The Good, The Bad and The AAG

Welcome back to my blog, I hope you’ve all been well and forgive me for the terrible pun in the title but I just couldn’t resist. You’ll be pleased to know that my lecturers didn’t plummet my grades after the last post and that I can resolve last weeks cliff-hanger by telling you all about AAG.

So, what is AAG and its role? (AKA, The Good)

Figure 1.
This figure shows the process of base-excision-repair.
The damaged base is removed by glycosylases (such as AAG) before an AP endonuclease removes the corresponding section of DNA backbone lacking a base. Then polymerase replaces the DNA strand that was cut for several bases before the old strand is cleaved away by flap endonuclease. The new strand of DNA is then bound to the surrounding older strand by ligase.
Adapted from “BER basic pathway” by Amazinglarry [Public domain] and licensed under CC BY 2.0

AAG stands for 3-alkyladenine DNA glycosylase and it is present in both the cell and the mitochondria where it acts as one of the first enzymes in the base-excision-repair (BER) pathway (Krokan and Bjørås, 2013). As seen in figure 1, the process of base excision repair is to replace damaged DNA bases, such as those produced by the alkylating agents covered last week. To do this AAG cuts out the damaged base from the DNA backbone, this site which contains the DNA backbone but lacks a DNA base is known as an abasic or apurinic site (AP site). Then AP endonuclease cuts this section of backbone out from the surrounding healthy DNA to produce a single strand break (SSB). Following this polymerase produces a section of a DNA strand to replace the missing section and neighbouring sections of the DNA strand. The old DNA strand is then cut away by flap endonuclease before ligase then binds the new DNA strand into place within the surrounding healthy strand (Krokan and Bjørås, 2013).  

This BER activity is key in correcting the DNA damage that occurs through alkylation and oxidation, such as in chronic inflammation (Meira et al., 2008; Calvo et al., 2012). The impact of this is especially evident in research by Meira et al., 2008 which identified that without AAG and the downstream BER activity, chronic inflammation similar to that seen in patients with Ulcerative Colitis or Crohns Disease can lead to cancer. This was because twice as many polyp growths were found in models lacking AAG than those with AAG.

Therefore, AAG should protect cells from Alkylation damage? (AKA, The Bad)

Well, in all honesty not always. Current research suggests that AAG can protect cells from minor alkylation induced DNA damage, but excessive alkylation appears to cause AAG to damage the cell itself.

Imagine AAG is like an energetic puppy, like the one my family just got, and cell damage is our stress levels. Ordinary levels of puppy energy are good because playing with them is fun and destresses us. But too much energy and the puppy can start chewing the furniture, chasing the cat and barking all hours, which drives our stress levels through the roof.

Now away from puppies and back to the science, a prime example of this was identified in research by Meira et al., 2009 in which alkylation damage, induced by MMS mentioned last week, to the retina caused retinal degeneration and apoptosis of the cells in the outer nuclear layer of the retina, seen figure 2.

Figure 2.
Retina’s of wildtype (WT) (Normal) and AAG lacking (AAG -/-) models prior to and following MMS alkylating agent treatment.
The outer nuclear layer (ONL) of the retina was shown to decrease in size, due to cellular apoptosis, following MMS treatment in the wildtype model. Whereas the ONL of AAG lacking model was not affected by MMS treatment.
Taken from (Meira et al., 2009)

However, the cells of the outer layer of the retina in AAG lacking model were completely unaffected by the MMS, with little to no retinal degeneration. This suggested that AAG is kind of like the Dr Jekyll and Mr Hyde of the DNA repair pathway, as it’s dual natured with both cell protective and cell damaging roles (Meira et al., 2009).

But how does AAG lead to cell toxicity? (AKA, The Interesting)

The current theory is that the toxicity of AAG is due to the creation of toxic intermediates during the BER and the activation of PARP1.

This is because the BER intermediates, seen in figure 3, are more toxic than the initial alkylated DNA base. In fact, until the DNA damage is fully repaired, each step is considered to be more toxic than the last with the toxicity increasing from alkylated base to apurinic site to single strand break (Wyatt and Pittman, 2006).

Both the alkylated bases and the apurinic sites can prevent transcription and replication. If key genes cannot be transcribed into proteins vital for cellular function then it leads to cell death, similarly if DNA replication stalls it can lead to incomplete DNA replication and cell death. Even if cell death doesn’t occur then trans-lesion polymerases (that replicate or repair DNA) can replace the missing bases with random ones which leads to mutation (Boiteux and Guillet, 2004; Calvo et al., 2013).

