We’ve highlighted seven common issues routinely seen with restriction digestion and have explained how to identify and solve these issues, therefore, helping you improve your results.

Issue 1. Incomplete or no digestion due to inactive enzyme

Since a restriction enzyme may lose activity due to improper storage or handling, it is important that you always check the expiration date, verify that the enzyme has been stored at -20°C, and check the temperature of your freezer (do not allow temperatures to exceed -20°C, as multiple freeze-thaw cycles (more than 3 cycles) may result in reduced enzyme activity).

Test the enzyme for activity by setting up a control reaction with 1 µg of standard control DNA (e.g., lambda DNA), where you know that the DNA quality is high and the expected banding pattern (Figure 1).

Avoid storing enzymes in frost-free freezers that undergo temperature fluctuations. It is also recommended to keep the enzymes in a cold rack in the freezer, as this helps to stabilize the storage temperature.

Figure 1. λ DNA digested with BamHI, 0.7% agarose, 5 cleavage sites

Issue 2. Incomplete or no digestion due to suboptimal reaction conditions

If there are no issues with digesting the control DNA, there may be something else wrong with your reaction set up. To assess the appropriate reaction conditions, follow these steps:

  1. Verify that you are using the recommended reaction buffer, including any additives specified in the product support material.
  2. Make sure you are using molecular biology grade water.
  3. Ensure that the restriction enzyme is the final component added to the reaction mix, and that the final glycerol content is below 5%, so that the enzyme volume represents less than one tenth of the total reaction volume.
  4. Double check the optimal reaction temperature for the enzymes being used, and control for evaporation during incubation as an increased salt concentration in the buffer can inhibit enzyme performance.

If the reaction parameters are all correct, consider the template DNA:

  1. The optimal DNA concentration is between 20-100 ng/µL in the final reaction mixture.
  2. The DNA substrate should be free of contaminants or reaction components like SDS, EDTA, protein, salts, and ethanol.

Issue 3. Incomplete or no digestion due to enzyme activity blocked by DNA methylation

If your enzyme is active and digests the control DNA and the reaction is set up using optimal conditions, but you still see issues with digestion, it might be because the enzyme is inhibited by methylation of the template DNA. Several endogenous methylases site-specifically methylate adenine (DAM) or cytosine (DCM) residues and can affect enzyme activity (Figure 2).

For example, methylation by Deoxyadenosine methylase (DAM methylation) occurs normally in E. coli at GATC sequences. This sequence overlaps with the recognition sites of some enzymes, like BamHI and BclI. In this case, BamHI cuts the DNA in the presence or absence of methylation, while BclI cannot cleave methylated DNA.

To overcome this restriction, you can transform your plasmid DNA into a dam-minus, dcm-minus strain, such as E. coli GM2163. These methylation minus strains do not interfere with methylation (Figure 3).

Figure 2. Methylation sites for BamHI and Bcll recognition sequences.

Figure 3. Genotype of a methylation minus strain.


Issue 4. Incomplete or no digestion due to the structure of substrate DNA

PCR fragments:

Incomplete or no digestion of PCR products may be due to the proximity of the recognition site to the end of the DNA fragment. Some restriction enzymes require additional flanking bases for efficient DNA binding and cleavage (Figure 4).

Because recognition sites are often introduced at the ends of PCR fragments and/or primers, it is important to understand how many bases flanking a site are needed for optimal cleavage.

Enzyme suppliers often provide tables that illustrate how many bases from the end of a recognition site should be present for optimal activity.

For example, PasI can cleave DNA even if the recognition site is at the very end of the fragment, while PaeI requires at least 5 additional bases for optimal digestion (Figure 5).

Figure 4. Cut PCR products close to the DNA template end. Restriction site is usually at the 5′ end of the PCR primer.
Figure 5. Relative cleavage efficiencies for PaeI, PagI, and PasI as a function of the number of residues flanking their recognition sites.

Plasmid DNA:

If you are trying to perform a double digest with two enzymes in the multiple cloning site, efficient cleavage may be difficult if the two recognition sites are too close together. One enzyme may physically block access of the second enzyme to its respective site. Inefficient cleavage is also related to the previously described proximity of the recognition site to DNA ends. After one enzyme cuts, there may not be enough bases flanking the second site for the second enzyme to bind and effect cleavage.

For these reasons, where possible, select restriction sites that are further apart for cloning. However, if you must use sites in close proximity, consider doing a sequential digestion. And, before you set up the reaction determine which enzyme is more effective at cutting close to the end of a DNA fragment, and use that enzyme second.

