| Nucleic Acids Research | Pages |
Design of modified oligodeoxyribonucleotide probes to detect telomere repeat sequences in FISH assays
Introduction
Materials And Methods
Synthesis of modified oligodeoxyribonucleotides
FISH experiments
Image acquisition
Results And Discussion
Design of probes to bind to (T2AG3)n repeats
Design of probes to bind to (C3TA2)n repeats
Properties of probes designed to bind to (T2AG3)n repeats
Properties of probes designed to bind to (C3TA2)n repeats
Potential FISH-related applications for modified oligonucleotide probes
References
Design of modified oligodeoxyribonucleotide probes to detect telomere repeat sequences in FISH assays
Received June 25, 1999; Accepted August 20, 1999
ABSTRACT A series of dye-labeled oligonucleotide probes containing base and sugar modifications were tested for the ability to detect telomeric repeat sequences in FISH assays. These modified oligonucleotides, all 18 nt in length, were complementary to either the cytidine-rich (C3TA2)n or guanosine-rich (T2AG3)n telomere target sequences. Oligonucleotides were modified to either increase target affinity by enhancing duplex stability [2[prime]-OMe ribose sugars and 5-(1-propynyl)pyrimidine residues] or inhibit the formation of inter- or intramolecular structures (7-deazaguanosine and 6-thioguanosine residues), which might interfere with binding to the target. Several dye-labeled oligonucleotide probes were found that could effectively stain the telomeric repeat sequences of either cytidine- or guanosine-rich strands in a specific manner. Such probes could be used as an alternative to peptide nucleic acids for investigating the dynamics of telomere length and maintenance. In principle, these relatively inexpensive and readily synthesized modified oligonucleotides could be used for other FISH-related assays.
INTRODUCTION
The multiple roles telomeres play in cellular and organismal biology are just now being elucidated (1-3). In mammals and birds, telomeres typically consist of multiple copies of the hexameric repeat sequence (T2AG3)n (4-6). The length of this repeat sequence varies with the stage of the cell cycle, the age and type of cell, as well as other environmental factors (7-8). In general, human telomeres tend to be 5-15 kb in length, corresponding to ~800-2400 hexameric repeats (9). Recently, there has been evidence that human telomeres terminate in single-stranded G-rich overhangs ~100-200 bp in length (10) and that these single-stranded segments may fold back on and invade upstream repeat sequences to form `t-loop' circular structures (11). Sensitive detection and accurate measurement of these repeats will help to shed further light on the dynamics of telomere maintenance.
Oligonucleotide analogs with enhanced affinity towards complementary nucleic acids have been used in a wide range of biophysical and biochemical studies (12,13). Those containing 5-(1-propynyl)pyrimidine bases (14,15) or 2[prime]-O-alkyl-modified ribose sugars (16-19) have been especially well characterized. Peptide nucleic acids (PNAs) are a class of oligonucleotide analogs in which the natural phosphodiester backbone is replaced by amide linkages (20-22). The most commonly used PNAs typically consist of achiral charge neutral N-(2-aminoethyl)glycine units and have high affinity for complementary nucleic acids under a wide range of hybridization conditions (20-26). Dye-labeled PNAs have proven to be efficient stains for telomeric (23,26), centromeric (24), and expanded trinucleotide repeat sequences (25) in fluorescence in situ hybridization (FISH) experiments (PNA-FISH). The reported quantitative binding of PNAs to (T2AG3)n repeat sequences has allowed telomere repeat lengths from individual chromosomes to be estimated (26). PNA-FISH experiments have also been used to characterize the products of irradiation-induced chromosomal damage in cell culture (27,28).
In this study, we investigated using oligonucleotides with modified bases and sugars to detect telomeric repeat sequences in FISH experiments. Specifically, we screened for dye-labeled oligonucleotides that could stain either the cytidine-rich (C3TA2)n or guanosine-rich (T2AG3)n strands of telomeric repeat sequences from human cell lines. Cost-effective and readily synthesized modified oligonucleotides were found which provide an alternative to PNA-based telomere stains, and may offer significant advantages in some sequence contexts. In addition, they are readily amenable to further dye and ligand modifications using established protocols.
