Phenotypic Mutation 'dazzle' (pdf version)
Mutation Type nonsense
Coordinate94,093,624 bp (GRCm38)
Base Change C ⇒ A (forward strand)
Gene Spag9
Gene Name sperm associated antigen 9
Synonym(s) syd1, JIP4, Mapk8ip4, 4733401I23Rik, JLP, 3110018C07Rik, 4831406C20Rik
Chromosomal Location 93,996,091-94,126,085 bp (+)
MGI Phenotype FUNCTION: [Summary is not available for the mouse gene. This summary is for the human ortholog.] This gene encodes a member of the cancer testis antigen gene family. The encoded protein functions as a scaffold protein that structurally organizes mitogen-activated protein kinases and mediates c-Jun-terminal kinase signaling. This protein also binds to kinesin-1 and may be involved in microtubule-based membrane transport. This protein may play a role in tumor growth and development. Alternate splicing results in multiple transcript variants. [provided by RefSeq, Oct 2011]
PHENOTYPE: Male mice homozygous for a null mutation display reduced fertility with oligoasthenozoospermia. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_001199203.1; MGI: 1918084

Mapped Yes 
Limits of the Critical Region 88990137 - 99052085 bp
Amino Acid Change Cysteine changed to Stop codon
Institutional SourceBeutler Lab
Gene Model not available
SMART Domains Protein: ENSMUSP00000024979
Gene: ENSMUSG00000020859
AA Change: C636*

transmembrane domain 7 29 N/A INTRINSIC
PDB:2W83|D 253 305 1e-25 PDB
low complexity region 306 339 N/A INTRINSIC
coiled coil region 572 606 N/A INTRINSIC
low complexity region 735 751 N/A INTRINSIC
SCOP:d1kb0a2 823 969 3e-5 SMART
Blast:WD40 924 964 8e-18 BLAST
low complexity region 1132 1150 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000042271
Gene: ENSMUSG00000020859
AA Change: C774*

Pfam:Jnk-SapK_ap_N 24 179 2e-61 PFAM
Pfam:JIP_LZII 390 460 5.3e-32 PFAM
coiled coil region 710 744 N/A INTRINSIC
low complexity region 873 889 N/A INTRINSIC
SCOP:d1kb0a2 961 1107 1e-5 SMART
Blast:WD40 1062 1102 1e-17 BLAST
low complexity region 1270 1288 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000075115
Gene: ENSMUSG00000020859
AA Change: C635*

transmembrane domain 7 29 N/A INTRINSIC
PDB:2W83|D 253 305 1e-25 PDB
low complexity region 306 339 N/A INTRINSIC
coiled coil region 571 605 N/A INTRINSIC
low complexity region 734 750 N/A INTRINSIC
SCOP:d1kb0a2 822 968 3e-5 SMART
Blast:WD40 923 963 7e-18 BLAST
low complexity region 1131 1149 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000090452
Gene: ENSMUSG00000020859
AA Change: C636*

transmembrane domain 7 29 N/A INTRINSIC
PDB:2W83|D 254 306 1e-25 PDB
low complexity region 307 340 N/A INTRINSIC
coiled coil region 572 606 N/A INTRINSIC
low complexity region 735 751 N/A INTRINSIC
SCOP:d1kb0a2 823 969 3e-5 SMART
Blast:WD40 924 964 7e-18 BLAST
low complexity region 1132 1150 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000099457
Gene: ENSMUSG00000020859
AA Change: C631*

transmembrane domain 7 29 N/A INTRINSIC
PDB:2W83|D 249 301 1e-25 PDB
low complexity region 302 335 N/A INTRINSIC
coiled coil region 567 601 N/A INTRINSIC
low complexity region 730 746 N/A INTRINSIC
SCOP:d1kb0a2 818 964 3e-5 SMART
Blast:WD40 919 959 8e-18 BLAST
low complexity region 1127 1145 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000118850
Gene: ENSMUSG00000020859
AA Change: C424*

transmembrane domain 7 29 N/A INTRINSIC
coiled coil region 360 394 N/A INTRINSIC
low complexity region 523 539 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000117502
Gene: ENSMUSG00000020859
AA Change: C355*

PDB:2W83|D 1 25 4e-8 PDB
low complexity region 26 59 N/A INTRINSIC
coiled coil region 291 325 N/A INTRINSIC
low complexity region 454 470 N/A INTRINSIC
SCOP:d1kb0a2 542 688 3e-5 SMART
Blast:WD40 643 683 1e-17 BLAST
low complexity region 864 882 N/A INTRINSIC
Predicted Effect probably null
SMART Domains Protein: ENSMUSP00000115864
Gene: ENSMUSG00000020859
AA Change: C623*

Pfam:JIP_LZII 240 310 1.1e-32 PFAM
coiled coil region 559 593 N/A INTRINSIC
low complexity region 723 739 N/A INTRINSIC
Predicted Effect probably null
Meta Mutation Damage Score Not available question?