In this example, XbaI and SalI are next to each other in the pUC19 multiple cloning site (MCS) (Figure 6). If you cut with XbaI first, SalI would only cut with up to 20% efficiency.

However, if you cut with SalI first, XbaI can cut near the end of the strand, with up to 100% efficiency. For this reason, you should perform sequential reactions and digest with SalI followed by XbaI to  prepare the plasmid for cloning (Figure 7).

Figure 6. pUC19 MCS.

Figure 7. Cleavage efficiency helps to determine the best order for sequential digestion.


Issue 5. Unexpected cleavage pattern due to star activity (off-target cleavage)

Star activity, also known as off-target cleavage, is an undesirable but intrinsic attribute of restriction enzymes. It is the loss of enzyme specificity resulting in cleavage at sites that are similar, but not identical to, the canonical recognition site for an enzyme.

Most enzymes will not exhibit star activity when used under recommended conditions in optimal buffers. However, under suboptimal or extreme conditions star activity can occur. These conditions include:

  • High glycerol concentration (greater than 5% in the final reaction). Most commercial enzymes are formulated with glycerol for stability and to prevent freezing at -20°C.
  • High enzyme:DNA ratio, or overdigestion
  • High pH or low ionic strength
  • Presence of organic solvents such as DMSO or ethanol
  • Use of a divalent cation other than magnesium in the reaction mix
  • Inclusion of other non-optimal buffer conditions
  • Prolonged incubations, such as overnight digestions

How can you reduce the possibility of star activity in your digest reactions?

  • Choose an enzyme supplier that has addressed star activity as follows:
    • Optimizes their enzymes and buffers to minimize star activity,
    • Offers isoschizomers with low or no intrinsic star activity, and
    • Offers engineered or modified enzymes to eliminate star activity
  • Follow the recommendations provided with each enzyme for optimal activity including the use of the correct buffer, enzyme amoun, and reaction time for the enzyme.
  • Learn how to recognize star activity, and to differentiate it from incomplete digestion and gel shift effects.

Incomplete digestion results in additional bands above the expected bands on a gel. These bands disappear when the incubation time or amount of enzyme is increased, as seen when comparing sample in lanes 2 and 3 to the completely digested sample in lane 4 (Figure 8).

Star activity, as seen in lanes 5 and 6, results in additional bands below the smallest expected size. These bands will generally become more intense with increasing enzyme dose or time, while the expected bands become less intense (Figure 8).

Figure 8. Banding patterns for incomplete and complete digestion and star activity.


Issue 6. Unexpected cleavage pattern due to gel-shift effect

Gel-shift is the result of another enzyme attribute and can result in an unexpected banding pattern when viewing digested samples on a gel. It is typically more apparent when high enzyme doses are used and can have minor or significant impact on visualizing samples (Figure 9A).

Gel-shift effects can be minimized by heat inactivation of enzymes after digestion, typically by incubation at 65°C or 80°C for 10 to 20 minutes. This varies by enzyme, and some enzymes cannot be heat inactivated efficiently.

A second method to reduce gel-shift is to add SDS into the loading buffer prior to loading on the gel. These methods will denature the enzyme, releasing it from the DNA fragment (Figure 9B).

Figure 9. Panel A: Gel shift effect. Panel B: Gel shift effect +/- SDS in the loading buffer.


Issue 7. Unexpected cleavage pattern—other

If you see an unexpected cleavage pattern on the gel and have confirmed that it is not from incomplete digestion, star activity, or gel-shift effects, there are a few more possible causes.

  1. The restriction enzyme tube or reaction buffer tube may be contaminated with a second enzyme. This can happen where the same reaction buffer is used for multiple different enzymes. The only way to test this is to try a fresh tube of enzyme or reaction buffer.
  2. Another cause might be contamination of the DNA substrate. Check for this by using a fresh DNA preparation.
  3. In rare cases, it may be possible that there are unexpected recognition sites in the substrate DNA. You can check for mutations that may have been introduced during PCR amplification. There is also potential to generate new restriction sites after ligation of DNA fragments.
  4. Finally, some restriction enzymes have degenerate recognition sites. For example, XmiI cuts at GTMKAC, where M is either A or C, and K is either G or T. Make sure to check your substrate sequence for all potential sites. (Figure 10).

Figure 10. Some restriction sites are degenerate and may result in unexpected banding patterns.