Protein folding and misfolding
Protein Folding and Misfolding
Protein folding involves the folding of a polypeptide to a functional three-dimensional structure. The original configuration of a protein is stated by the amino-acid sequence (or primary structure). Each amino acid has a blueprint placed by base triplets in coding regions of genes that are conceded by ribosomes, the amino acids originate and are consecutively joined by these protein building sites of the cell (Ison et al., 2005).
The protein molecule folds during or after synthesis. The folding process for the most part depends on the solvent, concentration, temperature and presence of molecular chaperones (Dinner et al., 2000). Folded proteins have hydrophobic cores which stabilise the folded state, and charged or polar side chains that occupy the exposed solvent surface where they interact with water in the surrounding (Pace et al., 1996). Reduction of the sum of hydrophobic side chains in contact with water is important in the folding process.
Folding in vivo begins co-translationally, so that the protein begins to fold at the N-terminus while C-terminal part of the protein is synthesised by the ribosome. Proteins that are specialised (chaperones) help in the folding of other proteins (Lee and Tsai, 2005). Globular proteins can achieve their native state without assistance. However, chaperones help with folding in case there is a change in the environment i.e. increased temperatures to prevent incorrect folding and aggregation. These chaperones gird to the evolving protein chain and defend the newly formed protein responsive to reactions with the environment (Lee and Tsai, 2005). When the new protein chain is let go by the chaperones, it is usually ready to fold.
Folding involves the establishment of secondary structures, in particular alpha helices and beta sheets. These structures are formed by small numbers of amino acids close together which combine, fold and coil resulting in the formation of a tertiary structure containing functional regions called domains (Alexander et al., 2007). Formation of quaternary structures usually involves the assembly and co-assembly of already folded subunits. The regular beta sheet and alpha helix structures can fold quickly because they are stabalised by intramolecular hydrogen bonds first characterised by Linus Pauling. Protein folding may also involve covalent binding in form of disulphide bridges formed between two cysteine residues in the protein that hold the protein structure together (Alexander et al., 2007).
Proteins may not fold into functional forms when exposed to some conditions. Temperatures above or below the range of 37°C which cells tend to live in will cause thermally unstable proteins to unfold or denature. High concentrations of solutes, extremes of pH, mechanical forces, and presence of chemical denaturants can also do the same (Shortle, 1996). A fully denatured protein lacks both its tertiary and secondary structures. Failure to fold into the intended shape produces inactive proteins with different properties including toxic prions. Several neurodegenerative and other diseases result from the accumulation of these misfolded proteins (Dobson, 2006). Proteins that fail to achieve their native states instead form large aggregate structures by associating with other polypeptide chains that are not folded. These may also occur once a protein acquires a mutation, this being a genetic change leading to the alteration of one of the sites on the chain of the amino acid (Dobson, 2006).
Amyloid formation and Disease
The most common feature found in many neurodegenerative disorders such as Alzheimer's, Parkinson's, and Huntington's disease is the presence of protein aggregates in fibrillar structures called amyloid found inside and outside of brain cells (Uversky and Fink, 2004). The conversion of normal soluble protein into amyloid may result in a gain in toxic function, or a loss in function, or result on occlusion of normal cellular function. The appearance of the aggregates indicates the molecular chaperones are not normally functioning to prevent the accumulation of misfolded proteins. Amyloid formation arises from partially unfolding and exposure of hydrophobic surfaces that are buried inside a folded protein, thus increasing attractive forces among protein molecules (Sipe and Cohen, 2000). These amyloid fibers consist of β strands despite structural origins of precursors. For instance, islet amyloid precursor polypeptide (IAPP) from type two diabetes is natively unfolded (Higham et al., 2000), β-2 microglobulin (β2m) from dialysis related amyloidosis is a β - sandwich (Bjorkman et al., 1987). These systems adopt a β - strand secondary structure after conversion to fibers.
Figure 1.Shows a globule-like form of the protein that is partially folded (b), not the same as the denatured (c) and native (a), β-domains are self associated (d) initiating oligomer function. This leads to further deposition of the protein and development of β-sheet, a core structure of the protofilaments within the amyloid fibril (e).
