Detection of a 7SL RNA-derived small non-coding RNA using Molecular Beacons in vitro and in cells

2.1 Molecular Beacon design and characterization in vitro

To unveil the cellular localization of endogenous snc7SL RNA we designed and characterized several Molecular Beacons (MB). The principle of a MB assay lays in the formation of a stem formed between the 5′- and 3′-end of a DNA oligonucleotide (Tyagi and Kramer 1996). In case of MBs described herein, the 5′-end is linked to a fluorophore (FAM or Cy5) whereas the 3′-end is linked to a quencher (DAB or BHQ2). As long as the stem is formed, the 5′- and 3′-ends are in close proximity, the quencher absorbs the energy of the excited fluorophore and prevents emission of fluorescence light upon excitation. Hence, the stability of the stem is a critical parameter. In case the stem is very weak, spontaneous strand separation occurs and fluorescence light emission causes a strong background fluorescence signal (F _back). On the other hand, if the stem is very stable the fluorescence light emission in presence of the target molecule is diminutive indicating a low sensitivity. To calculate the assay window (AW = F _max–F _back) for the various MBs tested herein, we determined the maximal fluorescence (F _max) by incubating the MBs in 50 % formamide and subtracted the background fluorescence signal. Hence, an optimal MB generates a low background signal, has a large assay window (F _max–F _back) and is highly sensitive and specific in detecting the target sequence.

Considering the MB criteria mentioned above, we designed seven MBs of different stem stability (MBsnc7SL, MBsnc7SL-S, MBsnc7SL-Mut, MBcon, MBcon-S, MB7SL, MB6145-S) and complementarity to four target RNA molecules: (1) snc7SL RNA (piR-has-1254; DQ571003), (2) synthetic control RNA (con-target, reverse sequence of snc7SL RNA), (3) 7SL RNA (RN7SL2; NR_027260.1), and (4) as positive control piR-has-6145 (DQ575882), another endogenously expressed sncRNA (Figure 1A and B). Bold capital letters indicate MB sequences complementary to the respective target RNA sequence (Figure 1B). The number of guanine:cytosine (G:C) base pairs of the stem varies between six (MBsnc7SL-S, MBcon-S, MB6145-S), five (MBsnc7SL-Mut), and four (MBsnc7SL, MB7SL, MBcon) (Figures 1B and 2). In case of MBsnc7SL-Mut, five G:C base pairs are closing the structure, two additional adenosines flank the G:C base pairs, five nucleotides were deleted to shorten the complementary sequence to snc7SL target RNA and three cytosines were substituted by adenosines to weaken the base pairing with the snc7SL target RNA. As mentioned above, snc7SL RNA is derived from the larger 7SL RNA and the target sequences for MBsnc7SL and MB7SL are shown in the scheme of the predicted folding of RNA7SL (Figure 1C; Andersen et al. 2006; Bovia and Strub 1996).

Figure 1:

Sequences of Molecular Beacons and their target RNAs. A) RNA sequences targeted by Molecular Beacons (MB). B) MBs are 5′-labeled with fluorophores (FAM or Cy5) and quenchers (DAB or BHQ) are linked to the 3′-end. MB sequences (DNA) complementary to the respective target RNA (bold). In MBsnc7SL-Mut five nucleotides are deleted (−) and three nucleotides are changed (small letters underlined). MB with –S (stable) = more guanine:cytosine (G:C) base pairs. FAM, fluorescein; DAB, dabcyl quencher; Cy5, cyanine 5; BHQ, black hole quencher. C) Structure of the 7SL RNA molecule and binding sites of the six signal recognition particle proteins 72, 68, 54, 19, 14, 9 (adapted from (Bovia and Strub 1996)). The target sequences for MB7SL and MBsnc7SL are indicated by black lines.