Is this an essential gene? Probably essential (E-score: 0.842) question?
Phenotypic Category
Phenotypequestion? Literature verified References
Candidate Explorer Status CE: no linkage results
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI
All alleles(6) : Targeted, knock-out(1) Gene trapped(4) Chemically induced(1)
Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00422:Spag9 APN 11 94097866 missense probably benign 0.02
IGL01776:Spag9 APN 11 94116727 splice site probably benign
IGL02095:Spag9 APN 11 94108582 missense probably damaging 1.00
IGL02307:Spag9 APN 11 94102160 critical splice donor site probably null
IGL02417:Spag9 APN 11 94116741 missense probably benign 0.27
IGL02480:Spag9 APN 11 94108587 nonsense probably null
IGL02864:Spag9 APN 11 94106661 missense probably damaging 1.00
IGL02976:Spag9 APN 11 94083953 missense probably benign 0.30
IGL02979:Spag9 APN 11 94097364 missense probably benign
IGL03349:Spag9 APN 11 94093509 missense possibly damaging 0.51
R0128:Spag9 UTSW 11 94093539 missense probably damaging 1.00
R0418:Spag9 UTSW 11 94091753 splice site probably benign
R1463:Spag9 UTSW 11 94116837 missense probably damaging 1.00
R1593:Spag9 UTSW 11 94097233 missense probably damaging 1.00
R1605:Spag9 UTSW 11 94048539 missense probably damaging 0.99
R1649:Spag9 UTSW 11 94108452 splice site probably null
R1697:Spag9 UTSW 11 93996565 missense probably benign 0.00
R1952:Spag9 UTSW 11 94097358 missense possibly damaging 0.77
R2011:Spag9 UTSW 11 94092375 nonsense probably null
R2012:Spag9 UTSW 11 94092375 nonsense probably null
R2351:Spag9 UTSW 11 94092900 missense probably damaging 1.00
R2367:Spag9 UTSW 11 94116757 missense probably damaging 1.00
R3027:Spag9 UTSW 11 94086377 missense probably null 1.00
R3766:Spag9 UTSW 11 94060283 intron probably benign
R3777:Spag9 UTSW 11 94099026 critical splice acceptor site probably null
R3937:Spag9 UTSW 11 94044417 missense possibly damaging 0.94
R3937:Spag9 UTSW 11 94044479 missense possibly damaging 0.92
R4417:Spag9 UTSW 11 94060346 intron probably benign
R4445:Spag9 UTSW 11 94097253 missense possibly damaging 0.95
R4711:Spag9 UTSW 11 94114351 critical splice donor site probably null
R4799:Spag9 UTSW 11 94048516 missense possibly damaging 0.87
R4799:Spag9 UTSW 11 94048517 missense probably damaging 0.96
R4816:Spag9 UTSW 11 94048599 intron probably benign
R4843:Spag9 UTSW 11 94097818 missense probably damaging 1.00
R5020:Spag9 UTSW 11 94097786 missense probably benign 0.08
R5119:Spag9 UTSW 11 94122722 missense probably damaging 1.00
R5298:Spag9 UTSW 11 94100135 missense probably damaging 1.00
R5304:Spag9 UTSW 11 94069012 missense probably damaging 1.00
R5305:Spag9 UTSW 11 94069012 missense probably damaging 1.00
R5395:Spag9 UTSW 11 94091751 splice site probably null
R5636:Spag9 UTSW 11 94069012 missense probably damaging 1.00
R5638:Spag9 UTSW 11 94069012 missense probably damaging 1.00
R5654:Spag9 UTSW 11 94090712 missense probably damaging 1.00
R5779:Spag9 UTSW 11 94114253 missense probably benign 0.20
R5814:Spag9 UTSW 11 94082828 missense possibly damaging 0.94
R5912:Spag9 UTSW 11 94044425 missense probably damaging 0.98
R6038:Spag9 UTSW 11 94112092 missense probably damaging 1.00
R6038:Spag9 UTSW 11 94112092 missense probably damaging 1.00
R6269:Spag9 UTSW 11 94044507 missense probably benign 0.05
R6294:Spag9 UTSW 11 94093485 critical splice acceptor site probably null
R6389:Spag9 UTSW 11 94086311 missense probably damaging 1.00
R6420:Spag9 UTSW 11 94086302 missense probably damaging 1.00
R6460:Spag9 UTSW 11 94068975 missense probably damaging 1.00
R6482:Spag9 UTSW 11 94093502 missense possibly damaging 0.94
R6860:Spag9 UTSW 11 94081370 missense probably benign 0.25
R7086:Spag9 UTSW 11 94097864 missense probably benign
R7179:Spag9 UTSW 11 94089432 splice site probably null
R7225:Spag9 UTSW 11 94097358 missense probably damaging 0.98
R7351:Spag9 UTSW 11 94092976 missense probably benign 0.00
R7366:Spag9 UTSW 11 94108521 missense possibly damaging 0.56
