WO1997030152A1

WO1997030152A1 - Separation of single stranded and double stranded nucleic acid materials

Info

Publication number: WO1997030152A1
Application number: PCT/NL1997/000062
Authority: WO
Inventors: Jaap Goudsmit; Cornelis Johannes Andreas Sol; Marcellinus Gualbertus Hubertus Maria Beld; Willem René BOOM
Original assignee: Amsterdam Support Diagnostics B.V.
Priority date: 1996-02-14
Filing date: 1997-02-14
Publication date: 1997-08-21
Also published as: AU1676597A

Abstract

The invention relates to methods for easily separating single stranded nucleic acid material from double stranded nucleic acid material in a sample containing both. By the right choice of at least one chaotropic agent, preferably a guanidine salt, at a selected concentration and other suitable conditions such as chelating agents, pH and the like, it is possible to bind double stranded material to a solid phase such as silica particles, whereas single stranded material will not bind under those circumstances. By separating the silica particles from the sample the double stranded nucleic acid material is removed. It can easily be eluted from the silica particles. In a second step the single stranded material may be bound to a solid phase by selecting a different set of conditions. The particles can again be separated from the sample and the single stranded material may now be eluted. For very efficient separations the process may be repeated.

Description

Title: Separation of single stranded and double stranded nucleic acid materials.

The invention relates to the field of purification and separation of nucleic acids from nucleic acid containing starting materials, especially from biological materials such as urine, faeces, sperm, saliva, whole blood, serum or other body fluids, fractions of such fluids such as leucocyte fractions (buffy coats), cell cultures and the like, but also samples from the environment such as soil, water and the like.

Until recently isolation and/or purification of nucleic acids from complex mixtures as described above was a laborious, multi-step procedure. In EP 0389063, incorporated herein by reference, a simple and rapid purification of nucleic acid material from a complex mixture is disclosed. This procedure comprises treating the complex mixture, such as whole blood with "a chaotropic agent in the presence of a nucleic acid binding silica solid phase material under conditions that allow for binding of all nucleic acid material to said solid phase and separating said solid phase from the mixture. The reference shows that both single stranded and double stranded nucleic acids are bound to the solid phase if present in a mixture.

Thus said reference teaches a simple and rapid detection method for known nucleic acids suspected to be present in a sample.

In many cases the nature of the target nucleic acid (double stranded or single stranded) may not be known beforehand, or there may be many different targets necessary to be analyzed. In these cases the rapid but rather crude method described above may not be sophisticated enough and further separations of the crude material may be wanted. Fractionation of mixtures of double- (ds) and single-stranded (ss) nucleic acids (NA) into single- and double-stranded forms is frequently needed e.g. in the separation of labelled ss-NA probes from ds-hybrids, in the separation of in vitro transcripts from ds-DNA templates, and in the separation of genomic DNA from mRNA. Currently, the separation of different kinds of nucleic acids can be accomplished by several techniques. Electrophoresis can be used to fractionate different forms of nucleic acids, because of differences in size and shape (1-3). Centrifugation takes advantage of differences in density (4), and more recently the technology of high-performence liquid chromatography (HPLC) has been applied to separate and purify single- and double-stranded DNA and RNA molecules (5-8).

RNA purified from eukaryotic cells by the currently most widely used procedure (9) appears to contain significant amounts of genomic DNA, an adaptation which reduces genomic DNA contamination of the ss-RNA fraction has recently been described (10).

It is not possible to look at single stranded and/or double stranded material separately using the method of EP 0389063 because the method does not discriminate between the two. The present invention therefor provides a method for separating single stranded nucleic acid material from double stranded nucleic acid material comprising contacting a mixture of the both with a liquid comprising a chaotropic agent and a nucleic acid binding solid phase, whereby the liquid has a composition such that double stranded nucleic acid binds to the solid phase and a substantial amount of single stranded nucleic acid does not and separating the solid phase from the liquid. Suitable circumstances to arrive at such a separation can be determined by the person skilled in the art. Circumstances under which double stranded material binds to the solid material and single stranded material will vary, however important parameters to obtain such differential binding are the concentration of the chaotropic agent, which should roughly be between 1-10 M, preferably between 3-6 and particularly about 5 M; the concentration of chelating agent, which in the case that EDTA is applied should be equal to or greater than 10 mM and preferably not higher than 1 M; the pH of the aqueous solution in which the separation is carried out should be above 2 when a thiocyanate is used as chaotropic agent and it should be below 10 because otherwise there is a risk that the ds material will become ss. The temperature at which the process is carried out seems to be non-critical, however, it is probably best to keep it between 4°C and 60°C. An important aspect of the process is of course that the ds material remains double stranded during the separation. Under the circumstances as disclosed above this will normally be the case if the ds nucleic acid is at least 50 bp long at 40% GC basepairs . The skilled artisan knows how this length may vary with lower or higher GC content. In Van Ness et al (17) and/or Thompson et al (18) it is shown that the whole process depends on intricate interactions between a.o. the factors mentioned above. Using this disclosure and the cited references the skilled artisan will be able to adjust the circumstances to his or her particular process.

