Epistatic and Synergistic Interactions Between Circadian Clock Mutations in Neurospora crassa

Corresponding author: Jerry F. Feldman, Department of Molecular, Cellular, and Developmental Biology, University of California, Santa Cruz, CA 95064., feldman{at}biology.ucsc.edu (E-mail)

Communicating editor: R. H. DAVIS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We identified a series of epistatic and synergistic interactions among the circadian clock mutations of Neurospora crassa that indicate possible physical interactions among the various clock components encoded by these genes. The period-6 (prd-6) mutation, a short-period temperature-sensitive clock mutation, is epistatic to both the prd-2 and prd-3 mutations. The prd-2 and prd-3 long-period mutations show a synergistic interaction in that the period length of the double mutant strain is considerably longer than predicted. In addition, the prd-2 prd-3 double mutant strain also exhibits overcompensation to changes in ambient temperature, suggesting a role in the temperature compensation machinery of the clock. The prd-2, prd-3, and prd-6 mutations also show significant interactions with the frq⁷ long-period mutation. These results suggest that the gene products of prd-2, prd-3, and prd-6 play an important role in both the timing and temperature compensation mechanisms of the circadian clock and may interact with the FRQ protein.

CIRCADIAN rhythms are endogenously controlled daily oscillations that persist in the absence of environmental stimuli with a period length close to 24 hr. In addition, the period length of circadian rhythms is temperature compensated, remaining nearly constant over a wide range of physiological temperatures. Genetic analysis is a powerful tool for dissecting the underlying mechanism of circadian clocks. In several organisms, including Drosophila, Neurospora, Chlamydomonas, Arabidopsis, and Synechococcus, multiple mutant strains with altered clock function were isolated (FELDMAN 1982 ; DUNLAP 1993 , DUNLAP 1996 ). The cloning of some of the identified genes has led to the identification of molecular components of the circadian clock, most notably in Drosophila and Neurospora (reviewed in DUNLAP 1999 ).

Neurospora crassa has a circadian rhythm of conidiation that has a period length of 21.5 hr at 25° (SARGENT et al. 1966 ). Several mutations that alter the period length of the clock were previously isolated (LAKIN-THOMAS et al. 1990 ; DUNLAP 1993 ; BELL-PEDERSEN 2000 ). The majority of the clock mutations map to separate loci, each of which is represented by a single mutant allele. These mutations are called either the period (prd) mutations (not related to the period gene of Drosophila), or the chrono (chr) mutation. The frequency (frq) gene is the only Neurospora locus for which several mutant alleles have been isolated and has been implicated as part of the central mechanism of the circadian clock (GARDNER and FELDMAN 1980 ; ARONSON et al. 1994B ).

The frq gene has been cloned and studied at the molecular level (MCCLUNG et al. 1989 ; ARONSON et al. 1994A ; LEWIS et al. 1997 ; DUNLAP 1999 ). Both short-period and long-period frq mutants have been isolated; the long-period frq mutations also affect the temperature compensation of the circadian rhythm (GARDNER and FELDMAN 1981 ). Strains carrying a frq-null allele exhibit loss of rhythmicity under certain conditions on race tubes and loss of temperature compensation when rhythmicity is expressed (LOROS and FELDMAN 1986 ; ARONSON et al. 1994A ). Loss of rhythmicity is also evident in mutant white-collar (wc-1; wc-2) strains (CROSTHWAITE et al. 1997 ). Molecular analysis of the frq, wc-1, and wc-2 gene products led to a model in which the FRQ and WC proteins are believed to interact physically in a feedback loop as part of the central mechanism of the clock (CHENG et al. 2001 ; DENAULT et al. 2001 ; MERROW et al. 2001 ).

Double-mutant strains have been constructed in Neurospora, Chlamydomonas, and Drosophila to look for genetic interaction between the clock mutations. However, in strains carrying multiple clock mutations the effects of the mutations have been cumulative, and no cases of epistasis were found in the early reports (BRUCE 1974 ; FELDMAN et al. 1979 ; JACKSON 1993 ). Strains carrying two long-period mutations or two short-period mutations had a period length more severe than each of the parental strains; strains carrying both a long-period and a short-period mutation had a period length intermediate between the period lengths of the parental strains. The period length of the double mutant strains was close to values predicted from assuming an additive or multiplicative effect from each mutation (FELDMAN and ATKINSON 1978 ; LAKIN-THOMAS and BRODY 1985 ).

