A Cryptochrome adopts distinct moon- and sunlight states and functions as sun- versus moonlight interpreter in monthly oscillator entrainment

l-cry mutants show higher spawning synchrony than wild-type animals under non-natural light conditions

In order to test for a functional involvement of L-Cry in monthly oscillator function, we generated two l-cry mutant alleles (Δ34 and Δ11bp) (Fig. 1a) using TALENs²⁸. In parallel, we generated a monoclonal antibody against Platynereis L-Cry. By testing mutant versus wildtype worms with the anti-L-Cry antibody in Western blots (Fig. 1b) and immunohistochemistry (Fig. 1e–j), we verified the absence of L-Cry protein in mutants. Furthermore, we confirmed that the staining of the antibody in wildtype worms (Fig. 1e–h) matches the regions where l-cry mRNA is expressed (Fig. 1d). These tests confirmed that the engineered l-cry mutations result in loss-of-function alleles. In turn, they validate the specificity of the raised anti-L-Cry antibody.

We next assessed the circalunar maturation timing of wild types and l-cry mutant populations in conventional culture conditions, i.e. worms grown under typical indoor room lighting (named here artificial sun- and moonlight, Supplementary Fig. 1b).

We expected either no phenotype (if L-Cry was not involved in circalunar clock entrainment) or a decreased spawning precision (if L-Cry was functioning as moonlight receptor in circalunar clock entrainment). Instead we observed an increased precision of the entrained worm population:

We analyzed the maturation data using two statistical approaches, linear and circular statistics. We used the classical linear plots⁵ and statistics to compare the monthly spawning data distribution (Fig. 2a–c, i). This revealed a clear difference between mutant animals, which exhibited a stronger spawning peak at the beginning of the NM phase, compared to their wildtype and heterozygous counterparts (Fig. 2a–c, Kolmogorov–Smirnov test on overall data distribution, Fig. 2i).

We then analyzed the same data using circular statistics (as the monthly cycle is repeating, see details in Methods section), which allowed us to describe the data with the mean vector (defined by the direction angle µ and its length r, shown as arrows in Fig. 2e–g). The phase coherence r (ranging from 0 to 1) serves as a measure for synchrony of the population data. As expected for entrained populations, all genotypes distributed their spawning across a lunar month significantly different from random (Fig. 2e–g, p values in circles, Rayleigh’s Uniformity test²⁹). In line with the observed higher spawning peak of the l-cry−/− mutants in the linear plots, the circular analysis revealed a significant difference in spawning distribution (Mardia–Watson-Wheeler test, for details see Methods section) and higher spawning synchrony of mutants (r = 0.614) than in wild types and heterozygotes (r = 0.295 and r = 0.222) (Fig. 2i). The specificity of this phenotype of higher spawning precision for l-cry homozygous mutants was confirmed by analyses on trans-heterozygous l-cry ^(Δ34/Δ11) mutants (Supplementary Fig. 2), and by the fact that such a phenotype is not detectable in any other light receptor mutant available in Platynereis (r-opsin1³⁰: Supplementary Fig. 3a, b, e, f, i; c-opsin1³¹: Supplementary Fig. 3c, d, g, h, i, Go-opsin: refs. 32, 33).

The higher spawning synchrony of l-cry mutants under artificial light mimics the spawning precision of wild-type at its natural habitat

This increased spawning precision of l-cry mutants under artificial (but conventional indoor) laboratory light conditions let us wonder about the actual population synchrony of the worms under truly natural conditions. The lunar spawning synchrony of P. dumerilii at the Bay of Naples (the origin of our lab culture) has been worked on for more than 100 y. This allowed us to re-investigate very detailed spawning data records from the worms’ natural habitat published prior to environmental/light pollution. For better accessibility and comparability we combined all months and replotted the data published in 1929³⁴ (Fig. 2d, h, I; see details in Methods section; r = 0.631). This analysis revealed that the higher spawning synchrony in l-cry^–/– worms mimics the actual spawning synchrony of P. dumerillii populations in their natural habitat³⁴ (compare Fig. 2c, g with 2d, h.)

