Giuseppe Donato Ciccotosto

Background: We have previously shown that repeated oral inoculation of mice with bacteria that cause the chronic oral disease periodontitis resulted in brain infiltration and induction of Alzheimer's disease (AD)‐like pathology. Here, we examined whether purified bacterial membrane vesicles (BMVs) from the oral pathobiont Porphyromonas gingivalis or oral commensal Neisseria oralis could enter the brain and cause pathology, without the need for an infection of the brain. Methods: Three experimental groups of 36 female C57BL/6 mice, were injected intravenously with 10 µg of purified BMVs from P. gingivalis or N. oralis , or an equal volume of vehicle alone (Control), through the tail vein on Days 1, 5, 8 and 11. Six mice from each group were killed on Days 2, 6, 12, 26, 40 and 55, then PBS perfused before their brains dissected. Immunohistochemistry analyses of 3 hemibrains per group, per timepoint, were probed for the presence of the BMVs, neuronal damage, neuroinflammation, AD‐like pathology and immune system activation. Result: BMVs from both P. gingivalis and N. oralis were found within the brains after one injection. Detection of P. gingivalis MV positive cells peaked at two timepoints, at Days 6 and 26, whereas detection of N. oralis MV positive cells peaked at Day 26, 2 weeks after the final intravenous injection. By Day 55, the detection of cells positive for both BMVs had decreased to levels consistent with the BMVs being removed from the brains. There were similar levels of neuronal degeneration, Il‐6 and IL‐1β in the hippocampus between the three groups of mice over time, but only P. gingivalis MVs were able to induce microgliosis and astrogliosis in the hippocampus. There were no amyloid plaques detected in the brains, but phosphorylated Tau peaked significantly at Day 26 in the mice that received BMVs. Conclusion: BMVs from both N. oralis and P. gingivalis entered the brain from the bloodstream but caused different host responses. Upon cessation of BMV injection, the BMVs were naturally removed from the brain over time, suggesting that treatment of the source of the BMVs, namely periodontitis, would reduce exposure to brain pathology causing agents.

Autophagy is a highly-conserved mechanism that ensures the lysosomal degradation of disposable or potential toxic cytosolic entities in support of cell survival and adaptation to stress. Autophagy is deregulated in various pathological conditions including cardiovascular, neurological, neoplastic-autoimmune and degenerative disorders. However, clinically relevant pharmacological modulators of autophagy remain elusive, calling for the development of novel screening approaches that are amenable to high throughput/content applications. Here, we describe a simple method to detect autophagy in cultured mammalian cells based on the pH-sensitive fluorescent conjugate CalRexin™:pHrodo™ Red and high throughput/content imaging. CalRexin™:pHrodo™ Red is promptly taken up by endosome-amphisome-autolysosome (low pH ∼4.5) pathway, which is connected to canonical autophagy, culminating with a bright red fluorescence. Importantly, this system allows for the discrimination between autophagic responses coupled with the endosome-amphisome-autolysosome pathways and the endosome-lysosome pathway as elicited by receptor-driven endocytosis and phagocytosis. In human cervical carcinoma sHeLa cells, the accumulation of CalRexin™:pHrodo™ Red as elicited by canonical autophagy activators like the mTOR inhibitor rapamycin or serum starvation was suppressed by conventional inhibitors 3-methyladenine or chloroquine. Similarly, RB1CC1-/-(FIP200) as well as ATG5-/- sHeLa cells (which bear genetic defects in two different steps of autophagy) were unable to accumulate CalRexin™:pHrodo™ Red upon exposure to autophagy activators. Thus, CalRexin™:pHrodo™ Red provides a novel approach for imaging autophagy on high throughout/content platforms to screen large libraries for the identification of novel autophagy-targeting agents.

Periodontitis is a chronic inflammatory disease driven by dysbiosis in subgingival microbial communities leading to increased abundance of a limited number of pathobionts, including Porphyromonas gingivalis and Treponema denticola. Oral health, particularly periodontitis, is a modifiable risk factor for Alzheimer disease (AD) pathogenesis, with components of both these bacteria identified in postmortem brains of persons with AD. Repeated oral inoculation of mice with P. gingivalis results in brain infiltration of bacterial products, increased inflammation, and induction of AD-like biomarkers. P. gingivalis displays synergistic virulence with T. denticola during periodontitis. The aim of the current study was to determine the ability of P. gingivalis and T. denticola, grown in physiologically relevant conditions, individually and in combination, to induce AD-like pathology following chronic oral inoculation of female mice over 12 weeks. P. gingivalis alone significantly increased all 7 brain pathologies examined: neuronal damage, activation of astrocytes and microglia, expression of inflammatory cytokines interleukin 1β (IL-1β) and interleukin 6 and production of amyloid-β plaques and hyperphosphorylated tau, in the hippocampus, cortex and midbrain, compared to control mice. T. denticola alone significantly increased neuronal damage, activation of astrocytes and microglia, and expression of IL-1β, in the hippocampus, cortex and midbrain, compared to control mice. Coinoculation of P. gingivalis with T. denticola significantly increased activation of astrocytes and microglia in the hippocampus, cortex and midbrain, and increased production of hyperphosphorylated tau and IL-1β in the hippocampus only. The host brain response elicited by oral coinoculation was less than that elicited by each bacterium, suggesting coinoculation was less pathogenic.

