Method: Denaturing Gradient Gel Electrophoresis (DGGE)

C. Helms

Purpose:

Denaturing gradient gels are used to detect non-RFLP polymorphisms. The small (200-700 bp) genomic restriction fragments are run on a low to high denaturant gradient acrylamide gel; initially the fragments move according to molecular weight, but as they progress into higher denaturing conditions, each (depending on its sequence composition) reaches a point where the DNA begins to melt. The partial melting severely retards the progress of the molecule in the gel, and a mobility shift is observed. It is the mobility shift which can differ for slightly different sequences (depending on the sequence, as little as a single bp change can cause a mobility shift). Alleles are detected by differences in mobility.

Most of the following is taken from a procedure written in June of 1987 by Mark Gray while he was a post-doctoral student in Welcome Bender's Laboratory (Harvard). There are many variations of the protocol; some of them are mentioned here. Refer to the literature citations list for more information.

Principle:

Denaturing gradient gel electrophoresis has been shown to detect differences in the melting behavior of small DNA fragments (200-700 bp) that differ by as little as a single base substitution. When a DNA fragment is subjected to an increasingly denaturing physical environment, it partially melts. As the denaturing conditions become more extreme, the partially melted fragment completely dissociates into single strands. Rather than partially melting in a continuous zipper-like manner, most fragments melt in a step-wise process. Discrete portions or domains of the fragment suddenly become single-stranded within a very narrow range of denaturing conditions. The rate of mobility of DNA fragments in acrylamide gels changes as a consequence of the physical shape of the fragment. Partially melted fragments migrate much more slowly during electrophoresis through the polyacrylamide matrix than completely double-stranded fragments. When a double-stranded fragment is electrophoreses into a gradient of increasingly denaturing conditions, it partially melts and undergoes a sharp reduction in mobility because it changes shape. In practice, the denaturants used are heat (a constant temperature of 60 degrees C) and a fixed ratio of formamide (ranging from 0-40%) and urea (ranging from 0-7 M). The position in the gradient where a domain of a DNA fragment melts and thus nearly stops migrating is dependent on the nucleotide sequence in the melted region. Sequence differences in otherwise identical fragments often cause them to partially melt at different positions in the gradient and therefore 'stop' at different positions in the gel. By comparing the melting behavior of the polymorphic DNA fragments side-by side on denaturing gradient gels, it is possible to detect fragments that have mutations in the first melting domain.

Many fragments can be analyzed simultaneously on a single denaturing gel in which the direction of electrophoresis is perpendicular to that of the denaturing gradient. When a large number of different fragments is electrophoresed, the fragments can be identified by their molecular weight in the low denaturant side of the gel. By following the S-shaped curves, the characteristic denaturant concentration at which the first domain melts can be determined. When two nearly identical sets of fragments are mixed together and electrophoresed into a 'perpendicular' denaturing gradient gel, the melted domains that have sequence differences between each other will melt at slightly different positions and produce double bands.

Sequence differences are often easily detected in DNA fragments when nearly identical digests are electrophoresed in the same direction as that of the denaturing gradient. These 'parallel' gels permit the simultaneous comparison of as many sets of fragments as there are lanes on the gel, unlike the perpendicular gels. The procedures below refer almost entirely to parallel denaturing gradient gels.

DGGE blotting

Denaturing gradient gel blots of genomic DNA can be used to detect most single base differences (which may occur as frequently as every 400 bp). Genomic DNA is digested with an enzyme combination that cuts the DNA into fragments of average size 200-700 bp. Enzymes with 4-base recognition sites, such as AluI, HaeIII, HhaI, MspI, and RsaI are the most useful. The samples are electrophoresed long enough for many fragments to reach a position in the gradient at which the first melting domain denatures. After electrophoresis, the DNA fragments are transferred to a nylon blot. The blots are hybridized overnight with a radioactive probe, washed, and exposed to X-ray film for 1-5 days. Each lane is examined for fragments that have mobility shifts.

Many polymorphisms are present in portions of a restriction fragment not in the first melting domain. Sequence differences are undetectable unless they are within the first melting domain. To avoid this problem, it is often possible to 'move' the mutation into the first melting domain by using different restriction enzyme combinations. Frequently, at least one of 4-5 different digests achieves this result. Occasionally, a mobility shift is observed in more than one restriction digest. By aligning on a restriction map the partially overlapping fragments that have altered melting behavior, the region containing the the base difference can be narrowed down to 100 bp of less.