R7378:Spag9 UTSW 11 94114351 critical splice donor site probably null
R7401:Spag9 UTSW 11 94097689 missense probably benign
R7506:Spag9 UTSW 11 94108464 missense probably damaging 1.00
R7507:Spag9 UTSW 11 94068080 missense probably benign 0.00
R7513:Spag9 UTSW 11 94112083 missense probably damaging 1.00
R7655:Spag9 UTSW 11 93996563 missense possibly damaging 0.56
R7656:Spag9 UTSW 11 93996563 missense possibly damaging 0.56
R7664:Spag9 UTSW 11 94102160 critical splice donor site probably null
R7665:Spag9 UTSW 11 94013654 missense probably damaging 0.98
R7862:Spag9 UTSW 11 94112066 missense possibly damaging 0.69
R8074:Spag9 UTSW 11 94112051 missense probably damaging 1.00
R8085:Spag9 UTSW 11 94099044 missense probably benign
Mode of Inheritance Autosomal Recessive
Local Stock Sperm, gDNA
MMRRC Submission 032803-UNC
Last Updated 2018-04-26 9:10 AM by Anne Murray
Record Created unknown
Record Posted 2010-04-15
Other Mutations in This Stock Stock #: 3370 Run Code:
Validation Efficiency: 17/31

GeneSubstitutionChr/LocMutationPredicted EffectZygosity
Clca2 C to T 3: 145,077,977 A626T probably damaging Homo
Dido1 A to G 2: 180,671,542 M979T probably benign Homo
Mnx1 A to G 5: 29,474,887 C241R unknown Homo
Rab38 A to G 7: 88,490,651 H176R probably benign Homo
Tap2 A to G 17: 34,209,279 probably null Homo
Tmem167 A to C 13: 90,098,466 K36N probably damaging Homo
Phenotypic Description

 The dazzle phenotype was first observed in G3 mice homozygous for mutations induced by N-ethyl-N-nitrosourea (ENU) (Figure 1).  Homozygous mice have lightened coat color and pale skin. The strain is subfertile as homozygous crosses breed inefficiently, and mutant phenotypes are observed at less than the expected ratio. Male mutant sperm has 50% normal motility.

Nature of Mutation
Figure 2. LOD score of 6.4 maps the dazzle mutation to Chromosome 11.
The dazzle coat color phenotype was mapped to Chromosome 11 between markers d11mit284 and d11mit124 (Figure 1), which corresponds to a critical region between 88,990,137 – 99,052,085 base pairs using NCBI m37 mouse assembly (Build 37.1). Subsequent whole genome sequencing of a homozygous dazzle mouse using the SOLiD technique (1X coverage of 78.5%) identified a single mutation within the critical region corresponding to a C to A transversion at base pair 93,954,938 located in the Spag9 gene.  The mutation was confirmed using standard Sanger sequencing (Figure 3), and corresponds to nucleotide 2544 of the Spag9 transcript in exon 22 of 34 total exons using NCBI record NM_001199203.1. Multiple Spag9 transcripts are displayed on Ensembl. Like the conserved human gene, the mouse Spag9 gene may contain as many as 38 exons (1-3).
783  -D--S--F--T--V--C--N--S--H--V--L-
The mutated nucleotide is indicated in red lettering, and converts a cysteine to a premature stop codon at amino acid 788 of the encoded protein.
Protein Prediction
Figure 4. Domain structures of Spag9 isoforms. The SPAG9 proteins belong to the JIP group of scaffold proteins. JLP is the most characterized mouse isoform and contains an LZ domain (LZI) a region that is missing from SPAG9. All JIP scaffold proteins have a conserved JNK binding domain (JBD). There is a second LZ domain (LZII) that is found in JLP and SPAG9 proteins. A coiled-coil (CC) region is found in SPAG9 proteins. The C-terminus of JLP contains a highly conserved region known as the C-terminal domain (CTD).  Many proteins have been reported to interact with the different domains of the SPAG9 isoforms to form multiprotein complexes. The JLP isoform was first cloned, and the proteins it interacts with are noted on its domain. JLPL, contains a 13 amino acid insertion in the C-terminal region of JLP. The 1321 amino acid isoform 1 contains a 14 amino acid insertion and remains uncharacterized. The position of the dazzle mutation is shown with a red asterisk and converts a cysteine to a premature stop codon at amino acid 788 in isoform 1.
