Effects of Low Intensity Focused Ultrasound on Liposomes Containing Channel proteins

Abstract

The ability to reversibly and non-invasively modulate region-specific brain activity in vivo suggests Low Intensity Focused Ultrasound (LIFU) as potential therapeutics for neurological dysfunctions such as epilepsy and Parkinson's disease. While in vivo studies provide evidence of the bioeffects of LIFU on neuronal activity, they merely hint at potential mechanisms but do not fully explain how this technology achieves these effects. One potential hypothesis is that LIFU produces local membrane depolarization by mechanically perturbing the neuronal cell membrane, or activating channels or other proteins embedded in the membrane. Proteins that sense mechanical perturbations of the membrane, such as those gated by membrane tension, are prime candidates for activating in response to LIFU and thus leading to the neurological responses that have been measured. Here we use the bacterial mechanosensitive channel MscL, which has been purified and reconstituted in liposomes, to determine how LIFU may affect the activation of this membrane-tension gated channel. Two bacterial voltage-gated channels, KvAP and NaK2K F92A channels were also studied. Surprisingly, the results suggest that ultrasound modulation and membrane perturbation does not induce channel gating, but rather induces pore formation at the membrane protein-lipid interface. However, in vesicles with high MscL mechanosensitive channel concentrations, apparent decreases in pore formation are observed, suggesting that this membrane-tension-sensitive protein may serve to increase the elasticity of the membrane, presumably because of expansion of the channel in the plane of the membrane independent of channel gating.

Figure 1

LIFU stimulation duty cycle effect studies involved varying duty cycle (60%- light grey, 80%- Dark grey and CW-Black) within three groups (Control vesicle or MscL); analysis using mixed effects linear regression **p-value < 0.001. (Control vesicle group: 60% DC 8.6 ± 1.1 (n = 30), 80% DC 13.1 ± 1.1 (n = 30), CW DC 14.9 ± 0.7 (n = 85); MscL group: 60% DC 8.4 ± 1.6 (n = 33), 80% DC 15.3 ± 1.6 (n = 28), CW DC 17.6 ± 1.3 (n = 90)).

Figure 2

Calcein efflux from proteoliposomes reconstituted with different concentrations of MscL, NaK2K F92A and KvAP treated with and without ultrasound. Lightest grey bar line per group (MscL, Nak2K F92A, KvAP) shows control vesicle (no-protein) % efflux in respond to LIFU stimulation. Black bar line per group presents % efflux for the highest protein concentration level possible per group (MscL 380 pmol/mg lipids, NaK2K F92A 467 pmol/mg lipids, KvAP 77 pmol/mg lipids). The three grey bar lines between the lightest grey and the black bar represent liposomes reconstituted with 0.125, 0.25 and 0.5 of the highest concentration of the corresponding protein. Each well was modulated once at a time, for 10 minutes. As protein concentration increased, significant increase in efflux was noticed among all three groups except the highest MscL concentration (n values from left to right equals, MscL group: 85, 90, 49, 36, 17; NaK group: 56, 55, 22, 15, 13; KvAP group: 26, 30, 18, 18, 18) (Mixed effect linear regression, *p < 0.05, **p < 0.01).

Figure 3

LPC induced calcein dye efflux from liposomes reconstituted with different protein channels: MscL (a), NaK2K F92A (b), KvAP (c) with varying protein concentrations per protein channel group. Calcein efflux increases for all three channels as the protein concentration increases. There is a significant difference (*P < 0.05, **P < 0.01, ***P < 0.001) between sham (no-protein liposomes) and proteoliposomes per protein channel group based on Mann-Whitney test. For no-protein and each protein concentration n equals 16, 5, 8, 12, 16 (a, MscL) 9, 5, 5, 7, 9 (b, NaK2K F92A) and 5, 4, 4, 5, 5 (c, KvAP).