Abstract
Purpose:
The light/dark cycle is the predominant synchronizer of circadian rhythms in humans. The clock-controlled secretion of melatonin from the pineal gland is widely used as a circadian marker. Photic suppression of nocturnal melatonin release in humans has been shown to be relatively insensitive, with a threshold of about 10 lux or 13 log photons cm-2 s-1. However, all previous studies were performed under light-adapted conditions, which could reduce photosensitivity. I aimed to reexamine the threshold intensity using fully dark-adapted subjects.
Methods:
Subjects were kept in a room that was completely dark except during stimulus light presentation. The stimulus was a 468nm blue LED whose intensity was adjusted using neutral density filters, and it was presented continuously for 1 hr in each trial through a Ganzfeld dome. Saliva was collected periodically from each subject and analyzed by the Bühlmann melatonin radioimmunoassay.
Results:
In darkness, melatonin level in all subjects rose monotonically for 2 - 3 hours at early subjective night before reaching a plateau that lasted several hours. All prior studies tested for melatonin suppression during the plateau, and initially I also used this period. But similar to previous researchers, I found that melatonin level fluctuated greatly during the plateau, making it difficult to detect small changes in melatonin secretion. Thus, I did all subsequent testing during the early-night rising phase, as the nearly linear increase in melatonin improved the signal-to-noise ratio in the measurements. One-hour exposure to 10.3 log photons cm-2 s-1 suppressed melatonin level by nearly 20% (p-value < 0.001). The suppression induced by 1-hr 9.2 log photons cm-2 s-1 was smaller (~7%) but still significant (p-value = 0.03). The color of the 9.2 log photons cm-2 s-1 light was barely discernable, suggesting this intensity was near the S-cone threshold.
Conclusions:
I have found the human circadian system to be at least four orders of magnitude more sensitive to light than previously demonstrated, and the threshold intensity reported here is within about 2 log units of that for intracellularly recorded primate melanopsin ganglion cells (Dacey et al., Nature 2005). Light at night can disrupt sleep, cause depression, impair learning, and even increase cancer risks. The high photosensitivity of the circadian system suggests that at night, even the dim light from a television or street lights could be harmful.
Keywords: 590 melatonin •
458 circadian rhythms •
531 ganglion cells