The mouse Spag9 gene encodes multiple protein isoforms generated by alternative splicing (2-4). Many isoforms have also been reported for human SPAG9 (Figure 4) (1). The longest isoform, designated as isoform 1 (Uniprot Q58A65), contains 1321 amino acids in both mouse and human with 96% identity. At least five of these isoforms have been characterized (1-5). The first SPAG9 isoform discovered was the human testis-specific antigen SPAG9 (sperm antigen 9). SPAG9 contains 766 amino acids, and is missing the N- and C-terminal regions of the full-length protein (6). Homologous testis-specific antigens have since been reported for non-human primates (7;8). A variant of SPAG9 containing a 14 amino acid insertion near the N-terminus is known as protein highly expressed in testis (PHET) (1). Subsequent to the discovery of SPAG9, additional Spag9 isoforms have been characterized in the mouse including the 1307 amino acid c-Jun N-terminal kinase (JNK)-associated leucine zipper protein (JLP) (4), the 1142 amino acid JNK-interacting protein 4 (JIP4) missing the N-terminus (2), and JLPL, which contains a 13 amino acid insertion in the C-terminal region of JLP (3). The 1321 amino acid isoform 1 contains the same 14 amino acid insertion found in PHET, but remains uncharacterized. Most of the variants found in the mouse are conserved with other species, but a 1334 amino acid splice variant incorporating both the N-terminal and C-terminal inclusions was found only in monkeys (3)
The SPAG9 proteins belong to the JIP group of scaffold proteins comprising JIP1, JIP2, and JIP3, and are most structurally related to JIP3. Whereas JIP1 and JIP2 contain Src-homology (SH3) and phospho-tyrosine binding (PTB) domains, JIP3 and SPAG9 isoforms contain coiled-coil regions and leucine zipper (LZ) domains. All JIP scaffold proteins have a conserved JNK binding domain (JBD) (5;9). Using the amino acid numbering for the most characterized mouse isoform, JLP, an LZ domain (LZI) occurs roughly at amino acids 66-170, a region that is missing from JIP4 and SPAG9. The JBD occurs at amino acids 200-208, and a second LZ domain (LZII) is found at amino acids 392-462. The C-terminus of JLP contains a highly conserved region known as the C-terminal domain (CTD). This region is missing from the SPAG9 isoform (3-5;10). LZ domains are characterized by the presence of heptad repeats containing a conserved hydrophobic residue, usually a leucine, at every seventh residue in a coiled-coil α-helical structure (11). LZI contains seven of these repeats while LZII contains nine (4;10). A leucine zipper will often form a dimer with the leucine zipper of another polypeptide, and SPAG9 isoforms are able to form dimers or oligomers (2;5;10;11). Hydropathy plots suggest that SPAG9 isoforms contain a putative transmembrane domain at residues 820-836 (2;6).      
Figure 5. The ARF6–JIP4-LZII complex. ARF6–GTP is shown in teal (the switch regions in gray), and the JIP4-LZII monomers are shown in magenta and purple. The UCSF Chimera structure is based on Isabet et al. The EMBO Journal 28, 2835–2845 (2009). Click on the image to view it rotate.
Many proteins have been reported to interact with the different domains of the SPAG9 isoforms to form multiprotein complexes (3;4;12-14). The JLP isoform was first cloned and characterized as interacting with the transcription factors Myc and Myc-associated factor X (MAX) (4). JLP binds to MAX through the LZI domain, while two regions of the protein, amino acids 1-110 and 160-209 (spanning the JBD), are each able to bind to Myc as well as the mitogen-activated protein kinases (MAPKs) p38α/β and JNK1-3 (4;5). JLP also binds to kinases upstream of the MAPKs, MKKs (or MAP2Ks) and MEKKs (or MAP3Ks). Specifically, JLP binds to MKK4 and MEKK3. MKK4 is able to activate both p38 and JNK (4). Neither SPAG9 nor JIP4, which are missing the N-terminal region, are able to bind stably to p38 (2;5), although the JIP4 isoform was demonstrated to be a p38 substrate and potentiates p38 activation (2). JIP4 and JLP are also able to interact with the MAP3K, apoptosis signal-regulating kinase 1 (ASK1) (2;15). The LZII domain can either bind to the motor protein kinesin light chain 1 (KLC1), or to the GTP-binding protein ADP-ribosylation factor 6 (ARF6), thus promoting an interaction with the dynactin complex (2;12;16). Amino acids 465-647 of JLP interact with the promyogenic cell surface protein Cdo (or Cdon for cell adhesion molecule-related/down-regulated by oncogenes) (13), while the C-terminal region binds to the FYVE-finger containing phosphoinositide kinase (PIKfyve), the α-subunits of the heterotrimeric G12 and G13 proteins (Gα13, Gα12), and the microtubule-destabilizing factor SCG10 (superior cervical ganglia clone 10) (3;14;17). JLP is reported to contain several SH2 and SH3 binding sites, and has been shown to associate with the tyrosine kinase ABL, which contains an SH3 domain and also binds to Cdo (18).
