Rigor, reproducibility, and in vitro cerebrospinal fluid assays: The devil in the details

Abstract

Divergent results and misinterpretation of non-significant findings remain problematic in science – especially in retrospective, hypothesis generating, translational research. When such divergence occurs, it is imperative that the cause of the divergence be established.

In their recent paper in Annals of Neurology, Dauvilliers et al challenged our earlier finding that cerebrospinal fluid (CSF) from some patients with unexplained excessive daytime sleepiness enhances the activation of GABA_A receptors (GABA_A-R). They present data from 15 subjects in which they were unable to find evidence of enhanced activation of GABA_A receptors. Here we: 1) establish how flaws in Dauvilliers’ experimental design account for this difference; 2) present new data demonstrating the robustness and reproducibility of our methods and 3) summarize the clinical promise of GABA_A-R antagonism in treating IH and related disorders.

Figure 1

A. Example trace of currents recorded from α₁β₂γ₂ GABA_ARs in response to 10–300μM GABA ± 50% CSF. Dotted lines and arrows mark the degree/direction of modulation. Notice that low ECs result in enhancement while high ECs result in desensitization and inhibition. Scale bar: 5sec, 1000pA. B. Average enhancement (%) measured from peak currents (I): (I_GABA+CSF – I_GABA)/(I_GABA)*100 for each GABA concentration shown in Figure A. (In Figure 1A, notice how enhancement is nearly zero at 30μM and negative beyond that as peaks desensitize, making measurements of enhancement unreliable. C. Enhancement comparison of 32 CSFs in 2 populations of receptors: enhancement of α₂β₂ receptors determined using planar patch clamp electrophysiology at the University of Queensland (Lynch Lab) and enhancement of α₁β₂γ₂ receptors determined using single electrode patch clamp electrophysiology at Emory University (Jenkins Lab). The line represents the line of identity. D. Average desensitization of peak currents from α₁β₂γ₂ (●) and α₂β₂γ₂ (◊) receptors, calculated as the difference of peak amplitude to the amplitude at the end of each GABA exposure. Linear regressions to calculate desensitization (d%) as a function of log[GABA] for each receptor were: α1: d%=8.67*log[GABA]-11.53 and α2: d%=18.12*log[GABA]-16.68. Effective concentrations for each GABA concentration could be back calculated using Hill=1.36 (α1) and 1.53 (α2) and EC₅₀= 60μM (α1) and 8.8μM (α2) and the Hill equation: I/I_max = [GABA]^nH/([GABA]^nH + EC₅₀^nH). Trace inset of a 2 sec exposure of saturating (300μM) GABA to α₂β₂γ₂ receptors with dotted lines and arrow indicating the degree of desensitization. Scale bar: 1sec, 1000pA. n=24 cells (α1), n=35 cells (α2). Where not shown, the error bars are smaller than the symbol.