It should be noted that prior to the patch clamp in 1976, Bean in 1969 and Hladky in 1970 had already measured the first currents from individual ion channels using only voltage clamping. They measured the conductances of these purified bacterial ion channels which are made up of small peptides and placed in purified lipid bilayers which imitate the cell membrane. Gramicidin was one of these and is composed of several amino acids joined into a small circular peptide. It was determined that water molecules were also able to travel thru the channel in single file between the sodium ions. (note: these kinds of ion channels are not generally thought of when discussing ion channels and won't be considered further here in part because they are not large enough proteins to be regulated in a way necessary for any kind of physiological control). These highly specialized ion channel peptides are used by the host as antibiotics, capable of "punching holes" in the cell membranes of other types of cells and are therefore used as a defense. However, they have proven useful for learning about simple mechanisms of ion travel thru a pore in a lipid membrane. More recently, it has been found that animals (including human beings) also produce these peptides. In fact, it has been postulated by some that the repeated bacterial infections present in CF patient's lungs occurs in part because of the inactivation of these natural ion channel peptides produced in the lung.
Even though there would be no known sequence of an ion channel until the 1980s, it was still possible to do experiments in order to uncover their underlying structures and functions. For example, Armstrong in 1973 proposed that the structure of sodium channels in the squid neuron allow for opening and closing of their pores by a "ball and chain" model. He showed that there was a portion of the channel which could be cleaved off by protease enzymes in his lab, and that this missing piece normally acted as a kind of "ball" which was present just below the channel and could come up and plug the pore when the voltage on the cell membrane changed, as during transmission of the message. When the ball was not present, the channel remained in the open at all times.
An improvement on the patch clamp method called the "gigaseal" occurred serendipitously in 1980. A researcher is said to have accidentally applied a slight suction to the capillary tube on an ordinary patch clamp set-up which resulted in the formation of an approximately 100 times better seal with the cell membrane. It was named the gigaseal because it provided a reduction in the noise level due to formation of a resistance to extraneous ion flow of around 1 gigaohm. It also allowed for different configurations of the technique to be used including the "inside-out" and "cell attached" configuarations.
In the 1980s, it was discovered that the previously held belief that all ion channels were regulated directly was false. Metabotropic receptors were discovered which activate ion channels indirectly by way of second messengers, among these are cGMP, cAMP, and IP3. Many of these receptors consist of seven transmembrane helices and activate G-proteins which can then activate the ion channels directly (the other way is via second messengers). An example is the protein found in eye cells (rods) called rhodopsin. There are also over 1000 of these type of receptors that activate ion channels indirectly in the olfactory epithelium and are responsible for our sense of smell. An advantage (or disadvantage, depending on how you look at it) is that these receptors allow for longer opening times of ion channels, somewhere in the neighborhood of seconds to minutes, while directly activated ion channels can gate in thousands of a second. In fact, some of these channels activated by metabotropic receptors have been known to remain active for days and are probably important for long term memory formation.
The next major revolution in ion channel research occurred because of the advent of "recombinant DNA technology" in the mid to late 1970s. This provided the means for obtaining sequence information about genes and therefore the proteins they coded for, as well as the production of large amounts of the protein in easy to grow organisms such as the bacteria E. coli. No longer would researchers be dependant on natural tissue sources as the only available source to obtain ion channels to work with. Using cross-linking and a variety of other studies in the previous three decades, biochemists knew by 1980 that some ion channels consisted of more than a single protein subunit. For example, the nAChR, which is a sodium channel present on muscle cells and binds to and opens when acetylcholine is present was known by this time to consist of 5 subunits (pentameric) with a total molecular weight of about 290 KDa (note: one KDa, or kilodalton, is equivilent to the weight of one thousand hydrogen atoms). The exact amino acid sequences of these subunits had also been determined by 1982, some via their corresponding mRNA sequences. With the amino acid sequences known (also called the "primary structure" of the protein) it was possible to begin predictions of what ion channels should look like in 3-dimensional space (secondary and tertiary structure). The advent of computer programs capable of positioning coordinate atoms in 3 dimension was also beginning to be of some help about this time. Another important advantage of being able to work with the gene of an ion channel was that the sequence could be changed deliberately in order to understand how it works. Called site-directed mutagenesis, it is one of the most useful techniques known to molecular biology. The nAChR ion channel was also the first ion channel to be purified and reconstituted back into a lipid membrane bilayer and have its activity measured. This was accomplished by Montal in 1986. In 1984 they were the first ion channels to be expressed in foreign cells via injection of mRNA (by Mishina). They were also the first ion channel to have their unitary conductance measured (Neher and Sakmann, 1976) by patch clamp.