Since double mutants had a period length close to values predicted by simply combining the effect of each mutation, it was argued that these mutations have independent effects on the period length of the circadian rhythm. More recently, however, some genetic interactions were found. In Drosophila melanogaster some mutant alleles of the timeless gene (tim^SL, tim^rit) were shown to have an allele-specific genetic interaction with the period^L (per^L) mutation (RUTILA et al. 1996 ; MATSUMOTO et al. 1999 ). In N. crassa, the first reported case of epistasis among clock mutations was found: strains carrying both the 18-hr prd-6 mutation and the 25.5-hr prd-2 mutation have an 18-hr period length (MORGAN and FELDMAN 1997 ).

The epistasis of prd-6 to prd-2 suggested the possibility that additional interactions between Neurospora clock mutations might exist. Therefore, in this study all combinations of rhythmic clock double mutant strains were analyzed. The genetic behavior of the clock double mutant strains is consistent with a central group of proteins that act together to maintain clock periodicity and temperature compensation.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and culture conditions:
The strains used in this study were either obtained from the Fungal Genetics Stock Center (Kansas City, KS) or isolated in this lab and are listed in Table 1. Each strain carried the band (bd) mutation, which allows for clear expression of the circadian rhythm in conidiation (SARGENT et al. 1966 ). The phenotypes of the single mutant strains are shown in Table 2. The period-5 (prd-5) mutation was previously identified as causing an altered period length (19.5 hr) following UV mutagenesis of the bd strain (LEWIS 1995 ). Strains were maintained on Vogel's minimal media (VOGEL 1957 ) and handled using standard procedures (DAVIS and DESERRES 1970 ). Scoring of the inv marker was as described (PERKINS et al. 1982 ).

Isolation of mutant strains:
The double mutant strains isolated for use in this study were obtained from analysis of either random ascospores (DAVIS and DESERRES 1970 ) or unordered tetrads (STRICKLAND 1960 ). Double mutant strains were identified as the nonparental, nonwild-type recombinant class of progeny. In cases where phenotype alone was not sufficient to identify the double mutants, tetrads (including both nonparental ditypes and tetratypes) were used to make a preliminary identification. In crosses involving clock mutants with slowed growth rates (see Table 2), linear growth rate was used to aid in identification of progeny genotypes. In some crosses segregation of the prd-6 mutation was followed using the linked (1–5 cM) inv marker (MORGAN and FELDMAN 1997 ). The inv mutation has no effect on circadian rhythms (SARGENT and WOODWARD 1969 ). Each double mutant strain used in this study was crossed with the wild-type strain to confirm the genotype.

Race tube assay:
Race tube assays were performed on either glucose-arginine media or acetate-casamino acid media as previously described (SARGENT and KALTENBORN 1972 ; FELDMAN and HOYLE 1973 ). Unless otherwise specified, race tube assays were performed at 25°. Period length was calculated as the average period of several cycles. Linear growth rate (millimeters per day) was measured as an average of 5–7 days of growth on race tubes containing glucose-arginine media. Period length and growth rate data were collected from all strains of the specified genotype; data from individual assays were pooled together for analysis (details given in figure and table legends).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The prd-6 mutation is epistatic to the prd-3 mutation:
The prd-3 mutation lengthens the circadian period length to 25.1 hr at 25°. The prd-6 mutation has a temperature-sensitive effect on circadian period length with a short period of 18 hr above 21° and a near wild-type period length at temperatures below 21°. The prd-6 mutation was previously shown to be epistatic to the long-period (25.5 hr) prd-2 mutation. At temperatures above 21° prd-2 prd-6 mutant strains have an 18-hr period length. At temperatures below 21° the period length of prd-2 prd-6 mutant strains is also 18 hr, indicating that the presence of the prd-2 mutation causes a loss of the temperature-sensitive nature of the prd-6 phenotype. This suggested that proper function of the mutant PRD-6 protein requires a functional PRD-2 protein (MORGAN and FELDMAN 1997 ).