Amyloid diseases are identified by a change in the absorbance intensity of dyes i.e. congo red or Thioflavin T. These dyes intercalate between the beta-strands. Congo red binds to amyloid and can be demonstrated as a yellow-green birefringence under crossed polarizer's (Benhold and Divry, 1920). This birefringence can be used as a diagnostic for amyloid fibrils, a scan shift in the maximum absorbance.
Thioflavin T (ThT) is a cationic benzothiazole dye that shows enhanced fluorescence upon binding to amyloid. The detection of amyloid with ThT after differentiation in acidic solutions was demonstrated to be highly specific (Vassar and Culling, 1959) thus it is more specific compared to congo red. Due to the specificity of binding of thioflavin T to amyloid fibrils, it has found many applications such as diagnosis of amyloid in tissue sections using fluorescence microscopy, monitoring extracted amyloid and in vitro amyloid fibril formation using fluorescence spectroscopy (LeVine, 1993) and direct observation of amyloid fibril growth using total internal reflection fluorescence microscopy (Khurana et al., 2005).
Insulin is a 51-residue hormone with a largely α-helical structure and exists as a mixture of monomer, dimeric, and monomeric states in solution depending on the conditions; the protein is monomeric in 20% acetic acid, dimeric in 20Mm HCL and hexameric in the presence of zinc (Uversky, 2005).
Insulin is a globular protein and tends to assume and maintain a three dimensional structure where its hydrophobic surfaces are buried by folding and gathering of individual molecules. However, changes of the native form can be stimulated when destabalised using various conditions such as temperature, low pH, organic solvents, or hydrophobic surfaces leading to formation of amyloid-like fibrils (Brange et al., 1997). Formation of insulin fibrils is a physical process by which partially unfolded insulin molecules interact with each other to form linear aggregates. Shielding of the hydrophobic domains is the main driving force but formation of the β-sheet structure may further stabalise the fibrillar structure (Brange et al., 1997).
Amyloid is characterised by a cross-beta sheet quaternary structure i.e. the beta strands arrange themselves perpendicular to the axis of the fibril (Goto et al., 2008) these beta strands are from seperate protein monomers. In each amyloid disease, a particular protein has polypeptide aggregates and forms insoluble fibrils i.e. insulin fibrillation is an example of amyloid fibrillation where amyloid deposits are observed in patients with diabetes after repeated injection and in normal aging, as well as after subcutaneous insulin infusion (Kahn et al., 1999). Aymloid deposition is a pathogenic feature in type 2 diabetes. The extent of amyloid deposition is associated with both loss of β-cell mass and impairment in insulin secretion and glucose metabolism, suggesting a causative role for amyloid in the lesion of type 2 diabetes (Kahn et al., 1999).
Insulin fibers can be observed using an electron microscope and are seen as long fibers. Fibrous insulin has also been studied using other techniques such as spectroscopic and diffraction techniques (Waugh, 1954). Using an infrared technique, insulin fibrils consist of extended β-chains lying perpendicular to the fibril axis, with a layer structure involving interchain hydrogen bonds (Waugh, 1954) .
There are different versions of the insulin protein. These include; human, bovine and porcine among others. Bovine insulin differs from human insulin in only three amino acid residues and porcine in one (Dunn, 2005). Different structural proteins (insulin) form amyloid fibrils when subjected to stress conditions like high temperature, agitation, chemical additives and extreme pH (Dobson, 2001). These environmental stresses trigger off an alternative folding pathway for amyloidogenic proteins, leading to partial unfolding of proteins followed by formation of amyloid having cross β - sheet structure (Dobson, 2001).
The formation of aggregated proteins results in conformational diseases such as type II diabetes. Type II diabetes is characterised by two factors: insufficient ability to secrete insulin and/or decreased sensitivity of peripheral tissues (increased insulin resistance) (Kahn et al., 1999). These two factors are present in varying degrees in different patients. Increased circulating insulin concentrations are a marker for decreased insulin sensitivity occurring in obesity whereas β-cell dysfunction results in lack of appropriate insulin secretion in relation to glucose concentrations (Kahn et al., 1999).
