Crayfish P. Clarkii Retina And Nervous System Exhibit Antioxidant Circadian Rhythms (fanjul-moles Et Al, 2009)

Photochemistry and Photobiology, 2009, 85: 78–87

Crayfish Procambarus clarkii Retina and Nervous System Exhibit Antioxidant Circadian Rhythms Coupled with Metabolic and Luminous Daily Cycles Marı´a Luisa Fanjul-Moles*, Julio Prieto-Sagredo, Dario Santiago Lo´pez, Ramo´n Bartolo-Orozco and Hugo Cruz-Rosas Laboratorio de Neurofisiologı´a Comparada, Facultad de Ciencias, Universidad Nacional Auto´noma de Me´xico, Mexico City, Me´xico Received 10 December 2007, accepted 5 May 2008, DOI: 10.1111 ⁄ j.1751-1097.2008.00399.x

entrains the rhythms to seasonal photoperiod changes, maximizing metabolic and behavioral functions. In animals, circadian and exogenous daily variations including those related to locomotor and brain activity, as well as temperature and light fluctuations, result in corresponding daily patterns of reactive oxygen species (ROS) (1), which should lead to rhythmic oxidative damage in the absence of a rhythmic antioxidant system to counterbalance this. ROS and antioxidants are known to influence the expression of a number of genes and transduction pathways (2). It has recently been demonstrated that the redox process plays a central role in the function of the master clock of different organisms (3). In addition, some authors propose that circadian clocks organize metabolic functions into a coherent daily schedule, assuring the synchrony of this schedule with environmental changes (4). The crayfish is a nocturnal organism that exhibits a span of overt circadian rhythms for adaptation to daily and seasonal environmental light variation. In Procambarus clarkii, many authors have documented metabolic (5–7) and activity (8,9) circadian rhythms that should produce daily variations in the oxidative status (OS) of this animal’s internal environment. Of late, daily and circadian changes in OS produced by different light conditions and its counterbalance by an antioxidant circadian system—the glutathione system—have been documented in the midgut and hemolymph of P. clarkii (10,11). These works revealed differences between the midgut and hemolymph reduced glutathione (GSH) oscillatory system that suggest differences in the coupling strength of both organs with the crayfish circadian clock. These results led us to propose the hemolymph as a passive system responding to an oscillating driving force depending on various self-sustaining oscillators such as the midgut, and those proposed as putative pacemakers of crayfish and located in the brain, optic lobe and the retina (12). However, neither the existence of a GSH circadian system in these putative pacemakers nor the efficacy of this to counterbalance oxidative damage of internal and external ROS in these neural structures, as well as its timing, have been studied. Thus, in this work we investigated the interaction of metabolic, environmental, and GSH antioxidant rhythmic changes with changes in oxidative damage in the aforementioned structures and proposed, as putative pacemakers of the crayfish circadian system, retina and the complex optic lobe-brain (OL-B).

ABSTRACT Based on previous work in which we proposed midgut as a putative peripheral oscillator responsible for circadian reduced glutathione (GSH) crayfish status, herein we investigated the retina and optic lobe-brain (OL-B) circadian GSH system and its ability to deal with reactive oxygen species (ROS) produced as a consequence of metabolic rhythms and light variations. We characterized daily and antioxidant circadian variations of the different parameters of the glutathione system, including GSH, oxidized glutathione (GSSG), glutathione reductase (GR) and glutathione peroxidase (GPx), as well as metabolic and lipoperoxidative circadian oscillations in retina and OL-B, determining internal and external GSH-system synchrony. The results demonstrate statistically significant bi- and unimodal daily and circadian rhythms in all GSH-cycle parameters, substrates and enzymes in OL-B and retina, as well as an apparent direct effect of light on these rhythms, especially in the retina. The luminous condition appears to stimulate the GSH system to antagonize ROS and lipid peroxidation (LPO) daily and circadian rhythms occurring in both structures, oscillating with higher LPO under dark conditions. We suggest that the difference in the effect of light on GSH rhythmic mechanisms of both structures for antagonizing ROS could be due to differences in glutathionesystem coupling strength with the circadian clock.

INTRODUCTION The rotation movement of the Earth originates changes in environmental conditions over the 24 h period, the most prominent being the light–darkness (LD) cycle. The predictability of these changes allows organisms to keep track of time by anticipating these changes and adjusting their internal temporal order to the external time. In organisms, this ability underlies the capacity for endogenous temporal organization of cellular process over the course of the approximately 24 h period. Cellular machinery that generates this ability is known as the biologic clock, and its output, as circadian rhythm. The circadian system in its function as an endogenous clock *Corresponding author email: [email protected] (Marı´ a Luisa Fanjul-Moles)  2008 The Authors. Journal Compilation. The American Society of Photobiology 0031-8655/09

