DNA-aptamer gating membranes

This report describes a membrane barrier whose permeability is modulated through the recognition of a small-molecule target, adenosine triphosphate (ATP), by a DNA-aptamer.


(1) Surface Modification of the Anodisc Alumina Oxide (AAO) Membranes
The aptamer--modified AAO membranes were prepared by a procedure of sequential depositing the monolayers of propylamylethoxysilane, covalent attachment of BS4, and finally attachment of biotinylated aptamer sequences through an avidin sandwich. All reagents were obtained from Sigma--Aldrich and used as received, unless stated otherwise. The details of layer--by--layer procedure involved the following steps: i) Silanization of the alumina surface: The 20 nm pore--size AAO (Whatman) membranes were functionalized with amino groups through an ethoxysilane monolayer deposition. The membrane discs were cleaned in 37 % peroxide followed by UV--ozone treatment. The clean membranes were incubated for 30 min. at room temperature in a 4 % solution of 3-aminopropyltriethoxysilane in a 5 % water in ethanol mixture, acidified to yield pH=4.5 with acetic acid. The membrane was the washed several times with ethanol and dried under vacuum at room temperature.
ii) Surface functionalization with biotin: Biotinylated N--hydroxysuccinimide ester of a polyethylene glycol spacer arm (EZ--Link NHS--PEG4--Biotin, Piercenet) was attached to amino groups on the surface of the silanized Anodiscs through incubation for 30 minutes at room temperature and subsequent washing with PBS--buffer.
iii) Immobilization of the biotinylated ATP--binding aptamer: The biotinylated anodisc surface was incubated for 4 hours in avidin solution and subsequently another 4 hours in biotinylated aptamer solution in PBS--buffer. The aptamer--functionalized membranes were air--dried and kept at 4 °C until used for the experiments. Fig. E2 Schematic of the conformational change which the ATP--binding aptamer undergoes upon molecular recognition of ATP in the membrane pore. The aptamer consisted of the specific base sequence (grey), added nucleotides for creating a hair--pin structure 13 (yellow) and a biotin--linker (violet). Specific binding to ATP results in an opening of the hairpin and a significant conformational change of the aptamer within the membrane pore. The conformational change has previously been quantified to be at the order of about 1.6 nm. 12

ATP$ (3) AAO--membrane surface modification: QCM--D measurements
Immobilisation of avidin/ATP--aptamer on aluminum oxide The efficiency of the AAO--membrane surface modification through avidin and subsequent ATP--aptamer immobilization was confirmed by a quartz crystal microbalance with dissipation monitoring. All experiments were carried out in The average coverage of avidin followed by the ATP--aptamer is given below. The molar coverage is a rough approximation assuming a density of 1350 kg/dm 3 for both avidin and the ATP--aptamer 12 as well as a molar weight of 67 kDa and 11,26 kDa for avidin and the ATP--aptamer, respectively: Table1 The data in the last column indicate that the avidin/ATP--aptamer interaction was indeed in the range of the maximum number of possible interactions (four). It can be expected that within the porous structure of the AAO membranes the effective binding efficiency of the aptamer to the avidin sub--layer might be lower than during the QCM--D analysis which uses controlled flow conditions over a non-porous surface.

(4) Permeability measurements
A permeation flow cell of an efficient membrane diameter of 5 mm was employed.
The membrane was sandwiched between the upstream (feed) chamber (green chamber on top) and the downstream (

) Correlation between binding studies of the ATP--aptamer hairpin in solution and the pore--closing observed in the ATP--aptamer modified responsive membrane:
The KD of the ATP--aptamer hairpin employed in this study is around 345 μM in solution and presence of its target, adenosine--5'--triphosphate (ATP), while it is virtually non--responsive to guanosine--5'--triphosphate (ATP). 12 This specificity reflects in the responsiveness of the ATP--aptamer modified AAO--membrane: Fig.   E4 represents the pore closing observed in this study at defined ATP-concentrations with the normalized fluorescence observed previously with an ATP--molecular beacon at identical concentrations. As can be seen, the maximum pore closing is observed when the normalized fluorescence is maximum, i.e., when the ATP--aptamer beacon is opened most going along with a major conformational change. In contrast, at low fluorescence data (=molecular beacon closed = no target recognition = no conformational change) also the pores remain almost fully open.
Hence, pore closing is strongly correlated with ATP--binding (represented by increasing normalized fluorescence). Fig. E4 Correlation of the responsiveness of the ATP--aptamer modified AAO-membrane (represented as pore closing) and the ATP--binding as observed with an ATP--aptamer molecular beacon. The origin of the graph represents the reference situation of maximum pore aperture in absence of any target (ATP); violet: mutated ATP--aptamer; yellow: ATP--aptamer with GTP as non--specific target; green: ATP--aptamer with specific target ATP.