(G2019S) LRRK2 activates MKK4-JNK pathway and causes degeneration of SN dopaminergic neurons in a transgenic mouse model of PD

Abstract

(G2019S) mutation of leucine-rich repeat kinase 2 (LRRK2) is the most common genetic cause of both familial and sporadic Parkinson's disease (PD) cases. Twelve- to sixteen-month-old (G2019S) LRRK2 transgenic mice prepared by us displayed progressive degeneration of substantia nigra pars compacta (SNpc) dopaminergic neurons and parkinsonism phenotypes of motor dysfunction. LRRK2 is a member of mixed lineage kinase subfamily of mitogen-activated protein kinase kinase kinases (MAPKKKs). We hypothesized that (G2019S) mutation augmented LRRK2 kinase activity, leading to overphosphorylation of downstream MAPK kinase (MKK) and resulting in activation of neuronal death signal pathway. Consistent with our hypothesis, (G2019S) LRRK2 expressed in HEK 293 cells exhibited an augmented kinase activity of phosphorylating MAPK kinase 4 (MKK4) at Ser(257), and protein expression of active phospho-MKK4(Ser257) was upregulated in the SN of (G2019S) LRRK2 transgenic mice. Protein level of active phospho-JNK(Thr183/Tyr185) and phospho-c-Jun(Ser63), downstream targets of phospho-MKK4(Ser257), was increased in the SN of (G2019S) LRRK2 mice. Upregulated mRNA expression of pro-apoptotic Bim and FasL, target genes of phospho-c-Jun(Ser63), and formation of active caspase-9, caspase-8 and caspase-3 were also observed in the SN of (G2019S) LRRK2 transgenic mice. Our results suggest that mutant (G2019S) LRRK2 activates MKK4-JNK-c-Jun pathway in the SN and causes the resulting degeneration of SNpc dopaminergic neurons in PD transgenic mice.

Figure 1

Expression of mutant (G2019S) or wild-type LRRK2HA in the brain of transgenic mouse. (a) Western blot analysis using anti-HA antiserum indicated that mutant (G2019S) LRRK2HA or wild-type LRRK2HA (MW=∼280 kDa) was expressed in the SN, cerebral cortex (CTX), or cerebellum (CB) of a 10-month-old transgenic mouse expressing mutant (G2019S) (line 32) or wild-type LRRK2 (line 36). (b) Double immunofluorescence staining using anti-TH antiserum and anti-HA antibody showed that HA-tagged (G2019S) or wild-type LRRK2 was expressed in the cytoplasm of SNpc TH-positive dopaminergic neurons of a 10-month-old (G2019S) LRRK2 (line 32) and LRRK2 transgenic mouse (line 36), respectively. Scale bar is 10 μm

Figure 2

(G2019S) LRRK2 transgenic mice exhibit a progressive and selective degeneration of SNpc dopaminergic neurons. (a) Immunohistochemical TH staining showed that compared with a 12-month-old wild-type mouse, age-matched (G2019S) LRRK2 mice (line 32) exhibited a reduction in the number of SNpc TH-positive dopaminergic neurons. Neuronal death of SNpc dopaminergic cells progressively deteriorated in 16-month-old (G2019S) LRRK2 transgenic mice. Note that degeneration of SNpc dopaminergic neurons was not observed in a 16-month-old transgenic mouse expressing wild-type LRRK2 (line 43). Each bar represents the mean±S.E. value of six mice. (b) Western blot analysis using AT8 monoclonal antibody demonstrated an upregulated protein expression of phospho-tau^{Ser202/Thr205} in the SN of a 12-month-old (G2019S) LRRK2 mouse (line 32). Note that protein level of phospho-tau^{Ser202/Thr205} was not altered in the SN of LRRK2 transgenic mouse (line 43). Each bar shows the mean±S.E. value of five mice. (c) Immunocytochemical staining of neuronal marker NeuN showed that compared with an age-matched wild-type mouse, significant neuronal death was absent in the striatum of a 12-month-old (G2019S) LRRK2 transgenic mouse (line 51). Each bar represents the mean±S.E. value of five mice. Scale bar is 100 μm. *P<0.01 compared with wild-type mice

Figure 3

The density of dopamine transporters or TH staining is reduced in the striatum of (G2019S) LRRK2 transgenic mice. (A) In-vivo ^99mTc-TRODAT microSPECT imaging was performed to visualize the density of dopamine transporters in the striatum. The uptake of ^99mTc-TRODAT is normal (red color scale) in the striatum (ST) of a wild-type mouse (d–f), but it is markedly reduced (green color scale) in a 12-month-old (G2019S) LRRK2 transgenic mouse (line 32) (g–i) (a–c: VOI definition of bilateral striatum and cerebellum (CB); d–i: fusion imaging of microCT and microSPECT; a, d, g: coronal section; b, e, h: sagittal section; c, f, i: horizontal section). (B) Compared with 12-month-old wild-type mice, specific uptake ratios of ^99mTc-TRODAT were significantly reduced in (G2019S) LRRK2 transgenic mice at the same age. Each bar shows the mean±S.E. value of eight mice. *P<0.01. (C) Compared with 16-month-old control mouse or transgenic mouse expressing wild-type LRRK2 (line 43), density of TH-immunoreactive staining was decreased in the striatum of a (G2019S) LRRK2 mouse (line 32). (D) Optical density of striatal TH-positive terminals was significantly reduced in (G2019S) LRRK2 transgenic mice. Each bar represents the mean±S.E. value of four mice. *P<0.01