At temperatures above 21° prd-3 prd-6 strains have an 18-hr period length identical to the period length of strains carrying only the prd-6 mutation (Table 3), indicating that the prd-6 mutation is epistatic to the prd-3 mutation. The prd-3 prd-6 strain has an 18-hr period length at temperatures below 21° as well (Fig 1), suggesting that the mutant PRD-6 protein requires both functional PRD-3 and PRD-2 proteins for proper function.

View larger version (18K): In this window In a new window Download PPT slide   Figure 1. Period lengths of wild-type (F1859a), prd-6 (1616-53a), prd-3 (612-33a), and prd-3 prd-6 (1504-33a) strains at different temperatures. The SEM for the strains ranged from 0.3 to 1.0 hr.

The prd-2 and prd-3 mutations interact synergistically:
Since the prd-6 mutation shows epistasis to each of the prd-2 and prd-3 mutations, a prd-2 prd-3 strain was constructed and its phenotype was analyzed. At 25°, the period length of the prd-2 prd-3 strain is considerably longer than values predicted by either additive or multiplicative methods (Table 3). This suggests that the prd-2 and prd-3 mutations act in synergy as components of the circadian clock. The synergy between the prd-2 and prd-3 mutations is not evident in strains that also carry the prd-6 mutation since the prd-6 mutation is epistatic to the prd-2 prd-3 double mutant (Table 3).

The synergistic interaction between the prd-2 and prd-3 mutations appears to increase with temperature, with a near additive value for period length at 20° but pronounced synergy at 29° (Fig 2). Strains carrying the prd-3 mutation show slight overcompensation in response to changes in ambient temperature, and the prd-2 mutation does not alter temperature compensation by itself (GARDNER and FELDMAN 1981 ; Fig 2). Since the prd-2 prd-3 strain shows dramatic overcompensation in response to temperature, the synergy between these mutations appears to be due to a defective temperature compensation mechanism.

View larger version (20K): In this window In a new window Download PPT slide   Figure 2. Period lengths of wild-type (1313-3a), prd-2 (613-43a), prd-3 (612-33a), and prd-2 prd-3 (1480-21a) strains at different temperatures. The SEM for the strains ranged from 0.3 to 0.8 hr.

The prd-6 mutation genetically interacts with the frq⁷ mutation:
To test whether the prd-6 mutation shows genetic interaction with the frq mutations, double mutant strains between prd-6 and each of the frq alleles were constructed. The frq¹, frq², and frq³ mutations show noninteractive behavior in combination with the prd-6 mutation (Table 4). However, the frq⁷ allele shows a significant interaction with the prd-6 mutation. At 25° prd-6 frq⁷ mutant strains have a period length (20.5 hr) 5 hr shorter than predicted (Table 3). At temperatures above 25° the period length of prd-6 frq⁷ strains is 18 hr, and at temperatures below 25° the period length of prd-6 frq⁷ strains is 22 hr (Fig 3).

View larger version (20K): In this window In a new window Download PPT slide   Figure 3. Period lengths of wild-type (F1858A), prd-6 (1463-12.2A), frq7 (1493-37A), and prd-6 frq7 (1541-1.1A) strains at different temperatures. The SEM for the strains ranged from 0.2 to 1.1 hr.

These data suggest that the period length of these double mutants is temperature sensitive, similar to prd-6 single mutant strains, albeit with a breakpoint at 25° in the double mutant strains as compared to 21° in prd-6 single mutant strains. Within both temperature ranges (17°–25°, or 25°–34°) the prd-6 frq⁷ strain appears temperature compensated, and the frq⁷ mutation has only a small effect on period length of prd-6. This suggests that the prd-6 mutation is an allele-specific suppressor of both the period length and temperature compensation defects caused by the frq⁷ mutation.