78

Photochemistry and Photobiology, 2009, 85

MATERIALS AND METHODS Animals and experimental design. We used 72 P. clarkii crayfish of homogenous size and weight in intermolt stage and that were acclimatized to the laboratory for 1 month in aquariums placed under natural LD cycle conditions at a temperature of 20C, pH 7.9 and O2 concentration of 5.7 mg L)1. The aquariums were provided with polyvinyl chloride tubes simulating burrows, which allowed the animals to hide from the light. After acclimatization, the animals were divided into three groups and sub-divided into two batches each for two experiments under the following different light-cycle conditions: (1) 36 animals submitted to LD 12 h:12 h of low intensity (0.043 W m)2) for 15 days, and at the end killed, and (2) 36 animals treated as described previously and subsequently exposed to dark–dark (DD) for 72 h. At the end of each condition, six specimens from each group were selected at random six times daily in a 24 h cycle. At each experimental time point, each organism was anesthetized by placing it on ice in the dark and each was subsequently killed. Retina and OL-B were dissected and processed. The first group was processed to determine GSH and oxidized glutathione (GSSG), while the second group was employed to determine glutathione peroxidase (GPx), glutathione reductase (GR) and lipid peroxidation (LPO). In addition, a third group was utilized to determine L-glucose. Tissue sample. For GSH and GSSG determination, animals were killed by decapitation, and OL-B and retina were dissected and homogenated separately with a tissue grinder in 400 lL 10% trichloroacetic acid. Samples were centrifuged for 15 min (12 800 g) at 8C. The supernatant was stored at )70C until GSH and GSSG were determined. Retina and OL-B samples for enzyme, LPO and glucose determinations were homogenized in 500 lL phosphate buffer (pH 8), centrifuged as previously described, and stored at )70C until assayed. For hemolymph glucose determination, 25 lL samples were obtained with a micropipette from a small hole punched into the ventral membrane of the second or third abdominal segment, 100 lL phosphate buffer (pH 8) was added, and the sample was centrifuged and stored as described previously. GSH and GSSG high-performance liquid chromatography (HPLC) determination. Reduced glutathione and GSSG were determined simultaneously with the reverse-phase method previously reported by Harvey et al. (13) and modified at our laboratory for amperometric methods. All reagents were obtained from Sigma-Aldrich Co. (St. Louis, MO). GSH and GSSG standards were employed to construct calibration curves, and L-cysteine was used as internal standard. Stock solutions of these reagents were prepared each week, and dilutions for calibration curves were prepared daily. The mobile phase consisted of a sodium phosphate (monobasic) 10 lM solution adjusted to 2.7 pH with 85% phosphoric acid to which 0.05 mM octyl sulfate and 2% acetonitrile were added. The mobile phase was filtered with 0.22 lm nylon membrane and degassed in line with the LC26B model On-line Vacuum Degasser, BAS (all Bioanalytical Systems [BAS] equipment, Bioanalytical Systems, Inc., West Lafayette, IN). An isocratic pump model PM 80 Solvent Delivery System BAS was utilized at a 0.5 mL min)1 rate to deliver the eluent. A 100 lL volume of both standards and samples was injected using a Rheodyne I125 injector (20 lL loop, Rheodyne LLC, Rohnert Park, CA), and separation was performed with a C18 BAS column model MF-8954 (3 · 10 mm, 3 lm, ODS 80A˚). Detection was performed with a LC4C model Amperometric Detector BAS using a glassy carbon dual electrode and an Ag-AgCl reference electrode. Detection parameters were as follows: applied voltage, 1.35 V; generator voltage, 0.95 V; detection range, 0.5–2 lA and filter, 0.1 Hz. Detected currents were recorded in a BAS Dual Pen Recorder model 1202-9054. GSH and GSSH peak amplitudes were measured, and the concentration was determined interpolating data in the calibration curves; the GSH ⁄ GSSG ratio was calculated. GR and GPx activity and protein assay. GR activity was determined as previously reported (14) employing the method described by Bompart et al. (15). Briefly, 20 lL of the total 500 lL homogenate was placed in a cuvette containing GSSG 0.44 mmol L)1 in 0.1 M phosphate buffer 0.03 M EDTA at pH 7 and reduced 0.036 M nicotinamide adenine dinucleotide phosphate (NADPH) prepared and added immediately prior to the assay as the starting reagent. The assay was run at 340 nm for 4 min with absorbance readings taken every 30 s. Absorbance was assayed for all readings in an Ultraspec

79

2000 spectrophotometer (Pharmacia Biotech, Buckinghamshire, UK). Activity was expressed in millimoles of oxidized NADPH per g of protein per min using an extinction coefficient of 6.2 mM min)1. GPx activity was determined using the method described by Paglia and Valentine (16). Twenty microliters of the homogenate was placed in a cuvette containing 0.001 M sodium azide (NaN3) in 0.1 M phosphate buffer 0.001 M EDTA at pH 7.4; GSH 0.001 M, NADPH 0.2 mM and GR 1 U mL)1 were then prepared and added immediately prior to the assay; 30 lL of 0.25 mM perhydrol was also prepared and added immediately prior to assay initiation as the starting reagent. The assay was run at 340 nm for 10 min, with absorbance readings taken every 1 min. Absorbance was assayed for all readings in an Ultraspec 2000 spectrophotometer (Pharmacia Biotech). Activity was expressed in millimoles of oxidized NADPH per g of protein per min utilizing an extinction coefficient of 6.2 mM min)1. The protein assay was performed according to Bradford. Glucose determination. Hemolymph (50 lL) as well as retina and eyestalk tissue were collected from each animal at the base of a walking leg every 4 h over a 24 h period. Glucose levels were measured by Trinder’s glucose oxidase method (17) utilizing a Diagnostic Chemical Limited kit (Charlottetown, CA). Absorbance at 505 nm was assayed in an Ultraspec 2000 spectrophotometer (Pharmacia Biotech). Lipid peroxidation analysis. LPO levels were quantified utilizing a K-ASSAY LPO kit (Kamiya Biomedical Co., Seattle, WA) (18). Six animals from each group were randomly chosen and killed by decapitation. Retina and OL-B samples were homogenized in PBS 1:10 and stored at )70C until the day of the assay to measure LPO. The homogenates of each sample were centrifuged. The supernatant was removed and the enzyme reagent (ascorbic oxidase and lipoprotein lipase) was added. The mixture was incubated for 5 min at a temperature of 30C and the chromogen reagent (10-N-methylcarbamoyl-3,7-dimethylamino-10-H-phenothiazine) was added. The resulting mixture was incubated for 10 min at 30C. Finally, absorbance was measured at 675 nm in a spectrophotometer. A two-point calibration curve was done using the saline blank (0 nmol mL)1) and the 50 nmol mL)1 cumene hydroperoxidase standard provided with the kit. Results were then calculated using the kit instructions. Assessment of light parameters. The photoperiod was provided by neon lamps turned on and off by a timer at 0700 and 1900 h, and light quantum scalar irradiance was calibrated with a photoradiometer equipped with a spherical submarine sensor (LiCor models LI-189 and LI-193SA; LiCor, Lincoln, NE). Intensity values were set at the minimal values of light irradiance to which crayfish are exposed in their natural environment. Data analysis. Chronograms were constructed as group mean ± standard deviation (SD). To estimate circadian rhythms for each GSH parameter (GSH, GSSG, GSH ⁄ GSSG, GR and GPx) of the different groups of animals, a single cosinor analysis was performed. The COSANA software program was used. Based on a test period (s), cosinor analysis adjusts data to a cosinusoidal function and provides an objective test of whether the rhythm amplitude differs from zero, i.e. whether the rhythm is validated for an assumed s (19). This method provides descriptive estimators for a number of different rhythm parameters, i.e. acrophase, mesor, amplitude and percentage of rhythmicity (PR). Acrophase is the crest time of the best-fitting mathematical function approximating the data, while mesor comprises the value around which oscillation occurs; when the time interval between data samplings is constant, it equals the rhythmic oscillation’s arithmetic mean. Hence, in the present work it corresponds to the arithmetic mean of the rhythmic oscillation of the concentration or activity of the GSH parameters over a 24 h period. Mesor (M) is defined as the adjusted mean of the rhythm. When the interval of time between data sampling (Dt) is constant, M equals the arithmetic mean. The amplitude is equal to one half of the difference between highest and lowest oscillation values, and PR is the percentage of data included within the 95% confidence limits of the best-fitting cosine function. This test allows subjective examination of the hypothesis that rhythm amplitude differs from zero using different trial-period lengths. In the current work, several periods were tested to analyze whether temporal GSH-parameter profiles under different LD conditions were indeed circadian. In addition, analysis of variance (ANOVA) was followed by Scheffe´ post hoc comparisons for rhythms for which waveform did not adjust cosinor. For statistical analysis of differences in terms of the concentration and activity of all parameters between

