Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/unpaywall/sn-1-2-diacylglycerol-levels-in-the-fungusneurospora-crassa-display-MEcJ0sBXf4
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Unpaywall
- ISSN
- 0021-9258
- DOI
- 10.1074/jbc.m002911200
- Publisher site
- See Article on Publisher Site
Abstract
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 275, No. 36, Issue of September 8, pp. 27541–27550, 2000 © 2000 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. sn-1,2-Diacylglycerol Levels in the Fungus Neurospora crassa Display Circadian Rhythmicity* Received for publication, April 6, 2000, and in revised form, May 31, 2000 Published, JBC Papers in Press, June 19, 2000, DOI 10.1074/jbc.M002911200 Mark Ramsdale‡ and Patricia L. Lakin-Thomas§¶ From the ‡Department of Molecular and Cell Biology, Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, United Kingdom and the §Department of Plant Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EA, United Kingdom The fungus Neurospora crassa is a model organism for been described in a variety of macromolecules and enzyme activities, and in energy metabolism, ions, and small molecules investigating the biochemical mechanism of circadian (daily) rhythmicity. When a choline-requiring strain (reviewed in Ref. 5). A number of rhythmically expressed genes (chol-1) is depleted of choline, the period of the conidia- have now been identified (7), some that are dependent on the tion rhythm lengthens. We have found that the levels of output of the clock, including a glycolytic enzyme (8), and some sn-1,2-diacylglycerol (DAG) increase in proportion to that may be components of the clock itself. the increase in period. Other clock mutations that Recently, much attention has focused on a molecular model change the period do not affect the levels of DAG. Mem- for circadian rhythmicity based on an autoregulatory negative brane-permeant DAGs and inhibitors of DAG kinase feedback loop involving the products of the frq (frequency) locus were found to further lengthen the period of choline- and the wc-1/wc-2 signaling pathway (reviewed in Ref. 9). depleted cultures. The level of DAG oscillates with a Mutations at the frq locus alter period, temperature compen- period comparable to the rhythm of conidiation in wild- sation, and expression of rhythmicity (reviewed in Ref. 10). The type strains, choline-depleted cultures, and frq mutants, frq locus also affects sensitivity to light-induced phase reset- including a null frq strain. The DAG rhythm is present at ting (11). RNA and differentially phosphorylated protein prod- the growing margin and also persists in older areas that ucts of the frq locus accumulate rhythmically (12, 13), and the have completed development. The phase of the DAG FRQ protein is proposed to negatively regulate its own tran- rhythm can be set by the light-to-dark transition, but the scription (reviewed in Ref. 9), although this has not yet been level of DAG is not immediately affected by light. Our demonstrated. Null mutants at the frq locus, which produce no results indicate that rhythms in DAG levels in Neuros- functional gene product, are often arrhythmic under standard pora are driven by a light-sensitive circadian oscillator growth conditions, but rhythmic conidiation can be seen in that does not require the frq gene product. High levels of some circumstances (14 –17). Recent studies indicate that the DAG may feed back on that oscillator to lengthen its frq gene product may be a component of an input pathway period. transducing light signals to a core oscillator (16 –18). A number of other mutants exhibiting defects in their time- The ability to measure time on a daily basis is an important keeping have been described (reviewed in Refs. 5, 10, 18), aspect of the biology of most eukaryotes and probably many which include several amino acid auxotrophs, some mitochon- prokaryotes (reviewed in Refs. 1–3). Biological clocks track drial mutants, and mutations that affect glycerol metabolism. changes in the internal and external environment in such a There are also several mutations affecting circadian period for fashion that key events in an organism’s life can be appropri- which no other defect has yet been described. We have been ately orchestrated. Clocks that monitor events in a 24-h time working with two mutants defective in lipid synthesis, cel frame are termed circadian (circa, about; dies, day). True cir- (chain-elongation) and chol-1 (choline-requirer), which also af- cadian clocks exhibit a number of characteristic features, fect circadian rhythmicity (reviewed in Refs. 5, 19). namely, a period approximating to 24 h in constant conditions, The cel mutant is defective in fatty acid synthesis and re- daily resetting by environmental cues such as light and tem- quires saturated fatty acids for normal growth and rhythmic- perature cycles, and a period length that is relatively constant ity. This mutant displays a long period, poor temperature com- at different temperatures (temperature compensation). pensation, and slow growth when supplemented with The fungus Neurospora crassa has proven to be a valuable unsaturated fatty acids or short-chain fatty acids (reviewed in model for extending our understanding of the biochemical and Ref. 5). The chol-1 mutation partially blocks synthesis of the genetic basis of circadian rhythmicity (reviewed in Refs. 4 – 6). lipid phosphatidylcholine (PtdCho) (20) and is associated with Rhythmic conidiation (spore formation) is the most obvious a decrease in growth rate, a lengthening of the period of the manifestation of rhythmicity, and is assayed in cultures grow- conidiation rhythm to about 60 h, and a loss of effective tem- ing over the surface of solid agar medium (Fig. 1). In constant perature compensation on choline-depleted media (21, 22). environmental conditions, this rhythm has a free running pe- chol-1 strains remain light-sensitive and can be entrained to riod of about 21 h. In Neurospora, circadian rhythms have also The abbreviations used are: PtdCho, phosphatidylcholine; PtdEtn, * This work was supported by Grants 039696/Z/93 and 045355/Z/95 phosphatidylethanolamine; PtdOH, phosphatidic acid; CT, circadian from The Wellcome Trust (to P. L. L.-T.). The costs of publication of this time; DAG, sn-1,2-diacylglycerol; DAGKI, diacylglycerol kinase inhibi- article were defrayed in part by the payment of page charges. This tor; DDG, 1,2-didecanoyl-rac-glycerol; Me SO, dimethyl sulfoxide; article must therefore be hereby marked “advertisement” in accordance DOG, 1,2-dioctanoyl-sn-glycerol; DD, constant darkness; LD, light/dark with 18 U.S.C. Section 1734 solely to indicate this fact. cycle; LL, constant light; OAG, 1-oleoyl-2-acetyl-sn-glycerol; PKC, pro- To whom correspondence should be addressed: Tel.: 44-1223-333- tein kinase C; RDW, residual dry weight; TAG, triacylglycerol; HPLC, 954; Fax: 44-1223-333-953; E-mail: [email protected]. high pressure liquid chromatography. This paper is available on line at http://www.jbc.org 27541 This is an Open Access article under the CC BY license. 