Figure 2:

Predicted structures and ΔG values of Molecular Beacons. A) MBsnc7SL-S, B) MBsnc7SL, C) MBsnc7SL-Mut, D) MB7SL, E) MBcon-S, F) MBcon, G) MB6145-S. Structures of DNA Molecular Beacons (MBs) were predicted and ΔG values calculated using DNA fold (http://www.unafold.org (Zuker 2003)). CT files (connectivity table) containing secondary structure information were imported into jViz.RNA 4.0 (http://jviz.cs.sfu.ca/). S = more guanine:cytosine (G:C) base pairs, Mut = deletions or mutations in the snc7SL RNA target sequence. FAM, fluorescein; DAB, dabcyl quencher; Cy5, cyanine 5; BHQ, black hole quencher.

We used the DNA folding form at the UNAFold web server (http://www.unafold.org/; Zuker 2003) to calculate the two dimensional folding of each MB and the stability of those structures (ΔG values). The predicted structures with the lowest ΔG values are shown (Figure 2). In all structures the stem formed as expected, based on the G:C base pairs, positioning the fluorophore and quencher in close proximity. All MBs – except MB7SL– had a large single stranded loop area harboring the complementary sequence of the target molecule. Furthermore, all MBs with six G:C base pairs were more stable than those with just five or four G:C base pairs. MB7SL formed a longer stem due to of G:C base pairs formed by complementarity within the target sequence (Figure 2D).

Since it was planned to test the MBs in cell culture experiments, their stability at 37 °C was an important test parameter. We characterized the stability of the MBs by measuring the background fluorescence (F _back) in different reaction buffers and at different temperatures (data not shown). The lowest F _back value for all MBs was measured in the ICN buffer (10 mM Tris/HCl, pH 7.5, 100 mM KCl, 10 mM NaCl, 0.75 mM MgCl₂, 0.0001 % NP40) at 37 °C. Therefore, all in vitro experiments were performed in ICN buffer. Next, we determined F _back for all MBs at different concentrations (5000 nM, 1000 nM, 40 nM (data not shown)) and 200 nM in ICN buffer over time at 37 °C. After 10 min we measured a low and stable F _back signal for all MBs at a concentration of 200 nM (Figure 3A) or 100 nM (Supplementary Table 1A). To determine the maximal fluorescence (F _max) of 200 nM (Figure 3A, +F) or 100 nM (Supplementary Table 1A) FAM-labeled MBs were incubated in 50 % formamide (F) over time at 37 °C. By calculating the assay window (AW = F _max–F _back) after 40 min at 37 °C, a 3- to 4-fold increase in fluorescence was determined for all FAM-labeled MBs (200 nM). Although relative fluorescence units (RFU) determined for the Cy5-labled MB7SL were lower compared to the RFUs measured for the FAM-MBs, a low F _back and a 5-fold higher F _max was determined at a concentration of 200 nM (Figure 3B) or 100 nM MB7SL (Supplementary Table 1A and B).

Figure 3:

Background (F _back) and maximal fluorescence (F _max) of Molecular Beacons. A) To determine the background (F _back) and maximal (F _max) fluorescence, 200 nM of various Molecular Beacons (MBs) were incubated with and without 50 % formamide (+F) in ICN buffer over time at 37 °C. FAM-fluorescence was recorded at time point 0, 10, 20, 30, 40 min. n = 3; AW, assay window. B) To determine F _back and F _max of MB7SL, 200 nM were incubated with and without 50 % formamide (+F) in ICN buffer over time at 37 °C. Cy5-fluorescence was recorded at time point 0, 10, 20, 30, 40 min. n = 3; AW, assay window.

To test the amount of target molecules required for activation of the various MBs, we titrated concentration of the target molecules (0, 25, 100, 100, 400 nM) referred to as con-target RNA, 7SL-target DNA, or snc7SL-target RNA into reactions containing 100 nM MB. None of the MBs were activated at the lowest target molecule concentration of 25 nM. However, 100 nM of the control target RNA (con-target RNA) was sufficient to reach almost F _max levels (Figure 4A, Supplementary Table 1A and B). Higher concentrations of con-target RNA reached earlier a plateau but were not reaching higher fluorescence levels as with 100 nM. These findings show that 100 nM MBcon (ΔG = −3.95) is almost maximal open (F _max) at a concentration of 100 nM of con-target RNA. When testing the more stable MBsnc7SL (ΔG = −5.25) a concentration of 200 nM of snc7SL target RNA was sufficient to reach a plateau near F _max. A concentration of 100 nM snc7SL target RNA opened MBsnc7SL but without reaching F _max in the given period of time (Figure 4B, Supplementary Table 1A and B). In case of the stable MB7SL (ΔG = −8.62) even a 4-fold increased concentration of 7SL-target DNA (400 nM) was not sufficient to reach F _max in the given time period (Figure 4C, Supplementary Table 1A and B). Taken together, the calculated stability of the MBs correlated with the sensitivity of the MBs and a more stable MB required a higher target molecule concentration for a strong activation.

