Novel Anti-Neuroinflammatory Properties of a Thiosemicarbazone-Pyridylhydrazone Copper(II) Complex

Abstract

Neuroinflammation has a major role in several brain disorders including Alzheimer's disease (AD), yet at present there are no effective anti-neuroinflammatory therapeutics available. Copper(II) complexes of bis(thiosemicarbazones) (Cu^II(gtsm) and Cu^II(atsm)) have broad therapeutic actions in preclinical models of neurodegeneration, with Cu^II(atsm) demonstrating beneficial outcomes on neuroinflammatory markers in vitro and in vivo. These findings suggest that copper(II) complexes could be harnessed as a new approach to modulate immune function in neurodegenerative diseases. In this study, we examined the anti-neuroinflammatory action of several low-molecular-weight, charge-neutral and lipophilic copper(II) complexes. Our analysis revealed that one compound, a thiosemicarbazone-pyridylhydrazone copper(II) complex (CuL⁵), delivered copper into cells in vitro and increased the concentration of copper in the brain in vivo. In a primary murine microglia culture, CuL⁵ was shown to decrease secretion of pro-inflammatory cytokine macrophage chemoattractant protein 1 (MCP-1) and expression of tumor necrosis factor alpha (Tnf), increase expression of metallothionein (Mt1), and modulate expression of Alzheimer's disease-associated risk genes, Trem2 and Cd33. CuL⁵ also improved the phagocytic function of microglia in vitro. In 5xFAD model AD mice, treatment with CuL⁵ led to an improved performance in a spatial working memory test, while, interestingly, increased accumulation of amyloid plaques in treated mice. These findings demonstrate that CuL⁵ can induce anti-neuroinflammatory effects in vitro and provide selective benefit in vivo. The outcomes provide further support for the development of copper-based compounds to modulate neuroinflammation in brain diseases.

Keywords: Alzheimer’s disease; copper; inflammation; microglia.

Figure 1

Chemical structures of tested copper(II) compounds Cu^II(gtsm), Cu^II(atsm), CuL¹⁻⁵.

Figure 2

CuL⁵ treatment exhibited inflammation-modulating effect. (a) Workflow of the screening pipeline to identify leading compound(s). (b,c) M17 cultures were incubated for 24 h with new-generation metal compounds (2–10 μM). (b) Cell viability and (c) cell–copper association were determined by LDH assay and ICP–MS, respectively. (d) Primary murine microglial cultures were stimulated with TNFα (10 ng/mL) and IFNγ (15 ng/mL) and concurrently treated with 20 μM minocycline or varying concentrations of CuL¹, CuL^2, CuL³, CuL⁴, or CuL⁵ for 24 h. MCP-1 secretion by microglia was measured by ELISA. Data are expressed as (b) percentage LDH released relative to Triton X-100 (TX 100)-treated positive control, (c) μg copper per g cellular protein, and (d) percentage of MCP-1 released relative to TNFα- and IFNγ-stimulated positive control and presented as the mean ± S.E.M. Dotted line represents (b) average % LDH release of the media-only negative control and (d) average % MCP-1 released when TNFα- and IFNγ-stimulated microglia were treated with minocycline. Determined by one-way ANOVA, compared to (b,c) media-only or (d) TNFα- and IFNγ-stimulated controls ***/**** = p ≤ 0.001/0.0001. Numbers on x-axes represent μM compound. n/a = concentration not assessed.