27542 DAG Rhythms in Neurospora entrained to a 12:12 h or 24:24 h light/dark (LD) cycle. Manipulations of the cultures in darkness were performed under a red light, which has been previously shown to have no measurable effect upon the clock in Neurospora. Light phases were achieved by illuminating cultures with a single cool-white fluorescent strip source suspended above the plates to give a maximum photon flux density at the culture surface of 25 mmol/m /s. Light intensity was measured with a PAR quantum sensor (Skye-SKP 200, Llandridod Wells, Powys, Wales, UK). Assay of Conidiation Rhythm Period—Periods and growth rates were determined on solid agar medium, either directly on sampled culture plates or on parallel control cultures, using daily growth front positions marked on the reverse of colonies as timing marks. Daily growth rates were then used to calculate the timing of the peaks of conidiation in FIG.1. Rhythmic conidiation in Neurospora crassa. A culture of each band, the difference in sidereal hours between band peaks giving chol was grown on solid agar medium in a 14-cm diameter Petri dish the period. Typically, three to four periods were determined per plate. at 22 °C. The dish was inoculated at the center and left in constant light Circadian time (CT) calculations were based upon a normalization of for 24 h before transfer to constant darkness for 3 more days. As the the individual periods to a standardized 24-h day with CT 12 defined as colony expands, the margin (growth front) alternates between the pro- the time of the light-to-dark transition. duction of regions that will later conidiate, and regions that are deter- DAG Analysis—For analysis of sn-1,2-diacylglycerols (DAG), sectors mined not to conidiate, forming a spatial pattern represented by bands of mycelium 1 cm wide by 0.5–1 cm long (approximately 0.1–2.0 mg of of conidiospores. Dense areas correspond to bands of conidiation, and residual dry weight, RDW) were scraped from the surface of the cello- sparse areas are non-conidiating interband regions. Maximal conidia- phane (either at the margin or within a differentiated region as re- tion (the peak of the band) occurs at about CT 0, or the subjective dawn quired), frozen in liquid N , and stored at 270 °C. Triplicate samples (the anticipated dark-to-light transition). In the wild-type strain, or the 2 chol-1 strain on high choline (200 mM), the average period (time elapsed were removed from locations on the mycelium of the same age and between successive peaks of conidiation) is about 21.5 h. phase and were spatially well separated from one another. This strat- egy was adopted to reduce the possibility that DAG levels in a partic- ular region of the colony could be affected by the localized cellular light/dark cycles near to their intrinsic period (21). Addition of damage or stress that accompanies the sampling procedure. exogenous supplies of choline restores normal clock functions Following a number of different trial protocols, lipid extractions were in a more or less dose-dependent manner. Studies of the inter- performed on the frozen samples using a method adapted from Bligh action between frq alleles and the cel or chol-1 mutants indicate and Dyer (28). Sample mycelium was extracted with 1.5 ml of ice-cold that membrane lipids may be involved with the temperature chloroform/methanol (1:2), vortexed, and kept on ice for 10 –15 min. The compensation of the circadian rhythm (17, 22, 23). samples were then mixed with 1 ml of 2.4 N KCl/1 mM EDTA (1:1) and 0.5 ml of chloroform. After 1 min of vortexing, incubation on ice for Blocks in the synthesis of PtdCho in choline-requiring mu- 10 –15 min and centrifugation at 822 3 g for 5 min, the lower phase was tants affect the levels of a number of important lipid metabo- transferred to a fresh tube containing 2 ml of ice-cold methanol/1 N KCl lites, including membrane phospholipids and neutral lipids (24, (1:1). The original tube and contents were then extracted twice more 25). The work described in this paper began as a search for the with chloroform at room temperature, and the lower phase at each step biochemical basis of the lengthened period in chol-1. An obser- was added to the methanol solution. The combined organic extracts vation that the levels of sn-1,2-diacylglycerol (DAG) increase in were centrifuged for 5 min at 822 3 g, and then the lower phase was transferred to a fresh glass tube. Following a final wash with metha- direct proportion to the increase in period led to an exploration nol/1 N KCl, the lower phase was transferred to a screw-capped glass of the relationship between DAG and the circadian clock in vial and stored under N gas at 220 °C. wild-type and mutant frq strains. We report that the level of Mycelial debris remaining in the original sample tube were washed DAG is rhythmically modulated as an output of the circadian once in 15 ml of water and twice in 15 ml of methanol. During the clock and that increased levels of DAG may feed back on the second methanol wash, the samples were sonicated for 5 min. The clock mechanism to lengthen its period. residual mycelium was then transferred, in a minimum volume of methanol, to a preweighed aluminum foil boat and dried overnight at EXPERIMENTAL PROCEDURES 110 °C. The residual dry weight (RDW) of the mycelium was deter- Strains and Culture Methods—All experiments were performed with mined on a Cahn electrobalance calibrated in the range 1 mg to 200 mg. strains with a csp-1; bd genotype as described previously (26). The csp-1 We have chosen to calculate DAG levels relative to RDW, with the (conidial separation) mutation reduces contamination by preventing knowledge that RDW may itself be varying rhythmically as the cultures the release of conidiospores, and the bd (band) mutation allows the differentiate. Our observations that DAG levels continue to display expression of conidiation at high CO concentrations in closed culture rhythmicity in old areas of the culture that have completed differenti- vessels. Construction of the csp-1; chol-1 bd strain and the csp-1; chol-1 ation, and that the DAG rhythms are not always in phase with the bd; frq series strains has previously been described (21, 22). For sim- conidiation bands, give us confidence that the rhythms in DAG we plicity, the csp-1 and bd mutations are not included in the strain names observe are not artifacts of rhythmic differentiation. in the text: the csp-1; bd strain is referred to as chol or wild-type (with Before assessing DAG levels, lipid samples were dried under a con- respect to the chol-1 locus), and the csp-1; chol-1 bd strain is referred to stant stream of N at 60 °C and then redissolved in 500 ml of chloroform. as chol-1. The csp-1; chol-1 bd; frq strains are referred to as chol-1 frq Typically a 100-ml subsample was removed for an absolute mass deter- double mutants, again ignoring the csp-1 and bd mutations carried by mination of DAG using the DAG assay reagents system (Amersham all strains. Pharmacia Biotech). This assay is based on the method of Preiss et al. Standard methods (27) were used for stock keeping and crosses. For (29) and relies on the quantitative conversion of sn-1,2-diacylglycerols 33 33 all experiments, cultures were incubated at 22 6 1 °C. All cultures were to [ P]phosphatidic acids by DAG kinase in the presence of [g- P]ATP. grown on solid agar medium (Vogel’s N, with 0.5% maltose and 0.01% Radiolabeled phosphatidic acid (PtdOH) can then be separated from arginine; 2% agar) in 9-cm diameter Petri dishes (as starter cultures) other products by thin layer chromatography (20- 3 20-cm Whatman and/or on NUNC (245 3 245 mm) BioAssay dishes (in assay cultures). Silica Gel 60 plates; 75-min separation at room temperature in chloro- Cultures requiring choline supplementation were supplied with choline form/methanol/acetic acid, 65:15:5). Consistent lipid markers were vi- chloride at concentrations varying from 0 to 200 mM choline as sualized with iodine vapor, and the location corresponding to PtdOH appropriate. was removed and quantified by liquid scintillation counting. In prelim- Under routine conditions, the starter cultures were initiated with inary experiments, the location of PtdOH on the TLC plates was conidial inoculations from frozen stocks followed by 48-h incubation in determined by cochromatography, iodine staining, and autoradiogra- constant light. Plugs of 6-mm diameter were cut from the margins of phy of samples with radiolabeled PtdOH, lyso-PtdCho, PtdCho, and actively