The hormone insulin Protect The Expression Of Ideas

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The hormone insulin, discovered by Banting and Best in 1921, is produced in low amounts in diabetics by the pancreas when compared to normal individuals. For a long time, such diabetics have been treated by insulin derived from the pancreases of abattoir animals. Insulin, which is comprised of A and B chains, is secreted by the beta cells of the islets of Langerhans in the pancreas, and regulates the use and storage of carbohydrates in particular. Although bovine and porcine insulin are similar to the human form there are small differences that can lead to antibody formation in some patients. This can cause inflammatory reactions at injection sites and can neutralise the action of the insulin. It was also thought that the long-term injection of a foreign substance could be harmful.

The advent of cloning technology allowed genes of interest to be inserted into suitable vectors to produce a substance chemically identical to its naturally produced counterpart. This technique, often referred to as recombinant DNA technology, has been used to insert the insulin gene into the E. coli bacterial cell. The recombination process is as follows:

Plasmid is removed from the E. coli cell.

The plasmid is opened by a special enzyme.

The DNA coding for insulin (A or B chain) is inserted into the opened plasmid.

Recombination - the plasmid is closed using another special enzyme.

The recombined plasmid is introduced into E. coli host cell.

Insulin is a small protein with 21 amino acids comprising the A chain and 30 amino acids which comprise the B chain. Disulphide bonds link the two chains.

Figure 1 shows the insulin A and B DNA (created experimentally) inserted into the E. coli host. β-galactosidase is conjugated to the insulin A or insulin B DNA to control the transcription process in the bacteria. Use of truncated β-galactosidase genes has been shown to double the amount of proinsulin produced in E. coli (Guo et al. 1984). Restriction enzymes are used to 'cut out' the DNA coding for the required protein. An antibiotic resistance gene is also included as this allows for selection of clones by their resistance to different antibiotics. The plasmid is then sealed to form circular DNA using DNA ligase enzymes. The lac "promoter" region of E. coli allows transcription of the insulin A or B chain gene to produce the fusion protein. The fusion proteins are then extracted and purified, treated with cyanogen bromide to cleave the insulin chains, which are then mixed to form functional insulin. There are three disulphide bridges formed in appropriate media; two joining the insulin A and B chains, and an internal disulphide bridge in the A chain. The human form of insulin produced this way is called "Humulin".

The DNA sequence for human insulin is found at the top of the short arm of chromosome 11. When the DNA helix divides, one of the strands acts as a template and mRNA forms in the process of transcription. The mRNA moves from the nucleus to the cytoplasm where it attaches to a ribosome. By the process of translation, the triplet codons on the mRNA bind tRNA and add the amino acid coded by that triplet to the growing peptide chain. A weakened strain of E. coli is used to produce the human fusion protein. When the E. coli reproduces, the insulin gene is replicated together with the plasmid. Mutant strains of E. coli which do not produce enzymes that would destroy the insulin protein are employed.

Figure 1. Plasmid incorporation of insulin A and B DNA, amplification in the E. coli host, and process of purification of the protein chains.

The insulin produced by this process appears to have an identical structure to human pancreatic insulin, thus eliminating the formation of anti-insulin antibodies which may occur with bovine or porcine insulin.

Insulin production in 'bioreactors' is now used to produce large quantities of Humulin. Early efforts resulted in contamination by host cells or their unwanted products, but the introduction of several purification steps eliminated this problem. In one case the retention of the initiator methionine residue caused problems because the E. coli, or at least the strain used, did not remove this contaminant (Cheng et al. 1995). Several major drug companies now have large-scale facilities for the production of Humulin. Yields of 600 mg A chain/litre and 800 mg/l of B chain have been reported using glucose as the sole carbon and energy source (Schmidt et al. 1999). Although the description used E. coli as the host cell, yeast cells are now frequently used. The description of the insulin bioreactor process is as follows:

The reaction takes place in a specialised vessel which has agitation facilities as well as ports for sampling of the reaction mixture, air/gas exchange (carbon dioxide and oxygen levels) and product recovery. It is also equipped with detectors for the levels of oxygen, pH, temperature and degree of foaming. The reaction vessel is equipped with heating/cooling facilities in order to maintain an optimum temperature for the growth of the host.