The prd-2 and prd-3 mutations interact synergistically with long-period frq alleles:
Strains of the prd-2 frq⁷ or prd-3 frq⁷ genotypes have a period length at 25° that is considerably longer than predicted from noninteractive models (Table 3). This suggests that the prd-2, prd-3, and frq⁷ mutations act synergistically in their effect on the period length of the circadian rhythm. The prd-3 frq³ double mutant strain also has a period length at 25° that is longer than noninteractive models predict (FELDMAN et al. 1979 ; Table 3), suggesting synergy between the prd-3 and frq³ mutations. In contrast, the prd-2 or prd-3 mutations do not exhibit any noticeable genetic interaction with the frq¹ or frq² short-period mutations (FELDMAN et al. 1979 ; Table 4). Since the prd-6 mutation is epistatic to the prd-2 and prd-3 mutations, the synergy between the frq⁷ mutation and the prd-2 or prd-3 mutation is not evident in strains that also carry the prd-6 mutation (data not shown).

The prd-1 and prd-2 mutations have an epistatic relationship:
With one exception, all strains carrying two long-period mutations have a period length longer than each of their parental single mutant strains (see Table 4). The exception involves prd-1 prd-2 double mutant strains, which do not show a clock phenotype that is more severe than either of the single mutant phenotypes (Table 3). Strains carrying the prd-1 (25.8 hr) or prd-2 (25.5 hr) mutations have similar period lengths; prd-1 prd-2 double mutant strains have a period length (25.6 hr) that is indistinguishable from either of the single mutant strains. Since prd-1 and prd-2 strains have essentially the same phenotype, it is impossible to determine which mutation is epistatic to the other. The prd-1 mutation exhibits an additive effect on period length in combination with the prd-3, prd-6, or frq⁷ mutations (Table 4).

Other clock mutations show no genetic interaction:
Consistent with prior studies, the other clock mutations do not show genetic interactions with each other. In most cases, the measured period length of each double mutant strain is close to values predicted from noninteractive models (Table 4). This suggests that these other mutations (prd-4, prd-5, chr, frq¹, and frq²) act independently of other clock mutations in the manner in which they alter circadian timing.

Growth rates of the mutant strains:
Some of the Neurospora period mutations also cause a decrease in growth rate (GARDNER and FELDMAN 1981 ). It was previously suggested that the period length alterations caused by these mutations might represent an indirect effect of the lesion (i.e., slow growth = a slowed clock; DUNLAP 1993 ). However, in each case, clock double mutant strains have a growth rate that can be predicted by combining the effect of each mutation (data not shown). Strains of the genotypes prd-3 prd-6 or prd-1 prd-2 have a growth rate slower than each parental strain, despite the epistasis evident in their clock phenotypes. The prd-2 prd-3 strain exhibits a synergistic interaction between the mutations for their clock defect, but the effects of the mutations on growth rate are additive. These data indicate that the slow growth phenotype associated with some clock mutations is a separate consequence from the defect in period length.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Although many clock mutations in Neurospora seem to act independently of each other, it is evident that there is a group of mutations that mutually interact in their effect on the circadian clock. The combinations of the prd-2, prd-3, prd-6, and frq⁷ mutations show significant genetic interactions. Each of prd-2, prd-3, and frq⁷ mutations show synergistic effects with each other in their period lengthening effects. The prd-6 mutation is epistatic to the prd-2 and prd-3 mutations and suppresses the frq⁷ mutation. These data indicate that the prd-2, prd-3, and prd-6 loci operate in conjunction with frq as part of the central mechanism of the clock to regulate both the circadian period length and the temperature compensation of the clock.

The prd-2, prd-3, and prd-6 mutations seem to be in a pathway that feeds into the FRQ-based oscillator. Although there is an epistatic relationship between the prd-1 and the prd-2 mutations, prd-1 does not appear to belong in this pathway, since it does not genetically interact with the prd-3 or prd-6 mutations. The function and regulation of the products of these genes are unknown; therefore ordering the genes in a pathway is subject to several caveats (AVERY and WASSERMAN 1992 ). However, a simple heuristic diagram showing the genetic interactions and how they may fit into the FRQ/WC feedback loop is presented in Fig 4. An alternative model involving coupled oscillators also fits our data equally well and is presented elsewhere (MORGAN et al. 2001 ).

