Lipid bilayer

A lipid bilayer is a sheet two molecules thick, made of amphiphilic lipids whose water-loving heads face the solution on both sides and whose oily tails meet in the middle. It is about 4 nm across, it assembles by itself when the lipids are put in water, and it is the structure of the boundary around every living cell. In nanopore research a flat bilayer stretched over a hole separates two chambers of salt solution and holds one protein pore, so that the current through that single pore can be recorded. The bilayer is a good electrical insulator and a poor mechanical one, and the second fact explains much of the history of biological nanopores.

A phospholipid has a polar head and two hydrocarbon tails. Placed in water, the molecules arrange themselves so that the tails are shielded from the solution, and the lowest-energy arrangement for most lipids is a double layer with the tails inside. The hydrocarbon core is about 3 nm thick and the whole sheet, heads included, about 4 nm.^[1]

The double layer was inferred in 1925. Gorter and Grendel extracted the lipid from red blood cells, spread it on a water surface and found that it covered about twice the area of the cells' surface, which they took to mean that the membrane was two molecules thick.^[2] Singer and Nicolson's fluid mosaic model of 1972 added the proteins, floating in a bilayer in which the lipids themselves move freely in the plane.^[3]

The planar bilayer, a membrane made on purpose across a small hole, dates from 1962. Mueller, Rudin, Tien and Wescott painted a lipid solution across an aperture in a partition between two aqueous compartments. The film thinned until it was a single bilayer, and they showed that it could be made to behave as an excitable membrane.^[4] The films became known as black lipid membranes, because a layer far thinner than the wavelength of light reflects almost none of it and appears black.^[5]

Painted films keep some of the hydrocarbon solvent they were made in. In 1972 Montal and Mueller described a second method that leaves less of it: two lipid monolayers on the surfaces of the two solutions are raised over the hole in a partition and meet there as a bilayer.^[6] Nanopore laboratories use this method.^[7] Later variants form the bilayer where two lipid-coated water droplets touch in oil, or between two monolayers at an interface in a microfluidic chip.^[8][9]

The lipid most often used is diphytanoylphosphatidylcholine (DPhPC), whose methyl-branched chains resemble those of archaeal membranes. A study of 1979 found no gel-to-liquid transition in it anywhere between −120 °C and 120 °C, so a DPhPC bilayer stays fluid across the whole range of temperatures a laboratory is likely to use.^[10] The bilayer in the 1996 experiments that first detected single DNA strands in alpha-hemolysin was DPhPC across a hole about 0.1 mm wide in a Teflon sheet 25 µm thick.^[11]

A bilayer is a capacitor with an unusually thin dielectric. Montal and Mueller measured 0.9 µF/cm² for their solvent-poor films, about twice the 0.45 µF/cm² of conventional painted films, and a resistance of 10⁶ to 10⁸ Ω cm² for both.^[6] A film across a hole 100 µm wide has an area of 7.85 × 10⁻⁵ cm², so its capacitance is about 35 pF for a painted film and 70 pF for a solvent-poor one, and its resistance runs from 13 GΩ upward. That capacitance matters, because the voltage noise of an amplifier acting on the capacitance at its input produces a current noise that grows as the bandwidth to the power 3/2. With an amplifier of 3 nV/√Hz, 70 pF adds about 0.8 pA rms in a band of 10 kHz and 24 pA in a band of 100 kHz; 35 pF adds half of that (see Nanopore noise). Laboratories that want wide bandwidth therefore try to shrink the aperture.

The resistance suits a recording whose signal is a few hundred picoamperes. At 200 mV across 4 nm, the electric field in the bilayer is 5 × 10⁷ V/m, and a field of that size is part of the reason the film can fail.

The same thinness makes the bilayer hard to keep. A study of 2006 that set out to make planar bilayers easier to form begins with their reputation: they are "a powerful tool for the functional study of membrane proteins" that "have not been widely used due to their low stability and reproducibility".^[9] The 1996 experiments needed one film and one pore, which inserted itself from solution. A 2012 report on the announcement of the MinION said that the first nanopore membranes were lipid bilayers, which "proved far too fragile to use in a production sequencer".^[12]

Oxford Nanopore Technologies says that its pores sit in "an electrically-resistant polymer membrane", each on its own microscaffold.^[13] Synthetic membranes of this kind have a long academic pedigree: in 2000 Meier, Nardin and Winterhalter reconstituted channel proteins in membranes made from an ABA triblock copolymer.^[14]

A bilayer can also be used to improve a solid-state nanopore. In 2011 a group coated silicon nitride pores with a fluid lipid bilayer, an idea taken from the olfactory hairs of insect antennae. The coating let the pore diameter be tuned in subnanometer steps, prevented clogging and non-specific binding, and, with mobile ligands in the bilayer, slowed targeted proteins enough to time-resolve single translocations.^[15]

Molecular Reality Corporation builds its pores in solid supports and keeps bilayers out of the instrument. The pore on its live molecular stream is a hole about 200 µm wide drilled in a membrane by laser, with no protein in it, and the company's public guide puts its view of the alternative in a sentence: "Protein pores are delicate flowers."^[16] A 1994 paper that used the channels of a peptide in a lipid bilayer to count single polyethylene glycol molecules began with a description of the Coulter counter.^[17]

• Biological nanopore
• Alpha-hemolysin
• Patch clamp
• Ion channel
• Nanopore noise