80

Marı´a Luisa Fanjul-Moles et al.

LD and DD in retina and OL-B, a Student’s t-test was performed. Significant clustering of phases in both LD and DD were determined using circular statistics (Rayleigh and V-tests). Both tests examine whether there is statistical evidence of directional phase preference in the population from which the sample is drawn (20).

RESULTS Analysis of crayfish retina and OL-B complex extracts revealed the presence of two substances that oxidized at 1.35 V and eluted at the same time as the GSH and GSSG standard (mean elution time = 4 and 8 min, respectively). In addition, a single symmetrical and additive peak was observed for both substances when GSH and GSSG external standards were added to acid extracts in each injection (Fig. 1). The voltage– amperage curve generated for GSH and GSSG standards and presumptive peaks in the sample were also identical. Based on these criteria, these two peaks were identified as GSH and GSSG. Results of the current study demonstrate the presence of all GSH-system components in crayfish retina and OL-B. These structures show lower GSH and GSSG concentration values than those reported for hemolymph and midgut in the same species (10) as follows: LD retina GSH = 42.6 ± 3.7 lM and GSSG = 8.6 ± 1.0 lM; DD retina GSH = 14.96 ± 1 lM and GSSG = 6.12 ± 0.8 lM; LD OL-B GSH = 26.4 ± 2.0 lM and GSSG = 2.35 ± 0.2 lM; and DD OL-B GSH = 12.5 ± 0.8 lM and GSSG = 7.3 ± 1.2 lM. Glutathione

reductase and GPx also demonstrated lower activity than that reported in the same work for hepatic tissue as follows: LD retina GR = 11.7 ± 2.5 lM NADPH ox g)1 prot min)1; GPX = 2.4 ± 0.4 lM NADPH ox g)1 prot min)1; DD retina GR = 17.1 ± 4.2 lM NADPH ox g)1 prot min)1; GPx = 10.2 ± 1 lM NADPH ox g)1 prot min)1; LD OL-B GR = 37.4 ± 4.9 lM NADPH ox g)1 prot min)1; GPx = 8.3 ± 1.9 lM NADPH ox g)1 prot min)1; DD OL-B, GR = 6.7 ± 1.3 lM NADPH ox g)1 prot min)1; and GPx = 15.33 ± 3.8 lM NADPH ox g)1 prot min)1. The amount and activity of all glutathione system parameters changed with the luminous condition. The Student’s t-test revealed significant differences in all parameters except GSSG and GR between LD and DD in brain and retina, respectively (Table 1). Daily and circadian rhythm determination Chronograms showing GR and GPx temporal activity in OL-B are depicted in Fig. 2a,b. In LD, the GR activity peak is at 0800 h, apparently increasing immediately after lights on at 0700 h, with a second peak appearing at 2000 h after lights off at 1900 h. Cosinor analysis revealed a GR bimodal significant daily rhythm with a 12 h period value (Table 2). Both peaks are coincident with maximal peaks of GPx bimodal activity, although GPx showed lower average activity throughout 24 h (M = 8.6) than GR (M = 36.9 ± 3.8) (Table 2). After 72 h of darkness, the mean GPx activity level increases (M = 16.1),

Figure 1. (a) Chromatograms from 150 and 300 ng external standards of GSH and GSSH. Retention times are 4 and 8 min, respectively. (b) HPLC of an acid extract of crayfish eyestalk-brain (OL-B). (c) Separation of the same acid extract with 100 and 200 ng GSH and GSSG external standards. Inset shows the mobile phase and TCA chromatograms. Vertical arrows indicate the injection time, diagonal arrows show GSH, GSSG and TCA peaks. For details, see Materials and Methods.