Given that recent, non-inbred isolates from the same habitat as our lab inbred strains (which is the same habitat as the data collected in ref. 34) exhibit a broad spawning distribution under standard worm culture light conditions (which includes the bright artificial moonlight)³⁵, we hypothesized that the difference in spawning synchrony between wildtype laboratory cultures and populations in their natural habitat is caused by the rather bright nocturnal light stimulus typically used for the standard laboratory culture (Supplementary Fig. 1a vs. b).

Lunar spawning precision of wild-type animals depends on naturalistic moonlight conditions

We next tested the resulting prediction that naturalistic moonlight should increase the spawning precision of the wildtype population, using naturalistic sun- and moonlight devices we specifically designed based on light measurements at the natural habitat of P. dumerilii³¹ (Supplementary Fig. 1a, c). We assessed the impact of the naturalistic sun- and moonlight (Supplementary Fig. 1a, c) on wildtype animals, maintaining the temporal aspects of the lab light regime (i.e. 8 nights of “full moon”). Indeed, merely adjusting the light intensity to naturalistic conditions increased the precision and phase coherence of population-wide reproduction: After several months under naturalistic sun- and moonlight, wildtype worms spawned with a major peak highly comparable to the wildtype precision reported at its natural habitat (Fig. 2d, h vs. j, k), and also exhibited an increased population synchrony (r = 0.398 compared to r = 0.295 under standard worm room light conditions). This increased similarity to the spawning distribution at the natural habitat (“Sea”) is confirmed by statistical analyses (Fig. 2l): The phase difference (angle between the two mean vectors) is only one day (corresponding to 12°). In contrast, the spawning distribution of wild types under standard worm room light versus naturalistic light conditions is highly significantly different in linear and circular statistical tests and has a phase difference of 7.7 days (Fig. 2l).

These findings show that it is the naturalistic light that is critical for a highly precise entrainment of the monthly clock of wild-type worms. Given that l-cry^–/– animals reach this high precision with the artificial light (i.e. standard lab light) implies that in wildtype L-Cry blocks artificial, but not naturalistic full-moonlight from efficiently synchronizing the circalunar clock. This block is removed in l-cry^–/– animals, leading to a better synchronization of the l-cry^–/– population. This finding suggests that L-Cry’s major role could be that of a gatekeeper controlling which ambient light is interpreted as full-moonlight stimulus for circalunar clock entrainment.

l-cry functions as a light signal gatekeeper for circalunar clock entrainment

A prediction of this hypothesis is that mutants should entrain better to an artificial full-moonlight stimulus provided out-of-phase than their wild type counterparts (in which L-Cry should block the “wrong” moonlight at least partially from re-entraining the circalunar oscillator).

We thus compared the spawning rhythms of l-cry^+/+ and l-cry^–/– worms under a re-entrainment paradigm, where we provided our bright artificial culture full-moonlight at the time of the subjective new moon phase (Fig. 3a). In order to compare the spawning data distribution relative to the initial full moon (FM) stimulus, as well as to the new full moon stimulus (i.e. new FM), we used two nomenclatures for the months: months with numbers are analyzed relative to the initial nocturnal light stimulus (i.e. FM), whereas months with letters are analyzed relative to the new (phase-shifted) nocturnal light stimulus (i.e. new FM, Fig. 3a). When the nocturnal light stimulus is omitted (to test for the oscillator function) we then refer to ‘free-running FM’ (FR-FM) or ‘new free-running FM’ (new FR-FM), respectively (Fig. 3a). Using these definitions, the efficiency of circalunar clock re-entrainment will be reflected in the similarity of spawning data distributions between month 1 and month D, i.e. the more similar the distribution, the more the population has shifted to the new phase.

When using the artificial nocturnal light conditions, the re-entrainment of l-cry^–/– animals was both faster and more complete than for their wildtype relatives, as predicted from our gate keeper hypothesis. This is evident from the linear data analysis and Kolmogorov–Smirnov tests when comparing the month before the entrainment (month 1) with two months that should be shifted after the entrainment (months C,D, Fig. 3b, c, f, g).

