Intrinsically Photosensitive Retinal Ganglion Cells (ipRGCs) Are Necessary for Light Entrainment of Peripheral Clocks

Abstract

Light is a powerful entrainer of circadian clocks in almost all eukaryotic organisms promoting synchronization of internal circadian rhythms with external environmental light-dark (LD) cycles. In mammals, the circadian system is organized in a hierarchical manner, in which a central pacemaker in the suprachiasmatic nucleus (SCN) synchronizes oscillators in peripheral tissues. Recent evidence demonstrates that photoentrainment of the SCN proceeds via signaling from a subpopulation of retinal ganglion cells (RGCs) which are melanopsin-expressing and intrinsically photosensitive (ipRGCs). However, it is still unclear whether photoentrainment of peripheral clocks is mediated exclusively by the ipRGC system or if signaling from RGCs that do not express melanopsin also plays a role. Here we have used genetic "silencing" of ipRGC neurotransmission in mice to investigate whether this photoreceptive system is obligatory for the photoentrainment of peripheral circadian clocks. Genetic silencing of ipRGC neurotransmission in mice was achieved by expression of tetanus toxin light chain in melanopsin-expressing cells (Opn4::TeNT mouse line). Rhythms of the clock gene Period 2 in various peripheral tissues were measured by crossbreeding Opn4::TeNT mice with PER2 luciferase knock-in mice (mPER2Luc). We found that in Opn4::TeNT mice the pupillary light reflex, light modulation of activity, and circadian photoentrainment of locomotor activity were severely impaired. Furthermore, ex vivo cultures from Opn4::TeNT, mPER2Luc mice of the adrenal gland, cornea, lung, liver, pituitary and spleen exhibited robust circadian rhythms of PER2::LUC bioluminescence. However, their peak bioluminescence rhythms were not aligned to the projected LD cycles indicating their lack of photic entrainment in vivo. Finally, we found that the circadian rhythm in adrenal corticosterone in Opn4::TeNT mice, as monitored by in vivo subcutaneous microdialysis, was desynchronized from environmental LD cycles. Our findings reveal a non-redundant role of ipRGCs for photic entrainment of peripheral tissues, highlighting the importance of this photoreceptive system for the organismal adaptation to daily environmental LD cycles.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1. TeNT expression doesn’t alter ipRGCs responses to light.

(A-C) Average ipRGCs responses to 1min light stimulations (480nm) of increasing irradiance (5.1010 photons/cm2/s to 5.1013 photons/cm2/s) recorded from retinas of Opn4cre/+, R26+/+ (A, n = 123), Opn4cre/+, R26+/TeNT (B, n = 87), Opn4cre/+, R26TeNT/TeNT (C, n = 158) mice and the corresponding dose response curves (D).

Fig 2. Synaptic “silencing” of ipRGCs and loss of pupillary light reflex (PLR) in Opn4::TeNT mice.

(A) Opn4Cre/+ mice were intercrossed with R26floxstop-TeNT mice that have the TeNT gene integrated into the Rosa26 locus downstream of a floxed transcriptional stop sequence. Expression of Cre recombinase in ipRGCs excises the stop sequence to activate TeNT expression irreversibly. (B) Representative images of pupillary constriction 20 s post-irradiation in control and Opn4::TeNT mice. White broken lines encircle pupillary diameters. (C) Relative pupil area upon light stimulation of control and Opn4::TeNT mice with low and high light intensities. Values denote means ± SEM (n  =  3–5). *p

Fig 3. Lack of c-FOS expression in the Opn4::TeNT suprachiasmatic nucleus following light pulse at CT 16.

(A) Immunocytochemical staining for c-FOS in the SCN of control and Opn4::TeNT mice exposed to a light pulse. Notice the diminished number of c-FOS positive cells in the mutant mouse. Scale bar = 100 um. (B) Quantification of c-FOS expression in the suprachiasmatic nucleus in control mice exposed to a light pulse (+) or no light pulse (-) and in mutant mice exposed to a light pulse. Expression is reflected by the number of c-FOS positive cells per suprachiasmatic nucleus per slice. n = 3–8. *P<0.004 by Kruskal-Wallis non-parametric ANOVA, followed by Dunn’s multiple comparison test.

