Phenotypic Mutation 'Kortiku' (pdf version)
Mutation Type missense
Coordinate91,423,829 bp (GRCm38)
Base Change A ⇒ G (forward strand)
Gene Per2
Gene Name period circadian clock 2
Synonym(s) mPer2
Chromosomal Location 91,415,982-91,459,324 bp (-)
MGI Phenotype FUNCTION: This gene is a member of the Period family of genes and is expressed in a circadian pattern in the suprachiasmatic nucleus, the primary circadian pacemaker in the mammalian brain. Genes in this family encode components of the circadian rhythms of locomotor activity, metabolism, and behavior. This gene is upregulated by Clock/Arntl heterodimers but then represses this upregulation in a feedback loop using Per/Cry heterodimers to interact with Clock/Arntl. Polymorphisms in this gene may increase the risk of getting certain cancers and have been linked to sleep disorders. [provided by RefSeq, Jan 2014]
PHENOTYPE: Homozygous null mutants have a partially functional circadian clock, exhibiting a short circadian period followed by loss of circadian rhythmicity in constant darkness. Mutants are also deficient in DNA damage responses and show increased sensitivity togamma radiation and tumor development. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_011066; MGI:1195265

Mapped Yes 
Amino Acid Change Leucine changed to Proline
Institutional SourceBeutler Lab
Gene Model predicted gene model for protein(s): [ENSMUSP00000066620]
SMART Domains Protein: ENSMUSP00000066620
Gene: ENSMUSG00000055866
AA Change: L985P

PAS 179 246 3.23e1 SMART
PAS 319 385 5.75e-2 SMART
PAC 393 436 1.6e0 SMART
low complexity region 475 488 N/A INTRINSIC
low complexity region 821 834 N/A INTRINSIC
low complexity region 996 1014 N/A INTRINSIC
Pfam:Period_C 1040 1234 2.7e-93 PFAM
Predicted Effect probably damaging

PolyPhen 2 Score 1.000 (Sensitivity: 0.00; Specificity: 1.00)
(Using ENSMUST00000069620)
Meta Mutation Damage Score 0.2556 question?
Is this an essential gene? Probably nonessential (E-score: 0.220) question?
Phenotypic Category Autosomal Semidominant
Candidate Explorer Status CE: potential candidate; human score: 0.5; ML prob: 0.376
Single pedigree
Linkage Analysis Data
Alleles Listed at MGI