Excitotoxicity is a neuronal death process initiated by over-stimulation of ionotropic glutamate receptors. Although dysregulation of proteolytic signaling networks is critical for excitotoxicity, the identity of affected proteins and mechanisms by which they induce neuronal cell death remain unclear. To address this, we used quantitative N-terminomics to identify proteins modified by proteolysis in neurons undergoing excitotoxic cell death. We found that most proteolytically processed proteins in excitotoxic neurons are likely substrates of calpains, including key synaptic regulatory proteins such as CRMP2, doublecortin-like kinase I, Src tyrosine kinase and calmodulin-dependent protein kinase IIβ (CaMKIIβ). Critically, calpain-catalyzed proteolytic processing of these proteins generates stable truncated fragments with altered activities that potentially contribute to neuronal death by perturbation of synaptic organization and function. Blocking calpain-mediated proteolysis of one of these proteins, Src protected against neuronal loss in a rat model of neurotoxicity. Extrapolation of our N-terminomic results led to the discovery that CaMKIIα, an isoform of CaMKIIβ undergoes differential processing in mouse brains under physiological conditions and during ischemic stroke. In summary, our findings inform excitotoxic neuronal death mechanism and suggest potential therapeutic strategies for neuroprotection. In Brief: Ameen, et al. used a proteomic method called N-terminomics to identify proteolytic events occurring in neurons during excitotoxicity. They found that most proteolytic processing is mediated by calpains, resulting in the generation of stable truncated fragments with the potential to induce synaptic dysfunction and loss, eventually leading to neuronal death. They further showed that some of these proteolytic processed proteins, such as the protein kinases Src and CaMKII, are potential targets for neuroprotection. Highlights: Identification of over 300 neuronal proteins cleaved by calpains to form stable truncated fragments during excitotoxicity. The calpain cleavage sites of these proteins unveil for the first time the preferred cleavage sequences of calpains in neurons. These pathological proteolytic events potentially induce synaptic dysfunction and loss, which likely contribute to excitotoxic neuronal death. Some of the neuronal proteins proteolyzed by calpains are potential targets of neuroprotection. Graphical Pathological proteolytic events in neurons during excitotoxicity unveiled by N-terminomic analyses: (A) N-terminomic and global proteomic analyses identified neo-N-terminal sites and neuronal proteins undergoing significant abundance changes during excitotoxicity. (B) Informatic analysis of the proteomic results predicted (i) the preferred sequences of proteolytic processing of neuronal proteins catalyzed by calpains during excitotoxicity and (ii) perturbation of synaptic organization and functions as the major consequence of calpain-mediated proteolytic events. (C) Validation of these predictions and further experimentations unveiled: (i) calpain-mediated cleavage of proteins associated with synaptic damage in excitotoxic neurons, (ii) a new mechanism of dysregulation of CaMKIIα and CaMKIIβ, which are key protein kinases governing synaptic dysfunctions and excitotoxic neuronal death and (iii) potential therapeutic targets such as the protein kinases Src and CaMKII for neuroprotection One Sentence Summary: Proteolytic events in neurons during excitotoxicity inform neuronal death mechanism and potential therapeutic strategies for neuroprotection.

IPF is a progressive and lethal disease and phase 2 trials have shown that inhibition of CTGF (Pamrevlumab/FG3019, Fibrogen Inc.) or Autotaxin (Ziretaxestat/GLPG1690 Galapagos Inc.) slow the decline in FVC experienced by IPF patients. Given the stabilization of FVC observed from combining Nintedanib and Pirfenidone, we investigated the potential benefit of combining FG3019 and GLPG1690 in a mouse model of IPF. Mice were subjected to intratracheal dosing of bleomycin to induce lung fibrosis and treated with FG3019 and/or GLPG1690 for 14 days. Changes in bodymass, survival, physiological measurements, plasma, BALF and lung tissue were recorded during the study. The progression of fibrosis was determined by Flexivent, lung hydroxyproline levels, Ashcroft-Hubner histopathological scores, as well as machine learning-based of lung sections or micro CT-scanned lungs respectively. FG3019 prevented the increase in lung hydroxyproline levels caused by bleomycin, while treatment with GLPG1690 did not have this effect. In contrast, GLPG1690 treated mice had better FVC and inspiratory capacity than vehicle controls, while mice receiving FG3019 did not. The combination of FG3019 and GLPG1690 reduced the accumulation of lung hydroxyproline and improved lung function parameters. The Ashcroft-Hubner scores were reduced in mice receiving FG3019 or GLPG1690 as a single treatment, but further reductions were not observed mice treated with FG3019 and GLPG1690. The combination of FG3019 and GLPG1690 treatment demonstrated an enhanced therapeutic efficacy by conferring the independent and parallel benefits of the single agents on experimental lung fibrosis endpoints.