MATERIALS AND METHODS
Synthesis of modified oligodeoxyribonucleotides
All the nucleoside phosphoramidites as well as Cy3-phosphoramidite were obtained from Glen Research (Sterling, VA). Oligonucleotides were made on a PE Biosystems 391 DNA synthesizer using coupling protocols recommended by Glen Research and labeled with Cy3 dye on their 5[prime]-ends using Cy3 phosphoramidite. PNA 2 (Table 1) was obtained from PE Biosystems (Framingham, MA). Oligonucleotides were de-protected according to the manufacturer's recommended protocols and lyophilized. The oligonucleotide pellet or oil was resuspended in formamide, heat denatured at 90°C for 2 min, and then purified by PAGE (29). Oligonucleotides were eluted from the gel slices by the crush and soak method (29). Since some oligonucleotides could not be efficiently ethanol precipitated, all were desalted by dialysis against 10 mM Tris-HCl pH 7.5 using the Slide-A-Lyzer system from Pierce (Rockford, IL) with a 3500 mol. wt cut-off membrane. Oligonucleotide concentrations were determined by their absorbance at 260 nm using the following nucleoside extinction coefficients: A, 15 400; C, 7300; G, 11 700; T, 8800; U, 10 000; 5-(1-propynyl)U, 3200; 5-(1-propynyl)C, 5000; 7-deazaG, 12 500; 6-thioG, 10 000.
Table 1. Oligonucleotide probes tested in FISH experiments
1d, 2[prime]-deoxyribose sugar; r, 2[prime]-OMe-ribose sugar; C, 5-(1-propynyl)cytosine; U, 5-(1-propynyl)uracil; a, PNA amide linkages; S, 6-thioguanine; D, 7-deazaguanine.
2Concentration of probe tested is given next to the maximum assigned signal intensity. The highest and lowest concentrations tested were 0.1 and 10 µM, respectively. -, no more than two faint telomere signals per metaphase spread; +/-, several faint telomere signals per metaphase spread; +, consistently moderate telomere signals; ++, consistently strong telomere signals.
FISH experiments
FISH experiments were performed as previously described (23). Slides containing human metaphase chromosome spreads were obtained from Vysis. Slides were rinsed in phosphate-buffered saline, 10 mM MgCl2 (PBSM) for 15 min prior to fixation in 4% formaldehyde in PBSM (F-PBSM). Slides were then washed three times in PBSM (5 min each) prior to treatment with 1 mg/ml pepsin in 10 mM HCl (37°C for 10 min). The slides were rinsed in PBSM and then treated in F-PBSM for 2 min. Slides were washed three times in PBSM for 5 min each prior to room temperature ethanol dehydration and air drying. A 15 µl volume of hybridization buffer [10 mM Tris-HCl pH 7.2, 70% formamide, 1% blocking reagent (Boehringer-Mannheim)] containing the indicated oligonucleotide was added to the slides which were then heat denatured at 80°C. Slides were incubated at room temperature for 8-12 h. Afterwards, slides were briefly washed in PBSM, with FT buffer (70% formamide, 10 mM Tris-HCl pH 7.2) for 15 min, and finally for 15 min with TNT buffer (10 mM Tris-HCl pH 7.2, 150 mM NaCl, 0.05% Tween-20). Slides were then ethanol dehydrated, air dried, and stained with DAPI II solution (Vysis).
Image acquisition
Digital images wereacquired using a Nikon Microphot FXA microscope equipped with a Quantix cooled CCD camera. Image data was processed using IP Lab v.3.2 (Scanalytics, Vienna, VA) and printed using Adobe Photoshop v.3.0.