All Mutations and Alleles(20) : Chemically induced (other)(1) Radiation induced(1) Targeted(10) Transgenic(8)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL01306:Per2 APN 1 91448833 missense probably damaging 0.98
IGL01350:Per2 APN 1 91430861 missense probably damaging 1.00
IGL01865:Per2 APN 1 91421517 missense probably benign 0.10
IGL01974:Per2 APN 1 91423718 missense probably benign 0.02
IGL02118:Per2 APN 1 91424309 missense probably damaging 0.99
IGL02271:Per2 APN 1 91445610 missense probably damaging 1.00
IGL02533:Per2 APN 1 91431002 missense possibly damaging 0.92
IGL02707:Per2 APN 1 91450728 missense possibly damaging 0.94
IGL02972:Per2 APN 1 91423981 missense possibly damaging 0.50
IGL03118:Per2 APN 1 91444619 nonsense probably null
IGL03125:Per2 APN 1 91450611 missense probably benign 0.00
IGL03375:Per2 APN 1 91424228 missense possibly damaging 0.76
IGL03388:Per2 APN 1 91444789 splice site probably benign
obst UTSW 1 91445539 missense probably benign 0.00
rhythm UTSW 1 91429382 critical splice donor site probably null
ANU23:Per2 UTSW 1 91448833 missense probably damaging 0.98
R0029:Per2 UTSW 1 91423712 missense possibly damaging 0.58
R0029:Per2 UTSW 1 91423712 missense possibly damaging 0.58
R0542:Per2 UTSW 1 91438332 critical splice donor site probably null
R0764:Per2 UTSW 1 91429420 missense probably damaging 1.00
R1370:Per2 UTSW 1 91445557 missense possibly damaging 0.94
R1655:Per2 UTSW 1 91448768 missense probably damaging 1.00
R1688:Per2 UTSW 1 91423829 missense probably damaging 1.00
R1997:Per2 UTSW 1 91440859 missense probably damaging 1.00
R2891:Per2 UTSW 1 91445603 missense probably damaging 1.00
R2893:Per2 UTSW 1 91445603 missense probably damaging 1.00
R2894:Per2 UTSW 1 91445603 missense probably damaging 1.00
R3109:Per2 UTSW 1 91445575 missense probably benign 0.02
R4125:Per2 UTSW 1 91429450 missense possibly damaging 0.71
R4997:Per2 UTSW 1 91450783 missense probably benign 0.02
R5110:Per2 UTSW 1 91429515 missense possibly damaging 0.57
R5478:Per2 UTSW 1 91432868 missense probably benign 0.09
R5590:Per2 UTSW 1 91427856 nonsense probably null
R5634:Per2 UTSW 1 91444707 missense probably benign 0.02
R5654:Per2 UTSW 1 91445501 splice site probably null
R5928:Per2 UTSW 1 91444651 missense probably damaging 1.00
R6241:Per2 UTSW 1 91421529 missense probably damaging 0.97
R6295:Per2 UTSW 1 91449872 missense unknown
R6345:Per2 UTSW 1 91448722 missense probably damaging 1.00
R6480:Per2 UTSW 1 91429382 critical splice donor site probably null
R6502:Per2 UTSW 1 91427763 missense probably benign 0.01
R6702:Per2 UTSW 1 91427949 missense probably damaging 1.00
R6703:Per2 UTSW 1 91427949 missense probably damaging 1.00
R6790:Per2 UTSW 1 91445539 missense probably benign 0.00
R7043:Per2 UTSW 1 91419408 missense probably benign
R7092:Per2 UTSW 1 91421431 missense probably damaging 1.00
R7430:Per2 UTSW 1 91423983 nonsense probably null
R7555:Per2 UTSW 1 91435135 missense probably damaging 1.00
R7860:Per2 UTSW 1 91444759 missense probably damaging 0.99
R7943:Per2 UTSW 1 91444759 missense probably damaging 0.99
R8046:Per2 UTSW 1 91435703 missense possibly damaging 0.56
X0011:Per2 UTSW 1 91420589 missense possibly damaging 0.85
Z1176:Per2 UTSW 1 91421493 missense possibly damaging 0.94
Mode of Inheritance Autosomal Semidominant
Local Stock
Last Updated 2019-09-04 9:42 PM by Diantha La Vine
Record Created 2016-08-15 1:53 PM
Record Posted 2017-03-31
Phenotypic Description
Figure 1. Kortiku mice display a shortened free-running period in constant darkness. Raw scores of period (tau) are plotted. Abbreviations: WT, wild-type; REF, homozygous reference mice; HET, heterozygous variant mice; VAR, homozygous variant mice. Mean (μ) and standard deviation (σ) are indicated.

The Kortiku phenotype was identified among G3 mice of the pedigree R1688, some of which showed a shortened free-running period in constant darkness and an advanced phase of entrainment compared to wild-type controls (Figure 1).

Nature of Mutation
Figure 2. Linkage mapping of the circadian period phenotype using a semidominant model of inheritance. Manhattan plot shows -log10 P values (Y-axis) plotted against the chromosome positions of 74 mutations (X-axis) identified in the G1 male of pedigree R1688. Raw period (tau) data were used for single locus linkage analysis with consideration of G2 dam identity.  Horizontal pink and red lines represent, respectively, thresholds of P = 0.05, and P = 0.05 after applying Bonferroni correction.