The ability to store and release insulin in a regulated manner is due to pancreatic islets β cells. Deposits of amylin fibrils in pancreatic islets are associated with the disease diabetes (Carrell and Lomas, 1997). Islet amyloid polypeptide (IAPP or amylin) is a 37 aa-residue peptide hormone, islet β-cells co-express and secrete these protein (Butler et al., 2003). In diabetes (late onset), it includes the major fibrillar component of the islet amyloid deposits, a hallmark in type II diabetes mellitus (found in over 90% of patients). Note that amyloid is formed only in type II diabetic subjects because in type 1 diabetes, there is a destruction of the islet β-cells which removes the source of IAPP (Carrell and Lomas, 1997).
Aims and Ojectives
This project will analyse the mis-folding of bovine and human insulin in vitro. The two batches of protein will be destabilised using various conditions such as temperature, low pH, organic solvents or hydrophobic surfaces to form fibrils or amorphous aggregates. Other physiological conditions will also be used i.e. insulin in presence of glucose to investigate fibrillation. The protein fibrils will be stained with two specific dyes congo red and thioflavin T. The absorbance and fluorescence will then be determined using techniques such as using a spectrophotometer or fluorimeter. The amyloid fibers are visualised using an electron microscope.
2.0MATERIALS AND METHODS
Human Insulin (batch no: CAS 11061-68-0) and Bovine Insulin (batch no: CAS 11070-73-8) were purchased from Sigma and used without further purification. Congo red (CR) was obtained from British Drug Houses LTD and Thioflavin T (Th T) obtained from Difco (C.I. no: 42005). Trizma base (batch no: 017k54121) obtained from Sigma-Aldrich. Sodium Chloride (batch no: 0305373) obtained from Fisher Scientific. Glucose (anhydrous) batch no: CAS 492-62-6 obtained from Lancaster and Acetone (E.C no: 200-662-2) was obtained from BDH laboratory supplies.
2.2.1. Calculation of protein (bovine insulin) concentration.
The molecular weight of bovine insulin is 5, 733.49kDa.
Therefore, 1M of bovine insulin protein = 5733.49g/L
1mM = 5. 73349g/L
1mM = 0.005734g in 1000µl
1mM = 0.00401g in 700µl
2.2.2. Preparation of samples.
Stock solutions of insulin protein were made up by weighing out the required amount of insulin and dissolving the dry powder in distilled water adjusted to pH 2.0, pH 4.0, or pH 7.4. These stock solutions of water at pH 2.0, pH 4.0, or pH 7.4 were prepared using distilled water and different dilute solutions of dilute hydrochloric acid (HCL) and sodium hydroxide (NaOH).
2.2.3. Preparation of Insulin Solutions and Incubation conditions.
Insulin samples were prepared from fresh protein stock solutions of 2mM monomeric bovine insulin. The stock protein solutions were freshly prepared each time to prevent inconsistencies in fibril formation due to nuclei formation in a previously prepared stock. 2mM insulin solution was prepared by adding 700µl distilled water at pH 2.0, pH 4.0 or pH 7.4 to 0.008g bovine insulin, respectively.
100µl of 2mM bovine insulin stock solution in known pH water (pH 2.0, 4.0 and 7.4) was incubated at temperatures of 37˚C or 65˚C in order to study the effect of pH and temperature on insulin fibrillation. The samples were incubated for one day, three days and seven days to induce amyloid fibre formation. All samples had a total volume of 100µl of the 2mM insulin solution and were prepared in triplicate. Each of the 3 triplicate samples was analysed to increase precision of the data and the mean of the three values are shown in Section 3. The standard deviation of the mean value was used to produce error bars. The same procedure was repeated using human insulin.