DGGE- finding polymorphisms with PCR amplified DNA

Although we have yet to try it in our lab, the DGGE should provide an excellent method of finding polymorphisms when DNA samples amplified from different individuals are run on parallel denaturing gradient gels. The mobility shifts of these amplified fragments are visible with ethidium staining, eliminating the need for autoradiography. The polymorphisms detected in such a screen could be pursued with genotypic data collection in the CEPH pedigrees. Richard Myers advocates the PCR amplification of larger fragments, e. g. 2- 3 kb in size, followed by restriction digestion of the amplified products with two different frequent cutting enzymes such as HaeIII and Sau3AI, then running the samples on two different denaturing gradient gels (0-50% denaturant and 40-80% denaturant) with different running times. The gels are then examined under UV illumination after staining with ethidium bromide. Myers estimates as many as 75% of the base pair changes in a 2-3 kb fragment could be detected in this screen.

1. DGGE apparatus (Green Mountain Scientific Supply)
2. Electroblotter (Hoefer)
3. BioRad gradient maker

1. 0% denaturing acrylamide stock
2. 100% denaturing acrylamide stock
3. 50X TAE buffer
4. BstN I digest of pBR322 DNA (gel standards)

After the preparation of restriction digests/ or PCR amplification:
1. Day 1: 2-4 hours to prepare, load, and start the electrophoresis
2. Day 2: 1-2 hours to stain and photograph
3. 4 hours for electrotransfer
4. 2 hours for baking the blots

Procedures:

General Notes:

The 25 ml gels are poured with the aid of a gradient maker. The general strategy for detecting and following polymorphisms is first to use a DGGE gel which has wide range denaturing conditions, typically 20% to 80% denaturant. Once a polymorphism is detected, the distance separating the mobility shifts is amplified by narrowing the range of the denaturing conditions on the gel to an appropriate range of denaturing conditions, e. g. 20-50% or 40-80% denaturant, and the rest of the data collection is done under the optimized conditions.

The denaturing gradient gels are electrophoresed in an apparatus purchased from Green Mountain Lab Supply, Inc. (86 Central Street, Waltham, MA 02154). Parallel denaturing gels can be poured in the gel frame supplied (perpendicular gradient gels must be prepared outside the gel frame; this procedure does not cover perpendicular DGGE). The glass plates (one notched and one unnotched), spacers and gel combs are supplied with the basic apparatus, and will produce gels approximately 5 x 5 inches. After the gel is poured and the acrylamide polymerized, it is immersed in the tank (filled with 1X TAE buffer equibrated to temperature of 60 degrees C with the Polyscience constant temperature regulator/circulator). The buffer is circulated through the tank with a peristatic pump. The anode is permanently mounted in the gel tank; the cathodes are graphite rods which are immersed in the upper electrophoresis chamber while electrophoresis proceeds.

Restriction digestion of genomic DNA samples:

Choose restriction enzymes that will produce an average fragment size in the 200-700 bp range. The most reliable and inexpensive 4-base recognition site enzymes include AluI, HaeIII, HhaI, MspI, and RsaI. Cut 7 to 10 ug human genomic DNA for each lane to be run. Be sure to include an extra ug of genomic DNA for the test digest (with 1 ug bacteriophage lambda DNA, see procedure "human genomic DNA digests" for more details on restriction digestion). Use the manufacturer's recommended restriction buffer and digest the samples overnight.

Run the test digest and lambda controls on a high percentage agarose gel (e. g. 1.5%) to determine whether or not the human sample was completely digested. Once complete, concentrate the DNA (by ethanol precipitation) to approximately 1 ug DNA/µl TE, and add the appropriate amount of 10 X Stop Mix (the lab's standard glycerol-based loading dye).

For a DGGE polymorphism screen of unknown DNA sequences, prepare digests of at least 4 unrelated individuals with 3 or more different restriction enzymes. The DNA 'molecular weight 'standard used to monitor the progress of fragments in the acrylamide gels is a BstN I digest of pBR322 DNA [fragment sizes are: 1857, 1060, 929, 383, 121, and 13 bp].