Whole exome HiSeq sequencing of the G1 grandsire identified 74 mutations. The circadian period phenotype was linked to a mutation in Per2: an A to G transition at base pair 91,423,829 (v38) on chromosome 1, or base pair 35,574 in the GenBank genomic region NC_000067 encoding Per2. The mutation corresponds to residue 3,128 in the NM_011066 mRNA sequence for Per2, in exon 19 of 23 total exons. Linkage was found with a semidominant model of inheritance, wherein 5 variant homozygotes and 10 heterozygotes differed phenotypically from 5 homozygous reference mice (P = 6.877 x 10-16; Figure 2).



980   -S--S--P--L--Q--L--N--L--L--Q--L-


Nucleotide numbering corresponds to NC_000067; the mutated nucleotide is indicated in red.  The mutation results in substitution of leucine 985 with proline (L985P) in the PER2 protein.

Protein Prediction

Figure 3. Domain organization of PER2. The Kortiku mutation results in a leucine to proline substitution at position 985. Binding sites for CK-1ε and CRY1/2 are shown. CLD, cytoplasmic localization domain; NLS, nuclear localization signal; NES, nuclear export signal. The fragment used for X-ray crystallography (structure shown in Figure 4) is indicated (aa170-473). This image is interactive. Other mutations found in the protein are noted in red. Click on each allele for more information.

The PERIOD2 (PER2) protein is one of three mammalian PER proteins (PER1, PER2, and PER3) that are key components of the core feedback loop controlling circadian rhythm in mammals.  In the positive arm of the feedback loop, the transcription factors CLOCK (see uhr), BMAL1, and NPAS2 activate the expression of the PER proteins and the two cryptochromes (CRY1 and CRY2) (1).  PER and CRY proteins inhibit their own transcription, completing the circle of negative feedback. 


Similar to several other circadian clock components (CLOCK, BMAL1, BMAL2, NPAS2), PER proteins contain two Per-Arnt-Sim (PAS) domain repeats (PAS-A and PAS-B; Figure 3). The PAS domain is approximately 100-120 amino acids in length.  The PAS-A domain has been proposed to primarily mediate protein-protein interactions with other PAS-containing proteins (2).  It has been suggested that the PAS-B domain is important as a sensor of environmental signals through binding to proteins that transduce sensory information (2;3).  Sequence analysis of 34 mouse PAS domain-containing proteins suggests that PAS-A and PAS-B domains have distinct consensus sequences (PAS-A: LxALDGFxxV VxxxxGxxxY xSExVxxxLG xxQxxLxxxG xSxxxxxHPx DxxExxxxL; PAS-B: IxxxxxxFx xRHxxDxxFx xxDxRxxxxx GYxPxxLxGx xxYxxxHxxD xxxxxxxH) (2).  However, they exhibit a similar three dimensional fold in a variety of proteins, with four to five α-helices surrounding a sheet of five antiparallel β-strands (4;5).  The crystal structure of a fragment of mouse PER2 (aa170–473) encompassing the two PAS domains showed that both PAS domains exhibited the canonical PAS fold (Figure 4) (6).  The structure contained a homodimer of two PER2 peptides stabilized by antiparallel interactions between the PAS-B β-sheets of each molecule, and by interactions of the PAS-A domain of one molecule with helix αE (C-terminal to PAS-B domain) and the PAS-B domain of the second molecule (6).  It has been reported that the PAS domain homodimer interface differs between mouse and fly PER2 (6;7), and between mouse PER1, PER2, and PER3 (8).

Figure 4. Crystal structure of the PER2 PAS domains (PDB 3GDI). The two molecules of the homodimer are colored in blue/green tones and pink/purple tones.