growing colonies and were then used to inoculate the center of 1-stearoyl-2-arachidonyl-sn-glycerol standards. Levels of DAG in sam- bioassay plates overlaid with uncoated cellophane membranes (gener- ples were calculated by comparison to a series of standard assay reac- ously donated by UCB Films plc, Bridgewater, Somerset, UK). After a tions performed using known amounts of 1-stearoyl-2-arachidonyl-sn- further 24 h under white light, the cultures were either maintained in glycerol. DAG levels are expressed as picomoles of DAG per milligram constant light (LL), allowed to free-run in constant darkness (DD), or of RDW (mean 6 1 S.E., n $ 3 per sample time). In some cases, one of DAG Rhythms in Neurospora 27543 a set of replicate samples gave a DAG value several times higher or lower than that of the other replicates. This could usually be attributed to very small or very large residual mass values, due to errors in recovery or drying of extracted mycelial residue, and that value was removed from the calculations. Time Series Analysis of DAG Levels—Unless otherwise indicated, cultures were grown on NUNC BioAssay plates overlaid with cello- phane. Each time series combines data from several different colonies, each providing a short series of samples. Inoculation times and light- to-dark transitions were staggered by 11 or 12 h so that two colonies could be harvested simultaneously to provide data for different phases. Sampling times were arranged so that some overlap of time series occurred between colonies. Individual time points represent the mean 6 1 S.E. of triplicate samples from a single colony. Analysis of Phospholipids by HPLC—Cultures of chol-1 were grown in Petri dishes overlaid with sterile dialysis tubing on solid agar me- dium supplemented with [ H]palmitic acid at 5 mCi/ml and various levels of choline to manipulate the period. Samples were harvested from the combined band and interband regions after 1.5 cycles in DD. Lipids were extracted with acidic chloroform/methanol as described previously FIG.2. Lipid levels in the chol-1 strain as a function of period. (30). Lipids were separated by HPLC using a silica column (Waters Period was manipulated by altering the choline concentration in the mPorasil, 10 mm) and a gradient of hexane:propane-2-ol:H O from 6:8: growth medium, from 0 choline (longest period) to 100 mM choline 0.75 to 6:8:1.5 (v/v). Phospholipid peaks were quantitated by on-line (shortest period). Samples were harvested after 1.5 cycles in constant radiometric detection of H and were identified by cochromatography darkness, at different ages for different choline concentrations. A, indi- with known standards detected by UV absorption. vidual phospholipids, as percentage total phospholipids. Data points Effects of Light on DAG Levels—Cultures of chol-1 were grown in are the means of two independent experiments. Triangles, phosphati- mM choline in 1-ml microtiter dylethanolamine (PtdEtn). Squares, phosphatidylcholine (PtdCho); B, liquid medium with 2% glucose and 200 wells, as described previously (30). Cultures were labeled during DAG, as pmol/mg of RDW. Data points are the means 6 S.E. of three independent experiments. growth with [ H]palmitic acid at 10 mCi/ml. After an initial 24 h in constant light, cultures were transferred to constant dark and har- vested after 30 h in darkness. Mycelial disks were exposed to cool-white choline supplementation, but PtdCho levels did not correlate mmol/m fluorescent light at approximately 20 –25 /s for 0, 2, 5, and 60 s with period. Neither did levels of PtdEtn (Fig. 2A), nor did any before freezing rapidly in liquid N . Lipid extracts were performed as of the other phospholipids assayed (phosphatidylserine, phos- described above. Samples of lipid extracts were separated on HPLC phatidylinositol, and cardiolipin, data not shown). Similar re- using the method of Bocckino et al. (31) for DAG purification. The DAG peak was identified by UV absorption of a standard DAG sample. DAG sults were obtained if phospholipid levels were calculated as peaks from labeled cultures were quantitated by on-line radiometric dpm per milligram of RDW instead of percentage of total phos- detection of H. pholipids (data not shown). Effects of DAG Modulators on Period—Cultures were inoculated onto In contrast to the phospholipids, the level of the neutral lipid solid agar medium in “race tubes” made from sterile disposable plastic DAG was found to correlate with period (Fig. 2B). The data for pipettes (32). Tubes were capped with plastic culture caps at one end, DAG in Fig. 2B were obtained by assaying DAG levels in lipid and the cotton plugs were left in the pipette tops at the other end. Growth medium was supplemented with one or more of the following: extracts as described under “Experimental Procedures” and are DOG, 1,2-dioctanoyl-sn-glycerol; OAG, 1-oleoyl-2-acetyl-sn-glycerol; reported as mass of DAG per milligram of RDW. Similar re- DDG, 1,2-didecanoyl-rac-glycerol; DAGKI-I, diacylglycerol kinase in- sults were obtained by labeling cultures with [ H]palmitic acid hibitor I, R59022; DAGKI-II, diacylglycerol kinase inhibitor II, R59949. and separating neutral lipids by HPLC. Triacylglycerol (TAG) All chemicals were obtained from Sigma. DAG analogs were diluted in levels are also known to increase with choline deprivation in dimethyl sulfoxide (Me SO) and added to autoclaved growth medium Neurospora (24), although the levels of TAG were not assayed immediately before filling race tubes, to give a final Me SO concentra- tion of 0.1% (v/v). DAGKIs were dissolved in ethanol and added to in these experiments. growth medium to give a final ethanol concentration of 0.4% (v/v). The samples assayed in Fig. 2 were taken from large areas of Control tubes contained the solvents (Me SO and/or ethanol) at the mycelium, encompassing both older and newer regions of the same concentrations. Cultures were incubated at 22 °C in constant light fungal colony and both band and interband regions. To deter- for 24 h before transfer to constant darkness. Periods were calculated as mine whether DAG levels differ between different regions of a described previously (26). colony, samples were harvested from defined regions. In Fig. 3, RESULTS DAG levels are reported for six regions: three bands and three Effects of Choline Deprivation on DAG Levels—Choline dep- interbands, of increasing age. (Band 1 is the youngest region of rivation of the choline-requiring chol-1 mutant of Neurospora the colony, and Interband 3 is the oldest.) In all regions, levels lengthens the period of the conidiation rhythm and alters the of DAG were higher on the lower level of choline supplemen- lipid composition. In an attempt to find the cause of the length- tation, and there were no significant trends in DAG level with ened period, we looked for correlations between the period of age or band/interband identity. the conidiation rhythm and the levels of lipids in a chol-1 strain Time Series Analysis of DAG Levels—DAG levels were ob- grown on various levels of choline. Phospholipid levels were served to fluctuate when samples were harvested at various assayed by labeling cultures to equilibrium during growth with times after the transition to DD. In Figs. 2 and 3, samples were [ harvested at a single time after cultures were transferred into H]palmitic acid and separating a lipid extract on HPLC. Fig. 2A shows results for the two most abundant phospholipids constant darkness (Fig. 3) or at a single phase of the circadian in Neurospora, phosphatidylethanolamine (PtdEtn) and Ptd- rhythm for cultures with different periods (Fig. 2). In Fig. 4, Cho, reported as % total phospholipids. Note that the two data samples were taken at the growing margin of the colonies, points at approximately 20.5 h represent two