Most notably, while l-cry−/− worms were fully shifted in month D (Fig. 3c: compare boxes and see complete lack of spawning at the light blue arrowhead indicating the old NM/new FR-FM phase versus massive spawning at new NM phase around dark blue arrowhead), wildtype animals were still mostly spawning according to the initial lunar phase (Fig. 3b: compare boxes and see spawning at the light blue arrowhead versus almost lack of spawning at dark blue arrowhead). The faster re-entrainment of l-cry^–/–, compared to l-cry+/+ animals is also confirmed by the Mardia–Watson-Wheeler test (see Methods section for details). For l-cry^+/+ animals, the comparisons of the spawning distributions before and after re-entrainment show a 1000-fold (months 1 versus C) and tenfold (months 1 versus D) higher statistical significance difference than the corresponding comparisons for l-cry−/− worms (Fig. 3f, g). Consistently, the phase differences in days calculated from the angle between the two mean vectors from the circular analysis is smaller in the mutants than in the wild types when comparing the phase of the month before the entrainment (month 1) with two months after the entrainment (months C, D) (Fig. 3d–g). The fact that there are still differences in the mutant population before and after entrainment is likely due to the fact that even the mutants are not fully re-entrained. However, they have shifted more robustly in response to an artificial nocturnal light stimulus than the wild types. This provides further evidence that in wildtype worms L-Cry indeed blocks the “wrong” light from entering into the circalunar clock and thus functions as a light gatekeeper.

L-Cry functions mainly as light interpreter, while its contribution as direct moonlight entraining photoreceptor is (at best) minor

We next tested to which extent L-Cry is itself a sensor for the re-entrainment signal under naturalistic light conditions. Based on the finding that l-cry−/− worms can still re-entrain the circalunar oscillator (see above), it is clear that even if L-Cry also directly contributed to the entrainment, it cannot be the only moonlight receptor mediating entrainment. With the experiments below, we aimed to test if L-Cry has any role as an entraining photoreceptor to the monthly oscillator.

Thus, we tested how the circalunar clock is shifted in response to a re-entrainment with naturalistic moonlight in Platynereis wt versus l-cry−/− worms. For this, animals initially raised and entrained under standard worm room light conditions of artificial sun- and moonlight (Supplementary Fig. 1b, e) were challenged by a deviating FM stimulus of 8 nights of naturalistic moonlight (Fig. 4a, Supplementary Fig. 1c, e). This re-entraining stimulus was repeated for three consecutive months (Fig. 4a).

The resulting spawning distribution was analyzed for the efficacy of the naturalistic moonlight to phase-shift the circalunar oscillator. In order to test if the animals had shifted their spawning to the new phase, we again compared the spawning pattern before the exposure to the new full moon stimulus (months with numbers: data distribution analyzed relative to the initial/old FM, see Fig. 4a for an overview) to the spawning pattern after the exposure to the new full moon stimulus (months with letters: data distribution analyzed relative to the new FM, Fig. 4a). The more similar the data distributions of month 1 is to the months C, D, the more the population was shifted to the new phase.

The first re-entraining full moon stimulus (Fig. 4b, first dark green box) is given in the middle of the main spawning period. The nocturnal light itself does not cause immediate effects on the number of spawning worms (Fig. 4b, see also Fig. 2b, c), but the repeated exposure resulted in a noticeable shift of the spawning distribution indicating a phase shift of the monthly oscillator in wildtype. Already at the third re-entraining full moon stimulus, wildtype animals exhibited a completely shifted spawning pattern (Fig. 4b, d-d″, month 1, 2 vs. month C). This is supported by statistical analyses: When comparing the months 1 and 2 (relative to the old FM before the shift) to the month C (relative to the new FM after the shift), both the Kolmogorov–Smirnov test (Fig. 4b: gray rectangles, 4f) and the Mardia–Watson–Wheeler test of the same data were non-significant (Fig. 4f), indicative of the population shifting to the new phase. Consistently, the direction angle (µ) of the mean vectors before and after the shift was highly similar, resulting in a phase difference of only 0.2 days between months 1 and C and 0.5 days between month 2 and month C (Fig. 4f, for details see methods). The month under circalunar free-running conditions (month D) supports this observation, albeit with lower statistical support (Fig. 4b, d″, f).

Of note, wild-type worms would eventually reach the high spawning precision found under naturalistic moonlight only after several more months based on independent experiments (Fig. 2j, k).