Chaotropic agents are a very important feature of the present invention. They are defined as any substance that can alter the secondary, tertiary and/or quaternary structure of nucleic acids . They should have no substantial effect on the primary structure of the nucleic acid. If nucleic acids are present associated with other molecules, such as proteins, these associations can also be altered by the same or different chaotropic agents. Many chaotropic agents are suitable for use in the present invention, such as sodium iodide, potassium iodide, sodium (iso)thiocyanate, urea or guanidinium salts, or combinations thereof. A preferred class of chaotropic agents according to the invention are guanidinium salts, of which guanidinium thiocyanate is most preferre .

By serendipity we found that ss-nucleic acid did not bind to silica particles or diatomeous earth in the presence of buffer Lll (see examples), whereas ds nucleic acid did. Experiments with different circumstances showed that addition of Mg2⁺ or other positive (bivalent) ions to the unbound fraction was of great importance. The best results were obtained with a concentration of bivalent ion (Mg2⁺) about equal to the concentration of the chelating agent (EDTA).

The solid phase to be used is less critical . Important is that it should bind nucleic acids reversibly. Many such materials are known, of which a number are silicium based, such as aluminium silicate and the like, preferably silica. Silica is meant to include Siθ₂ crystals and other forms of silicon oxide, such as diatom skeletons, glass powder and/or particles and amorphous silicon oxide. The solid phase may be present in any form, it may even be the vessel which contains the nucleic acid mixtures or a part of such a vessel. It may also be a filter or any other suitable structure. Apart from silicium based materials other materials will also be suitable, such as nitrocellulose (filters), latex particles and other polymeric substances. A preferred form of the solid phase is a particulate form, which allows for easy separation of bound and free material, for instance by centrifugation. The particle size of the solid phase is not critical. Suitable average particle sizes range from about 0.05 to 500 μm. Preferably the range is chosen such that at least 80, preferably 90 % of the particles have a size between the values just mentioned. The same holds true for the preferred ranges of which the average particle sizes are between 0.1 and 200 μm, preferably between 1 and 200 μm. The binding capacity of a given weight of the particles increases with decreasing size, however the lower limit of the size is when particles cannot easily be redispersed after separation through for instance centrifugation. This will be the case in starting material rich in nucleic acid containing many nucleic acids of a higher molecules weight. The particles and the nucleic acids may form aggregates in these cases . The person skilled in the art will be able to choose the right particle size for the particular application envisioned. The formation of aggregates may be avoided by using fractionated silica or diatomaceous earth in a number of applications.

A further embodiment of the present invention is a method for isolating single stranded nucleic acid material from a mixture of nucleic acid material, comprising the steps of subjecting the mixture to a method as described hereinabove and treating the supernatant containing the single stranded nucleic acid material with a second liquid comprising a chaotropic agent and a second nucleic acid binding solid phase, whereby the second liquid has a composition such that the resulting mixture of supernatant and second liquid allow for binding of the single stranded nucleic acid material to the second solid phase. This way the double stranded nucleic acid material is removed from the crude mixture and the single stranded nucleic acid is purified from the remaining still crude mixture in another single step. Both the double stranded material and the single stranded material are reversibly bound to the respective solid phases, so that they may be easily eluted from said solid phases to undergo further analysis or other treatments. A very useful further treatment is the amplification of the (double or single stranded) nucleic acid material. The invention will now be explained in further detail in the following detailed description.

Separation / Isolation

MATERIALS AND METHODS

Source of nucleic acids.