View larger version (8K): In this window In a new window Download PPT slide   Figure 4. Genetic interactions in the circadian clock in Neurospora crassa. The epistasis between the prd-1 and prd-2 mutations puts them in the same pathway. The epistasis of the prd-6 mutation to the prd-2 and prd-3 mutations suggests that prd-6 mediates the involvement of these genes in the clock. The FRQ/WC feedback loop is adapted from LEE et al. 2000. Boxes around genes indicate involvement in temperature compensation. See the text for further discussion.

The prd-2 and prd-3 mutations cause a similar phenotype (25-hr period length, slowed growth) and show genetic synergy, suggesting that PRD-2 and PRD-3 may affect the same process, share overlapping functions, or operate in convergent signaling pathways. The effects of PRD-2 and PRD-3 appear to be mediated by PRD-6, since the prd-6 mutation is epistatic to the prd-2 and prd-3 mutations. Since mutation of prd-2 or prd-3 has an opposite effect than mutation of prd-6, they may act as negative regulators of prd-6 (Fig 4).

The restoration of wild-type function of PRD-6 in strains carrying the mutant prd-6 allele appears to be dependent on functional proteins produced by the prd-2 and prd-3 loci, possibly due to physical interaction between their encoded proteins (MORGAN and FELDMAN 1997 ). The suppression of the frq⁷ mutation by the prd-6 mutation may also be due to a physical interaction since allele-specific suppression is often due to conformational suppression. Alternatively, the suppression may be a result of phenotypic compensation (MANSON 2000 ). The molecular phenotype of frq⁷ strains includes a slower decay rate of the FRQ7 protein and elevated mRNA levels (ARONSON et al. 1994B ; GARCEAU et al. 1997 ; RUOFF et al. 1999 ). The suppression of the frq⁷ mutation by the prd-6 mutation may be due to a correction in the frq⁷ molecular phenotype. Likewise the synergy of the prd-2 or prd-3 mutations with the frq⁷ mutation may represent exaggerations of the frq⁷ molecular phenotype.

It is apparent that the mutant alleles of the frq locus show differential behavior in combination with the other clock mutations. The frq³ and frq⁷ mutations each cause a partial loss of temperature compensation; the frq⁷ mutation is a more severe defect (GARDNER and FELDMAN 1981 ). Even more severe in this series of alleles is the frq⁹ mutation, which has the null phenotype and exhibits a complete loss of temperature compensation (LOROS and FELDMAN 1986 ). The interactive behavior of the long-period frq alleles in combination with other clock mutations may be due to a defect in the role of the frq locus in the temperature compensation mechanism of the clock. Consistent with this, the frq⁷ mutation shows a strong genetic interaction with the prd-2, prd-3, or prd-6 mutations, whereas the frq³ mutation shows weaker or no interactions with these mutations. In addition, the frq¹ and frq² short-period alleles, which do not alter temperature compensation, show no genetic interaction with other clock mutations.

While the nature of the temperature compensation mechanism of the clock is unknown, "opposing reaction" models for temperature compensation have been proposed (HASTINGS and SWEENEY 1957 ; DUNLAP and FELDMAN 1988 ; RUOFF et al. 1997 ). Our data suggest that a temperature-dependent clock component has lost function in the prd-2 prd-3 mutant strain, leading to a net slowing of the clock rate (increased period length) with increasing temperature as the result of the opposing activity of the functional compensation mechanism of the clock. The frq mutants with altered temperature compensation (e.g., frq⁷ or frq⁹) have elevated mRNA levels (ARONSON et al. 1994B ; MERROW et al. 1997 ), suggesting that regulation of frq transcription may be a possible mechanism of temperature compensation.

The epistasis between the prd-1 and prd-2 mutations suggests that they represent some loss of function in the same pathway since they cause similar phenotypes (slowed growth, 26-hr period length). The prd-1 mutation is also epistatic to the period lengthening effects of the cel mutation (a fatty acid-requiring auxotroph) (LAKIN-THOMAS and BRODY 1985 ). This is of particular interest since the cel mutation is epistatic to the loss of rhythmicity caused by the frq¹⁰ null mutation or by wc-1 or wc-2 mutations (LAKIN-THOMAS et al. 1997 ; LAKIN-THOMAS and BRODY 2000 ). This suggests that prd-1 and prd-2 may be genetic links between a temperature-dependent metabolic "frq-less oscillator" and the frq-dependent oscillator that includes wc-1 and wc-2 (IWASAKI and DUNLAP 2000 ; MORGAN et al. 2001 ).