Figure 4:

Sensitivity of Molecular Beacons. A) 100 nM of MBcon plus con-target RNA, B) 100 nM of MBsnc7SL plus snc7SL-target RNA, and C) 100 nM of MB7SL plus 7SL-target DNA. MB specificities were tested by incubation of 25, 100, 200, 400 nM of respective target RNA or DNA, without any RNA (F _back) or in presence of 50 % formamide (+F, F _max). All reactions were assembled on ice, placed in the plate reader preheated to 37 °C. Fluorescence was recorded at 0 and every 10 min for 2 h. n = 2, AW, assay window.

Next, we tested the specificity of the MBs. Therefore, we incubated MBs (100 nM) with 500 nM target molecules. As shown, the fluorescence signal of MBs was only activated by specific target molecules, indicating a high specificity of all MBs tested (Figure 5). The snc7SL-target RNA activated strongly MBsnc7SL (ΔG = −5.25) but only to a small extend MBsnc7SL-S likely because of the increased stability (ΔG = −9.66) of the stem extended by two additional G:C base pairs (Figures 2 and 5A). Interestingly, reducing the complementarity of MBsnc7SL-Mut (ΔG = −5.36) to snc7SL-target RNA by a deletion of five nucleotides and the change of three cytosine residues to adenosine residues was sufficient to prevent its activation by snc7SL-target RNA as also supported by the difference in ΔG values for the MBsnc7SL-Mut:sncRNA7SL-target interaction (ΔG = −6.51) when compared to the MBsncRNA:sncRNA7SL-target interaction (ΔG = −53.13, Figure 6A) underscoring the specificity of the MBsnc7SL (Figure 5A) Con-target RNA activated strongly both con-target RNA specific Molecular Beacons (MBcon-S (ΔG = −8.36), MBcon (ΔG = −3.95)), but activation was not overserved for any other MB (Figure 5B). Similar, MB7SL (ΔG = −8.62) was only activated by its 7SL-target sequence but not, as expected, the snc7SL-target sequence (Figure 5C).

Figure 5:

Target specificity of Molecular Beacons. A) 100 nM of MBs were incubated with 500 nM of snc7SL-target RNA. B) 100 nM of MBs were incubated with 500 nM of con-target RNA. C) 100 nM of MB7SL were incubated with 500 nM of 7SL-target DNA, snc7SL-target RNA or without any RNA. All reactions were incubated for 0, 10, 20, 30, 40, 50 min at 37 °C in ICN buffer. n = 3.

Figure 6:

Target specificity of MBsnc7SL. To further test the sequence specificity of MBsnc7SL, degenerated target sequences derived from Alu elements found in the 3′UTR of four mRNAs (Mdm2, Mdm4, eEF2K, KNL1) were chosen. A) ΔG values for the different MB:target sequence interactions are given. B) Sequence of degenerated target sequences: mismatches (mis), deletions (del) or extra (extra) nucleotides are indicated (bold, underlined). V _max, K _m and R squared values are given. Grey shaded = no binding, n = 2, ΔG [kcal/mol] for the base pairing between the loop sequence of MBsnc7SL and the target sequence. C) 100 nM MBsnc7SL were incubated without or with 200 nM target DNA in ICN buffer over time at 37 °C. Fluorescence was recorded at 0 time point followed by 10 min intervals for 2 h.