different choline following the margin as the colony diameter expanded. DAG concentrations, 20 and 100 mM choline: Although the growth levels were found to fluctuate in the wild-type chol strain on rate is slower at 20 mM, there is no change in the period as both minimal medium (0 choline, Fig. 4A) and 200 mM choline compared with 100 mM, demonstrating that period is independ- (Fig. 4B), and in the chol-1 strain growing at wild-type rates on ent of growth rate. As expected, PtdCho levels determined by 200 mM choline (Fig. 4C). HPLC were lower in cultures with below optimal levels of Changes in the level of DAG are not a product of active 27544 DAG Rhythms in Neurospora FIG.3. Spatial analysis of DAG levels. The chol-1 strain was grown on 0 or 200 mM choline. Samples were taken at one time point 170 h after the light-to-dark transition, after several bands had formed. Three band and interband regions from each set were sampled. Region 1 is the outer, most recently formed, region, and 3 is the inner, oldest region. Data points are means 6 S.E. of triplicate samples from one culture. differentiation per se, because fluctuations in DAG levels were also detected in older areas of the colony after the growth front had moved on. Fig. 5 presents the results of analyses of DAG in the chol-1 strain on 200 mM choline, comparing the growing margin (Fig. 5A) with an aging band (Fig. 5B) and an aging interband region (Fig. 5C). These results indicate that the DAG rhythm is not confined to actively growing areas and is there- fore not caused by developmental changes occurring in the harvested regions. The time between later peaks (after 40 h in DD) is approximately 20 h, similar to the 21-h free running period of the conidiation rhythm, indicating that this rhythm in DAG may be related to the circadian rhythm of conidiation. DAG levels are elevated in a chol-1 strain displaying a long period. In Fig. 6, DAG levels were assayed at the margin of the chol-1 strain growing on minimal medium (zero choline). Under FIG.4. Time series analysis of DAG levels at the margin of these conditions, the colonies show a conidiation period of growing colonies in constant darkness. Cultures were sampled at about 60 h, but replicate colonies are often not in synchrony various times after the transition from constant light to constant dark with one another (21). The colonies harvested for Fig. 6 were (DD), harvesting samples every 2 h from the growing margin. DAG levels are plotted against time of harvest, in hours after the transition subjected to 2.5 cycles (5 days) of a 48-h light/dark cycle, 24 h to DD. Data points are means 6 S.E. of replicate samples from one of light and 24 h of dark, before transfer to constant darkness colony, with N usually equal to three. The solid line is a three-point and harvesting. We have found that this procedure improves running average of the data points. Arrows mark the peaks of the DAG synchrony of colonies for the first few days in DD. Control rhythm, with positions estimated by visual observation. Stars at the plates that were not harvested typically produced an interband baseline indicate times at which band peaks were being formed at the margin. A, wild-type (chol ) on zero choline. The periods between DAG region shortly after the light-to-dark transition in this experi- peaks (to the nearest hour) were 15, 12, and 20 h. The mean period ment. The data in Fig. 6 show a single major peak of DAG after between band peaks was 21.5 h, with band peaks occurring at CT 1.0; the transition to DD; however, because samples were collected B, wild-type on 200 mM choline. The periods between DAG peaks (to the for little more than one cycle, we cannot say whether this major nearest hour) were 12, 13, 13, and 15 h. The mean period between band peaks was 21.2 h, with band peaks occurring at CT 2.0; C, chol-1 on peak would be repeated in a second cycle. The smoothed curve 200 mM choline. The periods between DAG peaks (to the nearest hour) from Fig. 5A (200 mM choline) has been replotted in Fig. 6 for were 22, 17, 13, and 18 h. The mean period between band peaks was comparison. Note that the level of DAG is higher at all time 21.7 h, with band peaks occurring at CT 1.6. points in the cultures on minimal medium as compared with 200 mm choline. locus, frq , a deletion mutant that makes no functional frq Effects of frq Mutations on DAG Levels—The data presented gene product and is arrhythmic under some conditions. Sam- above demonstrate that the levels of DAG are higher under ples were harvested at two different times in DD. The frq conditions that lengthen the conidiation period in the chol-1 mutations were assayed both in chol strains on zero choline strain. If the high level of DAG is a consequence of the length- and in chol-1 strains on 200 mM choline. ened period, then we would predict that other mutations that The correlation between DAG level and the period of the alter the period should also alter DAG levels. To test this conidiation rhythm apparent in the chol-1 mutant is not ap- prediction, we have assayed the levels of DAG in the series of parent in frq mutants with different periods. As shown in frq mutants. Mutations at the frq locus of Neurospora change Fig. 7, there were no significant trends in the average DAG the period of the circadian conidiation rhythm but have no levels with respect to the periods of the frq mutants. These effects on growth rate or viability (4). We have assayed DAG results indicate that the correlation between DAG levels and 1 2 levels in two short period frq mutants (frq and frq ), two long period (Fig. 2) is not a nonspecific consequence of changing 3 7 period frq mutants (frq and frq ) and a null mutant at the frq period but is specific for choline depletion. The two phases DAG Rhythms in Neurospora 27545 FIG.6. Time series analysis of DAG levels in chol-1 on minimal medium. Cultures of chol-1 on zero choline were grown on 15-cm diameter Petri dishes overlaid with cellophane and inoculated at the edge. Cultures were entrained to 2.5 cycles of a 24 h:24 h light/dark cycle before the final light-to-dark transition at 120 h after inoculation. The dashed bar between 25 and 68 h in DD indicates the time during which a band was being formed at the margin. The free running period of control cultures in DD was 63.8 h. Data were combined from a total of 60 plates, each sampled in triplicate at one time point. Data analysis was the same as for Fig. 4, except samples were harvested every 3 h. The thin line is the running average from Fig. 5A, chol-1 on 200 mM choline, replotted for comparison. Stars at the baseline indicate the times at which band peaks were being formed at the margin of the culture on 200 mM choline, as in Fig. 5A. Note the change in the y axis scale as compared with Figs. 4 and 5. FIG.5. Time series analysis of DAG levels in various regions of growing colonies. Cultures of chol-1 on 200 mM choline were har- vested at various times in DD. Data analysis as for Fig. 4. A, samples were harvested from the advancing margin. The periods between DAG peaks (to the nearest hour) were 14, 14, and 20 h. The mean period between band peaks was 21.5 h, with band peaks occurring at CT 0; B, initial samples were harvested at the margin of a colony during the formation of a band, and subsequent samples were harvested from within the same band region. The periods between DAG peaks (to the nearest hour) were 18 and 24 h. The mean period between band peaks was 21.1 h, with band peaks occurring at CT 2.4; C, same as for B, except samples were harvested from an interband region. The periods between DAG peaks (to the nearest hour) were 13, 15, 26, and 16 h. The mean period between band peaks was 20.5 h, with band peaks occurring at CT 2.0. assayed in Fig. 7 gave different DAG levels in most