All air entering the system is sterile-filtered.

The raw materials (media, salts etc) are continuously sterilised and added to the reaction vessel (and backups called feed 1 and feed 2) as required.

The 'seed' vials of recombinant DNA plasmid with the insulin genes in E. coli are inoculated as required into the reaction vessel.

Antifoam is added as required to prevent excess frothing of the reaction mixture.

Developments such as the NASA bioreactor which operates in microgravity (Earth orbit) requires only very slow rotation to maintain items in suspension. This has the advantages of reducing shear stresses on the cells and allows particles to remain in contact with each other for longer periods of time (Pellis, 1999). The cost of production of human insulin is an issue which pharmaceutical companies are always taking into account, and advances in oxidative cytoplasmic folding and cell-free protein synthesis may replace current expression methods as described here (Swartz, 2001). The production of human mini-proinsulin controlled by the T7 promoter in E. coli BL21(DE3)[pET-3a22] strain at high concentration has been described (Shin et al. 2008). They used a two-stage fed-batch fermentation process with mini-proinsulin in an induction fermenter and recombinant cells in a growth fermenter to enhance the production of mini-proinsulin which in turn can increase the yield of human insulin.


Chen JQ, Zhang HT, Hu MH and Tang JG. (1995). Production of human insulin in an E. coli system with Met-Lys-human proinsulin as the expressed precursor. Appl. Biochem. Biotech. 55: 5-15.

Guo L, Stepien PP, Tso JY, Brousseau R, Narang S, Thomas DY and Wu R (1984). Synthesis of human insulin gene VIII. Construction of expression vectors for fused proinsulin production in Escherichia coli. Gene 29: 251-254.

Pellis NR (1999). Novel approaches to cellular transplantation from the U.S. space program. Diabetes Tech. and Therapeut. 1: 85-87.

Schmidt M, Babu K, Khanna N, Marten S and Rinas U. (1999). Temperature-induced production of human insulin in high-cell density cultures of recombinant Escherichia coli. J. Biotech. 68: 71-83.

Shin CS, Hong MS, Soon C, Bae JL (2008). Enhanced production of human mini-proinsulin in fed-batch cultures at high cell density of Esherichia coli BL21(DE3)[pET-3aT2M2]. Biotech. Prog. 13: 249-257.

Swartz JR (2001). Advances in Escherichia coli production of therapeutic proteins. Curr. Opin. Biotech. 12: 195-201.


For a very long time antibodies for animal and human treatment have been produced by injecting other species (rabbit, horse etc) with the antigen together with an agent (called an adjuvant) to boost the immune response to that antigen. The initial immunisation was done with an adjuvant such as Freund's Complete which contains heat-killed bacteria which serve to increase the level of the antibody response. The serum fraction from bleeds (called antiserum) taken from these immunised animals is used for treatment. This antiserum is polyclonal i.e. it contain specificities directed against a wide range of antigens on the target molecule. The target molecule or antigen will have many sites or epitopes which are recognised by different clones of antibody-producing spleen B lymphocytes. When only one epitope on an antigen is recognised by an antibody such an antibody is said to be monoclonal.

The basic technique for monoclonal antibody (mAb) production (developed by Kohler and Milstein in 1975) is shown in Figure 2. A mouse or rat is immunised with antigen and the spleen removed to provide the antibody-producing B cells. These are then "fused" in a special mixture containing a membrane-disrupting agent with mouse or rat myeloma cells. The myeloma cells are used as they have the machinery to produce antibodies (but do not produce any) and are immortal, thereby providing immortality to the successful hybridoma (antibody-producing cells) formed by selection in appropriate media. The hybridomas are screened for antibody production by methods such as ELISA (enzyme-linked immunoabsorbent assay) or immunohistology. Repeated subcloning leads to a pure hybridoma which will produce a monoclonal antibody directed against a single epitope. Antibodies formed are of one class or subclass e.g. IgM, IgG2a.

Figure 2. Production of monoclonal antibodies.

