Neuronal Dysfunction and Behavioral Abnormalities Are Evoked by Neural Cells and Aggravated by Inflammatory Microglia in Peroxisomal β-Oxidation Deficiency

Abstract

It is becoming evident that microglia, the resident immune cells of the central nervous system (CNS), are active contributors in neurological disorders. Nevertheless, the impact of microgliosis on neuropathology, behavior and clinical decline in neuropathological conditions remains elusive. A mouse model lacking multifunctional protein-2 (MFP2), a pivotal enzyme in peroxisomal β-oxidation, develops a fatal disorder characterized by motor problems similar to the milder form of human disease. The molecular mechanisms underlying neurological decline in men and mice remain unknown. The hallmark of disease in the mouse model is chronic proliferation of microglia in the brain without provoking neuronal loss or demyelination. In order to define the contribution of Mfp2^-/- neural cells to development of microgliosis and clinical neuropathology, the constitutive Mfp2^-/- mouse model was compared to a neural selective Nestin-Mfp2^-/- mouse model. We demonstrate in this study that, in contrast to early-onset and severe microgliosis in constitutive Mfp2^-/- mice, Mfp2^+/+ microglia in Nestin-Mfp2^-/- mice only become mildly inflammatory at end stage of disease. Mfp2^-/- microglia are primed and acquire a chronic and strong inflammatory state in Mfp2^-/- mice whereas Mfp2^+/+ microglia in Nestin-Mfp2^-/- mice are not primed and adopt a minimal activation state. The inflammatory microglial phenotype in Mfp2^-/- mice is correlated with more severe neuronal dysfunction, faster clinical deterioration and reduced life span compared to Nestin-Mfp2^-/- mice. Taken together, our study shows that deletion of MFP2 impairs behavior and locomotion. Clinical decline and neural pathology is aggravated by an early-onset and excessive microglial response in Mfp2^-/- mice and strongly indicates a cell-autonomous role of MFP2 in microglia.

Keywords: behavior; metabolic disorder; microglia; microgliosis; neuron-microglia communication; neurotransmission; peroxisomes; β-oxidation.

Figure 1

Early onset and extensive microgliosis in Multifunctional protein-2 (MFP2^−/−) vs. limited and delayed microgliosis in Nestin-Mfp2^−/− mice. (A,B) Age-dependent progression of microgliosis in brainstem of Mfp2^−/− mice. (A) Morphology of Mfp2^−/− microglia (Iba1⁺) at different time points. (B) Timeline showing progression of microglial transformation (see A), proliferation (see E) and F4/80 activation (based on immunohistochemistry (IHC), see Figure 2A) in Mfp2^−/− brain. (C,D) Age-dependent progression of microgliosis in brainstem of Nestin-Mfp2^−/− mice. (C) Morphology of microglia in Nestin-Mfp2^−/− brain at different time points. (D) Timeline showing progression of microglial transformation (see C), proliferation (see E) and F4/80 activation (based on IHC, see Figure 2A) in Nestin-Mfp2^−/− brain. (E) Quantification of Iba1⁺ cells (microglia) in Mfp2^−/−, Nestin-Mfp2^−/− and control brainstem. w, weeks. *Mfp2^−/− vs. control: *p < 0.05, ***p < 0.001. ^ФNestin-Mfp2^−/− vs. control: ^ФФФp < 0.001, ns, not significant. Error bars indicate standard error of the mean (SEM). n = 6–9/group. Mfp2^−/− mice all died at this time point.

Figure 2

Microglia in Nestin-Mfp2^−/− brain acquire a milder inflammatory state than Mfp2^−/− microglia. (A) Representative immunohistochemical staining for P2ry12 in inferior colliculus of 17-weeks-old Mfp2^−/− mice and age-matched Nestin-Mfp2^−/− mice relative to control mice. (B) Quantification of P2ry12 in 17-weeks-old Mfp2^−/− mice and Nestin-Mfp2^−/− mice relative to age-matched control mice in three brain regions. *Mfp2^−/− or Nestin-Mfp2^−/− vs. control: ***p < 0.001, ns, not significant. ^ФNestin-Mfp^2−/− vs. Mfp^2−/−: ^ФФФp < 0.001. n = 3–6/group. COLL, Inferior Colliculus, CTX, Visual Cortex, BS, Brainstem. (C) F4/80⁺ cells were counted in brainstem of Mfp2^−/−, Nestin-Mfp2^−/− and control mice. Graphs from control mice are not shown as they approach zero. w, weeks. *Mfp2^−/− vs. control: ***p < 0.001. ^ФNestin-Mfp2^−/− vs. control: ^ФФФp < 0.001, ns, not significant. Mfp2^−/− mice all died at this time point. n = 6–9/group. (D–G) Flow cytometric analysis of CD11b⁺ microglia isolated from 17-week-old Mfp2^−/− mice and 34-week-old Nestin-Mfp2^−/− mice using antibodies to pro-inflammatory markers F4/80 (D) and CD11c (E), and anti-inflammatory markers CD204 (F) and CD206 (G). Representative figures of multiple independent experiments are shown (Total: n = 4/group). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.001, ns, not significant. Error bars indicate SEM.

