Light and magic : CPD Photolyase ✨

I first came across CPD photolyase around August 2025, when deciding on a topic for my coursework, and it has been a whirlwind romance every since. To me, CPD photolyase is light and magic.

Because of photolyase, I learned Samson, Pymol, etc. etc. to demonstrate the wonders of CPD photolyase. (I thank my peers for putting up with my dreary presentations.) My first Samson animation was CPD formation. 

This is an overall summary post, specific details will be described in future posts. 

CPDs are cyclobutane pyrimidine dimers. They can form between TT, CT and CC. TTs (Thymine dimers) are the most common, due to physical factors, which I'll elaborate more about in the future. CPDs form during UV ray exposure, especially UVB and UVC. UVC is mostly absorbed by the atmosphere, but is used in germicidal lamps and scientific experiments. In daily life, UVB exposure is the main culprit for thymine dimers. 

UVB (280-320nm wavelength) and UVC (200-280) correspond to the peak absorbance spectrum of DNA bases (260nm). Normally, thymine has a double bond between C5=C6. When adjacent thymines absorb UVB/C, electrons within the C5=C6 bond are promoted to an excited state. If the adjacent thymines are close enough, in favourable conformation, such as in ssDNA, the C5=C6 bonds break and reform into a cyclobutane. TT can also form in dsDNA, but at a lower quantum yield. 


Bond breaking and formation occur simultaneously in a concerted mechanism, within 1 picosecond (10-12). If you remember your quantum mechanics, these are π bonds converting to σ bonds. So here's a 3D tour in Samson. In this cartoonified animation, I broke it down into steps, but in reality, the dimer should form in one step. The point of this animation was to show the DNA backbone distortion. 
So this transforms normally parallel adjacent thymines into an asymmetrical butterfly shape that opens out towards the major groove. The pentose pucker becomes more planar, along the xy axis of the DNA helix. The phosphate molecules are dragged along, decreasing their usual angle of 36 degrees. If you use your hands to model the DNA bases (right-handed helix, 36 degrees rotation), and your forearms as the backbone : open your hands outwards, and you will see that your forearms move closer together. 

Now this changes the major/minor groove geometry of DNA. From the top view, the backbones are in the minor groove. Since the backbone molecules are pinched, the minor groove narrows. Since the dimer flares out towards the major groove, the major groove widens. But this is a general statement, specific geometric changes depends on the context, since the DNA helix is quite flexible. 
(created in Biorender with PDB 1BNA. By Delphine.)

You can imagine that this distortion isn't great for DNA. You're right. This local distortion can become large-scale distortions when poly-T tracts, such as those in human telomeres, dimerize during UV exposure. Different genomic contexts give different yields for thymine dimers for a given exposure of UVC. This study by Law et al., 2013 used different permutations of the tetrad XTTY to explore the various yields. Lu et al, 2021 also investigated this. A Phd student at the University of Huddersfield explored this topic as well. 

(From Lu et al., 2021. Y denotes T.)

Significant distortions cause replication fork stalling. DNA can bring in translesion polymerase (TLS) to replicate past the lesion, but this is more error-prone. Mutations build up with successive rounds of replication. If replication fork stalling is unresolved, DNA collapses into double-strand breaks (DSBs). The awful downstream effects include chromosomal rearrangements, a story for another day. 

 ✨Now here comes the light and magic : CPD photolyase.✨

Since most living things encounter UV rays (unless you're a vampire or a Somalian blind cavefish), there is a wonderful DNA repair mechanism known as CPD photolyase to repair CPDs efficiently. This is an ancient mechanism that emerged in response to high levels of UV radiation in early earth, due to a nascent, permeable atmosphere that did not have enough oxygen or ozone yet. It was free game for UVB and UVC. So modern descendants of ancient archaea still retain their superpowers of being extremely UV-resistant. Halophiles and thermophiles can survive intense UV exposure with minimal mutagenesis, due to robust and redundant DNA repair pathways. 

Photo - light, lyase - break. So photolyase breaks CPDs (usually thymine dimers) in the presence of light, within nanoseconds. We call this photoreactivation. This term was coined by biologist Albert Kelner in 1949, when he discovered that UV-killed bacteria was resurrected by exposure to visible light. Shocking! So the bacteria were photo-reactivated. The field of DNA repair was born. 

However, Kelner could not solve the mystery of photoreactivation. This would have to wait till 1958, when Claud S. Rupert identified CPD photolyase as the enzyme behind photoreactivation. Rupert compared photoreactivation-competent E. Coli vs photoreactivation-deficient H. influenzae, and identified that a light-dependent enzyme was reponsible for the resurrection and DNA repair. 

Rupert's student, Aziz Sancar, then went on to purify and isolate E. Coli CPD photolyase (CPDphr) via cloning and amplification of the CPDphr gene, beginning in 1975. Recombinant DNA cloning had just been invented in 1973 by Cohen and Boyer. Sancar then went on to elucidate the molecular mechanisms of CPD photolyase and NER, eventually winning the 2015 Nobel Prize in Chemistry for his work on DNA repair mechanisms. 