The 7SL RNA is the precursor of all Alu elements (Jurka 2004; Quentin 1994) and 7SL RNA sequences appear as Alu elements in sense or antisense orientation in the 3′UTR of mRNAs (Zhang et al. 2021). To further assess the specificity of the MBsnc7SL, we tested whether those Alu elements represent potential degenerated snc7SL target sequences. We selected six potential target sequences characterized by a relatively high complementarity (few mismatches only) to the snc7SL RNA (Figure 6B). The selected snc7SL target sequences are located in the 3′UTR of Mdm2 mRNA (one sequence), Mdm4 mRNA (two sequences, Mdm4-S1 and -S2), eEF2K mRNA (two sequences, eEF2K-S1 and -S2) and one sequence in KNL1 mRNA. Target molecules with those sequences were tested for their recognition by MBsnc7SL (Figure 6C). Interestingly, whereas the snc7SL target sequence strongly activated (K _m 37.68 nM) the MBsnc7SL, Mdm2- and Mdm4-S1-target sequences were unable to activate MBsnc7SL (Figure 6B and C) even at concentrations of a 4-fold excess (not shown). The second best target molecule was a sequence element derived from Mdm4 (Mdm4-S2, K _m = 77.43 nM) and the remaining three eEFK2-S1, eEFK2-S1 and KNL1 target molecules activated the MBsnc7SL only with high K _m suggesting a low affinity (Figure 6B and C). Using the program RNAcofold (http://rna.tbi.univie.ac.at//cgi-bin/RNAWebSuite/RNAcofold.cgi; Gruber et al. 2008), we calculated ΔG values for the base pairing between the loop sequence of MBsnc7SL and all potential target sequences (Figure 6B). A correlation between low ΔG values, strong MBsnc7SL opening and higher ΔG values and weak or no MBsnc7SL opening was found. In light of previous studies showing that MBs can even discriminate between target molecules based on a single nucleotide substitution (Bonnet et al. 1999; Tyagi et al. 1998), it is not surprising that degenerated snc7SL RNA target sequences let to no or only to a weak MBsnc7SL activation when compared to the 100 % complementary target sequence. Taken together, the in vitro studies established that MBsnc7SL has a large assay window and is highly sensitive to alterations in the target sequence indicating a high sequence specificity.

2.2 Detection of snc7SL RNA and 7SL RNA in cells

The MB-based detection of snc7SL RNA and 7SL RNA in vitro was highly specific. Next, we aimed to apply MBs to detect and study the subcellular localization of endogenous 7SL RNA and its potential derivate snc7SL RNA. Initially, we plated U2–OS cells into 24-well tissue culture plates, transfected 37.5 nM of MBcon (Figure 7A) or MBsnc7SL (Figure 7B) and took overview pictures 24 h later with the Bio-Rad ZOE cell imager, to test whether it is possible to apply a basic LED-light equipped cell imager to detect MB signals. The control transfection with MBcon showed only very weak signal derived from artificial spots located outside of cells (Figure 7A). In contrast, transfection of MBsnc7SL led to cells with green fluorescence signals mainly localized to the nucleus (Figure 7B, transfection efficiency ∼20 %). As clearly visible in the enlarged cells the green fluorescence signal is concentrated in the nucleus with some cytoplasmic staining (Figure 7C and D). In addition, in about ∼ 19 % of cells positive for the MBsnc7SL signal fluorescent dots were also found in the cytoplasm (Figure 7D, arrows, 2–3 dots/cell). The nature of these structures is currently under investigation.

Figure 7:

Endogenous sncRNA7SL is predominantly detected in the nucleus of cells. A) As a control U2–OS cells were transfected with 37.5 nM MBcon (arrow indicates some artificial fluorescence signal) or B) sncRNA7SL specific MBsnc7SL. C) and D) shows enlarged cells with a strong MBsnc7SL derived fluorescence signal diffuse in the nucleus and some cytoplasmic foci-like staining. Arrows indicate cytoplasmic foci detected in 19 % of MBsnc7SL positive cells. Scale bar = 100 µm. Pictures were processed using Adobe software package elements.

