Approximately 20% of familial amyotrophic lateral sclerosis (FALS) cases are caused by mutant superoxide dismutase type 1 (mtSOD1). Although the mechanisms of mtSOD1-induced toxicity remain poorly understood, evidence suggests that accumulation of misfolded SOD1 is fundamental to its toxicity and the death of motor neurons. Misfolded mtSOD1 can accumulate inside the endoplasmic reticulum (ER), leading to ER stress, with activation of the unfolded protein response (UPR). We have previously carried out genetic studies focused on PERK (which is an eIF2α kinase that is rapidly activated in response to ER stress and leads to a repression in translation) and GADD34 (which participates in the dephosphorylation of eIF2α). We reported that mtSOD1 transgenic mice that are haploinsufficient for PERK have a significantly accelerated ALS disease, while mtSOD1 mice that are haploinsufficient for GADD34 have a remarkably ameliorated disease. Guanabenz, a centrally acting oral drug approved for the treatment of hypertension, enhances the PERK pathway by selectively inhibiting GADD34-mediated dephosphorylation of eIF2α. We have now treated G93A mtSOD1 transgenic mice with guanabenz and found a significant amelioration of disease with a delay in the onset and prolongation of the early phase of disease and survival. Guanabenz-treated G93A mice have less accumulation of mtSOD1 and an enhanced phosphorylation of eIF2α at endstage. This study further emphasizes the importance of the PERK pathway in the pathogenesis of FALS and as therapeutic targets in ALS, and identifies guanabenz as a candidate drug for the treatment of ALS patients.

Closely related substrains of inbred mice often show phenotypic difzferences that are presumed to be caused by recent mutations. The substrains BALB/cJ and BALB/cByJ, which were separated in 1935, have been reported to show numerous highly significant behavioral and morphological differences. In an effort to identify some of the causal mutations, we phenotyped BALB/cJ and BALB/cByJ mice as well as their F1, F2, and N2 progeny for behavioral and morphological phenotypes. We also generated whole genome sequence data for both inbred strains (∼3.5× coverage) with the intention of identifying polymorphic markers to be used for linkage analysis. We observed significant differences in body weight, the weight of the heart, liver, spleen and brain, and corpus callosum length between the two substrains. We also observed that BALB/cJ animals showed greater anxiety-like behavior in the open field test, less depression-like behavior in the tail suspension test, and reduced aggression compared to BALB/cByJ mice. Some but not all of these physiological and behavioral results were inconsistent with prior publications. These inconsistencies led us to suspect that the differences were due to, or modified by, non-genetic factors. Thus, we did not perform linkage analysis. We provide a comprehensive summary of the prior literature about phenotypic differences between these substrains as well as our current findings. We conclude that many differences between these strains are unstable and therefore ill-suited to linkage analysis; the source of this instability is unclear. We discuss the broader implications of these observations for the design of future studies.