The wc-1 and wc-2 genes encode transcription factors containing PAS domains similar to the clk, cyc, and per genes of Drosophila (DUNLAP 1999 ) and act in a feedback loop with FRQ (LEE et al. 2000 ). It is possible that the PRD-2, PRD-3, and PRD-6 proteins may interact with the WC proteins to modify their activity and only affect FRQ indirectly. The biochemical function of the FRQ protein is not known, and little is known about other factors that regulate frq gene products or act as other components of the circadian clock. These genetic data suggest the possibility that the PRD-6, PRD-2, and PRD-3 proteins may have physical interactions with the FRQ or WC proteins or regulate their activity. Recent progress in this laboratory toward cloning the prd-2 and prd-6 genes may reveal the functions of their encoded proteins. This extends the possibilities of looking for interactions between Neurospora clock proteins and studying their method of regulation using molecular techniques.

	FOOTNOTES

¹ Present address: Biology Department, Texas A&M University, College Station, TX 77843.

	ACKNOWLEDGMENTS

We thank Amy Bass and Jennifer Compton for technical assistance. We also thank Barry Bowman, Manual Ares Jr., Deborah Bell-Pedersen, and Jennifer Compton for critical review of the manuscript. Research for this project was initiated with grants from the faculty research committee of the University of California, Santa Cruz and support by a gift from the Rosenbaum Medical Trust.

Manuscript received February 20, 2001; Accepted for publication June 1, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ARONSON, B. D., K. A. JOHNSON, and J. C. DUNLAP, 1994a  The circadian clock locus frequency: a single ORF defines period length and temperature compensation. Proc. Natl. Acad. Sci. USA 91:7683-7687[Abstract/Free Full Text].

ARONSON, B. D., K. A. JOHNSON, J. J. LOROS, and J. C. DUNLAP, 1994b  Negative feedback defining a circadian clock: autoregulation in the clock gene frequency. Science 263:1578-1584[Abstract/Free Full Text].

AVERY, L. and S. WASSERMAN, 1992  Ordering gene function: the interpretation of epistasis in regulatory hierarchies. Trends Genet. 8:312-316[Medline].

BELL-PEDERSEN, D., 2000  Understanding circadian rhythmicity in Neurospora crassa: from behavior to genes and back again. Fungal Genet. Biol. 29:1-18[Medline].

BRUCE, V. G., 1974  Recombinants between clock mutants of Chlamydomonas reinhardi. Genetics 77:221-230[Abstract/Free Full Text].

CHENG, P., Y. YANG, C. HEINTZEN, and Y. LIU, 2001  Coiled-coil domain-mediated FRQ-FRQ interaction is essential for its circadian clock function in Neurospora. EMBO J. 20:101-108[Medline].

CROSTHWAITE, S. K., J. C. DUNLAP, and J. J. LOROS, 1997  Neurospora wc-1 and wc-2: transcription, photoresponses, and origins of circadian rhythmicity. Science 276:763-769[Abstract/Free Full Text].

DAVIS, R. H. and F. J. DESERRES, 1970  Genetic and microbial techniques for Neurospora crassa. Methods Enzymol. 27A:79-143.

DENAULT, D., J. J. LOROS, and J. C. DUNLAP, 2001  WC-2 mediates WC-1-FRQ interaction within the PAS protein-linked circadian feedback loop of Neurospora. EMBO J. 20:109-117[Medline].

DUNLAP, J. C., 1993  Genetic analysis of circadian clocks. Annu. Rev. Physiol. 55:683-728[Medline].

DUNLAP, J. C., 1996  Genetic and molecular analysis of circadian rhythms. Annu. Rev. Genet. 17:579-601.

DUNLAP, J. C., 1999  Molecular bases for circadian clocks. Cell 96:271-290[Medline].