Photochemistry and Photobiology, 2009, 85 exhibiting a bimodal rhythm of s = 12.3 h, which peaks at 1200 h. Meanwhile, GR-activity dampens showing no statistically significant oscillation (M = 6.7, A = 3.14). Table 1. Mean and standard error for the variables in LD and DD conditions. Variable OL-B GSH1* GSSG* GSH ⁄ GSSG* GR2* GPx* LPO3* Glucose4 Total protein5* Retina GSH* GSSG GSH ⁄ GSSG* GR* GPx* LPO* Glucose Total protein* Hemolymph Glucose

LD 12:12

Constant darkness

26.42 2.35 13.35 37.4 8.3 5.76 1 245

± ± ± ± ± ± ± ±

2.04 0.2 1.35 4.9 1.9 0.62 0.12 16.5

12.57 7.33 2.83 6.7 15.33 26.82 1.13 801.1

± ± ± ± ± ± ± ±

0.85 1.27 0.42 1.3 3.8 0.96 0.18 15.4

42.67 8.64 6.2 11.7 2.47 31.51 0.52 230

± ± ± ± ± ± ± ±

3.75 1.14 0.58 2.5 0.38 1.32 0.17 20

14.96 6.12 3.31 17.1 10.2 509.87 0.44 658.3

± ± ± ± ± ± ± ±

1.08 0.82 0.31 4.2 1.08 23.31 0.06 6.5

19.61 ± 2.24

20.43 ± 3.29

Units of measure: 1GSH and GSSG, lM; 2GR and GPx, lM NADPH ox g)1 prot min)1; 3LPO, nmol mL)1; 4glucose, mM; 5protein, lg per structure. *Student’s t-test revealed statistically significant differences between LD and DD means for each variable (P < 0.01).

81

In the retina, changes in GR and GPx activity in LD and DD were observed (Fig. 2c,d). GR activity oscillates under LD, exhibiting a bimodal daily rhythm not statistically significant neither by cosinor or ANOVA (Table 2). GPx activity depicts a statistically significant bimodal rhythm (s = 12.3 h) that establishes a 4 h phase relationship (w) with GR rhythmic oscillation. Maximal GR peak activity takes place at photophase at the middle of the subjective day, after the second lower peak of GPx activity at 0800 h after lights on. However, under darkness, a GR statistically significant circadian rhythm emerges (t = 24.3); meanwhile GPx rhythmic activity seems to disappear (Table 2). Temporal GSH and GSSG concentration changes in OL-B complex are shown in Fig. 3a,b. Figure 3a depicts chronograms showing 24 h LD variations of GSH, GSSG and GSH ⁄ GSSG ratio; cosinor analysis detected no 24 h circadian significant rhythms either in GSH or in GSSG abundance, but did detect bimodal oscillations with period values of s = 12.5 and 11.9 h, respectively. The corresponding chronograms show GSH and GSSG damped bimodal oscillations, which shows an increasing GSH ⁄ GSSG ratio in midphotophase at the lowest GSSG oscillation value, followed by a second peak at 0400 h after GSH maximal peak at the subjective night. After 72 h in DD (Fig. 3c), GSH rhythm amplitude decreases, but this tripeptide abundance exhibits statistically significant bimodal rhythm with a 12 h period value (Table 2) with maximal peak at 2000 h and a second at 0800 h. In contrast, GSSG oscillation amplitude under this dark condition increases, indicating an increment of

Figure 2. Chronograms showing rhythmic enzymatic activity of optic lobe-brain (a, b) and retina (c, d) under 12:12 light–dark cycles and continuous darkness. All data are mean ± standard error (SE) (n = 6). Upper black and white bars in each graph denote light and dark phase.

82

Marı´a Luisa Fanjul-Moles et al.

Table 2. Cosinor analysis for the brain and retina variables. Variable ⁄ parameter OL-B LD GSH GSSG GSH ⁄ GSSG GR GPx LPO Glucose OL-B DD GSH GSSG GSH ⁄ GSSG GR GPx LPO Glucose Retina LD GSH GSSG GSH ⁄ GSSG GR GPx LPO Glucose Retina DD GSH GSSG GSH ⁄ GSSG GR GPx LPO Glucose Hemolymph LD Glucose Hemolymph DD Glucose

Period (h)

Mesor1

12.5* 11.9* 243 12* 12* 244 12.8*

26.97 ± 2.34 ± 13.31 36.9 ± 8.6 ± 5.37 1.01 ±

1.72 0.17

12* 26.85 11.46 127 12.3* 24.4* 12.3*

12.57 ± 7.72 2.82 6.7 16.14 ± 26.81 ± 0.98 ±

0.57

12* 24* 24* 128 12.3* 12.1* 24*

42.67 ± 8.65 ± 6.2 ± 11.5 2.48 ± 31.71 ± 0.35 ±

21.49 11.9* 12* 24.3* 1210 21.4* 12.6*

14.79 6.1 ± 3.31 ± 17.3 ± 10.3 505.47 ± 0.45 ±

3.8 1.7 0.11

3.43 0.85 0.08 2.35 0.9 0.39 0.32 0.95 0.04 0.66 0.27 3.2 21.9 0.05

Amplitude1

9.1 ± 0.81 ± 1.87 24.26 ± 6.13 ± 1.49 0.46 ±

2.45 0.24

3.39 ± 3.76 1.07 3.14 14.49 ± 3.74 ± 0.6 ±

1.06

22.74 ± 5.7 ± 3.36 ± 5.8 1.58 ± 7.09 ± 0.2 ± 3.21 3.92 ± 1.26 ± 17.7 ± 3.36 77.88 ± 0.24 ±

5.4 2.5 0.15

4.82 1.19 0.11 3.32 1.27 0.55 0.44 1.33 0.06 0.94 0.38 4.5 31.3 0.07

Acrophase2 (h)

PR (%)

P

34.7 30.26 3.46 43.7 18.9 9.5 25

0.01 0.01 0.63 0.01 0.05 0.26 0.05

27.37 14.73 11.75 9.2 29.2 26.83 54.19

0.01 0.12 0.22 0.2 0.02 0.01 0.01

63.44 42.74 57.61 16.7 35.67 52.11 26.32

0.01 0.01 0.01 0.2 0.01 0.01 0.01

1:20 0:35

14.75 39.12 28.69 36.3 16.62 18.7 35.13

0.1 0.01 0.01 0.01 0.09 0.05 0.01

02:52 ± 05:35 ± 11:29 07:40 ± 07:21 ± 05:43 02:30 ±

0:32 0:35

08:26 ± 23:51 03:34 08:45 10:16 ± 22:27 ± 9:17 ±

0:36

12:19 ± 10:23 ± 00:11 ± 11:23 08:20 ± 02:20 ± 23:07 ± 13:16 04:25 ± 09:18 ± 03:14 ± 4:39 12:43 ± 01:17 ±

