for

either Sense or Anti Sense promoter spanning RNA. (D) Ablation of endogenous priming

following periodate treatment and “hot” RT; Identification of noncoding RNA/s. Nuclear RNA

from SKBR3 breast carcinoma cells was subjected to periodate and DNase treatments,

followed by RT with either RT-AS2 or RT-S1 primer at 57ºC/59ºC, and PCR with primers F1 &

R1, leading to a 72 bp amplified product.

Fig.4. Identification of human CST6 S and AS promoter upstream transcripts; Periodate

treatment coupled to “hot” RT enables clearing of “no primer” endogenous RT priming. (A) Total

RNA from Hs578T cells was subjected to DNase followed by periodate treatment. RT reactions

were performed with either primer RT-AS1 or with primer RT-S1, both at 50ºC, followed by PCR

with CST6 primers F1 & R2 leading to a 116 bp size product. (B) Nuclear RNA from SKBR3

cells was subjected to periodate treatment, followed by DNase, and subjected to RT with either

primer RT-AS1 or with primer RT-S1 at 50ºC/55ºC. PCR reactions were then conducted with a

���

�

single pair of primers (F1 & R1) located at the promoter region, leading to a 76 bp amplified

product.

Figure 1

Figure 2

Figure 3

Figure 4