In addition to homodimerization, the PAS domains of PER2 mediate heterodimerization among the PER proteins (9-12), and interactions with other proteins including CLOCK, BMAL1 (also called Arntl, aryl hydrocarbon receptor nuclear translocator-like), and NPAS2 (5;13-15).  It has been reported that all three PER proteins also bind to heme; the PAS domains and a C-terminal region of PER2 mediate the interaction with heme, which may modulate PER2 stability and expression (16-18)


Important dimerization partners of the PER proteins are CRY1 and CRY2, which bind to a C-terminal region of PER2 (aa1157-1257) (19-23).  Crystal structures of the CRY-binding domain (CBD) of PER2 in complex with the photolyase-homology region (PHR) of CRY1 or CRY2 show a highly extended structure of the PER2 CBD, which folds into five α-helices that wrap around the CRY C-terminal α-helical domain (Figure 5) (24;25).  A zinc ion is coordinated within a CCCH-type intermolecular zinc finger motif at the PER2-CRY interface, stabilizing it.  Binding of the PER2 CBD conceals binding sites on CRY for FBXL3 and CLOCK/BMAL1 (24;25).  This finding is consistent with reports that PER2 competes with the ubiquitin ligase FBXL3 (see De Largo) for binding to CRY1/2; PER2-CRY interaction stabilizes CRY by preventing FBXL3-mediated ubiquitination and degradation by the proteasome (26-28).

Figure 5. Crystal structure of the PER2 CRY-binding domain in complex with the CRY2 photolyase-homology region (PDB 4U8H).


PER2 contains a nuclear localization signal (NLS; aa778-794) and three nuclear export signals recognized by the CRM1/Exportin1 system (NES; aa109-118, 460-469, 983-990) with consensus sequence LX1-3LX2-4LXL(V/I/M) (11;22).  The three NES can function independently but also additively (22).  The kortiku mutation affects a leucine residue within the most C-terminal nuclear export sequence.


Gene expression of Per2 is ubiquitous (29;30).  Most sub-regions of the brain displayed Per2 expression, with the exception of the pars tuberalis and Purkinje neurons (29).  In the suprachiasmatic nucleus of the hypothalamus, Per2 expression peaks during the subjective day (29;30).  PER2 localization cycles between the nucleus and cytoplasm (22).

Figure 6. Model of mammalian circadian clock. CLOCK-BMAL1 heterodimers bind to E-box DNA to drive the expression of PER and CRY genes in the core feedback loop of the clock.  CLOCK-BMAL1 also drives transcription of Rev-Erbα and RORα/β, which act as negative and positive transcriptional regulators, respectively, of a second feedback loop that controls rhythmic expression of BMAL1. The nuclear translocation and proteasome-mediated degradation of PER proteins are regulated by CK1δ/ε and PP1/5. CRY protein degradation is mediated by the SKP1-CUL1-FBXL3 E3 ubiquitin ligase complex in the nucleus, and by the SKP1-CUL1-FBXL21 E3 ubiquitin ligase complex in the cytoplasm. PER2 protein degradation is mediated by the SKP1-CUL1- β-TrCP1/2 E3 ubiquitin ligase complex.

Circadian rhythms are intrinsic daily cycles of behavioral and physiological changes driven by an endogenous “clock” or “oscillator” [reviewed in (31-33)].  Changes in sleep/wakefulness, locomotion, feeding, and temperature are all overt signs of the body’s circadian rhythm.  These rhythms, which follow a 24-hour cycle or period, are both self-sustained, occurring even in the absence of external inputs, and entrained (synchronized) by environmental cues such as the light-dark cycle.  In mammals, the master circadian clock is located in the hypothalamic suprachiasmatic nucleus (SCN) (34), where a network of interconnected and therefore synchronized neurons receives photic input from the retina and provides circadian timing information to the rest of the body (35).  This information is conveyed by direct neuronal projections to other brain and peripheral regions, as well as by diffusible factors (36;37) and systemic cues (e.g. body temperature) from the SCN (38-40).  Peripheral organs can sustain circadian rhythms in the absence of SCN input, and do so with tissue-specific differences in period and phase (41).  However, peripheral tissues rely on the master clock for long term synchronization to an internally coherent timing system (41-44).