Figure 4

(G2019S) LRRK2 transgenic mice display motor deficit of parkinsonism, which is rescued by ℒ-DOPA. (a) Compared with 12-month-old wild-type mice, (G2019S) LRRK2 transgenic mice at the same age exhibited a significant decrease in spontaneous ambulatory activity. Severity of hypoactivity progressively increased in (G2019S) LRRK2 mice at 16 months of age. In contrast, 12- or 16-month-old LRRK2 transgenic mice did not display the symptom of hypoactivity. Each bar shows the mean±S.E. value of 10–12 mice. (b) Pole test indicated that compared with wild-type mice or transgenic mice expressing wild-type LRRK2, 12- or 16-month-old (G2019S) LRRK2 transgenic mice took a significantly longer time to perform the pole test and exhibited an impaired motor performance. Each bar represents the mean±S.E. value of eight mice. *P<0.01 compared with wild-type mice. (c) Compared with saline-treated wild-type mice, 12-month-old (G2019S) LRRK2 mice injected with saline displayed a significant reduction in open-field locomotor activity measured by the distance traveled during the test period. Methyl ℒ-DOPA (2 mg/kg of body weight) was intraperitoneally injected into animals, and locomotor activity was measured 40 min after the injections. Note that methyl ℒ-DOPA rescued the hypoactivity exhibited by (G2019S) LRRK2 transgenic mice. Each bar represents the mean±S.E. value of seven mice. *P<0.01 compared with saline-injected wild-type mice. ^#P<0.01 compared with saline-injected (G2019S) LRRK2 mice

Figure 5

(G2019S) LRRK2 enhances the phosphorylation of MKK4 at Ser²⁵⁷. (a) Western blot study using anti-HA antiserum showed that a similar protein level of HA-tagged wild-type or mutant (G2019S) LRRK2 was stably expressed in HEK 293 cells. (b) FLAG-tagged MKK4 was transiently transfected into control 293 cells or HEK293 cells stably expressing HA-tagged wild-type or mutant (G2019S) LRRK2. Then, FLAG-tagged MKK4 was immunoprecipitated by anti-FLAG antibody, and phosphorylated MKK4 at Ser²⁵⁷ serine was visualized by performing immunoblotting analysis using anti-phospho-MKK4^Ser257 antibody. As a negative control, addition of FLAG peptide completely blocked immunoprecipitation of FLAG-tagged MKK4. Compared with control 293 cells or HEK 293 cells expressing wild-type LRRK2, immunoprecipitated protein level of phospho-MKK4^Ser257was significantly increased in 293 cells expressing mutant (G2019S) LRRK2. Note that anti-FLAG antibody also co-immunoprecipitated HA-tagged wild-type or (G2019S) LRRK2. Each bar shows the mean±S.E. value of five experiments. *P<0.01 compared with control 293 cells

Figure 6

(G2019S) LRRK2 activates MKK4-JNK-cJun pathway and upregulates mRNA expression of Bim or FasL in the SN of PD transgenic mice. (a) Western blot study showed an upregulated expression of active phospho-MKK4^Ser257 in the SN of a 12-month-old (G2019S) LRRK2 mouse (line 51). Protein levels of phospho-JNK^{Thr183/Tyr185} and phospho-c-Jun^Ser63, which act as downstream effectors of active MKK4, were also greatly increased in the SN of (G2019S) LRRK2 transgenic mouse. Note that protein expression of phospho-MKK4^Ser257, phospho-JNK^{Thr183/Tyr185} or phospho-c-Jun^Ser63 was not altered in the SN of a 12-month-old transgenic mouse expressing wild-type LRRK2 (line 36). Each bar represents the mean±S.E. value of six 12-month-old mice. (b) Real-time RT-PCR assay showed that compared with wild-type mice or transgenic mice expressing wild-type LRRK2, Bim or FasL mRNA expression was greatly upregulated in the SN of 12-month-old (G2019S) LRRK2 transgenic mice. Each bar shows the mean±S.E. value of 5–7 experiments. *P<0.01 compared with wild-type mice

Figure 7

Active caspase-9, caspase-8 and caspase-3 were expressed in the SN of (G2019S) LRRK2 transgenic mice. (a) Western blot analysis showed that protein expression of cleaved active caspase-9 (MW=37 kDa), active caspase-8 (MW=18 kDa) or active caspase-3 (MW=17 kDa) was significantly upregulated in the SN of a 12-month-old (G2019S) LRRK2 mice (line 32). Note that protein level of active caspase-3, caspase-8, or caspase-9 was not altered in the SN of LRRK2 transgenic mice (line 36) at the same age. Each bar shows the mean±S.E. value of six mice. *P<0.01 compared with wild-type mice. Double immunofluorescence staining was performed with anti-TH antiserum and anti-active caspase-9 (b) or anti-active caspase-3 (c) antibody. Confocal microscopy showed that active caspase-9 (b) or caspase-3 (c) was found in TH-positive SN dopaminergic neurons of a 12-month-old (G2019S) LRRK2 mouse (line 32). In contrast, active caspase-9 (b) or caspase-3 (c) was not observed in TH-positive SN dopaminergic cell of LRRK2 transgenic mice (line 36) at the age of 12 months. Scale bar is 10 μm