There are two characteristics of antibodies that make them so valuable. The first is specificity. In the case of monoclonal antibodies, they will recognise only one determinant (epitope) on an antigen. The part of the antibody that binds to the epitope is known as the idiotype. The second characteristic of antibodies is that when activated by a disease, immunological memory is retained in the form of B cell clones, which confers long-term immunity.

mAbs can be produced in large quantities of a high degree of purity. For example, purified immunoglobulin or antibodies can be isolated from serum.  Affinity chromatography is often used. Here, the binding of antibody to antigen on a solid matrix is exploited.  The antigen is bound to chemically reactive beads in a column, and the antiserum is added. Antibodies with specific activity will bind whilst all other molecules including non-specific antibodies pass straight through. After washing, the specifically bound antibodies are eluted, commonly be lowering the pH of the elution buffer.

Affinity chromatography can also be used to purify antigens.  In this case the mAb is attached to the solid matrix and the mixture of antibodies added. The antigen is bound to the immobilised mAb and is eluted again by lowering the pH to disrupt the mAb-antigen bonds which have formed.

Large numbers of mAb-producing hybridomas can be screened using a variety of different methods. Light and electron microscopy, using standard immunohistological techniques, can be used to determine the cellular and subcellular localisation of antigenic epitopes e.g. those directed against particular tumour cells within a tissue sample. However, these methods alone give no indication of the nature of the epitope recognised by the mAbs, and for this procedures such as ELISA, RIA (radioimmunoassay) and antibody microarrays are employed. Radioimmunoassay (RIA) and ELISA are similar binding assays for antibody or antigen, but the method of detecting specifically bound mAb is slightly different. 

RIA is often used to measure the levels of hormones whereas ELISA is frequently used in for the detection of viral antigens. Naturally, these methods are sufficiently adaptable in that mAb or antigen can be immobilised on the assay plates allowing for highly sensitive means of detection by use of "bridging" methods that exploit antigen-antibody interactions. A pure preparation of the antigen or antibody is used to calibrate the assay.  In RIA for an antigen, antibody against that antigen is radioactively labelled. For ELISA, an enzyme, commonly peroxidise, is linked chemically to the mAb.  The antigen is attached to a solid support, such as a plastic multi-well plate, which will absorb a small amount of any protein (mAbs are proteins). The labelled antibody binds to the unlabelled antigen (non-specific binding is blocked by application of unrelated serum) and unbound materials are washed away. Antibody binding in the RIA is measured by direct determination of the radioactivity emitted after washing. In the case of ELISA, binding is detected by a chemical reaction driven by peroxidise-bound antibody that converts a colourless substrate, such as diaminobenzidene (DAB) into a brown-coloured product.  The colour change is read directly in the reaction tray, making data collection very easy, and of course ELISA doesn't involve radioactivity.  ELISA is the preferred method for most binding assays: it is relatively cheap, quantifiable and results often can be seen immediately by the naked eye. It also offers the important advantage, not unique to it, of exploiting intermediate bridging antibody steps which can enhance greatly the level of sensitivity. Cells may contain very large numbers of different protein variants due to post-translational modifications such as phosphorylation and glycosylation (which can offer many variations because of the multiple branching afforded by sugars as opposed to the linear sequence of amino acids in proteins.  In any disease, one of these post-translational modifications may be important in the progression of that disease. Another detection method uses antibody microarrays where a slide ("chip") is coated with thousands of known antibodies, and binding to antibodies on the chip can be measured by fluorescent assays. Use of two- and three-colour fluorochromes can detect less specific families of changes.  Protein microarray chips are a fast method of detecting protein changes in the course of disease. The process is simple and the amount of data generated is quite large:

cellular proteins from samples is extracted

the extracted proteins are labelled with fluorescent dyes

remove unbound compounds (wash)

incubate the labelled proteins with the antibody microarray

Scan and analyse the fluorescent microarray results.