Day 1

Pouring and electrophoresis of parallel denaturing gradient gels:

1. Prepare 150 ml of 1.5% TAE agarose for each gel frame used: Weigh 2.25 g agarose, add 3 ml 50X TAE buffer and 150 ml dH2O. Boil in the microwave until dissolved, then cool to 50 degrees C in a water bath.
2. Wash the glass plates with hot water and soap, rinse well with tap water and deionized water, followed by a 95% ethanol rinse. Lay a clean and dry notched plate down, then place the lightly greased (with petroleum jelly) spacers along the length of the two sides. Carefully lay a clean un-notched plate on top. (The apparatus is designed to accomodate 2 more gels, prepared with layers of notched plates and spacers, with the last plate being the un-notched plate.)
3. Pick up the stacked plates and using a pasteur pipet, dribble 50 degrees C 1.5 % agarose (prepared in TAE buffer) down the side to help seal the spacers.
4. Place the plates in the gel frame so that the notched plate faces the upper buffer compartment. Center the plates so that the small gaps on the sides are equal. Insert the plexiglass supports between the screws in the gel frame and the glass plates. Twist the screws evenly but firmly enough so that the neoprene gasket between the first notched plate and the gel frame is compresses enough to form a waterproof seal.
5. Pour 1.5 % TAE agarose into the trough at the bottom of the gel frame so that the agarose is drawn up between the plates about 1 cm. Let the agarose plug harden, then fill with more agarose, and pour agarose down the side gaps of the gel frame to ensure the seal between glass plates and spacers is tight. It is important to be ready to pour the gel soon after the agarose hardens, because as the agarose dries, it shrinks, and gaps will form causing the acrylamide to leak before it can polymerize.

   The linear gradient is made by placing the high denaturant mix (e. g. 80%) in the downstream side of the gradient maker, and the low denaturant mix (e. g. 20%) in the upstream side of the gradient maker. When the gradient maker is allowed to flow, as long as the downstream chamber is continuously mixed (using a small magnetic stirrer), the 20% mix flows into the 80% mix forming the linear gradient of decreasing denaturant concentration.

   The gel volume will vary depending on the thickness of the spacers used. For 0.7 mm spacers and combs, each gel holds approximately 20 ml. Prepare 12.5 ml of the two gel mixes to be used for each gel using the 0% and 100% denaturing stocks (the total volume is 25 ml, an excess of 5 ml). The following table gives the proportions of each denaturant to use in preparing typical gel mixes:

   % denaturant	mls 0% denaturant stock	mls 100% denaturant stock	volume
   20%	10.0	2.5	12.5 ml
   40%	7.5	5.0	12.5 ml
   60%	5.0	7.5	12.5 ml
   80%	2.5	10.0	12.5 ml

   The urea in the 100% denaturant stock will drop out of solution at 4 degrees C, (be sure to warm the stock to redissolve the urea before mixing the desired denaturants).

6. Pipet the desired amount of stocks into two separate, labeled Erlenmeyer flasks. Degas and put the gel mixes on ice (the degassing can be omitted).
7. Set up the gradient maker with the stirrer in the downstream chamber and tubing leading from the gradient maker to the top of the gel plates. Attach a 25 gauge needle to the tubing and wedge the needle between the glass plates at the top left side. Close the valve separating the two sides of the gradient maker, and close the tubing valve.
8. Add 1/200 volume (63 µl) of a 20% ammonium persulfate (in dH20) solution and 1/2000 volume (6.3 µl) of TEMED (N, N, N', N'-tetramethyl-ethylenediamine) to each mix.
9. Pipet 12 ml of the high denaturant mix (e. g. 80%) into the downstream chamber. Open the valve to the upstream chamber so that a small amount of 80% mix enters the upstream chamber, then close the valve. With a pasteur pipet, remove the small amount that entered the upstream chamber and place it back in the downstream chamber. This eliminates the air pocket in the passageway between the two chambers. Pipet in 12 ml of the low denaturant mix (e. g. 20%) into the upstream chamber.
10. Open the valve between the two sides of the gradient maker. Be sure the stirrer is mixing well in the downstream side. Open the valve in the tubing that leads to the gel plates. The gel mix should begin to flow down the tubing to the gel. If it does not flow due to a bubble near the chamber, remove the needle for a short time to start the flow, (or use an adjustible flow rate pump to move the gel mix). The entire gel should be poured over a 5 minute period in order to avoid disruption og the gradient. During the time the gradient is being poured, check the gel for leaks, and for any disruption in the pouring.

    11. After the gel is poured, insert the comb without trapping small bubbles under the wells. If additional gels are being poured, refill the gradient maker as before, and repeat the entire process.

