Two types of non-radioactive DNA detection systems were optimized for use with nylon membranes in Southern transfers. A luminol substrate system (consisting of an enhanced chemiluminescent reaction utilizing luminol enzyme substrate) was used with peroxidase labeled probe DNA; a dioxetane based substrate was used with alkaline phosphatase/antibody and digoxigenin labeled probe DNA.

Chemiluminescence was detected by autoradiography. Methods for reprobing the membranes were also optimized for both systems; blots could be reprobed at least ten times. Results showed that excellent sensitivity and low background can be achieved on both amphoteric and positively charged nylon membranes using either detection system.

Many of the techniques of molecular biology rely on the ability to detect and identify RNA and DNA. One of the most important means of achieving this is via a hybridization process in which complementary target DNA (or RNA) is detected by attachment to a probe DNA, RNA or synthetic oligonucleotide [1].

In practice, solid-phase hybridization on a membrane surface, following electrophoresis to separate various molecular weight DNA species, is the most common approach. Many of the procedures included in this type of analysis are derived from the work of Southern [2], who developed a technique for transferring DNA from an agarose gel to a membrane, where it is then analyzed via hybridization with a probe DNA.

The first techniques for detecting DNA on membranes after a Southern transfer used a radioactive probe DNA. Typically, radioactive ³²P was incorporated into a probe DNA, generally by using an endonuclease in a nick translation system [3]. With this technique, it is possible to detect less than 1 pg of target DNA (a single copy of a gene complementary to the probe) on an autoradiogram [4]. Exposure times required to obtain this level of sensitivity are often 2-5 days.

Radio labeled probe DNA may be stored for only about one month, due to the half life of ³²P. Another drawback to radioactive analysis is the large amount of radioactive liquid waste generated during the hybridization and wash steps. Disposal of the low level wastes produced is becoming ever more expensive.

Non-radioactive alternatives to ³²P-labeled probes are now becoming popular for safety, environmental and cost concerns. The increased sensitivity that radioactive systems once enjoyed over early nonradioactive systems has now been reduced or eliminated [5], so more laboratories are turning to non-radioactive techniques.

Most non-radioactive techniques incorporate enzyme based detection systems. The systems used for these experiments use peroxidase and alkaline phosphatase. An enhanced chemiluminescent reaction utilizing luminol (Amersham, Arlington Heights, IL, USA) was used with the peroxidase system [6]; a dioxetane substrate was used with the alkaline phosphatase system. With either system, a very brief exposure of the developed blot to X-ray film (1 to 15 min) results in the typical "autorad" generated in a Southern transfer. Both of these systems will detect as little as 100 fg of DNA on amphoteric or positively charged nylon membranes.

An important feature of radioactive labeling systems is the ability to reprobe the blots. Reprobing is important for screening blots for more than one genetic marker [7]. A radiolabeled probe can be stripped from the blot. The ability to remove signal from the blot is necessary so that the signal from a second probe can be detected. Color precipitated on the membrane by colorimetric enzyme substrates cannot be completely removed. Signal produced by chemiluminescent substrates can be successfully stripped from the membranes.

In this article, two methods for chemiluminescent detection of DNA on amphoteric and positively charged nylon membranes are presented, along with stripping and reprobing procedures for each detection system. A modification of the standard Southern transfer procedure is used. The standard Southern protocol calls for pre-wetting the membrane with transfer buffer and placing it on a gel which has been neutralized with 1.5 M NaCl, 1M Tris-HCI, pH 8.0; our protocol calls for placing a dry membrane on an alkaline gel. A neutral transfer buffer (20X standard saline citrate [SSC]) was used, so that the initial transfer occurs under alkaline conditions, and then the DNA is gradually neutralized on the membrane by the transfer buffer. This technique has yielded better sensitivity in our tests than either alkaline or neutral transfer procedures.

Lambda HindIII fragments were applied to lanes on a 1% Seakem® GTG agarose (FMC Bioproducts, Rockland, ME, USA) gel in 1X TBE (90 mM Tris-HCI, 90 mM boric acid, 2mM EDTA, pH 8.35) containing 1ug/ml ethidium bromide. After electrophoresis, gels were depurinated with 0.1N HCl for 20 minutes and were then denatured with 0.5N NaOH/1.5M NaCl for 30 minutes.

Denatured gels were applied directly to wicks on a capillary transfer apparatus. The gels were not neutralized before placing onto wicks. 20X SSC was used as the transfer buffer. Dry amphoteric nylon membrane (Biodyne® A membrane from Pall, Port Washington, NY, USA) and positively charged nylon membrane (Biodyne B membrane from Pall) were placed directly onto the denatured gels for the transfers.

DNA was fixed onto the membranes by baking at 80 °C for 30 min. After fixation, DNA bands on the membranes were visualized using either an alkaline phosphatase or peroxidase enzyme detection system. Intact Lambda DNA was used as probe DNA for the alkaline phosphatase system. DNA was labeled with digoxigenin by nick translation using Klenow enzyme (Boehringer Mannheim, Indianapoils, IN, USA); Lambda HindIII fragments were used to prepare probe DNA for the peroxidase system. Peroxidase was directly coupled to probe DNA using gluteraldehyde, according to instructions from Amersham, Arlington Heights, IL, USA. Hybridization was allowed to proceed for 18 hours.

Hybridization

The protocols used for DNA labeling and hybridization were based on literature supplied with the Amersham enhanced chemiluminescence (ECL) DNA detection kit for the peroxidase system and on literature supplied with the Boehringer Mannheim Genius^TM DNA detection kit for the alkaline phosphatase system.