In the 2000s, Sancar collaborated with physicist Zhong DongPing to create groundbreaking femtosecond spectroscopy visualizations of photolyase. This allowed real-time observations of ultrafast dynamics. The field of femtobiology was born, beginning with photolyase. Building on the work of Zhong, the latest development is the 2023 TR-SFX visualization by Maestre-Reyna et al., which captured the full catalytic cycle for the first time at atomic resolution, since the recovery phase is spectrally silent in previous techniques. In 2025, Li et al. of Foulkes Lab, that study fish circadian rhythms, discovered novel, light-independent repair mechanisms of CPD photolyase (more on that in another post, this was the first paper that got me into photolyase).

From this wonderful history, we now have a clearer picture of how CPD photolyase works. Here is a brief, simple summary of the entire process. More details in future posts.
  1. Upon UVR exposure, adjacent thymines dimerise into a cyclobutane within 1 picosecond.
  2. This distorts the groove geometry of DNA, changing the distribution of negative charges on DNA.
  3. Even before light activation of photolyase, this electrostatic change is detected by CPD photolyase, which has positive charges on its surface. 
  4. CPD photolyase scans for, and binds CPDs without light, flipping the CPDs out of the DNA backbone.
  5. Arginine residues in the catalytic site of photolyase stabilises the DNA. 
  6. Blue light or UVA light activates photolyase, specifically the cofactor FAD, which normally exists in an oxidised, inactive state. 
  7. The chromophore FAD absorbs blue light/UVA, and is excited into FAD*, an unstable, oxidising state.
  8. An electron from the HOMO is promoted to the LUMO. (highest occupied molecular orbital to lowest unoccupied molecular orbital).
  9. There is now an electron deficiency in the HOMO, and FAD* is hungry for electrons. 
  10. So FAD* grabs an electron from the tryptophan-triad in the photolyase. Twice.
  11. After grabbing two electrons, FAD* becomes the stable FADH-.
  12. FADH- donates one electron to the thymine dimer. The second electron helps to stabilise the reaction. (More on electron tunneling in a future post.)
  13. C5-C5 breaks first, followed by C6-C6.
  14. The donated electron is returned to FADH-.
  15. Now the separated thymines occupy more space in the catalytic site, signaling to photolyase that the job is done, and so photolyase relaxes. 
  16. Repaired thymines flip back into their original positions. 
  17. The entire repair process completes within 500 nanoseconds. 
However, I have to end this story on a sad note. Humans do not have CPD photolyase, or any kind of photolyase at all (there different photolyases, more in a future post). In the Neoproterozoic era, photolyase evolved into cryptochrome in placental mammals. Cryptochrome conserves structural and functional features of photolyase, and regulates our circadian rhythm, but does not repair DNA. Partially because there are negative charges on its surface, compared to the positive charges of cryptochrome. 

Now the exact reasons for this are unknown. Many organisms have both photolyase and cryptochrome, so there's no good reason why placental mammals should not have both. 

There's this wacky theory though. It's called the Nocturnal Bottleneck Hypothesis (Gordon Lynn Walls, 1942). It goes something like : to avoid the dinosaurs🦖who hunted in the day, placental mammals became nocturnal, and developed nocturnal adaptations (loss of photolyase, nocturnal vision, etc.). Then the dinosaurs went extinct, so we didn't have to be nocturnal anymore. And so it was the dinosaurs' fault that we lost photolyase, and have to wear sunscreen now. 

So, wear your sunscreen, and don't suntan on your lawn (or rooftop), as many students in Kingston like to do. Also, the roof might collapse.

· · ─ ·✶· ─ · · 

Further reading (Photolyase research history in chronological order)

Kelner, A. (1949). Effect of visible light on the recovery of Streptomyces griseus from ultraviolet irradiation injury. Proceedings of the National Academy of Sciences, 35(2), 73-79. https://doi.org/10.1073/pnas.35.2.73
Rupert, C. S. (1958). Photoreactivation of ultraviolet-irradiated Escherichia coli, with special reference to the dose-reduction principle and to ultraviolet-induced mutations. Journal of General Physiology, 41(3), 451-471. https://doi.org/10.1085/jgp.41.3.451
Sancar, A., & Rupert, C. S. (1978). Cloning of the phr gene and amplification of photolyase in Escherichia coli. Gene, 4(3), 213-226. https://doi.org/10.1016/0378-1119(78)90018-4
Kao, Y. T., Saxena, C., Wang, L., Sancar, A., & Zhong, D. (2005). Direct observation of thymine dimer repair in DNA by photolyase. Proceedings of the National Academy of Sciences, 102(45), 16128-16132. https://doi.org/10.1073/pnas.0505257102
Maestre-Reyna, M., Essen, L. O., Carell, T., et al. (2023). Visualizing the DNA repair process by a photolyase at atomic resolution. Science, 382(6675), 1218-1223. https://doi.org/10.1126/science.add7795
Further reading (related topics)

Gerkema, M. P., Davies, W. I., Foster, R. G., Menaker, M., & Hut, R. A. (2013). The nocturnal bottleneck and the evolution of activity patterns in mammals. Proceedings. Biological sciences, 280(1765), 20130508. https://doi.org/10.1098/rspb.2013.0508

Mei, Q., & Dvornyk, V. (2015). Evolutionary History of the Photolyase/Cryptochrome Superfamily in Eukaryotes. PloS one, 10(9), e0135940. https://doi.org/10.1371/journal.pone.0135940