Figure 3.
Diagram detailing how base excision repair can lead to the repair of DNA base damage and how it can lead to toxicity.
The presence of both alkylated bases and apurinic sites can lead to the prevention of DNA replication which can inhibit cell growth and lead to apoptosis. However, trans lesion polymerases can replaces these methylated bases or apurinic sites to repair the DNA however the base replacement is random and so can lead to mutation.
Following the cutting of the DNA strand by AP endonuclease a single strand break (SSB) in the DNA is produced and Parp1 is recruited to it. Parp1 functions to produce polymers nearby to the SSB and in doing so recruit enzymes such as polymerases and ligases to repair the SSB. However, overactivation of Parp1 leads to excessive consumption of NAD+ and ATP, which can lead to energetic collapse of the cell and result in cell death.
Taken from (Calvo et al., 2013)

Furthermore, following the activity of AP endonuclease at the apurinic site a single strand break in the DNA is produced. If this strand is not repaired quickly enough before the cell replicates it can lead to the replication of the DNA collapsing and the single-strand break becoming a double strand break, this can lead to the cell’s DNA fragmenting and/or cell death (Kuzminov, 2001; Saleh-Gohari et al., 2005;Ubhi and Brown, 2019). However, part of the repair mechanism, seen figure 3, is to activate Parp1 at the single-strand break in order for it to recruit polymerases and ligases to repair the break (Calvo et al., 2013).

This would ordinarily be helpful, but Parp1 activation consumes high quantities of NAD+ and ATP which provide energy to the cell. If Parp1 is overactivated then the excessive Parp1 activity leads to depletion of NAD+ and ATP to the extent that the cell becomes exhausted, undergoes energetic collapse and dies (Calvo et al., 2013). To extend the puppy metaphor (or simile, I always forget which is which), Parp1 is the puppy and we are the cell. If the puppy is too active then it can wear us out, causing us to run out of energy and collapse on the sofa.

Finally, to link this blog post back to my first one, the activation of AAG has been thought to lead to the generation of endoplasmic reticulum stress. Specifically, this stress is the activation of the unfolded protein response that I alluded to all those weeks ago and shall finally cover within the next fortnight. Once I have covered the unfolded protein response, I will be in a prime position to draw all these blog posts together and explain what this means for my project and what I hope to achieve through it. So, thank you for sticking with me so far and I hope you can tolerate my writing a little longer now the finish line is in sight.

(As a little bribe to keep you hooked, here is a picture of the aforementioned one-year old puppy “Eoin” fast asleep next to his festive new older sister “Patsy”)


Boiteux, S. and Guillet, M. (2004). Abasic sites in DNA: repair and biological consequences in Saccharomyces cerevisiae. DNA Repair, 3(1), pp.1-12.

Calvo, J., Meira, L., Lee, C., Moroski-Erkul, C., Abolhassani, N., Taghizadeh, K., Eichinger, L., Muthupalani, S., Nordstrand, L., Klungland, A. and Samson, L. (2012). DNA repair is indispensable for survival after acute inflammation. Journal of Clinical Investigation, 122(7), pp.2680-2689.

Calvo, J., Moroski-Erkul, C., Lake, A., Eichinger, L., Shah, D., Jhun, I., Limsirichai, P., Bronson, R., Christiani, D., Meira, L. and Samson, L. (2013). Aag DNA Glycosylase Promotes Alkylation-Induced Tissue Damage Mediated by Parp1. PLoS Genetics, 9(4), p.e1003413.

Krokan, H. E., & Bjørås, M. (2013). Base excision repair. Cold Spring Harbor perspectives in biology5(4)

Kuzminov, A. (2001). Single-strand interruptions in replicating chromosomes cause double-strand breaks. Proceedings of the National Academy of Sciences, 98(15), pp.8241-8246.

Meira, L., Bugni, J., Green, S., Lee, C., Pang, B., Borenshtein, D., Rickman, B., Rogers, A., Moroski-Erkul, C., McFaline, J., Schauer, D., Dedon, P., Fox, J. and Samson, L. (2008). DNA damage induced by chronic inflammation contributes to colon carcinogenesis in mice. Journal of Clinical Investigation, 118(7), pp. 2516–2525.

Meira, L., Moroski-Erkul, C., Green, S., Calvo, J., Bronson, R., Shah, D. and Samson, L. (2009). Aag-initiated base excision repair drives alkylation-induced retinal degeneration in mice. Proceedings of the National Academy of Sciences, 106(3), pp.888-893.

Saleh-Gohari, N., Bryant, H., Schultz, N., Parker, K., Cassel, T. and Helleday, T. (2005). Spontaneous Homologous Recombination Is Induced by Collapsed Replication Forks That Are Caused by Endogenous DNA Single-Strand Breaks. Molecular and Cellular Biology, 25(16), pp.7158-7169.

Ubhi, T. and Brown, G. (2019). Exploiting DNA Replication Stress for Cancer Treatment. Cancer Research, 79(8), pp.1730-1739.

Wyatt, M. and Pittman, D. (2006). Methylating Agents and DNA Repair Responses:  Methylated Bases and Sources of Strand Breaks. Chemical Research in Toxicology, 19(12), pp.1580-1594.