The therapeutic utility of systemic “biologicals” (i.e. recombinant proteins), in particular monoclonal antibodies, has rekindled interest in development of diverse inhaled biological platforms to treat lung diseases. However, very little is known about the specific patterns and mechanisms of their uptake, epithelial transfer, clearance and redistribution. Here we compare conventional PK analysis with whole-of-lung biodistribution by 3D light-sheet microscopy (LSM) and FACS profiling of macrophages using a human NGAL lipocalin protein as a test probe. Methods: Unlabelled NGAL was microsprayed into male Balb/c mice lungs. At 1, 2, 4, 6, & 24h, blood, PBS lung washes and lung lobes were collected. Separately, fluoro tagged NGAL-647 was microsprayed into mouse lungs, harvested, optically cleared and imaged by LSM and lung macrophages analysed by FACS. Results: Instilled NGAL displayed first order elimination kinetics. Appeared in the blood in a dose proportional manner (T_max=2h,& C_max=1859 ng/ml). LSM confirmed progressive redistribution from the airway lumen into the interstitium; most likely by paracellular transport. As previously demonstrated, instilled protein macrophage uptake was inferred from a punctate cell biodistribution pattern and redistribution at later time points. This was confirmed by FACS as the majority of NGAL+ cells were alveolar monocyte/macrophages cells. Conclusion: Consistent with previous reports, NGAL delivered directly to the lung by microspray undergoes first order elimination, transit from lumen to submucosa in airways and lumen to interstitium and blood from the alveoli as well as macrophage uptake and clearance via the mucociliary escalator.

Motor neurone disease (MND) is a neurodegenerative disorder characterised by progressive destruction of motor neurons, muscle paralysis and death. The amyloid precursor protein (APP) is highly expressed in the central nervous system and has been shown to modulate disease outcomes in MND. APP is part of a gene family that includes the amyloid precursor-like protein 1 (APLP1) and 2 (APLP2) genes. In the present study, we investigated the role of APLP2 in MND through the examination of human spinal cord tissue and by crossing APLP2 knockout mice with the superoxide dismutase 1 (SOD1-G37R) transgenic mouse model of MND. We found the expression of APLP2 is elevated in the spinal cord from human cases of MND and that this feature of the human disease is reproduced in SOD1-G37R mice at the End-stage of their MND-like phenotype progression. APLP2 deletion in SOD1-G37R mice significantly delayed disease progression and increased the survival of female SOD1-G37R mice. Molecular and biochemical analysis showed female SOD1-G37R:APLP2−/− mice displayed improved innervation of the neuromuscular junction, ameliorated atrophy of muscle fibres with increased APP protein expression levels in the gastrocnemius muscle. These results indicate a sex-dependent role for APLP2 in mutant SOD1-mediated MND and further support the APP family as a potential target for further investigation into the cause and regulation of MND.

Summary: Excitotoxicity, a neuronal death process in neurological disorders, is initiated by over-stimulation of neuronal ionotropic glutamate receptors. The over-stimulated receptors dysregulate proteases, protein kinases and phosphatases, which in turn modify target neuronal proteins to induce cell death. To decipher this cell death mechanism, we used quantitative proteomics, phosphoproteomics and N-terminomics to identify modified proteins in excitotoxic neurons. Data, available in ProteomeXchange (identifiers: PXD019527 and PXD019211), enabled us to identify over one thousand such proteins with calpains, cathepsins and over twenty protein kinases as their major modifiers. These protein modification events can potentially perturb signalling pathways governing cell survival, synaptogenesis, axonal guidance and mRNA processing. Importantly, blocking the modification of Src protein kinase, a signalling hub in excitotoxic neurons, protected against neuronal loss in vivo in a rat model of neurotoxicity. Besides offering new insights into excitotoxic neuronal death mechanism, our findings suggest potential neuroprotective therapeutic targets for treating neurological disorders. Graphical Highlights: Multi-dimensional proteomic analysis identified proteins modified by proteolysis and altered phosphorylation in neurons undergoing excitotoxic cell death. Calpains, cathepsins and over twenty protein kinases are major modifiers of these proteins. These protein modification events are predicted to impact cell survival, axonal guidance, synaptogenesis and mRNA processing. Blocking modification of an identified protein Src, which acts as a major signalling hub in neurons, was protective against excitotoxic injury in vivo . In Brief: Using multidimensional proteomic approaches, Ameen, et al . mapped the changes of proteome, phosphoproteome and N-terminome of cultured primary neurons during excitotoxicity, a crucial neuronal death process in neurological disorders. These proteomic changes document new excitotoxicity-associated molecular events, and offer insights into how these events are organized to induce neuronal death. Potential therapeutic relevance of these molecular events is illustrated by the demonstration that in vivo blockade of one of these events could protect against excitotoxic neuronal loss.