11. Immediately after the gels are poured, rinse out the gradient maker with distilled water.
12. Check the top corners of the gels for evidence of leaks. If the level of the mix drops quickly, there is likely to be a leak between the agarose plug and the plates. If the leak is slow, the gel is salvageable because polymerization is usually fast (about 10 minutes). After polymerization, there is usually a small amount of shrinkage at the edges; this is of no concern. Leave the combs in place until the gel is ready to be loaded. Fill the upper compartment with TAE buffer. The gel can be loaded one hour after pouring or several days later, as long as the agarose does not dry out too much.
13. The temperature of the gel buffer in the tank should be brought to 60 degrees C and the pump should be recirculating the buffer before loading the samples. Fill the large plexiglass tank with enough TAE buffer to bring the level to within 3/4 of an inch from the top of the gel frame (an empty tank takes at least 16 liters). Turn on the heater before putting the gels into the tank. It may take as long as 45 minutes for the temperature to reach 60 degrees C.
14. After the temperature reaches 60 degrees C, remove the comb and examine the wells for any defects. Put the entire gel frame into the tank, attach the tubing for recirculating the buffer to the gel frame, and start the pump. Attach the leads, immersing the cathode graphite rods into the upper buffer chamber of each gel frame, then 'piggy back' the cathode leads together at the power supply if more than one gel frame is in use. Attach the anode lead, and turn on the current to check the circuit.
15. When everything is ready, load the samples using a pipetman with flat tips (made for sequencing gels). Change tips with each sample. Load all the samples before turning on the power. Turn the power to 65-75 volts and note the time; electrophorese overnight for 16-17 hours. For samples of approximately 400bp, try running the gel for 1200 volt-hours.

Day 2

1. Turn off the power in the morning and note the time again. Turn off the heater and pump. Disconnect the electrical leads and pump tubing, and remove the gel frames from the tank. Loosen the screws, remove the plexiglass supports, and pull the plates out of the frame. Lay the plates down, keeping track of the order and orientation of the gels. Separate the glass plates by wedging a spatula between the plates at a lower corner, and lift off the top plate.
2. Mark the gel for orientation purposes by slicing the upper left hand corner with a razor blade. Stain the gel on the plate (which should be placed in a tray) by pouring enough ethidium bromide stain solution over the gel to cover the surface. Stain for 5-10 minutes, then carefully pour off or aspirate the staining solution and destain with distilled water for several minutes. Examine the gel by UV illumination (first transfer the gel to a platform such as UV translucent plexiglass or Saran wrap). Genomic DNA digests should be visible as evenly loaded straight smears; PCR-amplified fragments should be visible as bands or tight smears. Any unusual features in some lanes should be noted before photographing the gel.

   Electrophoretic transfer of the DNA to nylon blots:

3. After photography, transfer the gel back to a staining tray and aspirate any excess liquid. Add enough 0.5 N NaOH to cover the gel, and mix once or twice over the next 5 minutes, then aspirate the NaOH. This denatures the DNA fragments.
4. Neutralize the gel by pouring in 0.5M Tris-HCl, pH8.0, mixing once or twice over the next 5 minutes, then remove the Tris solution.
5. Fill the tray with transfer buffer (20 mM Tris-HCl, pH8.0 + 1 mM EDTA) and let it equilibrate for 10 minutes.
6. Measure the size of the gel after this equilibration, because it will bave increased in size. Cut a sheet of nylon membrane large enough to cover the gel, label, and wet it in a tray of transfer buffer for at least 5 minutes.
7. Fill the electrotransfer chamber with enough transfer buffer to reach close to the top.
8. Assemble the sandwich in a tray filled with transfer buffer--bubbles can easily be displaced by buffer when assembling the immersed sandwich. In the tray, lay down the rigid plastic support screen, then one of the 'Scotchbrite' pads (supplied with the electrotransfer apparatus), and then one layer of Whatman 3MM paper. In order to lay down the fragile gel without tearing or stretching, attach it to a 3MM paper as follows: Using a piece of X-ray film, scoop the gel up out of the tray. Gently spread the gel out flat using a glass pipet; as long as the gel and the film are thoroughly wet, the gel will not tear easily. After the gel is flat and the orientation noted, remove all excess buffer by soaking it up on the edges with paper towels. Roll the glass pipet across the gel several times to drive the buffer to the edges. Lay down a piece of dry 3MM paper that is slightly larger than the gel, but small enough to fit into the electrotransfer apparatus. Lay the paper down in the middle of the gel first, then spread it toward the edges so that no air bubbles are trapped. Pick up the gel by the 3MM paper (the gel should be completely bound to the dry paper). Lay the gel down on the sandwich with the paper side down. Next lay down the labeled nylon membrane, watching for the appearance of bubbles. Put down 2 layers of wet Whatman 3MM paper, then the other thoroughly wet Scotchbrite pad. Place the second rigid support screen on top of the sandwich, and quickly insert the sandwich into the guides of the electrotransfer box. Try to dislodge any newly formed bubbles by tilting the box from side to side. Note carefully the polarity of the sandwich; the blot side should be facing the anode (+) side of the electrotransfer box.
9. Connect the electrodes and turn the current up to 600 mA (about 30 volts). Keep the current at 600 mA for two hours. Usually it is necessary to adjust the voltage slightly downwards during the two hours to maintain 600 mA. Do not allow the buffer to heat up too much; if it does, some plastic parts of the transfer appaatus may soften and warp. (Do the trasfer in the cold room if heat is too much of a problem.) If a power supply that can produce 0.6 amps is unavailable, electrophoresis at 200mA for 4 hours is sufficient.
10. After the electrotransfer, turn off the power, remove and disassemble the sandwiches, and put the nylon blots in a tray of 2 X SSC for 5 minutes. Let the blots air dry one pieces of Whatman 3MM paper before baking in a vacuum oven at 80 degrees C for 1-2 hours.