Because of the limited resolution of the ZOE instrument, we applied the KEYENCE BZ-X810 fluorescence microscope. Therefore, we plated cells in chambered coverslip wells and transfected the cells with 12.5 nM MBcon, MBsnc7SL, MBsnc7SL-S or MBsnc7SL-Mut. After 24 h, nuclei were stained with DAPI and pictures taken with the KEYENCE BZ-X810 fluorescence microscope. Again, the MBsnc7SL detected snc7SL RNA mainly in the nucleus (Figure 8B, transfection efficiency: ∼27 %) and only very weak fluorescence signals were found in control transfections applying MBcon, MBsnc7SL-S or MBsnc7SL-Mut (Figure 8A, C, and D). Those findings are in agreement with our in vitro studies showing that the MBsnc7SL only detected specifically the snc7SL target RNA (Figure 6).

Figure 8:

Detection of snc7SL RNA only by MBsnc7SL, but not by MBsnc7SL-S, MBsnc7SL-Mut or MBcon in cells. Cells were transfected with 12.5 nM of A) MBcon, B) MBsnc7SL, C) MBsnc7SL-S, or D) MBsnc7SL-Mut. An enlarged cell is shown in B). Arrows indicate cytoplasmic foci in MBsnc7SL transfected SK–N–SH cells. At 24 h after transfection pictures were taken as Z stacks with a KEYENCE BZ-X810 fluorescence microscope. Magnification = 40×, plan apochromat objective. Scale bar = 10 µM. All pictures were processed using Adobe software package elements.

To further confirm the predominant nuclear localization of snc7SL RNA, U2–OS cells were transfected with the MBsnc7SL and pictures were taken with a confocal microscope. As shown, the confocal pictures confirmed the predominant nuclear localization of snc7SL RNA (Supplementary Figure 1A). Interestingly, some cytoplasmic snc7SL RNA foci co-localize with DAPI stained DNA foci in the cytoplasm. Those foci might represent micro-nucleoli occurring upon chromosome segregation defects as described by others (De Marco Zompit and Stucki 2021; Naim and Rosselli 2009), however, the characterization of those foci requires a more detailed analysis in future studies. As additional positive control, we designed and tested in vitro MB6145-S directed against another endogenously expressed 32 nucleotides long sncRNA, referred to as piR-has-6145 (Figure 1A and B). Upon transfection of MB6145-S, green fluorescence signal was predominantly detected in the nucleus of the cells (Supplementary Figure 1B). Taken together, only MBs (MBsnc7SL, MB6145-S) with expected cellular target RNAs (snc7SL RNA, piR-has-6145) were opened in cells. Control MBs (MBcon) or MBs with stabilized stem structures (MBsnc7SL-S) or reduced target RNA complementarity (MBsnc7SL-Mut) were not activated in cells.

As mentioned above, the snc7SL RNA is considered to derive from the larger 7SL RNA, which is part of the signal recognition particle (SRP). The SRP localizes to the endoplasmatic reticulum (ER), which is associated with the nuclear envelope (Shibata et al. 2006). Because the snc7SL RNA sequence is homolog to the 3′-end of the 7SL RNA, the MBsnc7SL should detect not only snc7SL RNA but also the 7SL RNA (Figure 1B). In contrast, MB7SL should only detect 7SL RNA and not snc7SL RNA (Figure 1C). Hence, we co-transfected Cy5-labeled MB7SL and FAM-labeled MBsnc7SL into SK–N–SH cells. As shown, the MB7SL was opened and a red signal in close proximity to the nuclear envelope and only a weak nuclear staining was detected (Figure 9A and B, Supplementary Figure 1C). The colocalization of the fluorescence signal of an ER-specific dye with the MB7SL fluorescence signal strongly suggests that the detected 7SL RNA – as part of the SRP – colocalizes with the ER (Figure 9A and B, Supplementary Figure 1C). On the other hand, the MBsnc7SL detected a strong signal in the nucleus (Figure 9A and B). To demonstrate the clearly distinguishable Cy5- and FAM-signals but also the partial colocalization, enlarged cells are shown (Figure 9B). MBsnc7SL detected snc7SL RNA mainly as diffuse signal in the nucleus and some peaks also in the cytoplasm (Figure 9B). In contrast, MB7SL detected 7SL RNA mostly in the cytoplasm and only a faint signal in the nucleus was recorded (Figure 9B). However, the co-localization (yellow signal) of the FAM- and Cy5-signals indicates that MBsnc7SL detects also 7SL RNA to some extend in the cytoplasm (Figure 9B). This observation is further strengthen by line profiles measuring the fluorescence intensity along a drawn line. The main red signal (red line) derived from MB7SL is recorded in the cytoplasm in proximity to the nucleus (blue line) and faint red signal is also detected in the nucleus (Figure 9C). The green line profile is derived from MBsnc7SL and shows the strongest signal in the nucleus but also outside in proximity to the nuclear envelope, suggesting some recognition of the 7SL RNA.