strains, including the frq strains, indicating possible rhythmicity. Levels of DAG were not strictly correlated with the band or interband phase of the margin at the time of harvesting. Av- erage levels of DAG were lower for chol-1 strains on 200 mM choline (Fig. 7B) as compared with chol strains on zero choline FIG.7. DAG levels in frq mutants. Cultures were grown in con- (Fig. 7A), as was also seen in Fig. 4 (A and C). stant light for 60 h before transfer to constant darkness. Samples were harvested from the margins at two different times in DD. The band/ The period of the DAG rhythm is correlated with the period interband phase of the margin at the time of sampling is indicated by of the conidiation rhythm in frq mutants. DAG levels were the patterns in the legend. Times of sampling in hours after the light- 7 10 assayed in detailed time courses for both the frq and frq 1 1 to-dark transition (and periods for chol and chol-1 strains) were: frq , mutants. The frq time course (Fig. 8A) displayed a clear bi- 5.3 and 10.0 h (17.1 and 17.3 h); frq , 6.3 and 11.9 h (18.2 and 18.9 h); 1 3 modal rhythm in the first half of the experiment, with three frq , 7.0 and 13.1 h (20.9 and 18.4 h); frq , 8.0 and 15.0 h (20.4 and 7 10 25.5 h); frq , 9.7 and 18.1 h (29.8 and 30.9 h); frq , 7.0 and 13.1 h (11.2 peaks approximately 13 h apart. The third and fourth peaks and 14.1 h). DAG values are the means 6 S.E. for triplicate samples. A, were approximately 28 h apart, similar to the free running chol frq strains on minimal medium (zero choline); B, chol-1 frq double conidiation period of about 27 h. A short period rhythm was mM choline. *, significant difference between the mutant strains on 200 seen in the frq strain (Fig. 8B), with four peaks about 12 h DAG levels at the two sample times, p , 0.05. 27546 DAG Rhythms in Neurospora FIG.9. Phase relationships among DAG rhythms. The smoothed curves representing the running averages have been replotted from Fig. 5. Thin line, interband; thick line, band; dashed line, margin. Matching arrows mark the peaks of the DAG rhythm, estimated by visual obser- vation. A, data were plotted against hours in constant darkness after the light-to-dark transition; B, data were plotted against hours after inoculation of the colonies. FIG.8. Time series analysis of DAG levels in frq mutants. chol-1 by the transition from constant light to constant darkness, as is strains were grown on 200 mM choline. Samples were harvested from the phase of the conidiation rhythm. We have tested this hy- the growing margins at various times after the transition to constant pothesis by examining the phase relationships among cultures darkness. Arrows mark the peaks of the DAG rhythm, estimated by that were transferred to DD at different times after inocula- visual observation. Stars at the baseline indicate times at which band peaks were being formed at the margin. A, chol-1 frq . Data analysis tion. The three experiments presented in Fig. 5 meet this was the same as for Fig. 4. The periods between DAG peaks (to the criterion, and the smoothed curves have been replotted in nearest hour) were 13, 13, and 28 h. The mean period between band Fig. 9 to compare the timing of the peaks. In Fig. 9A, the curves peaks was 27.1 h, with band peaks occurring at CT 4.3; B, chol-1 frq . are compared by plotting the data against time in DD; in Fig. Data analysis was the same as for Fig. 4, except samples were har- vested every 3 h. The periods between DAG peaks (to the nearest hour) 9B, the data are plotted against time after inoculation. It can be were 12, 13, and 10 h. The mean period between band peaks was 14.4 h, seen in Fig. 9A that the peaks are more consistent with each with band peaks occurring at CT 7.6. A total of 12 bands were observed other when plotted against time in DD, indicating that the on two control plates. The line joins the data points and is not a running light-to-dark transition sets the phase of the DAG rhythm, as average. Because of the short period in this strain and the longer expected of a circadian rhythm. These results also indicate that sampling interval, a running average would obscure the rhythm. the DAG peaks are occurring synchronously in all regions of apart. This finding confirms the significant difference between the colony, including older areas that have completed DAG levels at two phases seen in frq in Fig. 7. A short period development. or bimodal DAG rhythm during the first 40 h in DD is also The rhythm of DAG in cultures entrained to a light/dark suggested in Figs. 4A,4B,5A, and 5C. Under similar growth cycle is presented in Fig. 10. Cultures were grown in a 12 h:12 9 10 conditions, null frq strains, both frq (14) and frq (15), do not h light/dark cycle for 2 days to synchronize the conidiation produce clear bands consistently, but short period conidiation rhythm with the light/dark cycle before samples were har- rhythms with periods of about 12 h can sometimes be seen vested on the third day. The peak values of DAG were approx- during the first few days of growth. Merrow et al. (16), using a imately the same as those seen in similar cultures in DD temperature entrainment protocol, also observed rhythmic (Fig. 5A). The DAG peak occurred near the end of the formation conidiation in frq and extrapolated a free running period of of the conidiation band, in the first half of the light phase. No about 12 h for this rhythm. Conidiation rhythms early during significant changes in DAG levels were seen at this time scale growth were also found in this experiment with the frq strain, in response to the dark-to-light transition. and the positions of the first five band peaks are indicated in DAG levels were also assayed in cultures grown in constant Fig. 8B. The mean period of the conidiation rhythm in this light (Fig. 11). Cultures of chol-1 were grown on various levels experiment was 14.4 h. The null phenotype of our frq strain of choline supplementation and were harvested at either a was confirmed by Northern analysis of total RNA, and no frq fixed time after inoculation or at a fixed colony size to control mRNA could be detected (data not shown). The rhythm in DAG for the possible effects of age on DAG levels. On minimal we have observed in the frq strain must, therefore, be inde- medium without choline, both LL and DD cultures were in the pendent of any frq-based oscillator. Taken together, these re- same range, but at other choline concentrations levels of DAG sults indicate that the product of the frq gene can affect the in constant light were lower than in constant darkness period of the DAG rhythm but is not required for DAG rhyth- (Fig. 11). This is consistent with the observation that DAG micity per se and has no effect on the overall level of DAG or the levels are generally low at the end of a prolonged light expo- amplitude of its rhythm. sure, as seen at the beginning of DD in Figs. 4, 8, and 10. Effects of Light on DAG Levels—Circadian clocks are reset by Because DAG is well known as a signaling molecule that light, and light effects on the conidiation rhythm are well responds rapidly in animal cells, its response to light in Neu- documented in Neurospora (reviewed in Ref. 5). If the rhythm rospora was examined in more detail. Cultures were labeled to in DAG levels is associated with, or driven by, a circadian equilibrium during growth with [ H]palmitic acid, which oscillator, then the phase of the rhythm should be determined should label all lipid fractions. Samples of labeled mycelium DAG Rhythms in Neurospora 27547 FIG. 10. Effect of a light/dark cycle on DAG levels. A culture of chol-1 on 200 mM choline was grown in constant light for 24 h, then transferred to the first dark period of a 12 h:12 h dark/light cycle for 2 more days. Samples were harvested from the