The interaction between GTP-bound ARF6 and the LZII domain of JLP has been structurally characterized (10). Crystallized ARF6 and JLP-LZII form a heterotramer with the JLP-LZII homodimer recruiting two ARF6 molecules to its center. However, a structure-derived model of the association of the ARF6/JLP complex with membranes suggests that only one ARF6 molecules is able to bind to the JLP-LZII homodimer. The JLP4-LZII domain complexed to ARF6 forms two long, parallel α-helices that wind around each other (Figure 5) (PDB 2W83). ARF6 interacts with both JLP-LZII monomers with the ARF6-binding site spanning three of the leucine repeats (amino acids 412-432). The interface between ARF6 and JLP involves both hydrophobic and hydrophilic interactions. Both JLP monomers are involved in the hydrophobic interactions, which include Ala 412, Leu 413, Ile 415, Val 416, Leu 420 and Lys 423 from one JLP molecule and Lys 417, Ile 421, Val 424 from the other molecule. Only residues from one of the JLP-LZII molecules form hydrophilic interactions with ARF6. These residues are Lys 417, Asn 418, Asp 425, Thr 428, Cys 429 and Asp 432.
The dazzle mutation alters an amino acid common to all known SPAG9 isoforms to a stop codon, and occurs prior to the putative transmembrane domain and the CTD. 
In humans, the SPAG9 and PHET isoforms are reported to be testis-specific (1;6)In situ hybridization of human testis sections revealed SPAG9 cDNA on round spermatids of stages I, II, and III of the human seminiferous cycle, while immunohistochemical staining showed that SPAG9 is associated with elongated spermatids and not with round spermatids, indicating posttranscriptional delay in expression (6). SPAG9 appears to be localized specifically on the acrosomal compartment of intact human spermatozoa (5). In addition, SPAG9 expression is associated with various human cancers including chronic myeloid leukemia (19), epithelial ovarian cancer (20), renal cell carcinoma (21), breast carninoma (22), cervical cancer (23), and thyroid cancer (24) by RT-PCR and antibody. PHET, or other SPAG9 isoforms containing the N-terminal 14 amino acid insertion, is overexpressed in dermal fibroblasts from patients with systemic sclerosis (SSc), an autoimmune disease characterized by excessive fibrosis of the skin and internal organs as well as microvascular injury. Patients with SSc develop autoantibodies to various proteins including PHET (1). One caveat with these studies is that the primers used in these RT-PCR analyses will recognize both SPAG9 and PHET isoforms, while polyclonal antibodies likely recognize most of the isoforms encoded by SPAG9. Examination of all SPAG9 cDNAs, suggests that most tissues and cells express some form of SPAG9 protein (1).
Similarily in mouse, JLP and JIP4 isoforms are reported to be expressed ubiquitously (2;4;25). Using Northern blot analysis, two major transcripts of 4.8 and 8.9 kilobases (kb) were detected. The 4.8 kb RNA was most abundant in the testis, while the larger transcript of 8.9 kb was more abundant in the brain, kidney, liver, heart, and other tissues (2). Western blot analysis found JLP protein expressed in most tissues, with highest expression level in the testis. Moderate or low expression was seen in the brain, lung, spleen, and ovary, and very low expression was seen in the heart, liver, kidney epididymis, and uterus. Immunopositive signals were detected within the seminiferous tubules, in spermatocytes and spermatids, with the most intense staining in elongated spermatids. The JLP protein was reported to be the most abundant isoform in the testis, although other isoforms are also detected (25).   