When we analyzed the spawning distribution of l-cry mutants in the same way as the wild types, we found that the data distribution exhibited significant differences in the linear Kolmogorov–Smirnov test when comparing months 1 and 2 before the shift to the months C and D after the shift (Fig. 4c: gray rectangles, Fig. 4g); as well as in the phase distribution in the circular analyses when comparing the months before the shift (months 1 and 2) with the last months of the shift (months C,D) (Fig. 4e, e′ versus e″, e‴, g). The populations also exhibited a noticeable phase difference of ≥3.5 days (Fig. 4g).

Based on the statistical significant difference in the re-entrainment of l-cry^–/–, but not wild-type populations under a naturalistic sun- and moonlight regime, we conclude that L-Cry also likely contributes to circalunar entrainment as a photoreceptor. However, as these differences are rather minor, compared to the much stronger differences seen under artificial light regime, we conclude that its major role is the light gatekeeping function.

In an independent study that focused on the impact of moonlight on daily timing, we identified r-Opsin1 as a lunar light receptor that mediates moonlight effects on the worms’ ~24 h clock³⁶. We tested if r-opsin1 is similarly important for mediating the moonlight effects on the monthly oscillator of the worm, analyzed here. This is not the case. r-opsin1^–/– animals re-entrain as well as wildtype worms under naturalistic light conditions (Supplementary Fig. 4). This adds to and is also consistent with our above observation that the spawning distribution is un-altered between r-opsin1^–/– and wildtype animals under artificial light conditions (Supplementary Fig. 3a, b, e, f). This finding also further enforces the notion that monthly and daily oscillators use distinct mechanisms, but both require L-Cry as light interpreter.

L-Cry discriminates between naturalistic sun- and moonlight by forming differently photoreduced states

Given that the phenotype of l-cry^–/– animals suggests a role of L-Cry as light gatekeeper, i.e. only allowing the ‘right’ light to most efficiently impact on the circalunar oscillator, we next investigated how this could function on the biochemical and cell biological level.

While we have previously shown that Pdu-L-Cry is degraded upon light exposure in S2 cell culture¹⁵, it has remained unclear if L-Cry has the spectral properties and sensitivity to sense moonlight and whether this would differ from sunlight sensation. To test this, we purified full length L-Cry from insect cells (Supplementary Fig. 5a–c). Multi-angle light scattering (SEC-MALS) analyses of purified dark-state L-Cry revealed a molar mass of about 130 kDa, consistent with the formation of an L-Cry homodimer (theoretical molar mass of L-Cry monomer is 65.6 kDa) (Fig. 5a). Furthermore, purified L-Cry binds Flavin Adenine Dinucleotide (FAD) as its chromophore (Supplementary Fig. 5d, e). We then used UV/Vis absorption spectroscopy to analyze the FAD photoreaction of purified L-Cry in presence of 1 mM TCEP to prevent protein oxidation. The absorption spectrum of dark-state L-Cry showed maxima at 450 nm and 475 nm, consistent with the presence of oxidized FAD (Supplementary Fig. 5f, black line). As basic starting point to analyze its photocycle, L-Cry was photoreduced using a LED (PerkinElmer ACULED Dyo) with a blue-light dominated spectrum and spectral peak at 450 nm (Supplementary Fig. 1d, d′, henceforth referred to as “blue-light”) for 110 s³⁷. The light-activated spectrum showed that blue-light irradiation of L-Cry leads to the complete conversion of FAD_ox into an anionic FAD radical (FAD^o-) with characteristic FAD^o- absorption maxima at 370 nm and 404 nm and reduced absorbance at 450 nm (Supplementary Fig. 5f, blue spectrum, black arrows). In darkness, L-Cry reverted back to the dark-state with time constants of 2 min (18 °C), 4 min (6 °C) and 4.7 min (ice) (Supplementary Fig. 5g–k).

We then investigated the response of L-Cry to ecologically relevant light, i.e. sun- and moonlight using naturalistic sun- and moonlight devices that we designed based on light measurements at the natural habitat of P. dumerilii³¹ (Supplementary Fig. 1a, c, e). Upon naturalistic sunlight illumination, FAD was photoreduced to FAD^o-, but with slower kinetics than under the stronger blue-light source, likely due to the intensity differences between the two lights (Supplementary Fig. 1c–e).

