Since its discovery by Banting and Best in 1921, insulin has been one of the most thoroughly studied molecules in the history of science. Insulin was first used in humans in 1922. Soon after that, insulin became widely available to diabetic patients in early 1923. In 1926, J. J. Abel at Johns Hopkins crystallized insulin (Beals and Kovach, 1997). Insulin became one of the first proteins to be crystallized in pure form. Almost thirty years later, Sanger and co-workers determined the amino acid sequence of insulin in 1955.
In fact, insulin was the first protein to be fully sequenced. Sanger was awarded the 1959 Nobel Prize in Chemistry for this work. (Beals and Kovach, 1997). The knowledge of the amino acid sequence made it theoretically possible to synthesize the molecule in a lab. In 1963, human insulin became the first protein to be chemically synthesized. However, the amount of human insulin obtained was not enough for clinical purpose. In 1979, scientists successfully converted porcine insulin to human insulin by enzymatic method. The resultant insulin was called semi-synthetic human insulin.
Also during the late l970s, recombinant DNA technology advanced to a new level and made biosynthesis of insulin possible. (Beals and Kovach, 1997). In September of 1978, Genentech, Inc. , City of Hope National Medical Center and hospital in Duarte, California announced the successful production of human insulin using recombinant DNA technology. For regulatory reasons, the A- and B-chains were produced separately in genetically engineered Escherichia coliandand then put together chemically to form complete human insulin (Beals and Kovach, 1997)..
Eli Lilly immediately licensed the recombinant human insulin technology from Genentech. For the first 60 years since the discovery of insulin, all the commercially available insulin used to treat diabetes was animal sourced, mainly bovine and porcine. The first human insulin on the market was the semi-synthetic human insulin manufactured by NovoNordisk, launched in 1982 (Beals and Kovach, 1997). Eli Lilly started clinical studies for the biosynthetic human insulin 1980. The product, Humulin, got approval from FDA in 1982.
This was the first product derived from recombinant DNA technology approved for human use by the FDA. From 1986, Lilly started a process of producing proinsulin in Saccharomyces cerevisiae and chemically converting it into insulin. Nowadays, there is almost no animal-sourced insulin product on the market in the US and Europe. (Beals and Kovach, 1997) Insulin is a small protein molecule, consisting of two polypeptide chains, referred to as A-chain and B-chain. The A-chain consists of 21 amino acids and the B-chain has 30 amino acids.
Two inter-chain disulfide bonds, between A7-B7 and A20-B19, connect the two chains. There is a third intra-chain disulfide bond located in the A-chain, between residues A6 and A11. The amino acid sequence of insulin varies among different species. For example, porcine insulin has different amino acid only in one position compared to human insulin (ThrB30> AlaB30), and bovine insulin has three (ThrA8>AlaA8, IleA10>ValA10, ThrB3O>AlaB30). The molecular weight of insulin is about 5800 Daltons, varying slightly among species (human insulin 5807. 58, porcine insulin 5777. 55 and bovine insulin 5733. 50) (Bowen, 1999).
In spite of the differences in the primary sequence of insulin among species, certain segments of the molecule are highly conserved, including the positions of the disulfide bonds, both ends of the A-chain, and the C-terminal residues of the B-chain. These similarities in the primary sequence confer a very similar three-dimensional conformation among species. As a result, insulin from one species is very possibly also biologically active in another species (Bowen, 1999). The insulin molecules have a strong tendency to self-associate into dimers and hexamers. Insulin exists as a monomer in solution only at very low concentration (< 0. 1 ? mol ~ 0. 6 (? g/ml).
It readily dimerizes at higher concentrations. The driving force of dimerization is predominantly the formation of four hydrogen bonds at the C-terminals of the B-chain between two monomers (Bowen, 1999). The dimer is a compact, stable unit in aqueous solution at pH ranging from 2 to 8. Three dimers will assemble into a hexamer in the presence of Zn2+ or other divalent metal ions, in the pH range of 4 to 8, when the concentration of insulin is above 10 ? mol. Each hexamer requires two Zn2+ ions. The interactions between dimers in the hexamer are relatively looser than those between the monomers in the dimer.
The hexamer has an almost spherical structure, with a diameter of 5 nm and a height of 3. 5 nm (Binder and Brange, 2003). Physiologically, insulin is synthesized only in the beta cells of islet of Langerhans in the pancreas. First, the gene encoding insulin is translated into a single chain precursor called preproinsulin. Preproinsulin is then transferred to the endoplasmic reticulum. A signal peptide is removed during the transfer, generating proinsulin. Proinsulin consists of 86 amino acids and has an amino-terminal B-chain, a carboxy-terminal A-chain and a connecting peptide in the middle known as the C-peptide (Binder and Brange, 2003).
Proinsulin is cleaved by some endopeptidases so that the C-peptide (31 amino acids) and two amino acids at both ends (a total of four amino acids) are removed to complete the insulin synthesis. Insulin is packaged in the Golgi apparatus into secretory granules, in the form of Zn2+ containing hexamers. Once released from the granules, insulin enters the portal vein and is delivered directly to the liver, where first-pass metabolism removes about 50% of the insulin. The insulin level in blood in a normal individual is 10-8 to 10-11 M, circulating and interacting with its receptor as a monomer (Binder and Brange, 2003).
Proinsulin is the main storage form of insulin in the beta cells. As a by-product of insulin synthesis, C-peptide is produced in equimolar amount of insulin. Measurement of C-peptide is useful as a biomarker of residual beta cell function, especially for diabetic patients who have already started insulin treatment. Since the half-life of C-peptide is much longer than that of insulin, the concentrations of C-peptide are 5-10 times higher than those of insulin. C-peptide was thought to have no physiological function.
However, some recently published studies revealed that C-peptide increased the parasympathetic nerve activity (Binder and Brange, 2003). Pharmaceutical Formulations Native (Unmodified) Insulin Preparations The early development of pharmaceutical formulations for insulin focused on improving the purity of the hormone. The first insulin formulation used in human was an acidic solution of insulin containing a significant amount of impurities, such as high molecular weight pancreatic proteins, proinsulin, and desamido insulin.
(Binder and Brange, 2003) The insulin formulation was frequently associated with allergic reactions due to impurities. The purity of insulin preparations was greatly improved when the method of crystallization of insulin with Zn + was introduced in the 1930s (Binder and Brange, 2003). Recrystallized insulin had higher purity and was considered essentially pure insulin until the late 1960s, when the introduction of new analytical methods allowed detection of other proteins in those insulin preparations.
In the 1970s, chromatographic purification reduced the impurities in insulin formulations by at least one order of magnitude (Binder and Brange, 2003). The insulin formulations available now have purity of more than 98%. There are three basic types of insulin preparations available for use in diabetic patients, with varying onset, peak and duration of action. Table 1 lists some of the common native insulin preparations available now. The onset, peak and duration of action are approximate for each product, depending on each individual, site of injection, dose, temperature and physical activity.