Purifying a recombinant protein is quite laborious and involves cloning the cDNA, expressing and purifying the protein.  Custom peptide synthesis can produce a peptide of known sequence of a protein, and then this peptide can be used to generate an antibody:

peptides for antibody production are usually 15-25 aa (amino acids) long

short peptides (<10 aa) should be selected with potential sequence homology, such as protein family grouping

Shorter peptides (<10 aa) may present a small number of epitopes but may increase the number of antigens detected on proteins.

Standard monoclonal antibody production methods as described above often cannot be used to treat patients suffering from cancer or other diseases because the mouse, or rat, part of the mAb may be recognised as foreign by the human immune system. The patient makes his/her own antibodies against the mAbs which may destroy the mAbs and can cause immune complex damage to the kidneys. (Monoclonal antibodies made in humans would help, but few people would want to be immunised in an attempt to make them!).

Other methods have been tried to produce mAbs that will not cause problems in patients. These include:

Chimaeric antibodies. The antibody combines the antigen-binding parts (idiotype) of the original mAb with the effector parts of a human antibody (the Fc portion).

Humanised antibodies. The antibody combines only the amino acids responsible for making the idiotype of the mouse (rat) antibody (as for chimaeric antibodies above) with the rest of a human antibody molecule.

In both cases, the new gene is expressed in mammalian tissue-culture cells because E. coli cannot add the saccharides that are an essential part for the functioning of these antibodies. Another approach is with transgenic mice that:

have had human antibody genes incorporated into their bodies (using embryonic stem-cell methods).

And have had their own rodent genes for making antibodies "knocked out".

The resulting mouse or rat:

can be immunised with the antigen

produces human, not rodent, antibodies against the antigen

Produces spleen B cells that can be fused with human myeloma cells to produce entirely human mAbs.

Phage display is another method to make human monoclonal antibodies in in vitro systems. Fragments of genes that encode the antigen-binding idiotype of the V (variable) domains of antibodies are fused to genes for the coat protein of a bacteriophage.  Phages containing such gene fusion products are used to infect bacteria, and the released phages have coats that express the antibody-like fusion protein in the correct orientation to bind antigen directly. 

A collection of recombinant phages, each displaying a different idiotype is called a phage-display library.  Phages which express specific antigen-binding domains can be isolated by selecting the phage that binds to that antigen.  These phages are recovered and used to infect further bacteria.  Each phage isolated will produce a monoclonal antigen-binding particle which functions just as a monoclonal antibody.

The genes for the antigen-binding site, unique to each phage, are extracted from the phage DNA and used to construct genes for a complete antibody molecule by joining them to parts of immunoglobulin genes that code for the invariant parts of the antibody.  If these reconstituted antibody genes are introduced into a host cell line, the new or transfected cells secrete antibodies with the characteristics of monoclonal antibodies of the host cell line.

In the same way that phages can display a wide range of antigen-binding sites, the phage can also be engineered to display a wide variety of antigens. This is known as an antigen-display library (compared to the phage-display library above).  The antigens displayed are often small peptides coded by chemically synthesised DNA that has mixtures of all four nucleotides in certain positions, so that all possible amino acids arrangements are produced. It is unusual for every position in a peptide to be varied in this way, because it would result in too many potential combinations (an eight amino acid sequence would result in over 2 x 10 possible sequences of nucleotides!).

mAbs achieve their therapeutic effect by different mechanisms. They can produce apoptosis directly (programmed cell death). They can also block growth factor receptors, thereby stopping the proliferation of tumour cells. In cells that express mAbs they can cause anti-idiotype antibody formation which may lead to a treatment for rheumatoid arthritis (Koopman et al. 1983). Indirect effects can include recruiting cells that are cytotoxic such as macrophages. This type of antibody-mediated cell killing is called antibody-dependent cell-mediated cytotoxicity. Monoclonal antibodies may also bind complement, leading to direct cell toxicity or complement-dependent cytotoxicity.

Transgenic plants have been investigated as a source of mAbs by plasmid transfer (During et al. 1990). Plants offer significant advantages for "molecular farming" - no animals are used, the potential of transfer of animal zoonoses (diseases) to man is reduced, and they can be grown easily and in vast quantities (Daniell et al. 2001).

As with insulin production, microgravity appears to increase synthesis of mAbs (Foster et al. 2003).