DUNLAP, J. C. and J. F. FELDMAN, 1988  On the role of protein synthesis in the circadian clock of Neurospora crassa. Proc. Natl. Acad. Sci. USA 85:1096-1100[Abstract/Free Full Text].

FELDMAN, J. F., 1982  Genetic approaches to circadian clocks. Annu. Rev. Plant Physiol. 33:583-608.

FELDMAN, J. F. and C. A. ATKINSON, 1978  Genetic and physiological characteristics of a slow-growing circadian clock mutant of Neurospora crassa. Genetics 88:255-265[Abstract/Free Full Text].

FELDMAN, J. F. and M. HOYLE, 1973  Isolation of circadian clock mutants of Neurospora crassa. Genetics 75:605-613[Abstract/Free Full Text].

FELDMAN, J. F., G. GARDNER and R. DENISON, 1979 Genetic analysis of the circadian clock of Neurospora, pp. 57–66 in Naito International Symposium on Biological Rhythms and Their Central Mechanism, edited by M. SUDA, O. HAYASHI and H. NAKAGAWA. Elsevier/North Holland Biomedical Press, Amsterdam.

GARCEAU, N. Y., Y. LIU, J. J. LOROS, and J. C. DUNLAP, 1997  Alternative initiation of translation and time-specific phosphorylation yield multiple forms of the essential clock protein FREQUENCY. Cell 89:469-476[Medline].

GARDNER, G. F. and J. F. FELDMAN, 1980  The frq locus in Neurospora crassa: a key element in circadian clock organization. Genetics 96:877[Abstract/Free Full Text].

GARDNER, G. F. and J. F. FELDMAN, 1981  Temperature compensation of circadian period length mutants of Neurospora crassa. Plant Physiol. 68:1244-1248[Abstract/Free Full Text].

HASTINGS, J. W. and B. M. SWEENEY, 1957  On the mechanism of temperature compensation in a biological clock. Proc. Natl. Acad. Sci. USA 43:804-811[Free Full Text].

IWASAKI, H. and J. C. DUNLAP, 2000  Microbial circadian oscillatory systems in Neurospora and Synechococcus: models for cellular clocks. Curr. Opin. Microbiol. 3:189-196[Medline].

JACKSON, F. R., 1993 Circadian rhythm mutants of Drosophila, pp. 91–121 in Molecular Genetics of Biological Rhythms, edited by M. W. YOUNG. Marcel Dekker, New York.

LAKIN-THOMAS, P. L. and S. BRODY, 1985  Circadian rhythms in Neurospora crassa: interactions between clock mutations. Genetics 109:49-66[Abstract/Free Full Text].

LAKIN-THOMAS, P. L. and S. BRODY, 2000  Circadian rhythms in Neurospora crassa: lipid deficiencies restore robust rhythmicity to null frequency and white-collar mutants. Proc. Natl. Acad. Sci. USA 97:256-261[Abstract/Free Full Text].

LAKIN-THOMAS, P. L., G. G. COTÉ, and S. BRODY, 1990  Circadian rhythms in Neurospora crassa: biochemistry and genetics. Crit. Rev. Microbiol. 17(5):365-416[Medline].

LAKIN-THOMAS, P. L., S. BRODY, and G. G. COTÉ, 1997  Temperature compensation and membrane composition in Neurospora crassa. Chronobiol. Int. 14:445-454[Medline].

LEE, K., J. J. LOROS, and J. C. DUNLAP, 2000  Interconnected feedback loops in the Neurospora circadian system. Science 289:107-110[Abstract/Free Full Text].

LEWIS, M. T., 1995 Molecular genetic analysis of circadian clock genes in Neurospora crassa. Ph.D. Thesis, University of California, Santa Cruz, CA.

LEWIS, M. T., L. W. MORGAN, and J. F. FELDMAN, 1997  Analysis of frequency (frq) clock gene homologs: evidence for a helix-turn-helix transcription factor. Mol. Gen. Genet. 253(4):401-414[Medline].

LOROS, J. J. and J. F. FELDMAN, 1986  Loss of temperature compensation of circadian period length in the frq⁹ mutant of Neurospora crassa. J. Biol. Rhythms 1:187-198[Abstract/Free Full Text].