While blue-light illumination led to a complete photoreduction within 110 s (Supplementary Fig. 5f), sunlight induced photoreduction to FAD^o- was completed after 5–20 min (Fig. 5b) and did not further increase upon continued illumation for up to 2 h (Supplementary Fig. 6a). Dark recovery kinetics had time constants of 3.2 min (18 °C) and 5 min (ice) (Fig. 1c, Supplementary Fig. 6b, c).

As the absorbance spectrum of L-Cry overlaps with that of moonlight at the Platynereis natural habitat (Supplementary Fig. 1a), L-Cry has the principle spectral prerequisite to sense moonlight. However, the most striking characteristic of moonlight is its very low intensity (5.8 × 10¹⁰ photons/cm²/s at −5m, Supplementary Fig. 1a–e). To test if Pdu-L-Cry is sensitive enough for moonlight, we illuminated purified L-Cry with our custom-built naturalistic moonlight, closely resembling full-moonlight intensity and spectrum at the Platynereis natural habitat (Supplementary Fig. 1a, c, e). Naturalistic moonlight exposure up to 2.75 h did not markedly photoreduce FAD, notably there was no difference between 1 h and 2.75 h (Fig. 5d). However, further continuous naturalistic moonlight illumination of 4 h and longer resulted in significant changes (Fig. 5d), whereby the spectrum transitioned towards the light activated state of FAD^o- (note peak changes at 404 nm and at 450 nm). This photoreduction progressed further until 6 h naturalistic moonlight exposure (Fig. 5d). No additional photoreduction could be observed after 9 h and 12 h of naturalistic moonlight exposure (Fig. 5e), indicating a distinct state induced by naturalistic moonlight that reaches its maximum after ~6 h, when about half of the L-Cry molecules are photoreduced. This time of ~6 h is remarkably consistent with classical work showing that a minimum of ~6 h of continuous nocturnal light is important for circalunar clock entrainment, irrespective of the preceding photoperiod⁵. The dark recovery of L-Cry after 6 h moonlight exposure occurred with a time constant of 6.7 min at 18 °C (Fig. 5f, Supplementary Fig. 6d). Given that both sunlight and moonlight cause FAD photoreduction, but with different kinetics and different final FAD^o- product/FAD_ox educt ratios, we wondered how purified L-Cry would react to transitions between naturalistic sun- and moonlight (i.e. during “sunrise” and “sunset”).

Mimicking the sunrise scenario, L-Cry was first illuminated with naturalistic moonlight for 6 h followed by 20 min of sunlight exposure. This resulted in an immediate enrichment of the FAD^o- state (Fig. 5g). Hence, naturalistic sunlight immediately photoreduces remaining oxidized flavin molecules, that are characteristic of moonlight activated L-Cry, to FAD^o-, to reach a distinct fully reduced sunlight state.

In contrast, when we next mimicked the day-night transition (“sunset”) by first photoreducing with naturalistic sunlight (or strong blue-light) and subsequently exposed L-Cry to moonlight, L-Cry first returned to its full dark-state within about 30 min (naturalistic sunlight: τ = 7 min (ice), Fig. 5h, Supplementary Fig. 6e; blue-light: τ = 9 min (ice), Supplementary Fig. 6f–h), despite the continuous naturalistic moonlight illumination. Prolonged moonlight illumination then led to the conversion of dark-state L-Cry to the moonlight state (Fig. 5i, Supplementary Fig. 6f). Hence, fully photoreduced sunlight-state L-Cry first has to return to the dark-state before accumulating the moonlight state characterized by the stable presence of the partial FAD^o- product/FAD_ox educt. In contrast to sunlight-state L-Cry, moonlight-state L-Cry does not return to the oxidized (dark) state under naturalistic moonlight (Fig. 5e), i.e. moonlight maintains the moonlight state, but not the sunlight state. We note, that a partially photoreduced L-Cry state may be formed transiently during dark-state recovery of the sunlight state under moonlight. However, this transiently occurring partially photoreduced L-Cry state would differ from the “true” moonlight state (e.g. by an allosteric change) preventing its accumulation (see discussion and Supplementary Fig. 6i).