Human SPAG9 has been reported to be a cell surface membrane protein (5;6), consistent with the presence of a putative transmembrane domain. However, other studies suggest that at least some of these isoforms are localized to the cytoplasm, and can redistribute to the perinuclear region in response to stress signals such as ultraviolet (UV) radiation (1;2;4). In the cytoplasm, SPAG9 isoforms colocalize with KLC1 in an LZII-dependent manner (12). In addition, some of this protein is associated with an endosomal compartment in transfected cells in association with ARF6 (16). SPAG9 proteins are upregulated during neuronal differentiation using the P19 cell model (12)
MAPK signaling cascades play critical roles in many cellular processes including proliferation, survival, locomotion, differentiation, as well as immune and stress responses. MAPKs are regulated by a cascade of protein phosphorylation with three distinct tiers of regulation known as MAPK modules [reviewed by (9;26)]. MAPK modules consist of three distinct kinases, an upstream serine/threonine MAP kinase kinase kinase (MAP3K), a MAP kinase kinase (MAP2K), and a downstream MAPK. MAP2Ks can phosphorylate tyrosine as well as threonine residues and specifically activate MAPKs by phosphorylating a TxY motif in their activation loops (27). In response to stimuli, the Ras-or Rho-families of small GTPases stimulate the activity of an upstream MAP4K, which in turn stimulates the MAPK module via the sequential phosphorylation of MAP3K, MAP2K and MAPK. Activated MAPKs can often translocate from the cytoplasm to the nucleus where they regulate target transcription factor activity through phosphorylation (28). Some well characterized MAPK modules include extracellular signal-regulated kinase (ERK), JNK, and p38 [reviewed by (9;28)].  MAPK modules have been shown to regulate distinctly different cellular responses in a cell type and/or context specific manner, and these responses may be facilitated by specific scaffold proteins that allow the assembly of a discrete set of signaling proteins into a complex (Figure 6) (9;26). By bringing MAPK components into close proximity, scaffolding proteins may enable the rapid transmission of signals. Scaffold proteins can also act as catalysts and activate different components in the signaling pathway. Interestingly, studies of scaffold proteins in the budding yeast Saccharomyces cerevisiae suggest that different scaffold proteins can recruit specific kinase modules and localize them to specific cellular regions to carry out an appropriate cellular response (29-31). As described above, SPAG9 isoforms are able to bind to the JNK and p38α/β and upstream kinases, and are able to enhance the activation of these kinases in several different cell lines (2;4;5;14;32).
Figure 6. JLP as a scaffolding protein in MAPK signaling. JLP is associated with both p38 and JNK signaling, while SPAG9 is only involved in the JNK pathway. MKK4 can stimulate both pathways and binds to JLP. JLP is the only mammalian MAPK scaffolding protein that is demonstrated to also bind to heterotrimeric G proteins along with JNK signaling components.
MAPK modules and scaffolding proteins were first discovered by characterizing the pheromone signaling pathway in yeast. In S. cerevisiae haploid cells, mating pheromones stimulate pheromone receptors, which then activate a heterotrimeric G protein by catalyzing the exchange of guanine nucleotides in the α-subunit. The GTP-bound G protein is then able to stimulate a kinase cascade culminating in the activation of a three-tier MAPK module defined by the MAP3K Ste11, the MAP2K Ste7 and a MAPK, Fus3. Phosphorylated Fus3 translocates to the nucleus where it can stimulate the transcription of genes necessary for the mating response. Signaling through these kinases required the function of a protein encoded by the Ste5 gene, a scaffolding protein that binds to Ste11, Ste7, and Fus3 through different domains. Ste5 was also demonstrated to associate with the heterotrimeric G protein that stimulates the kinase cascade [reviewed by (9;29;30)]. Interestingly, JLP is the only mammalian MAPK scaffolding protein that is demonstrated to also bind to heterotrimeric G proteins along with JNK signaling components (14). Using small interference RNA methodology (RNAi), JLP has been demonstrated to be necessary for Gα13 and JNK-mediated endodermal differentiation of murine P19 embryonic carcinoma cells (32). JLP also mediates the association of Gα13 with the proapoptotic MAPK ASK1, reducing the rate of ASK1 degradation (15). Finally, JLP has been demonstrated to be necessary for the Gα13 induction of cell migration in culture, a process also dependent on the Ras-related GTPase Rac1, which plays essential roles in cytoskeletal reorganization (33). Conversely, JLP appears to inhibit neurite outgrowth by decreasing the phosphorylation of SCG10 in neuronal cell cultures. SCG10 is a JLP-binding microtubule-destabilizing factor necessary for neurite outgrowth. Inhibition of JNK activity produced similar results (17).   
A number of studies have suggested that JLP, along with other JIPs, plays a role in vesicular or protein transport by its association with the motor protein KLC1, which is generally required for anterograde transport of cargo to the cell periphery along microtubules. The roles of JIP proteins in trafficking was first identified through studies of the Drosophila melanogaster homologue Sunday driver (Syd), a membrane-associated protein structurally related to JIP3 and JLP that is required for the functional interaction of kinesin with axonal cargo (34). Thus, the JIP proteins may act as adapter molecules for the transport of cargo by the kinesin motor protein. Alternatively, the association of JIPs to kinesin may allow the spatiotemporal regulation of JNK or p38 signaling.  Indeed, the distribution of JNK signaling modules is disrupted in hippocampal neurons in JIP1-deficient mice and affects JNK activation in response to stress (35). In mammalian cells, disruption of JLP expression using RNAi has been shown to affect the subcellular localization of p38 (33). JLP also associates with a number of other proteins involved in microtubule-associated trafficking such as PIKfyve, ARF6, and the dynactin complex (3;16). The interaction of JLP with the GTP-binding protein ARF6 appears to interfere with kinesin binding, but favors the association of JLP with the dynactin complex required for retrograde transport. ARF6 is located in the plasma membrane and endosomal compartments where it regulates the trafficking of plasma membrane receptors and lipids through the endocytic pathway. Both ARF6 and JLP are required to control the trafficking of recycling endosomes during cytokinesis, a mechanism required for the separation of daughter cells (16). PIKfyve, the enzyme that synthesizes phosphatidylinositol (3,5)-bisphosphate [PtdIns(3,5)P2], is also required during endocytosis and interacts with the Rab9 GTPase, which is necessary for transport from the late endosomes to the trans-Golgi network (TGN). Through PtdIns(3,5)P2 synthesis, PIKfyve plays a role in the fission and fusion events underlying vesicle formation (36;37). The association of JLP with PIKfyve links these vesicles to microtubule-associated motor proteins and enables their transport (3).   