Figure 3

CuL⁵ treatment modulated microglial phagocytic activity. Primary murine microglial cultures were stimulated with TNFα (10 ng/mL) and IFNγ (15 ng/mL) and concurrently treated with 10 µM cytochalasin D, 20 µM minocycline, 1 µM CuL⁵, or 2 µM CuL⁵ for up to 24 h. (a) 1 h post-stimulation and treatment, media were removed and replaced with 100 µg/mL pHrodo red E.coli bioparticles and fluorescence emission was measured using the IncuCyte imaging platform every 2 h. (aii–avi) Representative images from one experiment showing phase contrast and red channel signal with respective conditions. An actin inhibitor, cytochalasin D (10 µM) was incubated 1 h prior to the addition of E.coli bioparticles to validate pHrodo-labeled uptake. Scale bar = 200 µm. (b) The 24 h post-stimulation and treatment, RNA was extracted from cells for cDNA synthesis. qRT-PCR was used to determine mRNA expression levels of (bi) Tnf, (bii) Mt1, (biii) Trem2 and (biv) Cd33. Results obtained were normalised to mRNA expression level of Hprt. (ai) Data are expressed as the average red object count (1/image) of pHrodo-labeled microglial cells in each condition. Three images per well from three technical replicates were analysed. (b) Representative data set is expressed as fold induction relative to media-only control. Data are presented as the mean ± S.E.M. Determined by one-way ANOVA, compared to (bi) media-only or (ai,bii–biv) TNFα- and IFNγ-stimulated controls, */**/**** = p ≤ 0.05/0.01/0.0001.

Figure 4

CuL⁵ treatment increased microglial phagocytosis and Aβ load in 5xFAD mice. (a) Illustration of experimental workflow, created with BioRender.com. SSV or CuL⁵ (30 mg/kg) were delivered to non-transgenic (WT) or 5xFAD mice by oral gavage, once daily, for 1 or 8 weeks. (b) Brain samples collected from a mixed cohort of WT (n = 10) and 5xFAD (n = 5) animals treated for 1 week were analysed for copper content by ICP–MS (n = 5–10 per treatment group). Individual mice treated for 8 weeks were injected with Methoxy-X04 prior to the collection of brain samples for (c,d) ex vivo isolation of microglia and analysed for the proportion of X04⁺ and X04⁻ microglia by FACS (n = 7–8 per treatment group) and (e–g) the analysis of Aβ plaque load and microgliosis by immunohistochemistry (n = 7–8 per treatment group). (b) Brain copper content measured by ICP–MS is expressed as copper content (μg) per g brain tissue. (c) Representative FACS plot for microglia sorted from WT and 5xFAD mice, from each treatment group. (d) Data obtained are expressed as percentage of X04⁺ microglia population in total microglia sorted. (e) Representative immunofluorescence image of the hippocampus of WT and 5xFAD mice, from each treatment group, injected with Methoxy-X04 and stained for Iba1 (AlexaFluor 488). The (f) number of plaques and (g) percentage area of plaques in the hippocampal region were determined using ImageJ. (b,d,f,g) Data for each treatment group are presented as the mean ± S.E.M per treatment group. Determined by one-way ANOVA, */**/***/**** = p ≤ 0.05/0.01/0.001/0.0001.

Figure 5

Effect of CuL⁵ treatments on 5xFAD cognitive functions. (a) Timeline for behavioural testing of CuL⁵-treated mice. SSV or CuL⁵ (30 mg/kg) were delivered to non-transgenic (WT) or 5xFAD mice by oral gavage, once daily, for 8 weeks while also being to a battery of behavioural tests to assess the effect of treatment (n = 6–7 per treatment group). (b) Escape latency in the Morris water maze during the training phase. (c) Time spent in the former platform quadrant (P) and quadrats opposite (O), to the left (L), and to the right (R) of the platform during the probe trial. (d) Nest quality scores. (e) Choice accuracy in igloo test. (f) Perseverative errors in igloo test. Data for each treatment group are presented as the mean ± S.E.M per treatment group. Morris water maze data and choice accuracy in the igloo test were analysed with two-way ANOVA for repeated measure with Bonferroni pairwise comparisons. Perseverative errors in the igloo test and nest building scores estimated by Poisson regression as incidence rate ratios (IRRs), and by logistic regression as odds ratios (ORs), respectively. */**/*** = p ≤ 0.05/0.01/0.001.