growing margin every 2 h over 24 h of the third day in dark and then light, beginning with the third dark period. Data points are the means 6 S.E. of triplicate samples. The bars at the top indicate the times of darkness and light. The dashed bar at the bottom indicates the time during which the band of conidiation was being formed at the margin. The peak of conidiation is indicated by a star. FIG. 12. Effects of DAG modulators on period. The chol-1 strain was grown in “race tubes” on media containing membrane-permeant DAG analogs (DOG, OAG, and DDG) and/or inhibitors of DAG kinase (DAGKI). A, DAG analogs were added as Me SO solutions and control cultures (black bar) contained Me SO alone. *, significantly greater than control (black bar), p , 0.01; B, DAGKIs (20 mM) were added as ethanol solutions. Control cultures (black bar) contained both ethanol and Me SO. White bars represent cultures with added DAGKI (in ethanol) plus Me SO as control for DOG additions. *, significantly greater than control (black bar), p , 0.01. **, significantly greater than FIG. 11. Effects of constant light on DAG levels. The chol-1 strain control (black bar) and DAGKI alone (white bar), p , 0.01. was grown on various levels of choline in either DD or LL and samples were harvested from the margin. Open symbols, constant light: Data used membrane-permeant DAGs and inhibitors of DAG kinase points are the mean of duplicate samples from one experiment. Open to test this hypothesis. squares, samples were harvested when colonies reached 40-mm radius. Open circles, samples were harvested at 76 h after inoculation. Closed Cultures of chol-1 were grown on media without choline but symbols, constant dark: All samples were harvested after 1.5 conidia- containing various concentrations of artificial DAGs or DAG tion cycles in DD. Data were replotted from Fig. 2B, separating out one kinase inhibitors (Fig. 12). The membrane-permeant DAG an- experiment in which age was controlled. Closed squares, samples were alogs DOG and OAG significantly lengthened the period harvested at different ages. Data points are the mean of two separate (Fig. 12A) and inhibited the growth rate (data not shown), with experiments. Closed circles, samples were harvested at 100 h after inoculation. Data points are from a single experiment. the maximum effect of DOG at 250 mM. DDG had no effect on period or growth rate up to 1000 mM (Fig. 12A). Lower concen- were exposed to 0, 2, 5, or 60 s of light before freezing and trations of OAG and DDG also had no effect (data not shown). extracting lipids. DAG was separated by HPLC as described Inhibitors of diacylglycerol kinase are expected to increase under “Experimental Procedures,” and the relative amount of the endogenous levels of DAG by inhibiting its conversion to DAG was quantitated per milligram of RDW of mycelium ex- phosphatidic acid (33). Two such inhibitors were tested on the tracted. No significant change in the level of DAG was seen at chol-1 strain on zero choline medium, and both significantly any time point (data not shown). This result is consistent with lengthened the period (Fig. 12B). DAGKI-II was more effective the results in Fig. 10, which show no large change in DAG than DAGKI-I at the same concentration, as might be expected levels at the dark-to-light transition. from their respective effects on DAG kinase activity in isolated Effects of DAG Modulators on Period—The rhythmicity of platelet membranes (34). When DOG was added in combina- DAG, and the correlation between DAG levels and period, raise tion with DAGKIs, the effect on period was greater than either the possibility that DAG plays a direct role in the regulation of additive alone. This effect was statistically significant for circadian rhythmicity in Neurospora. The increase in DAG in DAGKI-I but not for DAGKI-II, possibly reflecting the relative chol-1 strains depleted of choline may be a cause, a direct effect, efficiencies of the two inhibitors at producing elevated levels of or merely a corollary effect of the change in period under these DAG. In no case was the period lengthened beyond 130 h. In conditions. If increased DAG levels are a cause of the increased the case of 250 mM DOG alone and DAGKI-II plus 100 mM DOG, period in choline-depleted chol-1, then artificially elevating further addition of DOG shortened the period but had no sig- DAG levels by other means should increase the period. We have nificant effect on growth rate, suggesting a ceiling for period 27548 DAG Rhythms in Neurospora effects. None of the DAG modulators tested had any significant conidiation rhythm (Fig. 8A), the later DAG peaks also display effect on the period of the chol strain (data not shown), al- a long period. In the chol-1 strain with a 64-h conidiation period though the growth rate was inhibited to approximately the (Fig. 6), a single major DAG peak was seen in one conidiation same degree as in chol-1. cycle. However, it must be stressed that the DAG rhythm is not Addition of DAG modulators reduced the growth rate under strictly linked to rhythmic conidiation. These two rhythms conditions where the period lengthened (data not shown). How- appear to be only loosely coupled, and may sometimes display ever, the lengthened period was not simply a consequence of different periods. This is indicated by the bimodal DAG peaks growth inhibition. This was shown by adding the detergent sometimes seen during the first 40 h of growth (Figs. 4A,4B, SDS to the medium to inhibit the growth rate by nonspecific 5A,5C, and 8A). In addition, band peaks are generally associ- mechanisms. The chol-1 strain was grown on zero choline with ated with DAG troughs, but this is not always the case, as seen the addition of 0.2% (w/v) SDS plus 150 mM KCl. The growth by the positions of the band peaks indicated in the time series rate was reduced on this medium, but the period shortened to figures (Figs. 4, 5, and 8). The lack of correlation between DAG about 24 h rather than lengthened (data not shown). It was levels and band/interband phases seen in the frq mutants previously shown (21) that this SDS growth medium reduces (Fig. 7) is also consistent with the conclusion that these two the growth rate but has no effect on the period of the chol rhythms are not tightly coupled. strain. During the formation of the pattern of bands of conidiation in Neurospora growing on solid medium, the regions that are DISCUSSION determined to conidiate (that is, to form bands) will subse- The experiments reported in this paper were initiated as an quently develop aerial hyphae and differentiate into conidios- attempt to find the biochemical basis for the lengthened period pores. The DAG rhythm could be a by-product of this process of of the conidiation rhythm in the chol-1 mutant of Neurospora. rhythmic differentiation. However, we found that the DAG We found a correlation between the level of the neutral lipid rhythm is clearly seen in an aging band region that has com- DAG and the period of the rhythm in cultures in which the pleted its program of differentiation (Fig. 5B) and is also found period had been manipulated by changing the choline concen- in an interband region that does not undergo differentiation tration (Fig. 2B). However, when the period was manipulated (Fig. 5C). This indicates that DAG is oscillating in all parts of by changing the allele at the frq locus, we found no correlation the culture simultaneously, and not just at the growing margin between DAG levels and period (Fig. 7). Changes in DAG level or in regions undergoing differentiation. In this way the DAG therefore are specifically associated with the choline defect and rhythm parallels the circadian oscillator, which is known to not with period changes in general, suggesting that DAG levels continue oscillating in old