Several studies suggest that JLP connects the cell surface protein Cdo to the p38 MAPK module, enhancing Cdo-induction of p38 activity (13;18;38;39). Cdo is a promyogenic transmembrane protein belonging to the immunoglobulin (Ig) superfamily that binds to a number of different proteins through its extracellular domain including N-cadherin. The interaction of N-cadherin with Cdo can triggers p38 MAPK activation leading to myogenic differentiation (39), a process that is dependent upon the interaction of both Cdo and p38 with JLP (13). The intracellular region of Cdo also binds to the tyrosine kinase ABL as well as the BCL2/adenovirus E1B 19 kDa protein-interacting protein 2 (Bnip-2), which recruits the Rho-related GTPase Cdc42. Both ABL and Cdc42 are necessary for differentiation-dependent increase in p38 activity (18;38)
Despite multiple studies that suggest SPAG9 isoforms are likely to play roles in various fundamental processes such as cellular transport, cytokinesis, or myogenesis, the generation of a JLP mouse knockout resulted in generally normal animals. As the exon targeted in this knockout model was the first ATG exon, only the longer SPAG9 isoforms are affected in knockout mice, leaving shorter isoforms such as JIP4 and SPAG9 intact. This suggests that the functions of these proteins in cellular transport, cytokinesis and myogenesis may not be affected by the specific loss of JLP or that JLP function can be compensated by the presence of other isoforms. JLP-deficient male mice are subfertile and produce approximately 60% as many spermatozoa as wild-type mice, while JLP-deficient sperm show reduced motility. JNK activity is reduced in the testes of these animals, but remains intact in the brain (25). The mechanisms underlying this phenotype are not well understood, but studies of the SPAG9 isoform during fertilization suggest that SPAG9 may have a role in spermatozoa adherence to the oocyte or in the subsequent fertilization process (5). Successful fertilization requires a precise series of cell-cell interactions between gametes, and the sperm surface is covered by a continuous plasma membrane that is divided into distinct domains in which functional molecules are distributed associated with motility, energy production and spermatozoa-egg interacting regions. Sperm surface molecules on the plasma membrane are involved in the recognition of the zona pellucida protein during the early process of spermatozoa-egg interaction, and the acrosome region to which SPAG9 is localized, participates in the acrosome reaction necessary for fertilization. During this reaction, the membrane surrounding the acrosome fuses with the plasma membrane of the sperm and exposes the acrosomal contents responsible for breaking through the oocyte surface [reviewed by (40)].
Putative Mechanism
The stop codon introduced by the dazzle mutation affects all known protein isoforms of the Spag9 gene, and is located prior to the putative transmembrane domain and the CTD. It is unknown if truncated protein is expressed by the mutated Spag9 gene as nonsense-mediated decay often occurs for nonsense alleles. If truncated protein is expressed, it is possible that it may retain some function or may even have dominant-negative effects as many of the protein-interacting domains occur prior to the introduced stop codon. No coat color phenotype has been described for the JLP-deficient mouse, but the knockout only affects certain SPAG9 isoforms (25). It is possible that unaffected isoforms may be functionally redundant with JLP or have a specific function in pigmentation that remains uncovered by JLP deficiency. An alternative explanation is that the expression of truncated SPAG9 isoforms may affect the function of other JIP proteins. However, studies using the JIP4 isoform in cell culture, which is unable to promote JNK activation, suggest that JIP4/JLP does not interact with other JIP family members and that dominant-negative isoforms would have little effect on the function of JIP1, 2 and 3 (2). Like JLP-deficient mice, dazzle animals appear to be subfertile.  