Phage MS-2 ss-RNA (3569 nt) , E. coli rRNA (16 and 23S; l,7kb and 3,5kb respectively), phage M13 ss-DNA (7599 nt) and Hindlll digested phage lambda ds-DNA were purchased from

Boehringer (Mannheim, Germany). Rotavirus ds-RNA was purified from feces of an infected individual by protocol Y/SC (11) . Plasmid DNA was purified from E. coli HB101 as described by Ish-Horowicz and Burke (13) followed by column chromotography with Sepharose CL2B (Pharmacia, Inc. Uppsala, Sweden). Total NA was purified from E.coli by protocol Y/D (11). Chemicals .

Guanidiniu thiocyanate (GuSCN) was obtained from Fluka (Buchs, Switzerland) . EDTA (Titriplex) and MgCl2.6H20 were obtained from Merck (Darmstadt, Germany) . TRIS was obtained from Boehringer (Mannheim, Germany) . The preparation of size-fractionated silica particles (silica coarse, SC) and diatom suspension has been described (11). Triton X-100 was from Packard (Packard Instrument Co., Inc., Downers Grove, 111).

Composition of buffers.

The lysis/binding buffer L6, washing buffer L2, and TE (lOmM Tris.HCI, 1 mM EDTA; pH=8.0) have been described (11). 0.2M EDTA (pH 8.0) was made by dissolving 37.2 g EDTA (Merck, Germany) and 4.4 g NaOH (Merck, Germany) in aqua in a total volume of 500 ml. Lysis/binding buffer Lll was made by dissolving 120 g of GuSCN in 100 ml 0.2M EDTA (pH=8.0). Binding buffer L10 was prepared by dissolving 120 g GuSCN in 100 ml 0.35M TRIS.HCI (pH 6.4); subsequently 22 ml 0.2M EDTA (pH 8.0) and 9.1 g Triton X-100 were added and the solution was homogenized; finally 11 g of solid MgCl₂-6H₂θ was added. The final concentration of MgCl₂ in L10 is about 0.25M. L10 is stable for at least 1 month when stored at ambient temperature in the dark.

Fractionation of ds-NA and ss-NA by protocol R.

The procedure is outlined in Figure 1. A 50μl specimen (containing a mixture of NA-types in TE buffer) was added to a mixture of 900μl Lll and 40μl SC in an Eppendorf tube and was subsequently homogenized by vortexing. After 10 min. binding at room temperature, the tube was centrifuged (2 min. at approx. 10.000 x g) which resulted in a silica/ds-NA pellet ("initial silica pellet") and a supernatant containing ss-NA. To recover ss-NA forms (protocol R-sup) , 900μl of the supernatant were added to a mixture of 400μl L10 and 40μl SC and ss-NA was bound during a 10 min. incubation at room temperature. The tube was subsequently centrifuged (15 sec. at approx. 10.000 x g), and the supernatant was discarded (by suction) . The resulting pellet was subsequently washed twice with 1 ml of L2, twice with 1 ml ethanol 70% (vol/vol) and once with 1 ml acetone. The silica pellet was dried (10 min. at 56°C with open lid in an Eppendorf heating block) and eluted in 50μl TE buffer (10 min. at 56°C; closed lid) . After centrifugation (2 min. at approx. 10.000 x g) the supernatant contains the ss-NA fraction. To recover ds-NA forms (protocol R-pellet) from the initial silica-pellet, the remaining supernatant was discarded, and the silica pellet was washed twice with Lll to remove unbound ss-NA. The resulting silica pellet was subsequently washed twice with L2, twice with ethanol 70%, once with acetone, dried and eluted as described above. After centrifugation (2 min. at approx. 10.000 x g) the supernatant contains the ds-NA fraction.

In the complete procedure (which takes about one hour) for fractionation of NA by protocol R, only two Eppendorf tubes are used.

Fractionation of genomic DNA and ss-NA.

Due to trapping of ss-NA into high-molecular-weight genomic DNA, protocol R as described above gives only low yields of ss-NA. This can be circumvented by first isolating total NA by protocol Y/D (11), which causes some shearing of the high-molecular-weight genomic DNA, sufficient enough to prevent trapping of the ss-NA. Total NA thus purified can subsequently be used as input for protocol R.

Gel electrophoresis.

In all experiments, NA was electrophoresed (8 to 10 V/cm) through neutral agarose slab gels containing ethidiumbromide (lμg/ml) in the buffer system (40mM TRIS-20 mM sodium acetate- 2mM EDTA adjusted to pH 7.7 with acetic acid; ethidium bromide was added to a concentration of lμg/ml of buffer) described by Aaij and Borst (14) . Hybridization.