areas of the mycelium (35). We do not may be related to the cause of the lengthened period in chol-1. know whether the DAG rhythm persists in cultures that are Does DAG play a directly causal role in lengthening the arrhythmic for conidiation. period, or is it merely a by-product of another process that is Mutations at the frq locus can affect the DAG rhythm, as responsible for period changes? We addressed this question by seen by the long period in the frq mutant (Fig. 8A), although manipulating DAG levels by means other than choline deple- a functional frq gene is not required for rhythmicity of DAG per tion to see if period changes could be induced. In the chol-1 se (Fig. 8B). Similar results are known for the effects of frq on strain growing under conditions that lengthened the period, rhythmic conidiation: frq affects the period, persistence, and addition of membrane-permeant DAGs or inhibitors of DAG temperature compensation of the conidiation rhythm (4), but kinase could further lengthen the period (Fig. 12). We do not rhythmic conidiation can still be found in null frq mutants have direct evidence that the inhibitors of DAG kinase can under some conditions, as described here and elsewhere elevate DAG levels in Neurospora, and indeed DAG kinase (14 –17). activity has not yet been characterized in this organism, but The persistence of the DAG rhythm in frq and the effects of the results with the inhibitors are consistent both with their frq on the period of the DAG rhythm might indicate the exist- predicted activities and with the effects of membrane-permeant ence of two mutually coupled oscillators. One oscillator would DAGs. directly involve the frq gene product, and its output would If elevated DAG can further lengthen the period in long influence the oscillator driving the DAG rhythm by altering its period cultures of chol-1, we would expect to see similar effects period and/or entraining it to match the period determined by in the chol strain, but we did not, even though growth rate the frq allele under normal conditions. The second oscillator was inhibited. The chol-1 strain may be susceptible to the would be sensitive to DAG and might run freely with a short effects of added DAG due to saturation by its high endogenous 12-h period in the absence of influence from the frq-dependent levels of the normal pathways for metabolizing DAG, whereas oscillator. Output from this second oscillator would be capable the chol strain may be able to metabolize the additional DAG of influencing the frq-dependent oscillator. Under conditions of and maintain lower levels. When the chol strain was grown on choline depletion and high DAG levels, the period of the DAG- medium containing 20 mM DAGKI-II, the period was length- sensitive second oscillator would lengthen beyond the range in ened by about 5 h during the first 2 days, but the effect did not which the frq-dependent oscillator could entrain it, and the persist throughout the remaining 4 days of growth (data not influence of the DAG-sensitive oscillator would drive the frq- shown). This may indicate a transient elevation of DAG and dependent oscillator with a long period. This model predicts subsequent adaptation. that the rhythmic expression of the frq gene would display a These results may point toward a causal role of elevated long period in the choline-depleted cultures and that manipu- DAG in lengthening the period in chol-1. Our second surprising finding is that DAG levels are rhythmic in Neurospora,as lating DAG levels would affect the expression of frq. Rhythmic DAG levels have not been reported in other organ- shown by Figs. 4, 5, 7, 8, and 10. This rhythmicity is associated with the circadian rhythm of conidiation, as shown by the isms. Circadian rhythms of DAG involved with the regulation of rhabdom shedding in the crustaceans Limulus polyphemus approximate 20-h period between DAG peaks (particularly af- ter 40 h in DD). The phase of the rhythm is set by the light- (36) and Leptograpsus variegatus (37) have been inferred from to-dark transition and not by the age of the culture (Fig. 9), as pharmacological studies. Work using the diacylglycerol lipase expected of a rhythm driven by the circadian oscillator that inhibitor U-57908 indicated that DAG released from a photo- drives conidiation. In the frq strain with a long period for the transduction cascade was regulated by a DAG lipase subject to DAG Rhythms in Neurospora 27549 circadian fluctuations. Unfortunately, no attempt was made to phosphorylation activity. They suggest that PKC may become verify that DAG levels were cycling under the conditions of activated around the time that light-induced gene expression is these experiments. maximal and that phosphorylation of WC-1 may be responsible We do not know what the origin and function of rhythmic for the adaptation response. Our results indicate that this light-induced activation of PKC is unlikely to be caused by a DAG is in Neurospora, but DAG is a known intermediate in fungal phospholipid and neutral lipid metabolism (reviewed in rapid increase in DAG: We have not seen any rapid changes in DAG in response to light (within 1 min), nor have we seen Ref. 38). Triacylglycerol (TAG) levels have previously been medium-term changes in DAG in response to the dark-to-light shown to be elevated in choline-depleted strains (24), and the transition in a light/dark cycle (Fig. 10). However, the effects of elevated DAG in chol-1 could be an intermediate on the TAG light on DAG may be phase-dependent and we might have synthetic pathway. Alternatively, the DAG we have assayed assayed the effects at the wrong phase of the DAG rhythm. If could be an intermediate in phospholipid synthesis, or could be DAG-activated PKC is involved in modulating the response to derived from phospholipase action on phospholipids. We do not blue light, as proposed by Arpaia et al. (52), then the rhythm in have information on the fatty acid composition of the DAG pool DAG we have described might cause a rhythm in light re- in Neurospora, which could help resolve these issues. Roeder et sponse. Such a rhythm in either light induction or the adapta- al. (39) reported a circadian rhythm in fatty acid composition in tion response has not yet been reported. both total lipid and phospholipid fractions of Neurospora. Another potential role for DAG as a rhythmic signaling mole- Therefore, it is possible that the DAG we have assayed may be cule in Neurospora has recently been described as the result of a rhythmic in fatty acid composition as well as in mass. The time-lapse analysis of rhythmic conidiation. Cultures undergo- precise molecular species of DAG important to circadian rhyth- ing rhythmic conidiation not only restrict their conidiation to a micity are therefore not known. spatial region (the band), but also restrict conidiation temporally DAG is well known as a signaling molecule in animal cells by controlling the timing of differentiation. Large regions of the (40), but a role in signaling pathways has not been demon- designated band differentiate simultaneously, as if responding to strated in Neurospora. Bocckino et al. (31) assayed DAG mass a signal that is synchronous throughout a large region of the in hepatocytes stimulated by various agents and found a range mycelium. This signal is associated with the circadian oscillator of approximately 100 –500 ng of DAG/mg of wet weight. Assum- that lays down the spatial pattern of bands, but like the DAG ing a ratio of 10:1 for wet weight:RDW, and an average molec- rhythm it may be partially independent of that oscillator. We ular weight of hepatocyte DAG of 609 (31), this is equivalent to have shown that DAG is synchronously rhythmic throughout the a range of approximately 1600 – 8200 pmol of DAG/mg of RDW mycelium in both old and new areas, and this makes it an ideal in hepatocytes. Cook et al. (41) assayed DAG mass in bombesin- candidate for the signal that times the differentiation program. stimulated Swiss 3T3 cells and found a range of approximately The effects of elevated DAG on the period of the banding rhythm 0.35– 0.85 nmol of DAG/100 nmol of phospholipid. Assuming may be an example of feedback from one oscillator onto another phospholipids are approximately 3% of mammalian cells by to which it is mutually coupled. weight, and a typical phospholipid has a molecular weight of about 788, this is equivalent to a range of approximately 1400 – Acknowledgments—Most of the work was carried out by M.R. in the Department of Zoology, University of Cambridge. The data for Fig. 12 3300 pmol of DAG/mg of RDW in Swiss 3T3 cells. Our values were collected by P.L.L.