RESULTS AND DISCUSSION
Design of probes to bind to (T2AG3)n repeats
Based upon the design of cytosine-rich PNA probes which efficiently stain (T2AG3)n telomeric repeat sequences in FISH assays (23), different classes of modified oligonucleotides were synthesized for similar applications (Fig. 1 and Table 1). Under the assay conditions (Materials and Methods), cytosine-rich oligonucleotides are not expected to contain significant inter- or intramolecular structures that may inhibit hybridization to complementary DNA (30). However, the guanosine-rich telomere strand has the potential to form intramolecular structures that may inhibit hybridization to telomeric probes (31). Therefore, our first efforts were geared toward enhancing the affinity of cytosine-rich oligonucleotides to the guanosine-rich target (Table 1). We constrained our efforts towards making oligonucleotides that could be synthesized with commercially available reagents and conventional oligonucleotide synthesis protocols to increase the accessibility of this technology to the common molecular biology laboratory.
Figure 1. Schematic of modified bases incorporated into oligonucleotide probes. meU, 5-methyluracil or thymine; pU, 5-(1-propynyl)uracil; pC, 5-(1-propynyl)cytosine; deazaG, 7-deazaguanine; sG, 6-thioguanine.
DNA oligonucleotide 1 contains the core sequence of all cytosine-rich probes tested. Probe 2 is the corresponding PNA with a charge neutral achiral N-(2-aminoethyl)glycine backbone. DNA oligonucleotide 3 contains modified 5-(1-propynyl)-uracil (pU) and 5-(1-propynyl)cytosine (pC) pyrimidine bases. These bases can significantly enhance affinity towards complement through favorable stacking interactions and possible entropic effects by affecting duplex hydration (14,15). The 2[prime]-OMe-ribose-modified phosphodiester backbone in oligonucleotides 4 and 5 [containing uracil (U) and thymine (T) bases, respectively] should also increase their affinities towards guanine-rich complement (16-19). Oligonucleotide 5 should have higher target affinity than oligonucleotide 4 due to the greater stability of A*T relative to A*U base pairs (32). Both 2[prime]-OMe-ribose residues and pC bases were incorporated into oligonucleotide 6. Due to increased stability of pC*G base pairs relative to C*G base pairs this oligonucleotide should have increased affinity towards the target relative to oligonucleotide 5.
Design of probes to bind to (C3TA2)n repeats
Unlike the case of cytosine-rich oligonucleotide probes designed to bind to(T2AG3)n repeats, there has been limited reported success in binding guanine-rich probes to (C3TA2)n telomeric repeats. This could be due to two non-exclusive reasons. First, the unmodified guanine-rich probes tested may lack sufficient affinity for the cytidine-rich target sequence. Second, higher order structures in either the guanine-rich probe (31) or target (30) could interfere with hybridization. However, the cytidine-rich telomere target strand is not expected to form stable structures under these assay conditions and therefore may be more accessible (30).
Analogs of guanine-rich DNA oligodeoxyribonucleotide 7 were designed using similar strategies employed for the cytosine-rich oligonucleotide series (Fig. 1). Since synthesis of the guanine-rich PNA analog was not recommended by the manufacturer (PE Biosystems) due to the formation of strong secondary structures, it was not tested in this study. DNA oligonucleotide 8 contains pU residues while oligonucleotides 9 and 10 (containing U and T residues, respectively) have 2[prime]-OMe-ribose backbones. Such modifications should increase the affinity of oligonucleotides 8-10 towards the target relative to oligonucleotide 7.
Oligonucleotides 11 and 12 were synthesized with guanine analogs designed to inhibit intra- and intermolecular structures while retaining affinity for complement. Oligonucleotide 11 contains 6-thioguanine, which has been shown to inhibit formation of guanine-based structures in oligonucleotides (33). Similarly 7-deazaguanine is frequently incorporated into DNA sequencing reactions to destabilize secondary structures that affect electrophoretic migration patterns (34); oligodeoxyribonucleotide 12 contains this substitution as well as thymidine residues.