0:26 0:46 0:41

0:39 1:15 0:21 0:17 0:51 0:38 0:35 0:22 1:06 0:27 0:35 1:00

22.8*

1.1 ± 0.11

0.44 ± 0.16

16:03 ± 1:19

21.02

0.05

23.5*

1.13 ± 0.17

0.61 ± 0.24

12:38 ± 1:27

19

0.05

PR, percentage of rhythm. P, cosinor statistical significance. 1Units of measure: GSH and GSSG: mM; GR and GPx: lM NADPH g)1 protein min)1; LPO: nmol mL)1; glucose: mg dL)1. 2External time. 3F = 2.16, P > 0.05. LSD post hoc: 1200 h vs 8, 16, 20, 24 h, P < 0.05. 4F = 2.05, P > 0.05. LSD post hoc: 800 h vs 12, 16 h, P < 0.05. 5F = 2.1, P > 0.05. LSD post hoc: 400 h vs 12, 16, 24 h, P < 0.05; 2000 vs 16, 24 h, P < 0.05. 6 F = 1.05, P > 0.05. 7F = 0.6, P > 0.05. 8F = 0.7, P > 0.05. 9F = 2.662, P < 0.05. LSD post hoc: 400 h vs 8, 12, 16, 20, 24 h, P < 0.05. 10 F = 1.011, P > 0.05. LSD post hoc: 400 h vs 8, 12, 20 h, P < 0.05. *Statistically significant by cosinor.

ROS. Although ANOVA and Scheffe´ post hoc tests revealed a statistically significant effect of time on GSSG 24 h variations, cosinor detected no statistically significant rhythm; the waveform of this bimodal rhythm does not adjust as expected to the cosine wave. In DD, as expected, the GSH ⁄ GSSG ratio diminished, exhibiting its minimal peak (at 0400 h); this is coincident after increasing GSSG levels and decreasing GSH levels. Figure 4a,b depicts chronograms demonstrating daily GSH and GSSG oscillations in retina. Reduced glutathione oscillates with a bimodal statistically significant daily rhythm (s = 12 h) whose maximal phases take place at 1200 and 2400 h, while GSSG oscillates and exhibits a 24 h significant daily rhythm whose zenith occurs at 1200 h in mid-photophase. GSH ⁄ GSSG-ratio oscillation shows a maximal value at 2400 h, when a maximal conversion of GSSG into GSH occurs, depicting a statistically significant daily oscillation. After 72 h of darkness, GSSG and GSH concentration values decrease, exhibiting GSSG statistically significant bimodal oscillation (s = 11.9 h) and GSH nonsignificant oscillation by cosinor, although it is statistically significant

by ANOVA (Table 2) and that peaks at the same external time (0400 h). In this work, we utilized 24 h glucose hemolymph, retina and brain variations as daily and circadian metabolic status crayfish markers, as well as LPO as a biomarker of oxidative damage in these structures. Figure 5 shows daily variations of both markers in hemolymph, retina and brain. The chronograms depict a statistically significant daily rhythm of hemolymph glucose concentration (s = 22.8 h), whose zenith is at 16 h, decreasing toward 20 h. Hemolymph glucose concentration increment coincides with progressively increasing glucose levels in both retina and brain. These structures present statistically significant daily uni- and bimodal variations (s = 24 and 12.8 h, respectively), peaking at 2400 and 1200 h. The glucose increment trend in both structures coincides with one of the two peaks of statistically significant retina bimodal daily rhythm LPO (s = 12 h). This rhythm exhibits two peaks—one at photophase (1600 h) and the other at scotophase (0400 h); after this second peak, both LPO and glucose levels decrease. The relationship of glucose and LPO rhythmic oscillations in the retina and OL-B in DD is shown in

Photochemistry and Photobiology, 2009, 85

83

Figure 3. Chronograms showing rhythmic changes of reduced (GSH) and oxidized glutathione (GSSG) concentration as well as GSH ⁄ GSSG ratio calculated from optic lobe-brain (OL-B) samples obtained after light–dark (LD) and dark–dark (DD) conditions. GSH and GSSG concentrations; (a, c), and GSH ⁄ GSSG ratio (b, d). All data are mean ± standard error (SE) (n = 6). Black and white bars as in Fig. 2. See text for further information.

Figure 4. Chronograms showing rhythmic changes of reduced (GSH) and oxidized glutathione (GSSG) concentration as well as GSH ⁄ GSSG ratio calculated from retina samples obtained after 12:12 light–dark cycles (a, b) and dark constant condition.(c, d). All data are mean ± standard error (SE) (n = 6). Black and white bars as in Fig. 2.

84

Marı´a Luisa Fanjul-Moles et al.

Figure 5. Chronograms showing rhythmic changes in glucose concentration and lipid peroxidation (LPO) in crayfish optic lobe-brain (OL-B) and retina under 12:12 light–dark cycles (upper panels a, b) and continuous darkness (lower panels c, d) In both graphs hemolymph glucose concentration data have been plotted for comparison purposes. All data are mean ± standard error (SE) (n = 6). Black and white bars as in Fig. 2. See text for further information.