-T. in the Department of Zoology with assist- for DAG levels in Neurospora fall in these ranges, consistent ance from Ingrid Wesley, Joelie Foster, and Mariam Orme. The data for with a signaling role for the DAG rhythm. Fig. 2 and the rapid effects of light on DAG levels were collected by If DAG acts as a signal in Neurospora, its target(s) is(are) P.L.L.-T. in the Department of Plant Sciences, University of Cam- bridge. P. L.-L.-T. gratefully acknowledges the hospitality and support unknown but could include protein kinase C (PKC). In animal of Dr. David Hanke, Department of Plant Sciences, University of cells, DAG can affect various cellular functions through effects Cambridge. on small G-proteins (42) and by direct effects on ion channels (43), or through activation of PKC (40). The three DAG analogs REFERENCES tested for effects on the period in Fig. 12 (DOG, DDG, and 1. Edmunds, L. E. (1988) Cellular and Molecular Bases of Biological Clocks, Springer-Verlag, New York OAG) have been shown to activate PKC in human platelets 2. Pittendrigh, C. S. (1993) Annu. Rev. Physiol. 55, 17–54 (44). In the yeasts Saccharomyces and Candida, PKC has been 3. Johnson, C. H., and Golden, S. S. (1999) Annu. Rev. Microbiol. 53, 389 – 409 4. Feldman, J. F., and Dunlap, J. C. (1983) Photochem. Photobiol. Rev. 7, shown to play a role in morphogenesis and cell cycle progres- 319 –368 sion (45, 46), and PKC genes have been identified in Saccha- 5. Lakin-Thomas, P. L., Cote ´ , G. G., and Brody, S. (1990) Crit. Rev. Microbiol. 17, 365– 416 romyces, Candida, and Schizosaccharomyces (45, 47– 49). PKC 6. Nakashima, H., and Onai, K. (1996) Cell Develop. Biol. 7, 765–774 activity and PKC-like genes have been characterized in several 7. Bell-Pedersen, D., Shinohara, M. L., Loros, J. J., and Dunlap, J. C. (1996) Proc. filamentous fungi (50). Recently, a putative PKC gene has been Natl. Acad. Sci. U. S. A. 93, 13096 –13101 8. Shinohara, M. L., Loros, J. J., and Dunlap, J. C. (1998) J. Biol. Chem. 273, identified in Neurospora by sequence homology, and the se- 446 – 452 quence appears to belong to the class of calcium-independent 9. Dunlap, J. C. (1999) Cell 96, 271–290 10. Dunlap, J. C. (1996) Annu. Rev. Genet. 30, 579 – 601 DAG-activated PKCs (51). 11. Lakin-Thomas, P. L., Brody, S., and Cote ´ , G. G. (1991) J. Biol. Rhythms 6, Arpaia et al. (52) have recently reported pharmacological 281–297 evidence for a blue-light activated PKC signaling pathway in 12. Aronson, B. D., Johnson, K. A., Loros, J. J., and Dunlap, J. C. (1994) Science 263, 1578 –1584 Neurospora. Blue light transiently induces the transcription of 13. Garceau, N. Y., Liu, Y., Loros, J. J., and Dunlap, J. C. (1997) Cell 89, 469 – 476 several genes, including al-3 (albino), after which the levels of 14. Loros, J. J., and Feldman, J. F. (1986) J. Biol. Rhythms 1, 187–198 15. Aronson, B. D., Johnson, K. A., and Dunlap, J. C. (1994) Proc. Natl. Acad. Sci. message decline due to adaptation of the light signaling path- U. S. A. 91, 7683–7687 way. Blue light responses in Neurospora require the WC-1 16. Merrow, M., Brunner, M., and Roenneberg, T. (1999) Nature 399, 584 –586 (white-collar) and WC-2 proteins. Arpaia et al. (52) have found 17. Lakin-Thomas, P. L., and Brody, S. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 256 –261 that inhibitors of PKC block the decline in al-3 mRNA levels, 18. Lakin-Thomas, P. L. (2000) Trends Genet. 16, 135–142 whereas a membrane-permeant DAG accelerates the decline, 19. Cote ´ , G. G., Lakin-Thomas, P. L., and Brody, S. (1996) in Membranes and Circadian Rhythms (Th. Vanden Dreissche, ed) pp. 13– 46, Springer-Verlag, suggesting that DAG-sensitive PKC activity down-regulates Berlin the blue light signaling pathway. These authors also demon- strated a PKC-like phosphorylation activity in Neurospora ex- tracts and found that the WC-1 protein is a substrate for this P. L. Lakin-Thomas and V. D. Gooch, manuscript in preparation. 27550 DAG Rhythms in Neurospora 20. Scarborough, G. A., and Nyc, J. F. (1967) J. Biol. Chem. 242, 238 –242 39. Roeder, P. E., Sargent, M. L., and Brody, S. (1982) Biochemistry 21, 21. Lakin-Thomas, P. L. (1996) Biol. Rhythm Res. 27, 12–30 4909 – 4916 22. Lakin-Thomas, P. L. (1998) J. Biol. Rhythms 13, 268 –277 40. Hodgkin, M. N., Pettitt, T. P., Martin, A., Michell, R. H., Pemberton, A. J., and 23. Lakin-Thomas, P. L., and Brody, S. (1985) Genetics 109, 49–66 Wakelam, M. J. (1998) Trends Biochem. Sci. 23, 200 –204 24. Juretic, D. (1977) Biochim. Biophys. Acta 469, 137–150 41. Cook, S. J., Palmer, S., Plevin, R., and Wakelam, M. J. (1990) Biochem. J. 265, 25. Hubbard, S. C., and Brody, S. (1975) J. Biol. Chem. 250, 7173–7181 617– 620 26. Lakin-Thomas, P. L. (1992) J. Biol. Rhythms 7, 227–239 42. Ebinu, J. O., Bottorff, D. A., Chan, E. Y., Stang, S. L., Dunn, R. J., and Stone, 27. Davis, R. H., and deSerres, F. J. (1970) Methods Enzymol. 17A, 79 –143 J. C. (1998) Science 280, 1082–1086 28. Bligh, E. G., and Dyer, W. J. (1959) Can. J. Biochem. Physiol. 37, 911–917 43. Hofmann, T., Obukhov, A. G., Schaefer, M., Harteneck, C., Gudermann, T., 29. Preiss, J. E., Loomis, C. R., Bell, R. M., and Niedel, J. E. (1987) Methods and Schultz, G. (1999) Nature 397, 259 –263 Enzymol. 141, 294 –300 44. Lapetina, E. G., Reep, B., Ganong, B. R., and Bell, R. M. (1985) J. Biol. Chem. 30. Lakin-Thomas, P. L. (1993) Biochem. J. 292, 805– 811 260, 1358 –1361 31. Bocckino, S. B., Blackmore, P. F., and Exton, J. H. (1985) J. Biol. Chem. 260, 45. Paravicini, G., Mendoza, A., Antonsson, B., Cooper, M., Losberger, C., and 14201–14207 Payton, M. A. (1996) Yeast 12, 741–756 32. White, B., and Woodward, D. (1995) Fungal Genet. Newsl. 42, 79–80 46. Madden, K., Sheu, Y.-J., Baetz, K., Andrews, B., and Snyder, M. (1997) Science 33. de Chaffoy de Courcelles, D., Roevens, P., and Van Belle, H. (1985) J. Biol. 275, 1781–1784 Chem. 260, 15762–15770 47. Levin, D. E., Fields, F. O., Kunisawa, R., Bishop, J. M., and Thorner, J. (1990) 34. de Chaffoy de Courcelles, D., Roevens, P., Van Belle, H., Kennis, L., Somers, Cell 62, 213–214 Y., and De Clerck, F. (1989) J. Biol. Chem. 264, 3274 –3285 48. Mazzei, G. J., Schmid, E. M., Knowles, J. K., Payton, M. A., and Maundrell, 35. Dharmananda, S., and Feldman, J. F. (1979) Plant Physiol. 63, 1049 –1054 K. G. (1993) J. Biol. Chem. 268, 7401–7406 36. Jinks, R. N., White, R. H., and Chamberlain, S. C. (1996) J. Photochem. 49. Toda, T., Shimanuki, M., and Yanagida, M. (1993) EMBO J. 12, 1987–1995 Photobiol. B Biol. 35, 45–52 50. Dickman, M. B., and Yarden, O. (1999) Fungal Genet. Biol. 26, 99 –117 37. Blest, A. D., Stowe, S., and Delaney, A. (1994) J. Comp. Physiol. A. 175, 51. Oeser, B. (1998) FEMS Microbiol. Lett. 165, 273–280 611– 617 52. Arpaia, G., Cerri, F., Baima, S., and Macino, G. (1999) Mol. Gen. Genet. 262, 38. Chopra, A., and Khuller, G. K. (1984) Crit. Rev. Microbiol. 11, 209 –271 314 –322
Journal
Journal of Biological Chemistry – Unpaywall
Published: Sep 1, 2000