1. Peroxidase system

Membranes were pre-hybridized for 30 min with 0.1ml/cm² of Amersham hybridization buffer, containing 5% blocking agent (Amersham) and 0.75M NaCl at 42 °C. 10ng/ml of peroxidase labeled Lambda DNA fragments was then added to the pre-hybridization solution. Blots were hybridized for 18 hours at 42 °C.

2. Alkaline Phosphatase system

Membranes were pre-hybrized for 60 min with 0.1ml/cm² of 5X SSC, 0.05% N-lauroyl sarcosine, 0.02% sodium dodecyl sulfate (SDS), 1% casein (BDH casein, Hammarsten grade; Hoefer scientific catalogue number 44020) at 68 °C. Digoxigenin labeled probe (20ng/ml) was added to the pre-hybridization solution, and the blots were hybridized for 18 hours at 68 °C. The probe concentration was calculated assuming 100% recovery of DNA during the labeling protocol (as suggested in the Boehringer-Mannheim Genius Kit literature).

Post-Hybridization

1. Peroxidase system

Blots were washed two times in 1ml/cm² of 6 M urea, 4% SDS, 0.5X SSC, 40 °C for 20 min per wash. Blots were then rinsed in 2X SSC at room temperature.

2. Alkaline phosphatase system

Blots were washed with 1ml/cm² of 2X SSC, 0.1% SDS, two times at room temperature, 5 minutes per wash, and then with 0.2X SSC, 0.1% SDS, two times at 68 °C, 15 minutes per wash. Blots were then rinsed with 2X SSC at room temperature.

Visualization

1. Peroxidase system

After post-hybridization washes, 0.5ml/cm² of luminol peroxidase substrate (Amersham) was added to the blot and incubated for 1 min with agitation. Excess fluid was drained from the membrane which was then sealed in a clear plastic folder (GIBCO BRL/Life Technologies, Gaithersburg, MD, USA) and exposed for 5 to 15 minutes with Kodak X-OMAT® AR film in an autoradiograph cassette.

2. Alkaline Phosphatase system

After post-hybridization washes, the blots were rinsed with 0.05 M NaHCO₃, 1mM MgCl₂, pH 9.5. The blots were then incubated with agitation in dioxetane substrate solution (Lumiphos^TM 480, from Lumigen Corp used at full strength, or AMPPD® from Tropix Corp used at 0.1mg/ml) for 5 min. Excess solution was removed from the blots, and the blots were sealed in plastic folders before placing against Kodak X-OMAT AR film in an autoradiograph cassette.

Stripping

Blots developed with the peroxidase system were incubated with 0.1% SDS, 0.1X SSC for 60 min at 80 °C. Membranes were then rinsed with 2X SSC.

Blots developed with the alkaline phosphatase system were incubated with 0.5 mg/ml proteinase K, 0.1% SDS, for 60 min at 65 °C to allow the stripping agent access to the hybrid DNA. Membranes were then washed with 0.1% SDS, 5X SSC three times, 5 min per wash, followed by incubation with 50% formamide, 10mM NaPO₄ , pH 6.5, for 60 min at 65 °C. Stripped blots were rinsed with 2X SSC.

Reprobing

Stripped blots were incubated with fresh solutions of pre-hybridization and hybridization solutions, and treated as for the original probe.

Both of the detection systems used in the experiments described above, used with amphoteric or positively charged membranes, can detect as little as 100 fg of DNA on the membrane in a dot blot test using chemiluminescent substrates. In a Southern transfer, all bands of a HindIII digest of Lambda DNA can be seen after loading the gels with as little as 3 ng of DNA per lane.

The systems described above and the use of chemiluminescence constitute convenient and sensitive methods for nonradioactive detection of DNA. A major advantage of chemiluminescent detection is the very short exposure times with X-ray film required to produce maximum signal, i.e. two to fifteen minutes as compared to between 6 hours and 6 days for ³²P labeled probes.

Given these benefits, more and more laboratories using radioactive DNA detection techniques may consider conversion to nonradioactive chemiluminescent detection. These can give equivalent sensitivity with faster, less expensive, and safer protocols. Using either of the nonradioactive systems evaluated in this report, we find that amphoteric and positively charged nylon membranes yield excellent results and have the strength and capacity for multiple reprobes of the same transfer.

Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.

Southern, E. M. (1975), "Detection of specific sequences among DNA fragments separated by gel electrophoresis," J. Mol. Biol. 98: 503-517.

Feinberg, A.P., and Vogelstein, B. (1983), "A Techique for Radiolabeling DNA Restriction Endonuclease Fragments to High Specific Activity," Anal. Biochem. 132: 6-13.

Hames, B.D. and Higgins, S.J. (1985) Nucleic acid hybridization, a practical approach. IRL Press Ltd., Oxford, U.K.

Bronstein, L., Voyta, J.C., Lazzari, K.G., Murphy, B.E., and Kricka, L.J. (1990), "Rapid and Sensitive Detection of DNA in Southern Blots with Chemiluminescence," BioTechniques 8: 310-313.

Pollard-Knight, D., et al (1990), "Nonradioactive Nucleic Acid Detection by Enhanced Chemiluminescence using Probes Directly Labeled with Horseradish Peroxidase," Anal. Biochem. 185, 1.

Gatti, R.A. Concannon, P., and Salser, W. (1984), "Multiple Use of Southern Blots," BioTechniques 2: 148-155.