Given that L-Cry forms a homodimer and moonlight photoreduces about half of the FAD molecules, we propose that the moonlight state corresponds to a half-reduced FAD^o- FAD_ox dimer, where FAD is only photoreduced in one L-Cry monomer, whereas in the sunlight state both monomers are photoreduced (FAD^o- FAD^o-) (Fig. 5j). This implies that the quantum yield for FAD_ox to FAD^o- photoreduction differs between the two L-Cry monomers. One monomer (referred to as “A” in Fig. 5j) acts as “very low intensity light sensor” with a high quantum yield Φ_A. Hence, the very low photon number provided after 6 h of moonlight illumination is sufficient to photoreduce its flavin co-factor, resulting in the partially photoreduced FAD^o- FAD_ox moonlight state (Fig. 5j).

For direct comparison, our naturalistic moonlight’s emission (in the main absorbance range of L-Cry: 330 nm–510 nm) is 5.4 × 10¹⁰ photons/cm²/s (Supplementary Fig. 1e), which accumulates to ~1.2 × 10¹⁵ photons/cm² in the 6 h required to reach the half-reduced moonlight state (Fig. 5d, e). For naturalistic sunlight, emitting ~7.5 × 10¹⁴ photons/cm²/s (330–510 nm), at least 5 min of sunlight illumination (i.e. > ~1.8 × 10¹⁷ photons/cm²) are required to photoreduce the flavin in both L-Cry monomers in order to reach the fully photoreduced FAD^o- FAD^o- sunlight state (Fig. 5b, j). Thus, the second L-Cry monomer (monomer “B” in Fig. 5j) has a significantly lower quantum yield Φ_B for FAD photoreduction (Φ_B < Φ_A). Of note, even 12 h of naturalistic moonlight (~2.3 × 10¹⁵ photons/cm²) are about 100-fold below the minimal number of photons required to reach the fully photoreduced sunlight state, consistent with our observation that even after 12 h of naturalistic moonlight exposure only about half of the FAD_ox molecules are photoreduced (Fig. 5e). Taken together, our results indicate the existance of L-Cry homodimers with very different light sensitivities of its two monomers, which enables the formation of a partially photoreduced moonlight state that is kinetically and structurally distinct from the fully photoreduced sunlight state of L-Cry (Fig. 5j, see Supplementary Fig. 6i and discussion for a more detailed model that also includes the possible formation of a short-lived [FAD^o- FAD_ox] intermediate upon dark-/sunlight state interconversion and possible allosteric effects).

Naturalistic sun- and moonlight differently affect L-Cry subcellular localization

In order to further investigate the response of L-Cry to naturalistic sun- and moonlight, we conducted Western blots and immunohistochemistry at different lunar and daily timepoints (Fig. 6a, a″). For the analyses of total protein levels via Western blots, we compared equal lengths of sun- versus moonlight illumination versus darkness, each having 8 h duration during their naturally occurring time (Fig. 6a, a″). L-Cry levels after 8 h of naturalistic sunlight (day before full moon = FM-1, diel time: zeitgeber time 8 = zt8, see Fig. 6a, a′) were significantly reduced compared to 8h under darkness at the same moon phase (FM-1, zt 0–10 min, Fig. 6b, c), in line with (canonical) L-Cry degradation in response to naturalistic sunlight.

In contrast to sunlight, exposure to an equal length (8 h) of naturalistic moonlight did not cause a reduction in L-Cry levels compared to an equivalent time (8 h) in darkness (FM-1, zt0–10 min versus FM7, zt0–10 min: Fig. 6b, c, Supplementary Fig. 7). Thus, any potential moonlight signaling via L-Cry occurs via a mechanism independent of L-Cry degradation.

We next examined the spatial distribution of L-Cry in worm heads (Fig. 6d) at lunar and diel timepoints (Fig. 6a–a″). After 8h of a dark night (i.e. NM, zt0–10 min), L-Cry is found predominantly in the nucleus of individual cells, (Fig. 6e–e‴, quantification as numerical data, i.e. nuclear/cytoplasmic ratio: Fig. 6h, for quantification as categorical data³⁸: Supplementary Fig. 8a′–c″, d–f, Supplementary Data 1). Given that an equivalent time of 8h of sunlight exposure results in strong degradation of L-Cry and hence loss of staining signal (see Western blots above), we analyzed L-Cry’s localization after a short exposure. After only 10 min of exposure to naturalistic sunlight (NM zt0 + 10 m, Fig. 6a, a′), the L-Cry nuclear localization strongly diminished, becoming predominantly cytoplasmic (Fig. 6f–f‴, numerical quantification Fig. 6h, categorical quantification Supplementary Fig. 8a′–c″, d–f, Supplementary Data 1, additional repetitions, image overviews and absolute cytoplasmic versus nuclear values: Supplementary Fig. 10). This suggests that naturalistic sunlight causes a shift of the protein to the cytoplasm, followed by degradation. (NM and FM-1 are identical in their illumination regime).