DNA fragments were transferred to nitrocellulose filters by the procedure of Southern (15) and hybridized with [alpha-³²P]dCTP labelled pHC624 (16) prepared by random labeling (Boehringer, Germany) . Hybridization conditions were as described previously (12).

RESULTS Comparison of different GuSCN-containing lysisbuffers with respect to the binding of different NA-types to silica particles revealed that only doublestranded forms were bound when using Lll (which is about 100 mM for EDTA) as binding buffer; on the other hand both double- and single-stranded forms were bound in binding buffer L6 (which is about 20 mM for EDTA) (Table 1) . These observations formed the basis for the development of a protocol (Protocol R) for the fractionation of single-stranded nucleic acids and double¬ stranded nucleic acids (Fig. 1) Once double-stranded nucleic acid is bound by silica particles in Lll, a brief centrifugation will separate the silica/ds-NA pellet from the supernatant containing the single-stranded forms. Addition of this supernatant to a mixture of silica particles and binding buffer L10 (which is about 250 mM for Mg²⁺) the binding of single-stranded nucleic acids to the silica particles is restored. Double-stranded and single-stranded forms can subsequently be purified by washing and eluting the silica-NA complexes (protocol R) . Double¬ stranded nucleic acid is recovered from the initial silica- pellet (protocol R-pellet) , whereas single-stranded forms are recovered from the initial supernatant (protocol R-sup) .

For optimization of protocol R we performed reconstruction experiments in which previously purified or commercially available, nucleic acids were mixed and subsequently fractionated by protocol R. Fractionation of a mixture of double-stranded DNA and single-stranded DNA.

The fractionation of a ds-DNA/ss-DNA mixture, into double stranded- and single stranded forms is shown in Figure 2. The recovery estimated from the band intensity of the ethidium bromide stained gel for ss-DNA was about 50%, the estimated recovery of ds-DNA in the range of 500 bp to 4,6 kb was 80%-90% [similar recoveries were obtained for ds-DNA fragments in the range of 100-500 bp (not shown)], larger fragments were significantly sheared as noted before (11). At the level of detection by UV-illumination, fractionation into ds- and ss- forms was complete.

Fractionation of a mixture of double-stranded RNA and single-stranded RNA.

Figure 3 shows the fractionation of a mixture of ds-RNA (human Rotavirus genome segments 1-11; for review see 14) and ss-RNA (phage MS2 RNA) into double stranded- and single stranded forms. The estimated recovery of ds-RNA and ss-RNA was at least 80% . At the level of detection by UV- illumination, fractionation into ds- and ss-forms was complete.

Fractionation of a mixture of double-stranded DNA and single-stranded RNA.

In Figure 4 it is shown that ds-DNA can also efficiently be separated from ss-RNA.

Again are the recoveries for both fractions at least 80%. Similar results were obtained when E.coli rRNA (23S and 16S) was used as ss-RNA input (not shown) .

In the experiments described above, fractionation of ds - and ss-NA forms (as judged by visual inspection of band intensities after ethidiumbromide staining and UV illumination) appeared to be complete . In order to establish the performance of the fractionation procedure for a mixture of ds-DNA and ss -RNA into ss - and ds -forms, NA purified by protocol R-sup from such a mixture was studied by Southern blotting and hybridization with a ³²P-labelled DNA probe, homologous to the ds -DNA used as input for fractionation . This experiment revealed that the ss-NA fraction contained less than 0, 1% of the ds -DNA input (figure 5) .

Fractionation of a mixture of genomic DNA and single- stranded RNA.

When we investigated the separation of high-molecular- weight (genomic) dsDNA and ss-RNA by direct fractionation using E. coli as input for protocol R, it appeared that the ds-DNA fraction was heavily contaminated with rRNA (Fig. 6, lanes 6 and 7), and ss-RNA recovery was low (Fig. 6, lanes 8 and 9) . This was likely due to trapping of RNA into high- molecular-weight (genomic) ds-DNA when silica/NA complexes were formed. On the other hand no genomic DNA was observed in the ss-RNA fraction. Total nucleic acid, which was first isolated using the standard protocol Y/D (11), and hereafter used as input material in protocol R showed significantly higher recoveries of the ss-RNA fraction (Fig. 6, lanes 2 and 5) .