    After the blots are baked, they can be treated as any other Southern blot.

Interpretation of results:

If the electrophoresis time was long enough, most of the fragments should be visible as sharp, focussed bands that do not trail up at the sides. Some bands will appear as almost invisible long smears because of the simultaneous melting of more than one domain in the fragment. Some particularly short fragments (less than 300 bp) and GC rich fragments will not form sharp focussed bands, usually in the most denaturing part of the gel. The most common difference caused by single base changes is a shift in the position of the fragment, either up or down with respect to a standard. Other types of changes include: a sharp focussed band will become very diffuse or almost a diffuse smear, a diffuse unfocused band becomes sharp and focused, the position of a smear is shifted up or down.

Often it is not clear whether or not a fragment is different from its counterpart in another lane. In these cases, it is easiest to reexamine them by repeating the electrophoresis in a denaturing gradient gel with a much narrower range of denaturant concentration. For example, if the ambiguous fragment is present at a position 60% of the distance from the top of a 20-80% gel, then its critical denaturant concentration is 56% (0.6 x 60 per cent units + 20%). Repeat the electrophoresis of this sample on a denaturing gradient gel with a range of 41-715 (15% on either side of that desired). Any real differences will often be greatly exaggerated in the less steep gradient. If more than one fragment is different, it is frequently an artifact. There are many ways to create artifactual results on denaturing gradient gel blots. Real differences are reproducible on different gels/blots.

Solutions:

• 40% Acrylamide stock
  (37.5: 1 Acrylamide: bisacrylamide)

  Caution: acrylamide is a neurotoxin, especially dangerous in powder form. Wear a mask when weighing out acrylamide.

  Dissolve 194.8 g electrophoresis grade acrylamide and 5.2 g bisacrylamide in enough deionized water to make a total volume of 500 ml. Store in the dark at 4 degrees C.

• 0% denaturant

  (6.5 % acrylamide in 1X TAE buffer)

  81 ml 40% acrylamide stock

  10 ml 50 X TAE buffer

  409 ml ddH2O

  Total volume = 500 ml; store in the dark at 4 degrees C.

• 100% denaturant

  (6.5 % acrylamide, 40% formamide, 7 M urea, in 1X TAE buffer)

  81 ml 40% acrylamide stock

  10 ml 50 X TAE buffer

  210 g urea

  200 ml formamide

  adjust final volume to 500 ml

  store in the dark at 4 degrees C.

• 50X TAE Buffer
  per liter:

  242 g Trizma base	2M Tris
  136.1 g sodium acetate	1M sodium acetate
  18.6 g Na2EDTA	50 mM EDTA

  Add ddH2O to bring volume close to 900 ml

  pH to 7.4 with concentrated HCl, then adjust volume to 1 liter

• Transfer Buffer (20 mM Tris-HCl, pH 8.0, 1 mM EDTA)
  per liter:
  20 ml 1 M Tris-HCl pH8.0

  2 ml 0.5 M EDTA

  978 ml ddH2O

• 1.5 % TAE agarose
  per 200 ml:
  3 g agarose

  4 ml 50 X TAE buffer

  196 ml ddH2O

• 20% ammonium persulfate (20 % APS)
  2 g ammonium persulfate

  + ddH2O to 10 ml

  store at 4 degrees C.

  Many scientists feel that making fresh 20 % APS every 1-2 weeks produces better results.

References:

Myers, R. M., Maniatis, T., and L. S. Lerman. (1987). "Detection and localization of single base pair changes by denaturing gradient gel electrophoresis." in Meth. Enzymol. 155: 501.

Gray, M., pers. comm. (1987). after his procedure notes "Detection and localization of single base pair changes by denaturing gradient gel electrophoresis".

Myers, R. M., pers. comm.