Given the degradation of L-Cry by naturalistic sunlight, we next asked the question if L-Cry is present at night timepoints, allowing for sufficient exposure to naturalistic moonlight to reach the moonlight state. We tested two diel timepoints of the first night lit by the naturalistic moonlight for circalunar entrainment (FM-1): at zt16 (just after the naturalistic sunlight is off and moonlight is on) and at zt20 (after 4 h of naturalistic moonlight exposure) (Supplementary Fig. 9a, a′). We observe that low levels of L-Cry can be detected at FM-1 zt16 (Supplementary Fig. 9b–b‴), and increase within the next hours (see FM-1 zt20, Supplementary Fig. 9c–c‴), with a predominantly nuclear L-Cry localization. At this timepoint still 4 h of moonlight illumination remain for the protein to biochemically reach the full moonlight state (ZT20 to ZT24). Based on these data we conclude that within the organism and under natural conditions (with the moon illuminating at least 8 h of the night under full moon conditions even during summer photoperiods), L-Cry has sufficient time to reach its moonlight state (by changing from sunlight to dark to moonlight state and/or by de novo synthesis of dark adapted L-Cry that reaches the moonlight state within 4 h- see biochemical kinetics, Fig. 5d–j, Supplementary Fig. 6f, g).

Upon further naturalistic moonlight exposure for seven continuous nights (FM7, zt0–10 min) L-Cry remained clearly nuclear (Fig. 6g–g‴, numerical quantification Fig. 6h, categorical quantification: Supplementary Fig. 8f, Supplementary Data 1, additional repetitions, image overviews and absolute cytoplasmic versus nuclear values: Supplementary Fig. 10). Thus, the sunlight and moonlight-states of L-Cry correlate with distinct subcellular distribution patterns.

We occasionally observed that L-Cry at FM7, zt0–10 min can be more nuclear restricted than at zt0–10 min under NM (Supplementary Fig 8f, Supplementary Data 1). This is likewise apparent from the cytoplasmic values in the absolute quantifications (Supplementary Fig 10a). However, the large set of numerical quantification repetitions (Supplementary Fig. 10) also shows that this aspect can be more variable, possibly due to additional metabolic and/or sexual differentiation differences at present outside of our control. Thus, the exact cellular consequences between dark and moonlight states of L-Cry remain to be determined.

Complementing the spawning analyses on genetically mutated animals, these findings however clearly show that the two different light signals- moonlight versus sunlight- impact differentially on L-Cry quantity and localization.

This allows us to put forward a model, in which L-Cry directly via its biochemical states and connected cellular signaling properties is able to discriminate between (naturalistic) sun- and moonlight and to function as a gate keeper for potentially entraining light stimuli for the circalunar oscillator (Fig. 7b). But why would it be required to do this in nature? As we expand in more detail in the discussion, we speculate that this is necessary to entrain to a specific moon phase, which is the full moon phase for Platynereis. This moon phase is specifically characterized by the long duration of detectable moonlight, i.e. moonlight during the entire night²⁶ (Fig. 7a). Interestingly, this matches the biochemical kinetics of at least 6 h of light exposure to acquire L-Cry’s biochemical moonlight state. However in nature, where the setting of the full and waning moons is immediately followed by sunrise (i.e. no darkness window, Fig. 7a²⁶,), measuring the duration of light exposure alone would not allow the worms to detect a specific moon phase (Fig. 7a). Thus, under the natural conditions of waning/waxing moon phases and sunrise/sunsets, being able to detect the switch from moonlight to sunlight is essential to determine the end of the moonlight phase and thus to discriminate between full moon and waning moon phases (Fig. 7a).

Furthermore, L-Cry’s gate keeping mechanism likely also makes the entrainment system more stable against irregular illumination as it could arise from thunderstorms.

Source: Ecology - nature.com