The Effect Of N-terminal Deletions And CADASIL-related Missense Mutations On Biological Properties Of Notch3

Notch3 signaling is essential for normal development and homeostasis in a wide rangeof tissues. In the adult, Notch3 is expressed primarily in arterial smooth muscle and isinvolved in signaling events that control cell fate decisions. Notch3 proteins areconserved typeⅠmembrane proteins that contain 34 epidermal growth factor-like repeats(EGFR) in their extracellular domains and many functional regions in their intracellulardomains. Extracellular domains specifically bind to ligands, and intracellular domainsregulate signaling. Notch3 proteins undergo a series of complex proteolytic processingevents which are prerequisites for signaling. After translation, Notch3 is targeted to theendoplasmic reticulum (ER) and Golgi, where it undergoes proteolytic processingresulting in cleavage at the S1 site in the Golgi apparatus by a furin-like convertase.Notch3 signaling induces further changes in the protein. Upon binding to classicligands (Delta and Jagged), Notch3 undergoes extracellular cleavage at the S2 site byTNF-α-converting enzyme (TACE). The C-terminal product of this event is anintermediate that undergoes further proteolysis at the S3 site within the transmembranedomain byγ-secretase to release the Notch3 intracellular domain (NICD), which translocates to the nucleus and, along with transcription factors such as CLS, regulatestranscriptional activity of target genes, such as the hairy/enhancer of split (HES) genes.Comparison of the putative signal peptide sequences of the Notch family demonstratesthat the Notch3 N-terminal sequence diverges from other Notch proteins and containsmultiple strings of arginines, prolines and leucines which are conserved across severalmammalian species, suggesting a significant biological function.Mutations in the Notch3 gene cause cerebral autosomal dominant arteriopathy withsubcortical infarcts and leukoencephalopathy (CADASIL), which is currently consideredthe most common hereditary form of vascular dementia. Other symptoms of CADASILinclude migraine and stroke. Histopathological examination of CADASIL patientsreveals a systemic arteriopathy, with vascular smooth muscle cell (SMC) degenerationand deposition of extracellular granular osmiophilic material (GOM) whose compositionremains to be characterized.CADASIL mutations are typically missense mutations leading to gain or loss of acysteine residue within an EGF repeat domain, thus altering the number of cysteineresidues within a given domain from six to either five or seven. However, the molecularmechanism of CADASIL pathogenesis remains unclear.Given the uncertain function of the N-terminal domain and unknown mechanism ofCADASIL, we carried studies with the following aims: (1) to elucidate the biologicalsignificance of the N-terminal sequence of Notch3 protein; (2) to address potentialmolecular mechanisms for CADASIL. For these purposes, N-terminal deletions andCADASIL-related mutants were constructed and tested for protein expression,processing, subcellular distribution and signaling.1 The effect of N-terminal deletions on biological properties of Notch31.1 Construction of△14-Notch3,△39-Notch3 and Sec-△39-Notch3 According to the characteristics of the N-terminus sequence of Notch3, we generatedthree mutant Notch3 cDNAs that were altered at the N-terminus. The first clone lacksthe N-terminal 14-aa (△14-Notch3) and deletes the string of arginine residues, utilizingthe second methionine of Notch3 for translational initiation. Comparisons between thisclone and the wild-type clone enable us to decipher the significance of the arginine-richfirst 14 amino acids. The second clone lacks the N-terminal 39-aa (△39-Notch3), whichis predicted to delete the entire predicted signal peptide including the string of leucinesclose to the predicted site of signal cleavage. Comparisons between this clone and thewild-type clone enable us to determine whether the first 39 amino acids are required forprotein targeting and signaling. Finally, we replaced the N-terminal 39 amino acidswith a consensus secretion signal from IgKappa(Sec-△39-Notch3). Comparison of thiscontruct to the wild-type allows us to determine the function of the wild-type N-terminalsequence relative to a consensus secretion signal.1.2 The effect of Notch3 N-terminal deletions on expression, processing andsubcellular distributionAnalysis of immunostaining, Western blotting and cytosolic distribution demonstratedthat, in comparison to WT-Notch3,△14-Notch3 had similar protein expression andprocessing, however, there was great increase in the cytosolic distribution;△39-Notch3notably decreased protein expression and processing, and had significantly morecytosolic Notch3; Sec-△39-Notch3 had similar protein expression, but had improvedprotein processing and less cytosolic Notch3.1.3 The effect of N-terminal deletions on signaling, pro-growth and anti-apoptoticactivityWe used a co-culture assay to determine the influence of N-terminal deletions onsignaling. In this assay, Notch3 plasmids were cotransfected with HES1-luciferase andcocultured with L cells stably expressing the Notch ligand Jaggedl or Deltal or withwild type of L cells. The amount of luciferase produced corresponded to the strength ofNotch3 signaling. The results showed that transfected△14-Notch3 and Sec-△39-Notch3 had ligand stimulated activities that were similar to wildtype Notch3. However,△39-Notch3 showed the weakest amount of activity relative to all other Notch3constructs.Previous studies showed that normal Notch3 signaling promoted the growth of SMCand was a critical determinant of SMC survival and vascular structure. To furthercompare the pro-growth activity of WT and its mutations in SMC, A7r5 cells weretransfected with WT-Notch3 or Notch3 mutants and subjected to quantitation of total cellnumber. The results indicated that WT-Notch3,△14-Notch3 and Sec-△39-Notch3significantly promoted cell growth, while△39-Notch3 did not have a significant effect oncell number after transfection.Notch3 signaling