MANSON, M. D., 2000  Allele-specific suppression as a tool to study protein-protein interactions in bacteria. Methods 20:18-34[Medline].

MATSUMOTO, A., K. TOMIOKA, Y. CHIBA, and T. TANIMURA, 1999  tim^rit lengthens circadian period in a temperature-dependent manner through suppression of PERIOD protein cycling and nuclear localization. Mol. Cell. Biol. 19:4343-4354[Abstract/Free Full Text].

MCCLUNG, C. R., B. A. FOX, and J. C. DUNLAP, 1989  The Neurospora clock gene frequency shares a sequence element with the Drosophila clock gene period. Nature 339:558-562[Medline].

MERROW, M., N. GARCEAU, and J. DUNLAP, 1997  Dissection of a circadian oscillation into discrete domains. Proc. Natl. Acad. Sci. USA 94:3877-3882[Abstract/Free Full Text].

MERROW, M., L. FRANCHI, Z. DRAGOVIC, M. GÖRL, and J. JOHNSON et al., 2001  Circadian regulation of the input pathway in Neurospora crassa. EMBO J. 20:307-315[Medline].

MORGAN, L. W. and J. F. FELDMAN, 1997  Isolation and characterization of a temperature-sensitive circadian clock mutant of Neurospora crassa. Genetics 146:525-530[Abstract].

MORGAN, L. W., J. F. FELDMAN, and D. BELL-PEDERSEN, 2001  Genetic interactions between clock mutations in Neurospora crassa: can they help us to understand complexity? Philos. Trans. R. Soc. Lond. B Biol. Sci. in press.

PERKINS, D. D., A. RADFORD, D. NEWMEYER, and M. BJÖRKMAN, 1982  Chromosomal loci of Neurospora crassa.. Microbiol. Rev. 46:526-570.

RUOFF, P., L. RENSING, R. KOMMEDAL, and S. MOHSENZADEH, 1997  Modeling temperature compensation in chemical and biological oscillators. Chronobiol. Int. 14:499-511[Medline].

RUOFF, P., M. VINSJEVIK, C. MONNERJAHN, and L. RENSING, 1999  The Goodwin oscillator: on the importance of degradation reactions in the circadian clock. J. Biol. Rhythms 14:469-479[Abstract/Free Full Text].

RUTILA, J. E., H. ZENG, M. LE, K. D. CURTIN, and J. C. HALL et al., 1996  The tim^SL mutant of the Drosophila rhythm gene timeless manifests allele-specific interactions with period gene mutants. Neuron 17:921-929[Medline].

SARGENT, M. L. and S. H. KALTENBORN, 1972  Effects of medium composition and carbon dioxide on circadian conidiation in Neurospora. Plant Physiol. 50:171-175[Abstract/Free Full Text].

SARGENT, M. L. and D. O. WOODWARD, 1969  Genetic determinants of circadian rhythmicity in Neurospora. J. Bacteriol. 97:861-866[Abstract/Free Full Text].

SARGENT, M. L., W. R. BRIGGS, and D. O. WOODWARD, 1966  The circadian nature of a rhythm expressed in an invertaseless strain of Neurospora crassa. Plant Physiol. 41:1343-1349[Abstract/Free Full Text].

STRICKLAND, W. N., 1960  A rapid method for obtaining unordered Neurospora tetrads. J. Gen. Microbiol. 22:583-588[Abstract/Free Full Text].

VOGEL, H. J., 1957  A convenient growth medium for Neurospora (medium N). Microbial. Genet. Bull. 13:42-43.

This article has been cited by other articles:

				L. Lombardi, K. Schneider, M. Tsukamoto, and S. Brody Circadian Rhythms in Neurospora crassa: Clock Mutant Effects in the Absence of a frq-Based Oscillator Genetics, March 1, 2007; 175(3): 1175 - 1183. [Abstract] [Full Text] [PDF]

				C. I. Hong, E. D. Conrad, and J. J. Tyson A proposal for robust temperature compensation of circadian rhythms PNAS, January 23, 2007; 104(4): 1195 - 1200. [Abstract] [Full Text] [PDF]