Figure 9:

Subcellular localization of 7SL RNA and snc7SL RNA in SKNSH cells. A) Cells were co-transfected with 12.5 nM MB7SL and 12.5 nM MBsnc7SL and the nucleus stained with DAPI. B) Enlarged cells with red signal derived from activated MB7SL show the location of 7SL RNA in distinct areas close to the nuclear envelop (grey arrows). The green FAM signal of activated MBsnc7SL indicates the predominant localization of snc7SL RNA in the nuclei. Colocalization of the FAM and Cy5 signal shows the distinct localization of 7SL RNA outside of the nucleus and the predominant localization of snc7SL RNA within the nucleus. However, some colocalization of both signals were detected (yellow). Stars indicating cytoplasmic foci of unclear nature. C) Line profile. 24 h after transfection pictures were taken as Z stacks with a KEYENCE BZ-X810 fluorescence microscope. Z stack were taken with a 40× magnification. Plan apochromat objective. Scale bar = 10 µM. All pictures were processed using Adobe software package elements.

Taken together, herein we report the application of different Molecular Beacons for the intracellular detection of an endogenous large precursor RNA (7SL RNA) and a small processing product (snc7SL RNA) by multiplex live cell microscopy. Interestingly, the MB-based detection revealed a distinct sub-cellular localization of both endogenous RNA molecules.

4.1 DNA and RNA oligonucleotides and Molecular Beacon (MB)

All DNA oligonucleotides were purchased from (Integrated DNA Technology (IDT)) and all RNA oligonucleotides from Horizone/Dharmafect.

Target RNA oligonucleotides:

RNA modifications: * = Phosphorothioate bond, m = 2′-O-methylation of RNA bases, P = phosphate

snc7SL-target RNA (32 nts):

5′-P-rA*rG*rC*rC*rU*rG*rA*rG*rC*rA*rA*rC*rA*rU*rA*rG*rC*rG*rA*rG*rA*rC

*rC*rC*rC*rG*rU*rC*rU*rC*rU*mA-3′

con-target RNA (32 nts):

5′-P-rA*rU*rC*rU*rC*rU*rG*rC*rC*rC*rC*rA*rG*rA*rG*rC*rG*rA*rU*rA*rC*rA

*rA*rC*rG*rA*rG*rU*rC*rC*rG*mA-3′

Target DNA oligonucleotides:

snc7SL-target DNA (wildtype target sequence):

5′-AGCCTGAGCAACATAGCGAGACCCCGTCTCTA-3′

7SL-target DNA (wildtype target sequence):

5′-AGTTCTGGGCTGTAGTGCGCTATGCCGATC-3′

Degenerated target sequences for snc7SL RNA located in the 3′ UTR of specific mRNAs (mismatches, extra nucleotides or deletions when compared to the wildtype target sequence are shown underlined and as capital, bold letters):

Mdm2: 5′- - gc T t AG g AG aca G a TG gag- A c TT gtctct -

Mdm4-S1: 5′-agcctga C c G acat G g A ga A acccc A t T t T ta

Mdm4-S2: 5′-agcctg G gcaaca A tagcga A a -- c T gtctc - a

eEF2K-S1: 5′-agcctg G g T aa A cat G gcgagaccc T gtctct -

eEF2K-S2: 5′-agcctg G gc G aca G ag T gagac T ccgtctc G -

KNL1: 5′-agcctg G gcaacat G gc A a A acccc AC ctcta

Molecular Beacon (MB) specifications:

FAM = fluorescein: excitation peak at 493 nm, emission peak at 517 nm

DAB = Dabcyl, quenching range: 400–550 nm

Cy5 = Cyanine 5: excitation peak at 651 nm, emission peak at 670 nm

BHQ2 = Black Hole Quencher 2, quenching range: 520–650 nm

MB-snc7SL:

5′-FAM-cgcgAGAGACGGGGTCTCGCTATGTTGCTCAGGCTcgcg-DAB-3′

MB-snc7SL-S:

5′-FAM-cgcgcgAGAGACGGGGTCTCGCTATGTTGCTCAGGCTcgcgcg-DAB-3′

MB-snc7SL-Mut (grey indicate nucleotide deletions or changes (small letters) when compared to wildtype snc7SL RNA):

5′FAM-cgcgca----ACGGGGTaTCGaTATGTTGCTCAGGC-agcgcg-DAB-3′

MBcon:

5′FAM-cgcgTCGGACTCGTTGTATCGCTCTGGGGCAGAGATcgcg-DAB-3′

MBcon-S:

5′FAM-cgcgcgTCGGACTCGTTGTATCGCTCTGGGGCAGAGATcgcgcg-DAB-3′

MB7SL:

5′Cy5-cgcgaGATCGGCATAGCGCACTACAGCCCAGAACTcgcg-BHQ2-3′

MB6145-S:

5′Fam-cgcgcgTGAGGCGAACGTGATAACCACTACACTACGGAcgcgcg-D-3′

4.2 In vitro Molecular Beacon (MB) assay

All reactions (20 µL) were performed in 384-well black plates (Perkin Elmer) and fluorescence was recorded over time at 37 °C in a BMG Caliber instrument (gain set at 1000). All reagents and 384-well plates were cooled on ice. Reactions comprised of MBs and target RNA or DNA oligonucleotides were assembled on ice in 1× ICN buffer (10 mM Tris/HCl, pH 7.5, 100 mM KCl, 10 mM NaCl, 0.75 mM MgCl₂, 0.0001 % NP40) and transferred to the pre-warmed (37 °C) multiplate reader.

4.3 Cells and cell culture

SK–N–SH and U2–OS cells were expanded and aliquots were stored in liquid nitrogen. New cell cultures were established every 2–3 months. SK–N–SH and U2–OS cells were purchased from DSMZ (German Collection of Microorganisms and Cell Cultures GmbH). SK–N–SH cells were cultured EMEM/Ham’s F12 (1:1) medium with 10 % fetal bovine serum, 2 mM glutamine, 1 % penicillin/streptomycin. U2–OS cells were cultured in RPMI with 10 % fetal bovine serum, 4 mM glutamine, 1 % penicillin/streptomycin.

4.4 Transfection of Molecular Beacons using RNAiMax

Molecular Beacons (MB) were transfected using the Lipofectamine RNAiMAX Transfection Reagent (ThermoFisher Scientific, #13778030).

For live-cell imaging with the Bio-Rad ZOE cell imager 24-Well plates were used. For a single transfection 100 μM MB stock solutions were diluted to 1 µM in H₂O and further diluted to 125 nM in 50 µL Opti-MEM. Furthermore, 1 µL Lipofectamine RNAiMAX was mixed with 49 µL Opti-MEM in a second Eppendorf tube. Contents of both tubes were mixed together and incubated at room temperature for 5 min. Afterwards transfection solution was mixed with 400 µL of medium containing between 50,000 and 150,000 U2–OS cells. Cells and transfection solution were mixed by pipetting and plated into one 24-Well plates. Final MB concentration 12.5 nM.

For live-cell imaging with the KEYENCE BZ-X810 fluorescence microscope 30,000 cells were plated into µ-slide 8 well chambered coverslip (Ibidi, #80826) and either cultured for 24 h prior to transfection at 37 °C in 5 % CO₂ atmosphere in medium (pre-plated) or mixed directly with the transfection mix and plated.

