Vanishing white matter: the integrated stress response in disease mechanisms, therapy targets and biomarkers

The leukodystrophy vanishing white matter (VWM) has a disease course that ranges from slow disease progression to rapid demise and may manifest a wide spectrum of neurological signs. All patients are affected by chronic neurological deterioration and the majority also experience stress-induced acute episodes of neurological decline [1]. The neuropathological hallmarks include white matter rarefaction and cystic decline, deficient myelin, lack of adequate astrogliosis, mislocalized Bergmann glia, and immature astrocytes and oligodendrocytes in the white matter [2-4]. VWM is caused by pathogenic variants in either of the five genes encoding the subunits of eukaryotic translation initiation factor 2B (eIF2B) [5, 6]. These pathogenic variants lead to dysregulation of the integrated stress response (ISR): increased levels of ATF4 and CHOP and their altered transcriptomes are accompanied by decreased eIF2α phosphorylation levels due to increased activity of the negative feedback loop component GADD34 [7-10] [Chapters 2-5]. In VWM mouse brain, the expression levels of ATF4- and CHOP-regulated transcriptomes positively correlate with chronic disease progression [7]. The ISR itself is a complex process with hundreds of regulated genes and with impacts on many cellular processes [11]. Understanding how the dysregulated ISR contributes to acute and chronic neurological decline in VWM may ultimately benefit treatment development [11-17]. Several preclinical studies show that the dysregulated ISR is a viable treatment target for VWM [4, 7, 10, 18] and promising treatment options are examined in clinical trials in patients with VWM [19]. Validated biologically relevant biomarkers reflecting VWM pathomechanisms or tissue damage in patient samples would be helpful to monitor target engagement of possible therapies [20], but are missing. This thesis has aimed to explore the consequences of a dysregulated ISR in VWM by aiming to 1) understand how ISR components up- and downstream of eIF2B contribute to VWM disease mechanisms underlying the chronic neurological deterioration by modifying these components with different drugs or genetic interventions; 2) understand if and how two commonly used anesthetics result in subacute deterioration that may be observed in patients after anesthesia; 3) and 4) identify potential biomarkers by quantifying ISR-regulated mRNA levels in blood samples from patients. Key findings include: GBZ significantly reduced ISR activation and improved symptoms in VWM mice, while Sephin1, lacking adrenergic effects, was less effective. TUDCA and 4-PBA reduce p-eIF2α levels but showed minimal effects in VWM models, reinforcing that targeting downstream of eIF2B is more effective. Pridopidine (PDPD) acts downstream of eIF2B, and mildly improved motor symptoms in VWM mice but was not effective enough for standalone therapy. Propofol might be favored over sevoflurane as anesthetic for VWM patients, based on our current data and the observation that there were no instances of acute neurological deterioration since propofol is our preferred anesthetic for VWM patients. The NanoString technology confirmed ISR dysregulation in the brains of VWM patients but did not identify an ISR blood biomarker that can be readily used in the clinic.