Fig. 5c,d. Although under this condition the OL-B complex shows lower lipid oxidation levels (26.8 nmol mL)1) than the retina (509.8 nmol mL)1), the corresponding chronograms depict OL-B and retina lipoperoxidative statistically significant circadian rhythms (s = 24.4 h, P < 0.05, and s = 21.4 h, P < 0.05, respectively), whose maximal peaks occur at 2400 and 0800 h. Both glucose and LPO circadian rhythms are antiphased, demonstrating a mirror image. Under DD conditions, these figures depict an evident glucose hemolymph circadian rhythm with higher amplitude and shorter period (s = 23.5 h, P < 0.05) than under LD conditions, showing a phase advance of ca 4 h. Internal synchrony between rhythms Figure 6a–c compares the internal phase angle between acrophase timing of all glutathione rhythms analyzed in this work with those of hemolymph and liver as analyzed in a previous work (11). After 72 h of DD, OL-B and retina glutathione enzymatic rhythms run freely, exhibiting similar phase angles between them (OL-B w GR-GPx = 1.7 h; retina w GR-GPx = 1.23 h) but dissimilar ones with their substrates (OL-B w GPx-GSH = 1.9.0 h, wGR-GSSG = 8.5. h; retina w GPx-GSH = 8.7 h; wGR-GSSG = 1.23 h). The Rayleigh test demonstrated statistically significant nonrandom clustering in OL-B (r = 0.7, P < 0.05), and retina (r = 0.7, P < 0.05), which means that the vector length differs significantly from zero, indicating a statistically significant mean direction.

In both structures, nearly all glutathione parameters appear to reset to LD, advancing or delaying and clustering with a nonstatistically significant vector in retina (r = 0.5, P > 0.05), and with a statistically significant vector in OL-B (r = 0.87, P < 0.05), which demonstrates phase preference toward dawn (u = 2.8, P < 0.05). Interestingly, under this condition OL-B and hemolymph GSH and GSSG acrophase are timed together, showing wGSH-GSSG = 0. In LD retina and OL-B glucose and LPO acrophase time at the dark phase showing wLPO-Glu = 3 h (Table 2). After 3 days of darkness, the phase relationship increased to wLPO-Glu about 12 h in both. On comparing these values with those obtained in midgut in a previous work (11) (Fig. 6c), GSH-system behavior in the midgut under DD appears to be more similar to OLB than to retina. The V-test demonstrated no statistically significant differences between midgut and OL-B preferred mean direction vector, but statistically significant differences between these organs’ mean direction vectors and that of retina (u = 2.5, P < 0.05).

DISCUSSION The results of this study demonstrate statistically significant bi- and unimodal daily and endogenous rhythms in all GSHcycle parameters, substrates and enzymes in OL-B and retina. The bimodality of some GSH and metabolic rhythms found in LD could be associated with bimodal characteristics of the crayfish activity rhythm. Although not recorded in the present

Photochemistry and Photobiology, 2009, 85

Figure 6. Circular phase maps produced by plotting the acrophases of glutathione daily and circadian rhythms of optic lobe-brain (OL-B) (upper graphs), retina (middle graphs), and midgut gland (lower graphs) under light–dark (LD) (left-hand side) and dark–dark (DD) (right-hand side) experimental conditions. Each point represents the acrophase calculated by cosinor analysis. Arrows denote statistically significant mean direction vectors (Rayleigh test). Midgut gland data reproduced with permission from Fanjul-Moles et al., Photochemistry and Photobiology (2003). See text for explanation.

work, this rhythm has been extensively studied (8,9). When the crayfish is maintained under LD cycles, this rhythm exhibits two peaks—lights-off endogenous circadian peak and lights-on exogenous one; thus, in the present work we are unable to discard a possible influence of locomotor activity on the rhythmic features of the parameters determined herein. Figures 2–5 demonstrate the clear bimodal patterns recorded in LD that change into unimodal circadian patterns after 72 h of darkness. Although the cosinor detected both bi- and unimodal patterns under both conditions (Table 2), there is a clear change of phase that demonstrates the circadian nature of these rhythms. Our results indicate an apparent strikingly direct effect of light on these rhythms that especially in the retina increase their amplitude greatly under 24 h LD cycles, exhibiting a greater than three-fold GSH increment. This is a larger increase than that found in the brain in this work, and in midgut and hemolymph previously (11). The cyclic luminous

85

condition appears to stimulate the entire GSH system in retina as a response to antagonize the oxidative effect of light on this organ. Light irradiation producing photo-oxidation is a factor that determines ROS in different animals (1), as has been previously demonstrated in crayfish (10). In this work, both retina and brain showed higher GSH ⁄ GSSG ratio mean values in LD than in DD. The increase in this parameter coincides with the photophase, indicating that oxidative stress produced by light is antagonized by a rapid transformation of GSSG into GSH. This could be the result of the direct effect of the LD light cycle on the glutathione system, but especially on GR activity. Light irradiance could produce an increase in peroxides and the subsequent production of GSSG concentration as a result of increased GPx activity to antagonize this oxidative stress. This reaction must consume a large amount of the GSH resulting from a possible direct effect of light on rhythmic GR activity. The participation of GSH as an antioxidant that either serves as a substrate for GPx peroxidase to reduce hydrogen peroxide, or participates directly in different types of oxidation–reduction reactions (21), one of these involving the hydroxyl radical, should be particularly important for GSH participation in the preservation of OL-B, retinal and neural cells. The hydroxyl radical is one of the most damaging free radicals, and neurons are particularly vulnerable to free radical damage (22). Our results are in agreement with some reports on mammals (23) concerning light-mediated retinal LPO, potentially an important mechanism of retinal degeneration. Nonetheless, and albeit that crayfish retina exhibits statistically significant circadian and daily LPO rhythms, we found an important increase of this marker in DD, as shown by the mean (Table 1) and by mesor and amplitude rhythm values (Table 2). In LD, crayfish retina and brain demonstrate only mild peroxidation, suggesting that the excitatory effect of light on the GSH system appears able to antagonize major LPO and potential retinal degeneration. Photoreceptor membranes, the site of light absorption, are composed of phospholipids highly enriched by unsaturated fatty acids (UFA). Interestingly, after the first 2 h of light exposition visual membranes of the darkadapted eye of P. clarkii exhibit a reduction in docosahexaenoic acid, one of the longest UFA present in crayfish visual membranes (24); this decrease is coupled with decreasing LPO levels (25). This fact could explain the apparent paradoxically statistical increase in LPO levels in DD with respect to LD. The resetting effect of light on its circadian clock could allow crayfish to anticipate changes in the environment for preventing luminous and metabolic oxidation by means of antioxidant systems and for preparation prior to the occurrence of both luminous and metabolic oxidation. It has been proposed that light—especially UVB-induced ROS—is involved in the activation of mitogen-activated protein kinase (MAPK) downstream antioxidant-response effectors (26), and certain MAPK signal-regulation regulations by GSH have been described recently (27). In mammals, much of the GSH synthesis in the body takes place in hepatocytes and is transported in the central nervous system via the blood–brain barrier transporters; notwithstanding this, it has been proposed that both neural and retinal cells are able to synthesize this thiol (28). In crayfish, GSH synthesis and distribution have been poorly studied. To our knowledge, this is the first work reporting the presence of GSH-cycle enzymes and substrates in crustacean nervous system and retina, although this is not the first work