REFERENCES

1. Serwer, P. (1989) Electrophoresis, 10 (5-6), 327-331.

2. Shain, D.H., Yoo,J., Slaughter, R.G., Hayes, S.E. and Tae H.J. (1992) Anal . Biochem . , 200, 47-51.

3. Sheer, D.G., Yamane, D.K., Hawke, D.H. and Yuan, P. (1990) Biotechniques, 9, 486-495.

4. Mirkes, P.E. (1985) Anal . Biochem. , 148 (2), 376-383.

5. Holstege, C.P., Pickaart, M.J. and Louters, L.L. (1988) J. Chromatogr. , 455, 401-405.

6. Thompson,J.A. (1986) Biochromatographry, 1, 68-80.

7. Stowers, D.J., Keim, J.M., Paul, P.S., Lyoo, Y.S., Merion, M. and Benbow, R.M. (1988) J. Chromatogr. , 444, 47-65. 8. Liautard,J.P. (1992) J. Chromatogr. , 584, 135-139.

9. Chomczynski,P. and Sacchi, N. (1989) Anal. Biochem . , 162, 156-159.

10. Siebert, P.D. and Chenchik, A. (1993) Nuclcleic Acids Res . , 21, 2019-2020. 11. Boom, R. , Sol, C.J.A., Salimans, M.M.M. , Wertheim van Dillen, P.M.E. and van der Noordaa, J.(1990) J. Clin . Microbiol . , 28, 495-503.

12. Boom, R. , Sol, C.J.A., Heijtink, R., Wertheim van Dillen, P.M.E. and van der Noordaa, J.(1991) J. Clin . Microbiol . , 29, 1804-1811.

13. Ish-Horowicz, D. and Burke, J.F.(1981) Nucleic. Acids Res . , 9, 2989-2998.

14. Aaij, C, and Borst, P. (1972) Biochim . Biophys . Acta . , 269, 192-200. 15. Southern, E.M.(1975) J. Mol . BioL, 98, 503-517.

16. Froussard, P. (1992) Nucleic Acids Res . , 20, 2900.

17. Van Ness et al. Nucleic Acids Research, vol. 9, NO. 19, pp. 5143-5151.

18. Thompson et al. Anal.Biochem. , vol. 3, pp. 281-291, 1987. NA-t pe binding in L6 binding in Lll

ds-DNA + +

ds-RNA . +

ss-DNA + -

ss-RNA + -

Table 1.

Binding of different NA-types to silica particles in different lysisbuffers; similar results were obtained using diatoms rather than silica particles (data- not shown).

Claims

Eur 3895CLAIMS

1. A method for separating single stranded nucleic acid material from double stranded nucleic acid material comprising contacting a mixture of the both with a liquid comprising a chaotropic agent and a nucleic acid binding solid phase, whereby the liquid has a composition such that double stranded nucleic acid binds to the solid phase and a substantial amount of single stranded nucleic acid does not and separating the solid phase from the liquid.

2. A method according to claim 1 whereby the liquid comprises at least 100 mM EDTA and comprising a guanidinium salt, preferably guanidinium thiocyanate as a chaotropic agent.

3. A method according to claim 1 or 2, whereby the solid phase is silicium based.

4. A method according to claim 3 whereby the solid phase is silica.

5. A method according to claim 4 whereby the silica is in the form of particles having a size between 0.05 and 500, preferably 0.1 and 200 μm.

6. A method according to anyone of the aforegoing claims whereby the solid phase is separated from the supernatant by centrifugation.

7. A method for isolating single stranded nucleic acid material from a mixture of nucleic acid material, comprising the steps of subjecting the mixture to a method according to anyone of the aforegoing claims and treating the supernatant containing the single stranded nucleic acid material with a second liquid comprising a chaotropic agent and a second nucleic acid binding solid phase, whereby the second liquid has a compositon such that the resulting mixture of supernatant and second liquid allow for binding of the single stranded nucleic acid material to the second solid phase.

8. A method according to any one of the aforegoing claims comprising a sequence of at least two of such methods being applied to a sample comprising nucleic acid material.

9. A method according to any one of the aforegoing claims whereby the single stranded nuclec acid material comprises mRNA.

10. A method according to claim 9 whereby the mRNA is converted into cDNA.

11. A method according to anyone of the aforegoing claims comprising a gel electrophoresis step.

12. A method according to anyone of the aforegoing claims followed by a sequencing step.