In 1987 a mutation in a gene in the fly Drosophila melanogaster was found to cause the insect to shake uncontrollably. The defect was subsequently traced to a gene which coded for a potassium channel and was therefore named the "shaker" ion channel. The gene itself is divided up into 23 different sections (exons) which can be put together in several different ways (called "splice variants") depending on which tissue it is found in. There are also at least 12 different known versions of the shaker ion channel in the mouse. And as if this were not enough diversity, it has since been found that the subunits can form different versions of the channel just by combining in different ratios with other subunits of the shaker ion channel. This diversity extends past its structure and into how the channel functions as well. Each version of the shaker channel will often have its own unique properties as to when and how it is activated, inactivated, etc.
The year 1982 saw the cloning of the first ion channel gene. Noda succeeded in isolating, cloning, and sequencing first the nAChR sodium ion channel from the electric marine ray Torpedo and then the voltage-gated sodium channel from the electric eel in 1984. Purified channels were obtained using a highly specific toxin, tetradotoxin obtained from puffer fish, which is known to bind nAChR channels tightly. Part of the amino acid sequence of nAChR was then easily obtained, and a nucleic acid probe corresponding to it used to screen a mRNA library in the form of cDNA. The mRNA sequence for the channel was obtained and used for determining the sequence of the entire protein. Prior to 1995, the largest fully sequenced genome was from a virus, bacteriophage lambda, which consists of 48,502 base pairs. Today over 25 bacteria and several eukaryotic genomes have been fully sequenced, including human.
In 1985, Levitan showed that ion channels could be altered in structure covalently, by the attachment of phosphate groups onto certain amino acids involved in their regulation. This "post-translational processing" altered the functioning of the channel itself.
In 1998, the first crystal structure of an ion channel (3.2 Angstroms resolution), the potassium channel KscA from a bacteria, was published by Doyle. This meant that it was now possible to determine the exact positions of nearly all of the individual atoms in the protein, which can still be said to be the ultimate in resolution of a protein's 3-dimensional structure. While this achievement provided a great deal of information about what a channel "looks" like, the next step will be to produce a series of crystal structures of an ion channel in all of the various modes of operation to determine how what it "acts" like, basically a motion picture of a protein in action, before any mechanism can be known for certainty. Crystal structures are an important step on the road towards making good assumptions about how ion channels do what they do. Since then, there have been 3? other ion channel structures solved to atomic resolution.
So by the time the gene for cystic fibrosis was discovered in 1989, and the protein the gene codes for given the name CFTR, the groundwork had already been laid theoretically and experimentally for going about the investigation of a newly discovered ion channel. Some of the immediate questions that needed to be answered were: if CFTR is indeed an ion channel, which ions are able to pass thru it and at what rate? Where is the pore located and how large is it? Under what conditions exactly is CFTR able to open and close? Which tissues is it found in? And many others.
References:
A History of Neurophysiology in the 19th Century Mary A. B. Brazier Raven Press NY, NY 1988
Nineteenth-Century Origins of Neuroscientific Concepts Edwin Clark and L.S Jacyna University of California Press 1987
Neuroscience Dale Purves, Et.al Sinauer Associates, Inc Publishers 1997
Ion Channels: Molecules in Action Aidley 1996
Ionic Channels of Excitable Membranes Bertil Hille Sinauer Associates Inc. Sunderland, MA 1992