86

Marı´a Luisa Fanjul-Moles et al.

on crustacean that identifies the relationship between metabolism LPO and an antioxidant system. In hepatopancreas and gills of the estuarine crab Chasmagnatus granulata (29), daily variations have been identified of metabolic rhythms coinciding with maximal activities of catalase, glutathione-S-transferase and LPO in the dark phase of the LD cycle. It could be interesting to identify these enzymatic activities in a new work based on the findings of the present one, during which we found two peaks of LPO—one at the photophase, and the other at the dark phase in a different species. Although all GSH parameter concentrations and activities are higher in the midgut and even in the hemolymph (11), the enhancing effect of light on these parameters in OL-B, and especially in the retina, is higher than in the midgut. This could suggest that although these compounds are distributed via the hemolymph, there is also de novo synthesis through the effect of light. In crayfish submitted to constant darkness, the previously mentioned circadian rhythms persist, but retinal sensitivity increases, and an increment in time-of-activity (alpha) takes place (30,31). Thus, we could expect an increase under this condition of pro-oxidant reactions linked with metabolism and consequently ROS formation. However, we found no statistically significant differences among glucose concentration in LD and DD in hemolymph, retina or OL-B that could support this increasing metabolic flux. We recognize that this parameter, although employed here as a metabolic marker, does not represent a good oxidative-flux marker. It appears plausible that the higher levels of LPO found under DD in the retina could be to due to an increase in ROS as a by-product of respiratory-chain flux that could not be scavenged by GPx, as indicated by the lower GSH ⁄ GSSG ratio found under this condition in both structures compared with LD (Table 2). Despite this decrease, OL-B showed a lower LPO increase than the retina. This fact may be due both to lower polyunsaturated acid content in the membranes of these structures compared with retina, and to the presence of other radical scavengers such as melatonin, as reported elsewhere in P. clarkii eyestalk (32). Nevertheless, another possibility could be one proposed elsewhere (11)—that this may be due to differences in the glutathione system internal coupling strength of both structures to the circadian clock. Retina, more susceptible to the effect of light, resets toward the middle of the subjective day, anti-phased with LPO and glucose rhythms, of which acrophase comprises the middle of the night. Conversely, in OL-B, GSH ⁄ GSSG increase at the subjective night is coupled with increasing levels of LPO and glucose concentration, resetting at dawn. Under darkness, liver and OL-B GSH enzymes and substrates show similar statistically significant mean vectors demonstrating a tendency around the same angle while the retina shows a dissimilar directional preference. This could suggest higher coupling strength between two possible selfsustained oscillators (midgut and OL-B) and a passive oscillator, i.e. retina; that appears to show a higher coupling strength with the zeitgeber, possibly due to a masking effect of light. The results of this work indicate the importance of the relationship between the circadian clock and the redox status resulting from the metabolism of organisms when they are exposed to potentially harmful exogenous periodic influences, such as LD cycles. Endogenous metabolic rhythms in accor-

dance with exogenous rhythms of light irradiation generate cycles in the redox state. The oxidative stress that occurs can only be antagonized by the anticipatory adaptive value of the circadian clock linked with enzymes and scavengers of antioxidant systems such as that of glutathione. The circadian clock should allow the crayfish, a nocturnal animal, by means of activity and metabolism, to synchronize their internal temporal order to 24 h LD cycles, avoiding oxidative stress in particular phases of the day–night cycle. Acknowledgements—We are grateful to Ing. Gero´nimo Bello and Dario Santiago-Lo´pez, M.Sc., for their technical support in the implementation of the HPLC technique. We are also in debt to Maggie Brunner, M.A., for the final English revision of the manuscript. This work was supported in part by CONACyT Me´xico grant 46193-Q and by PAPIIT IN 208405 and IN 207008 grants. We greatly appreciate the suggestion and commentaries of the anonymous reviewers that certainly improved this work.