For a single transfection MB stocks (100 µM) were diluted to 1 µM in H₂O and further diluted in Opti-MEM (Life Technologies #11058021) to a concentration of 125 nM, 250 nM, 500 nM, 750 nM and 1 µM in 30 µL. In a second Eppendorf tube, 0.6 µL RNAiMAX was mixed with 29.4 µL Opti-MEM. All contents were thoroughly mixed by pipetting. The liquids of the two Eppendorf tubes were combined, mixed thoroughly by pipetting and incubated for 5 min at room temperature, resulting in concentrations of 62.5 nM, 125 nM, 250 nM, 375 nM and 500 nM. Finally, 60 µL transfection mix was mixed with 240 µL medium and added to one well of pre-plated cells. Final MB concentrations are 12.5 nM, 25 nM, 50 nM, 75 nM or 100 nM.

Cells were cultured for 24 h at 37 °C in a humified incubator set to 5 % CO₂ atmosphere. Subsequently cells were washed once with 1× PBS (Life Technologies #14190-169), stained with DAPI (diluted 1 drop per mL in PBS, Thermos Fisher #R37605,) and incubated for 10 min at room temperature. Next the DAPI solution was replaced with 1× PBS and cells were washed once with 1× PBS. Finally, cells were covered with 300 µL 1× PBS and analyzed by fluorescence microscopy using the KEYENCE (BZ-X810) microscope or the Leica TCS SP5 confocal laser scanning microscope. In case of live-cell imaging with the Bio-Rad ZOE cell imager cells were not stained with DAPI.

4.5 Cell imaging and fluorescence microscopy

For live cell imaging with a Bio-Rad ZOE cell imager, a brightfield channel, a green channel (excitation: 480/17 nm; emission: 517/23 nm) and a LED light source was used. Live cell images were taken with a KEYENCE BZ-X810 fluorescence microscope equipped with a LED light source. Exposure times of 1/35 s to 3 s and magnifications between 20× (Objective: Plan Fluor 20× Na 0.45×), 40× (Objective: Plan Apo 40× NA 0.95×) and 100× (Objective: Plan Apo 100× 1.45, Oil) were applied. The following filters were used: BZ-X filter GFP (excitation 470 nM; emission 525 nM); BZ-X filter Cy5 (excitation 620 nM; emission 700 nM), BZ-X filter DAPI (excitation 360 nM; emission 460 nM). Z-stacks in the range of 0.2–0.8 µM were generated and line profiles were created by using the KEYENCE measurement module. For background signal adjustment, untransfected cells or control (MBcon) transfected cells were imaged. Exposure times were set so that control cells do not provide any signal. For data analysis, the BZ-X800 Analyzer software package (Version 1.1.2.4) was used.

The Leica TCS SP5 confocal laser scanning microscope with a laser as light source (Argon ion laser for 458 nm, 476 nm, 488 nm, 496 nm, 514 nm; Helium–neon laser for 633 nm; Diode laser for 561 nm and Pulsed UV laser (including B&H control system)) was used for live-cell imaging as well. Magnification of 40× (oil) and 63× (oil) were applied. Lasers were set on a range of 470 nM–525 nM for FAM detection, on a range of 620 nM–700 nM for Cy5 detection and on a range of 360 nM–460 nM for DAPI detection, respectively. Z-stacks were generated. For background signal adjustment, untransfected cells or control (MBcon) transfected cells were imaged. Exposure times were set so that control cells do not provide any signal. Pictures were processed with the Leica Application Suite (Version 2.7.3.9723) and with the Adobe software package Elements.

4.6 Staining of the endoplasmic reticulum (ER)

ER Tracker Green 100 µg, Thermo fisher #E34251, dye BODIPY^® FL (EX/EM 504/511) was dissolved in 128 µL DMSO upon arrival to a concentration of 1 mM. ER Tracker Green Stock solution (1 mM) was diluted to 100 nM in 1× PBS and 300 µL were added to the cells and incubated for 20 min at 37 °C. Thereafter, the liquid was replaced with fresh 1× PBS and pictures were taken with the KEYENCE BZ-X810 fluorescence microscope using the BZ-X filter GFP (excitation 470 nM; emission 525 nM).