Fig 4. Impaired photoentrainment and light negative masking in Opn4::TeNT mice.

(A) Representative activity records from individual animals initially maintained under a 12:12 h LD cycle regimen for >2 weeks and then transferred to DD conditions for 2 weeks. A control animal demonstrates light entrainment in LD as indicated by enhanced activity in the dark period while the Opn4::TeNT mouse shows activity that follows a “free running” pattern. Shaded gray regions indicate periods of darkness. (B) Summary of measured periods (Tau) in control and Opn4::TeNT mice under LD and DD conditions. n = 12–13. (C) Activity records of control and Opn4::TeNT mice kept in ultradian light cycles of 3.5:3.5 light dark periods (T7). (D) Control mice are active mostly during dark periods under T7 conditions while the Opn4::TeNT mice are active across the light and dark periods. n = 4–5. *p<0.01. **; **p<0.001.

Fig 5. PER2::LUC rhythms are maintained in peripheral tissues of Opn4::TeNT mice.

(A) Examples of locomotor activity in a control and a Opn4::TeNT mouse on a 12:12h LD cycle. Locomotor activity was recorded using running wheels and animals were sacrificed when activity was phase shifted by 8–12 hours between control and Opn4::TeNT mice. Tissues were then collected for ex-vivo bioluminescence. (B) Examples of cornea PER2::LUC bioluminescence rhythms in Opn4::TeNT (blue) and control (black) mice. Notice the antiphasic peak of bioluminescence in the mutant cornea compared to the cornea from a control mouse.

Fig 6. PER2::LUC rhythms are out of phase to LD cycles.

(A) Representative records of bioluminescence from various peripheral tissues in control and Opn4::TeNT mice with access to running wheels. (B) Phases for central and various peripheral circadian oscillators of control and Opn4::TeNT mice. The peak of the circadian oscillation was determined during the interval between 12 and 36 h in culture. The average times (± SEM) of peaks were plotted against the projected CT time. Notice that in the control mice the peak of PER2::LUC expression occurs in peripheral tissues during the subjective night (CT12-24). On the other hand, in the Opn4::TeNT mice the peak of PER2 expression is aligned to subjective day (CT0-12). (C) Period values of PER2::LUC rhythms were comparable in control and Opn4::TeNT mice with the exception of period lengthening in the adrenal. ADR, adrenal; COR, cornea; LIV, liver; LUN, lung; PIT, anterior pituitary; SPL, spleen. *P<0.05. ADR, adrenal (n = 12–17); COR, cornea (n = 14–22); LIV, liver (n = 12–14); LUN, lung (n = 13–14); PIT, anterior pituitary (n = 15–20).

Fig 7. PER2::LUC rhythms of peripheral tissues in Opn4::TeNT mice are not entrained by light.

Circular plots of peak bioluminescence rhythms in peripheral tissues presented in control (A) and mutant (B) mouse tissues. Mice were housed in the absence of running wheels. Notice the phases of peak bioluminescence rhythms are significantly dispersed among tissues in the Opn4::TeNT mice indicating their lack of entrainment to environmental light-dark cycles. ADR, adrenal (n = 12–17); COR, cornea (n = 15–20); LIV, liver (n = 12–14); LUN, lung (n = 14–16); PIT, anterior pituitary (n = 14–19; SPL, spleen (n = 7–11).

Fig 8. Circadian corticosterone secretion in Opn4::TeNT Mice is not entrained to LD Cycles.

Free corticosterone measured in extracellular fluid by in vivo subcutaneous microdialysis in 3 control mice and 3 Opn4::TeNT mice maintained under 12:12 h LD cycles. Shaded gray regions indicate periods of darkness. Arrows indicated projected CT12 as assessed by running wheel activity that preceded the microdialysis sampling procedure. Subcutaneous free corticosterone in control mice shows circadian rhythms that peak at the onset of the dark period and the projected CT12. In the Opn4::TeNT mice the corticosterone rhythms are maintained, but are not aligned to the LD cycles. Notice however that in the mutant mice the peak of the corticosterone rhythm is aligned to the projected CT12. Corticosterone CircWave peak phases are indicated with (*).