Properties of probes designed to bind to (T2AG3)n repeats
We compared the abilities of 5[prime]-Cy3-dye-labeledoligonucleotides 1-6 to detect telomere repeat sequences in FISH assays (Table 1). At 10 µM concentration, unmodified DNA oligonucleotide 1 produced only moderate telomere-specific hybridization signals (Fig. 2A). However, PNA oligonucleotide 2 produced bright telomere-specific signals at 100 nM concentration (Fig. 2B). Oligonucleotides 3-6 all produced bright telomeric signals comparable to PNA probe 1 at 100 nM concentration (Fig. 2C and D).No distinct non-telomeric locations showed that signal and general non-specific chromosome staining was low. Although these probes provide robust performance at this concentration, they will likely show different staining efficiencies at lower concentrations and different metaphase chromosome preparation qualities. Both the cost of reagents and the effort necessary to synthesize DNA oligonucleotide 1 and 2[prime]-OMe-modified oligonucleotide 4 are approximately the same. The cost per telomere stain reaction is comparable for oligonucleotide 4 and PNA 2 when the latter is purchased in bulk quantities. Collectively, these results show that a wide variety of modified bases, sugars, and internucleotide linkages can be incorporated into short cytosine-rich oligonucleotides to enhance both target affinity and specificity for telomeric repeat sequences.
A
 |
B
 |
C
 |
D
 |
Figure 2. Representative FISH experiments using telomere-specific oligonucleotide and PNA probes. Metaphase chromosome spreads were hybridized with 5[prime]-Cy3-labeled probe at the indicated concentration and subsequently stained with DAPI. Pseudo-colored images of fluorescent hybridization patterns are given with probe and DAPI staining depicted in green and red, respectively. (A and B) Hybridization patterns of cytosine-rich probes 2 and 6, both at 0.1 µM. (C and D) Hybridization patterns of guanine-rich probes 7 and 9 at 10 and 0.1 µM, respectively. Image processing and exposure times for all images are comparable, so differences reflect probe efficiency. Limitations associated with imaging in a single focal plane preclude detection of telomere-specific signals with equal efficiency in individual metaphase chromosome spreads.
Properties of probes designed to bind to (C3TA2)n repeats
5[prime]-Cy3-dye-labeled guanine-rich oligonucleotides were likewise evaluated for the ability to bind to telomere repeats in FISH assays (Table 1). Unmodified oligodeoxyribonucleotide 7 did not produce telomere-specific hybridization signals at 10 µM concentration (Fig. 2E). Oligodeoxyribonucleotide 8 produced faint telomeric signals only when incubated at 10 µM concentration and global non-specific staining could be discerned. At 100 nM concentration, 2[prime]-OMe-ribose-modified oligonucleotides 9 and 10 both produced strong telomere-specific hybridization signals comparable to that of PNA oligonucleotide 2 with little non-specific hybridization (Fig. 2F). Oligonucleotides 11 and 12, containing 6-thioguanine and 7-deaza-guanine bases, did not consistently produce hybridization signals at 10 µM concentration.
It is difficult to determine the main reason unmodified guanosine-rich oligodeoxyribonucleotide 7 is an ineffective telomere-specific stain. Oligonucleotides 9 and 10 should have higher affinities for the target sequence, but their base and sugar substitutions could also inhibit secondary structure formation. Oligonucleotides 11 and 12 did not stain telomeric sequences even though they should form less stable structures than unmodified oligodeoxyribonucleotide 7. However, this could be due to the lower stability of 6-thioguanine and 7-deaza-guanine base pairing to cytosine relative to guanine (35,36).
The short oligonucleotide probes 9 and 10 are cost-effective alternatives to PNA 2 which furthermore detect repeat sequences located on the alternate telomeric strand. This could become useful in studying the dynamics of telomere maintenance given recent evidence of higher order telomeric structures such as t-loops (11).