negatively modulate the sensitivity of SMC to apoptotic inducer ofFas ligand (Fas-L). We thus used the sensitivity of SMC to Fas ligand as another measureof functional Notch3 signaling. We exposed cells to Fas ligand after tranfection withwildtype and N-terminal mutants of Notch3 and determined the amount of cell death by anuclear staining. Apoptotic cells were stained using propidium iodide and quantitatedby fluorescent microscopy. The results indicated that the WT-Notch3,△14-Notch3 andSec-△39-Notch3 effectively protected A7r5 cells against apoptosis induced by Fas-L,while△39-Notch3 showed a much smaller protective effect.1.4 Cytosolic Notch3 interacts with PSMA1 and inhibits proteasome function.The highly conserved arginine-rich initial 14 amino acids are important for decreasingcytosolic Notch3 expression, which suggested that there may be important molecularramifications of ectopic expression of cytoplasmic Notch3. A yeast two-hybrid screenand coimmunoprecipitation identified a physical interaction between cytosolic Notch3and proteasome subunit PSMA, Analysis of proteasome function indicated that cytosolicNotch3 inhibits proteasome function. There was excellent correlation between theamount of cytoplasmic Notch3 produced, complex formation between PSMA1 andNotch3, and the level of proteasome inhibition.Conclusion: The principle findings in this study are: (1) The N-terminal 39 amino acid residues,which contain multiple strings of arginines, prolines and leucines, are required forNotch3 receptor expression, processing, subcellular distribution and biological activity;(2) The N-terminal 14 amino acid residues, which include seven arginines, do not appearto enhance function, yet are required to reduce ectopic cytoplasmic expression of Notch3;(3) Cytoplasmic Notch3 binds specifically to the proteasome subunit PSMA1, andincreases cytoplasmic expression of Notch3, resulting in inhibition of proteasome activity.Our findings support a multifunctional role for the conserved N-terminal sequence ofNotch3: targeting of the protein to the secretory pathway and reduction of cytoplasmicNotch3 expression may pathologically inhibit cytoplasmic functions.2 The effect of CADASIL-related missense mutations on biologicalproperties of Notch32.1 Construction of CADASIL-related mutantsTwo CADASIL-related mutants (C49Y and R90C), which represent removing andinserting cysteine residues within epidermal growth factor repeats in the extracellulardomain of the Notch3 receptor, were constructed and tested for expression, processing,subcellular distribution, signaling, pro-growth and anti-apoptotic activity.To examine the signaling capacity of wild-type Notch3 and CADASIL-associatedmutants in a more specific and sensitive way, we took advantage of a recently developedsystem in which Notch3 receptor activation can be measured by replacing the Notch3intracellular domain with GV signaling domain (Ga14 binding domain-VP16 activedomain). After the S3 cleavage during Notch3-CV activation, the GV translocates to thenucleus, where it specifically activates transcription of reporter gene (pFR-Luciferase),which contains a Ga14 responsive luciferase reporter.2.2 The effect of R90C and C49Y on expression, processing and subcellulardistribution of protein Immunostaining showed that WT-Notch3, C49Y and R90C expressed at comparablelevels. Quantification of the ratio of processed vs. unprocessed protein by Westernblotting indicated that WT-Notch, C49Y and R90C had similar expression and processingefficiency; cytosolic distribution assays showed that WT-Notch3, C49Y and R90C hadcomparable cytosolic proteins.2.3 The effect of R90C and C49Y mutations on Notch3-mediated signalingWe used a highly-specific system to determine the influence of CADASIL-associatedmutants on signaling. In this assay, Notch-GV constructs (WT-Notch3-GV, R90C-GVand C49Y-GV) were cotransfected with pFR-luciferase reporter which could beactivated only by GV signaling domain. The amount of luciferase produced correspondedto the strength of Notch3 signaling from transfection. The results indicated thatR90C-GV and C49Y-GV showed significantly weaker amount of activity relative toWT-Notch3-GV. R90C and C49Y inhibited notably the signaling activity ofWT-Notch3-GV after they were cotransfected into smooth muscle cells, which suggestedCADASIL-associated mutants had dominant-negative function on WT-Notch3 proteinwhen they co-expressed. The results provided a new clue for the elucidation ofCADASIL pathogenesis.2.4 The effect of R90C and C49Y on pro-growth and anti-apoptosis activity inSMCTo compare the effect of R90C and C49Y on the growth of SMC, we quantitated totalcell number of A7r5 transfected with WT-Notch3 or Notch3 mutants through LDHactivity assay. The result indicated that WT-Notch3 significantly promoted cell growth,while R90C and C49Y had less effect. The results provided a new clue for theelucidation of CADASIL pathogenesis.To determine the effect of R90C and C49Y on the sensitivity of SMC to apoptosis, weexposed A7r5 cells to Fas ligand after tranfection with WT-Notch3 or Notch3 mutantsand determined the amount of cell death by a nuclear staining. The results indicated thatthe WT-Notch3 could effectively protect A7r5 cells against apoptosis induced by Fas-L, while R90C and C49Y lost protective effect. In this study, two typicalCADASIL-associated mutations were studied. Considering the variation ofCADASIL-related mutations, it will be helpful to include more mutants for the completeelucidation of CADASIL pathogenesis.Conclusion:The principle findings in this study are: (1) CADASIL-associated mutants (R90C andC49Y) had no influence on receptor expression, processing, subcellular distribution; (2)R90C and C49Y decreased signaling activity. Moreover, these mutants displayed anegative-dominant function on WT-Notch3 signaling when co-expressed; (3) R90C andC49Y have impaired function, measured by the ability to stimulate A7R5 cellsproliferation and to protect A7r5 cells against apoptotic inducer of Fas-L. Our findingsprovided clues for the elucidation of CADASIL pathogenesis.