REFERENCES 1. Hardeland, R., A. Coto-Montes and B. Poeggeler (2003) Circadian rhythms, oxidative stress, and antioxidative defense mechanisms. Chronobiol. Int. 20, 921–962. 2. Allen, R. G. and M. Tresini (2000) Oxidative stress and gene regulation. Free Radic. Biol. Med. 28, 463–499. 3. Rutter, J., M. Reick, L. C. Wu and S. L. McKnight (2001) Regulation of clock and NPAS2 DNA binding by the redox state of NAD cofactors. Science 293, 510–514. 4. Wijnen, H. and M. W. Young (2006) Interplay of metabolic circadian clocks and metabolic rhythms. Annu. Rev. Genet. 40, 409–448. 5. Fingerman, M. (1955) The rate of oxygen consumption of the dwarf crawfish Cambarellus sufeltidi. Tulane Stud. Zool. 3, 103–116. 6. Kallen, J. L., S. L. Abrahamse and F. Van Herp (1990) Circadian rhythmicity of the crustacean hyperglycemic hormone (CHH) in the hemolymph of the crayfish. Biol. Bull. 179, 351–357. 7. Tilden, A. R., R. Brauch, R. Ball, A. M. Janze, A. H. Ghaffari, C. T. Sweeney, J. C. Yurek and R. L. Cooper (2003) Modulatory effects of melatonin on behavior, hemolymph metabolites and neurotransmitter release in crayfish. Brain Res. 992, 252–262. 8. Page, T. L. and J. L. Larimer (1972) Entrainment of the circadian locomotor activity rhythm in crayfish. The role of the eyes and caudal photoreceptor. J. Comp. Physiol. 78, 107–120. 9. Fanjul-Moles, M. L., M. Miranda-Anaya and J. Prieto (1996) Circadian locomotor activity rhythm during ontogeny in crayfish Procambarus clarkii. Chronobiol. Int. 13, 15–26. 10. Dura´n-Liza´rraga, M. E., J. Prieto-Sagredo, M. E. Gonsebatt and M. L. Fanjul-Moles (2001) Crayfish Procambarus clarkii shows circadian variations in different parameters of the GSH cycle. Photochem. Photobiol. 74, 350–355. 11. Fanjul-Moles, M. L., M. E. Dura´n-Liza´rraga, M. E. Gonsebatt and J. Prieto-Sagredo (2003) The effect of photoperiod and light irradiance on the antioxidant circadian system of two species of crayfish from different latitudes: Procambarus clarkii and P. digueti. Photochem. Photobiol. 77, 210–218. 12. Fanjul-Moles, M. L. and J. Prieto-Sagredo (2003) The circadian system of crayfish: A developmental approach. Microsc. Res. Tech. 60, 291–301. 13. Harvey, P. R., R. G. Ilson and S. M. Strasberg (1989) The simultaneous determination of oxidized and reduced glutathione in liver tissue by ion pairing reverse phase high performance liquid chromatography with a colorimetric electrochemical detector. Clin. Chim. Acta 180, 203–212. 14. Prieto-Sagredo, J., I. Ricalde-Recchia, M. E. Dura´n-Liza´rraga, M. E. Gonsebatt and M. L. Fanjul-Moles (2000) Changes in hemolymph glutathione status after variation in photoperiod and light-irradiance in crayfish Procambarus clarkii and Procambarus digueti. Photochem. Photobiol. 71, 487–492. 15. Bompart, G., D. Prevot and J. L. Bascands (1990) Rapid automated analysis of glutathione reductase, peroxidase and

Photochemistry and Photobiology, 2009, 85

16. 17. 18. 19. 20. 21. 22. 23.

24. 25.

S-transferase activity: Application to cisplastin induced toxicity. Clin. Biochem. 23, 501–504. Paglia, D. E. and W. N. Valentine (1966) Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J. Lab. Clin. Med. 70, 158–169. Trinder, P. (1969) Determination of blood glucose using 4-amino phenazone as oxygen acceptor. J. Clin. Pathol. 22, 246. Ohishi, N., H. Ohkawa, A. Miike, T. Tatano and K. Yagi (1985) A new assay method for lipid peroxides using a methylene blue derivative. Biochem. Int. 10, 205–211. Nelson, W., Y. L. Tong, J. K. Lee and F. Halberg (1979) Methods for cosinor-rhythmometry. Chronobiologia 6, 305–323. Batschelet, E. (1981) Circular statistics in biology. In Mathematics in Biology (Edited by R. Sibson and J. Cohen), pp. 52–83. Academic Press, London, UK. Kosower, N. S. and E. M. Kosower (1978) The glutathione status of cells. Int. Rev. Cytol. 54, 109–160. Halliwell, B. (1996) Cellular stress and protection mechanisms. Biochem. Soc. Trans. 24, 1023–1027. Dentchev, T., Y. Yao, J. D. Pratico and J. I. Dunaief (2007) Isoprostane F2alpha-VI, a new marker of oxidative stress, increases following light damage to the mouse retina. Mol. Vis. 13, 190–195. Meyer-Rochow, V. B. (2001) The crustacean eye: Dark ⁄ light adaptation, polarization sensitivity, flicker fusion frequency, and photoreceptor damage. Zool. Sci. 18, 1175–1197. Kashiwagi, T., V. B. Meyer-Rochow, K. Nishimura and E. Eguchi (1997) Fatty acid composition and ultrastructure of photorecep-

26. 27.

28. 29.

30. 31.

32.

87

tive membranes in the crayfish Procambarus clarkii under conditions of thermal and photic stress. J. Comp. Physiol. B. 167, 1–8. Assefa, Z., A. Van Laethem, M. Garmyn and P. Agostinis (2005) Ultraviolet radiation-induced apoptosis in keratinocytes: On the role of cytosolic factors. Biochim. Biophys. Acta 1755, 90–106. Limo´n-Pacheco, J. H., N. H. Herna´ndez, M. L. Fanjul-Moles and M. E. Gonsebatt (2007) Glutathione depletion activates mitogenactivated protein kinase (MAPK) pathways that display organspecific responses and brain protection in mice. Free Radic. Biol. Med. 43, 1335–1347. Shaw, C. A. (1998) Multiple roles of glutathione in the nervous system. In Glutathione in the Nervous System (Edited by C. A. Shaw), pp. 3–25. Taylor and Francis Publishers, Bristol, PA. Maciel, F. E., C. E. Rosa, E. A. Santos, J. M. Monserrat and L. E. M Nery (2004) Daily variations in oxygen consumption, antioxidant defenses, and lipid peroxidation in the gills and hepatopancreas of an estuarine crab. Can. J. Zool. 82, 1871– 1877. Ferna´ndez de Miguel, F. and H. Are´chiga (1994) Circadian locomotor activity and its entrainment by food in the crayfish Procambarus clarkii. J. Exp. Biol. 190, 9–21. Are´chiga, H. and L. Rodrı´ guez-Sosa (2002) Distributed circadian rhythmicity in the crustacean nervous system. In The Crustacean Nervous System (Edited by K. Weise), pp. 113–122. Springer, Heidelberg, Germany. Balzer, I., I. R. Espı´ nola and B. Fuentes-Pardo (1997) Daily variations of immunoreactive melatonin in the visual system of crayfish. Biol. Cell 89, 539–543.