At the molecular level, the circadian clock consists of several interacting positive and negative transcriptional feedback loops that drive recurrent cycling of RNA and protein levels of clock components (Figure 6) (1;32;45).  In the canonical model, the core feedback loop consists of BMAL1 and CLOCK, components of the positive arm of the loop, and three PER and two CRY proteins, components of the negative arm of the loop.  CLOCK-BMAL1 heterodimers activate the transcription of Per, Cry, and Rev-Erbα genes through interaction with E-box enhancers (46;47).  After heterodimerization in the cytoplasm, PER-CRY complexes translocate into the nucleus where they directly interact with CLOCK-BMAL1 and inhibit transcription (21;23;48-50).  Ubiquitin-mediated degradation of PER and CRY proteins gradually relieves their repression of CLOCK-BMAL1, and the cycle begins again (28;51-54)


Rhythmic changes in Bmal1 transcription are controlled by another transcriptional feedback loop involving the retinoic acid receptor (RAR)-related orphan receptors (RORα, RORβ, RORγ; see 4-limb clasper and chestnut) and Rev-Erbα/Rev-Erbβ (Figure 6).  RORα and RORβ have been shown to interact with several ROR response elements (RORE) in the promoter of Bmal1, cyclically competing with the transcriptional repressors Rev-Erbα and Rev-Erbβ for binding (47;55-57).  ROREs are also found in intron 1 of Cry1; these, together with E-box elements found in the Cry1 promoter, generate an essential delay in Cry1 expression relative to genes regulated strictly by CLOCK-BMAL1 (58;59).  Although rhythmicity of BMAL1 expression is not necessary for proper timing of the core feedback loop (60), the delay in feedback repression by CRY1 is required for mammalian circadian clock function (58).


Recent studies have explained more precisely the functions of PER and CRY proteins in the circadian core feedback loop.  Biochemical experiments demonstrated that CRY proteins bind to the CLOCK-BMAL1-E-box DNA complex independently of PER to repress transcription (61;62), a finding supported by genetic data (21;23).  By itself, PER1/PER2 had no effect on CLOCK-BMAL1-mediated transcription (62).  However, in the presence of CRY, PER1/PER2 entered the nucleus and displaced CLOCK-BMAL1-CRY from E-box DNA to either inhibit or promote transcription in a manner dependent on the regulatory elements present in the particular promoter (62;63).  A gene predominantly regulated by an E-box, such as Nr1d1 encoding Rev-Erbα, is repressed by PER2 when PER2 removes CLOCK-BMAL1 from the promoter.  A gene predominantly (negatively) regulated by Rev-Erbα, such as Bmal1, is de-repressed by PER2 indirectly through the repressive effect of PER2 on Nr1d1 transcription.  A third type of gene with multiple transcriptional regulatory elements in its promoter including an E-box, D-box, and RRE (e.g. Cry1) is de-repressed by PER2; PER2 displaces the repressive CLOCK-BMAL1-CRY from the promoter, and it is proposed that a binding site is for an unknown transcriptional activator is thereby uncovered to promote transcription (63).  Other reports have proposed different mechanisms for the positive effect of PER2 on transcription of some genes (64;65).


The 24-hour period of the circadian cycle is modulated by casein kinase 1δ (CK1δ) and CK1ε, serine kinases that interact with and phosphorylate PER proteins and thereby control their translocation into the nucleus (66;67) or their degradation (54).  PER1 and PER2 protein degradation is mediated by the SKP1-CUL1-F-box (SCF) E3 ubiquitin ligase complex containing the F-box protein β-TrCP1 or β-TrCP2 (54;68;69).  CK1δ/ε deficiency results in the constitutive cytoplasmic localization of PER proteins (66).  The activities of CK1δ and CK1ε are regulated by phosphatases PP1 and PP5 (66;67).