Figure 3

Microglia in Nestin-Mfp2^−/− brain are not primed. Nestin-Mfp2^−/− and control mice were challenged with i.p. lipopolysaccharides (LPS) or vehicle, and brainstem tissue was analyzed after 4 h. (A,B) Transcript levels of Tnfa (A) and Il1b (B) in 17-week-old Nestin-Mfp2^−/− and age-matched control mice after LPS treatment. (C,D) Transcripts levels of Tnfa (C) and Il1b levels (D) in 34-weeks-old Nestin-Mfp2^−/− brain compared to control brain after LPS treatment. *Nestin-Mfp2^−/− vs. control: *p < 0.05, **p < 0.01, ns, not significant. Error bars indicate SEM. n = 4–6/group.

Figure 4

Neuronal functioning is severely impaired in Mfp2^−/− vs. Nestin-Mfp2^−/− mice. (A,B) Transcript levels of neuronal Cx3cl and Cd200 markers in cortex (CTX) of 12-week-old Mfp2^−/− (A) and Nestin-Mfp2^−/− mice (B). (C,D) Transcript levels of neuronal Cx3cl and Cd200 markers in brainstem (BS) of 17-weeks-old Mfp2^−/− mice (C) and 34-week-old Nestin-Mfp2^−/− mice (D). Transcript levels were measured by qRT-PCR. n = 3–6/group. (E–H) Wave patterns of the auditory signal from different mouse models are shown. Arrow indicates appearance of acoustic stimulus. Scale bars are shown beneath the bottom tracing. (I–N) Brainstem responses on auditory stimulus in Mfp2^−/− and Nestin-Mfp2^−/− vs. control mice at 12 weeks of age. (I,L) Mean peak latencies show peaks assigned to different brain nuclei (peak 2–5). (J,M) Interpeak latencies between peak one to three and peak three to five. (K,N,Q) Mean peak amplitudes. (O–Q) Direct comparison between Mfp2^−/− and Nestin-Mfp2^−/− mice relative to respective control mice at 12 weeks of age. (O) Mean peak latency delays vs. control mice (i.e., 0.0 ms) at different stages of disease, (P) 1–3 and 3–5 interpeak latencies and (Q) mean peak amplitudes in Mfp2^−/− and Nestin-Mfp2^−/− mice (Controls are 100%). Significance level: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns, not significant. Error bars indicate SEM. n = 8–15 mice/group.

Figure 5

Early-onset disturbance in locomotor activity and exploration in Mfp2^−/− mice vs. normal activity and late-onset disturbance of explorative behavior in Nestin-Mfp2^−/− mice. (A,B,D,E) Locomotor activity shown by corner entries (A,D) and total path length (B,E) was analyzed at onset, mid and end stage of disease in Mfp2^−/− mice (A,B) and Nestin-Mfp2^−/− mice (D,E) compared to control mice. (C,F) Explorative behavior at different disease stages in Mfp2^−/− mice (C) and Nestin-Mfp2^−/− mice (F). (G–I) Direct comparison between Mfp2^−/− and Nestin-Mfp2^−/− mice relative to control mice. *Mfp2^−/− or Nestin-Mfp2^−/− vs. control (A–F), Nestin-Mfp2^−/− vs. Mfp2^−/− (G–I): *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.001, ns, not significant. Error bars indicate SEM. n = 8–24 mice/group. Mfp2^−/− mice all died at this time point.

Figure 6

Grip strength is decreased in Mfp2^−/− and Nestin-Mfp2^−/− mice. (A–D) Grip strength in front paws (A,C) and all paws (B,D) at different ages in Mfp2^−/− mice (A,B) and in Nestin-Mfp2^−/− mice (C,D). (E,F) Direct comparison between the knockout models relative to control mice. Significance level: *p < 0.05, **p < 0.01 ***p < 0.001; ****p < 0.0001, ns, not significant. Error bars indicate SEM. n = 4–9 mice/group.