Potential FISH-related applications for modified oligonucleotide probes
Oligonucleotide probes can be powerful tools in FISH-based experiments (37). DNA oligonucleotides ~40 nt in length have been used as probes for repetitive alphoid DNA sequences, the human 5S rDNA repeat, and a specific family of Alu repeats (38). A major challenge is to detect single copy sequence tracts. This entails both high oligonucleotide binding affinity and specificity as well as extremely sensitive hybridization detection. Novel strategies towards increasing hybridization detection sensitivity include using oligonucleotides conjugated to energy transfer dyes (39,40), dye-labeled dendrimers (41), and enzymes for tyramide signal amplification assays (37). In addition, long circularizable probes and DNA polymerase-mediated signal amplification schemes have been used in the padlock probe strategy (42,43). Oligonucleotides containing the general types of base and sugar modifications used in the present study should potentially be adaptable for all these approaches.
In the PRINS (primer in situ labeling) methodology, oligodeoxyribonucleotides are hybridized to chromosome spreads and used as primers to incorporate dyes or haptens (44). This procedure can be simplified by using labeled probes such as those described here to eliminate enzymatic steps needed to incorporate label. Furthermore, while PRINS could detect telomeric repeats using a cytosine-rich primer (45,46), it was inefficient for detectingthe same telomeric repeats with a guanosine-rich primer (45). This could be due to secondary structures or poor target affinity, as discussed herein for the guanosine-rich oligodeoxyribonucleotide 7. The high target affinity of 2[prime]-OMe-ribose- and 5-(1-propynyl)pyrimidine-modified probes (or similar chimeras with 3[prime]-ends consisting of DNA residues) could make them well suited for PRINS assays.
The relative merits of using both standard and modified oligonucleotides for detecting telomeric repeat sequences were examined in this study. A wide variety of modified oligonucleotides, all having increased affinity for the target sequence relative to standard oligonucleotide sequences, showed enhanced telomere staining potential. While the pyrimidine-rich PNA probe tested in this study is a robust reagent for staining telomere repeat sequences, comparable results could be obtained from 2[prime]-OMe-ribose-modified pyrimidine- and purine-rich oligonucleotides of equivalent length. The inclusion of 5-(1-propynyl)pyrimidines in DNA probes also enhanced the affinity of pyrimidine-rich probes. However, the fact that 2[prime]-OMe modifications can be incorporated at every nucleoside position makes this approach less sensitive to sequence composition than the 5-(1-propynyl)-modified pyrimidines. All of these modified oligonucleotide probes can be easily synthesized with the same equipment and PAGE purification protocols used for standard oligodeoxyribonucleotides. In contrast, specialized equipment and more involved chromatographic purification steps are needed for PNA synthesis. It seems likely that similar modified oligonucleotides could also be designed to recognize virtually any sequence tract in FISH assays, especially if tandem repeats were present. This would make oligonucleotide-FISH technology even more cost-effective and readily accessible to molecular biologists.
REFERENCES
*To whom correspondence should be addressed. Tel: +1 301 496 0844; Fax: +1 301 402 0837; Email: fc23a{at}nih.gov
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  V. V. Filichev, B. Vester, L. H. Hansen, and E. B. Pedersen Easily denaturing nucleic acids derived from intercalating nucleic acids: thermal stability studies, dual duplex invasion and inhibition of transcription start Nucleic Acids Res., December 23, 2005; 33(22): 7129 - 7137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. F. Riou, L. Guittat, P. Mailliet, A. Laoui, E. Renou, O. Petitgenet, F. Megnin-Chanet, C. Helene, and J. L. Mergny Cell senescence and telomere shortening induced by a new series of specific G-quadruplex DNA ligands PNAS, February 14, 2002; (2002) 52698099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Riou, L. Guittat, P. Mailliet, A. Laoui, E. Renou, O. Petitgenet, F. Megnin-Chanet, C. Helene, and J. L. Mergny Cell senescence and telomere shortening induced by a new series of specific G-quadruplex DNA ligands PNAS, March 5, 2002; 99(5): 2672 - 2677. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||