Although each of the three PER proteins can moderately inhibit transcription activated by CLOCK-BMAL1 in vitro (21), studies of single and double mutant mice suggest their functions in vivo are not equivalent.  Circadian behavior became arrhythmic in Per1-/- or Per2-/- mice after a period of time in constant darkness (10 days to 3 weeks) (70). In the interval before arrhythmicity, the circadian period of locomotor activity was slightly shortened (70);  ; notably, Per2-/- mice congenic with the C57BL/6J strain had a near-normal to normal period and did not develop arrhythmicity (71;72).  Analysis of clock gene and protein expression in these mice indicated that PER2 acted as a positive regulator of Per1, Per2, Cry1, and Bmal1 gene expression, while PER1 promoted the expression of PER2 and CRY1 proteins but not their gene expression (70;73;74).  Double knockout Per1-/-Per2-/- mice showed abrupt loss of behavioral rhythmicity immediately upon placement in constant darkness, a more severe defect than observed in either single knockout mutant (70).  In contrast, PER3 deficiency had no effect on gene expression of Per1, Per2, Cry1, or Bmal1, and only a subtle effect on circadian cycle length, reducing it by 0.5 hour (75)Per3-/-Per1-/- or Per3-/-Per2-/- mice showed circadian behavior phenotypes identical to single mutant PER1- or PER2-deficient mice, respectively, suggesting that PER3 is not a component of the circadian core feedback loop (70).  This is consistent with the finding that mRNA levels in the mouse SCN of Per1 and Per2, but not Per3, were acutely induced by light exposure during the night (76).


Humans with a heterozygous mutation of PER2 display familial advanced sleep phase syndrome 1 (FASPS1; OMIM #604348), in which a shortened circadian period leads to very early sleep onset and offset (77;78).  The mutation, S662G, resulted in impaired binding of CK1δ and CK1ε, consequent hypophosphorylation, and premature nuclear clearance of PER2 (72;79).  As a result of premature nuclear clearance, increased degradation of mutated PER2 occurred in the cytoplasm.

Putative Mechanism

The Per2kortiku mutation results in a leucine to proline substitution at position 985, within the most C-terminal nuclear export sequence of PER2.  The shortening of the free-running period of heterozygous and homozygous Kortiku mice is more severe than that observed in other PER2-deficient mice on a C57BL/6J background (70-72;74), suggesting that Kortiku is a gain-of-function allele. Another ENU-induced missense mutation of Per2 affecting a residue in the interdomain linker between the PAS domains resulted in a period phenotype similar to that observed in Kortiku homozygotes; that allele (Per2Edo; C57BL/6J congenic) was shown to cause a gain of function (80).

Primers PCR Primer

Sequencing Primer

PCR program

1) 94°C 2:00
2) 94°C 0:30
3) 55°C 0:30
4) 72°C 1:00
5) repeat steps (2-4) 40x
6) 72°C 10:00
7) 4°C hold

The following sequence of 544 nucleotides is amplified (chromosome 1, - strand):

1   actggcttca ccatgcctgt tgtgcctatg ggcacccagc ctgaattcgc agtgcagccc
61  ctgccattcg ctgccccttt ggctcctgtc atggccttca tgctgcccag ctacccgttc
121 ccaccagcaa ccccaaacct gcctcaggcc ttcctcccca gccagcctca ctttccagcc
181 caccccacac ttgcctccga aataactcct gcctcccagg ctgagttccc tagtcggacc
241 tcgacgctca gacagccgtg cgcttgccca gtcacccctc cagccggcac agtggccctg
301 ggcagagcct ccccaccgct cttccagtcc agaggcagta gtcccctaca acttaacctg
361 cttcagctag aggaggcgcc tgaaggcagc actggagccg cagggaccct ggggaccaca
421 gggacagcag cttctggtct ggactgcaca tctggcacat ctcgggatcg gcagccaaag
481 gcacctccaa cagtaaggct tctctgctgt gtctctttct caggcagtac acatggtagt
541 cagc

Primer binding sites are underlined and the sequencing primers are highlighted; the mutated nucleotide is shown in red.

Science Writers Eva Marie Y. Moresco
Illustrators Katherine Timer
AuthorsMarleen de Groot, Joseph Takahashi