As SPAG9 isoforms can interact with proteins such as KLC1, ARF6 and PIKfyve that are involved in types of vesicular trafficking, it is possible that they may also interact with proteins involved in regulating the trafficking of melanosomes or melanosomal proteins necessary for normal pigmentation. Melanins, the pigments for skin, hair and eyes, are synthesized in melanosomes, which are then transported along microtubule and actin filaments to the ends of melanocyte dendrites and exported to neighboring keratinocytes. Like ARF6, Rab27a (mutated in concrete),  Rab32 and Rab38 are GTPases, and are involved in the proper trafficking of melanosomes. Mice deficient in either Rab27a or Rab38 display reduced pigmentation similar to the phenotype observed in dazzle animals (41-43). Along with the actin-based motor protein myosin Va (mutated in new gray)  and melanophilin (mutated in koala), which is thought to link melanosome-bound Rab27a to myosin Va, Rab27a regulates the microtubule to actin filament transfer of melanosomes, a crucial step leading to melanosome exocytosis (44-46). Rab32 and Rab38 are partially functionally redundant and control melanosomal transport and docking, as well as the proper sorting of the melanogenic enzymes, Tyrp1 and tyrosinase (mutated in ghost), into melanosomes (42;43).  One of the incidental mutations identified by whole genome sequencing in a homozygous dazzle mouse is located within the Rab38 gene. However, the mapping data exclude the Rab38 alteration as the phenotype-causing mutation. A predictive tool (Polyphen) assessing the Rab38 amino acid change present in this mouse suggests that this change is not likely to impact protein function.
Although p38 and JNK have not been implicated in mouse pigmentation, p38 may have a role in inducing melanogenesis in human melanocytes.  Interestingly, lipopolysaccharide (LPS) obtained from pathogenic bacteria, is able to induce pigmentation in cultured melanocytes in a p38-dependent manner by upregulating the expression of the transcription factor microphthalmia-associated transcription factor (MITF) and its downstream target the melanogenic enzyme, tyrosinase  (47). The LPS receptor Toll-like receptor (TLR4; mutated in lps3) is able to bind JIP3 through its cytoplasmic domain, increasing the efficiency of LPS-induced JNK activation. JIP3 is also able to bind to TLR2 (see languid) and TLR9 (see CpG1), but not to TLR1 or TLR6 (see insouciant) (48). TLR interactions with SPAG9 isoforms have yet to be reported.
Primers Primers cannot be located by automatic search.
Dazzle genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide transition. 
PCR program
1) 95°C             2:00
2) 95°C             0:30
3) 56°C             0:30
4) 72°C             1:00
5) repeat steps (2-4) 29X
6) 72°C             7:00
7) 4°C               8
Primers for sequencing
The following sequence of 743 nucleotides (NCBI Mouse Genome Build 37.1, Chromosome 11, bases 93,954,784 to 93,955,526) is amplified:
ttcctctgcc tactcaggaa cagcagaaag aatttaaaaa tcaagaagag ttgtccagtc
aggtctggat ctgtaccagc acccactcaa ctacaaaggt tatcatcatt gatgctgttc
agcctggcaa catcctagat agtttcactg tttgcaattc tcatgttctg tgcattgcca
gtgtcccagg taaagcaaaa ctctgtctcc tccctccctg tgttccctgc acctcctttt
cggtagagat gggtgtggag ttatggaaat tgataagagt tgtgttcctg tgtttgtggt
tacagacttt atgttctagt aagagatgat ttaaaataga ctagaagcca ttgaataact
atgtttttta gctatttaaa aatcatgtga agagtatttt aaaaaatggc actatgattc
cgcgtggtag tggcacacac ctttaatccc agcacttggg aggtagaggc agggtctctg
tgagttcaaa gccctgtagt tcagggctac atagagaagc cctgtctcga aaacaaatgg
caatatgtag ggagagctgg ctgatagagg aggaggaagg ccagctgaag tcctggaatc
aatatgtaaa attaaaacag tagttgagaa atattagtca gccataaaag aaatgacact
cttccccgcc ccaccccctg ttatttcatg tggttgaaat actttttttt tttccccttg

agacaaagtt agcctgaacc aca         

Primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated C is indicated in red.
Science Writers Nora G. Smart
Illustrators Diantha La Vine
AuthorsAmanda L. Blasius Bruce Beutler
Edit History
2011-06-15 3:16 PM (current)
2011-06-15 3:16 PM
2011-04-19 10:18 AM
2011-02-01 9:29 AM
2010-11-30 11:12 AM
2010-11-12 8:37 AM
2010-11-08 11:15 AM
2010-11-08 9:55 AM
2010-09-14 12:12 PM
2010-05-12 9:03 AM
2010-05-12 9:03 AM
2010-05-12 9:02 AM
2010-05-12 8:58 AM
2010-05-12 8:54 AM
2010-05-12 8:51 AM
2010-04-19 1:31 PM
2010-04-18 10:23 AM
2010-04-15 11:54 AM
2010-04-15 11:47 AM
2010-04-15 11:07 AM
2010-04-15 10:41 AM