2.2.4. Effect of additives or agitation on insulin fibrillation.
The effect of an additive (i.e. glucose or acetone (at different concentrations)) or the effect of agitation on amyloid formation was also studied. The effect of glucose was investigated at a final concentration of 6mM and 12mM glucose. The study was carried out by incubating 50µl of 2mM insulin stock solution of known pH, 12µl of 100mM glucose stock solution and 38µl of distilled water at correct pH making up the final volume of 100µl. The same study was carried out but using a glucose concentration of 6mM final concentration.
The effect of acetone on insulin fibrillation was investigated at a final concentration of 1.5mM. 50µl of 2mM insulin stock solution of pH 2, pH 4 or pH 7.4 water was added to 1.5µl of 100mM acetone stock solution and volume made up to 100µl with addition of 48.5µl of distilled water at the same pH. All solutions were incubated at temperatures 37˚C and 65˚C for one day, three days and seven days.
The influence of agitation on amyloid fibre formation for both bovine and human insulin was investigated by incubating 2mM final concentration solutions without any additives. The 2mM insulin final concentrations were incubated at pH 2.0, 4.0 or 7.4 and 37˚C for 3days.
2.2.5. Congo Red (CR) Assay.
0.5g of CR was dissolved in 250ml sodium chloride to make up 2.87mM stock solution of congo red. 500ml of sodium chloride solution was prepared by dissolving 15g sodium chloride in 100ml distilled water and 400ml absolute alcohol. On the day of use, 20ml of diluted congo red stock solution was diluted by dissolving 0.697ml of congo red solution in 19.303ml distilled water making a 100µM final dilution.
All samples were spectrophotometrically analysed. The absorbance of each sample was measured by taking 10µl of the insulin solution (with or without additives) after incubation and adding it to 990µl of dilute conge red. A blank of 10µl distilled water with 990µl dilute congo red was used to calibrate the Jenway spectrophotometer at 540nm. The absorbance of each insulin sample was then measured at 540nm.
2.2.6. Thioflavin T (Th T) Assay for determining fibril formation.
1mM stock solution of Thioflavin T was prepared in distilled water, wrapped in tin foil and stored at room temperature (20˚C) because ThT is very light sensitive. A 400ml stock of 1M Tris buffer was prepared at a final pH of 8.5 and stored at room temperature.
0.4ml of 1mM ThT stock and 0.4ml of Tris buffer was added to 19.20ml distilled water to make up 20ml dilute Thioflavin T solution. 10µl aliquots of insulin solution samples after incubation were added to 2990µl dilute ThT and the cuvette inverted a few times before measuring the fluorescence emission on a fluorescence spectrophotometer (Jenway-6285). The excitation wavelength was 425nm with a 60% gain, and the emission was recorded at 540nm. A blank of 10µl distilled water plus 2990µl dilute ThT was measured and subtracted from each sample fluorescence signal.
The amyloid fiber formation of both bovine and human insulin in vitro was monitored under various conditions such as low pH and physiological pH, different temperature the presence of chemical additives (glucose or acetone) and agitation.
3.1. Effect of pH and Temperature on Insulin fibrillation. 2mM solutions of either bovine or human insulin at pH 2.0, pH 4.0 and pH 7.4 were incubated at two different temperatures of 37°C and 65°C in the absence of additives for one day, three days and seven days. Aliquots of insulin solutions were added to dilute solutions of Congo red used to measure the absorbance (Klunk et al., 1989) (data shown in appendix) and Thioflavin T (Th T) measuring the fluorescence intensity as arbitrary fluorescence units (LeVine, 1993) that indicated the presence of amyloid structures. All the insulin samples were incubated in triplicate and the data is expressed as the mean value. The standard deviation of each mean was used to plot the error bars (see Figure 3.1).
The data in figure 3.1 shows bovine insulin readily forms amyloid fibers at the lowest pH 2.0 (blue bars) compared with fibers formed at pH4.0 and pH 7.4 (red and green bars) when incubated at either 37°C and 65°C, However when the temperature was increased to 65°C, fiber formation was also promoted at pH 7.4. Human insulin follows a similar trend to the bovine protein, but is more prone to fibrillation at physiological pH 7.4 after 24 hrs when the temperature is raised to 65°C compared with bovine insulin (green bars in Figure 3.1 (B)).
3.2. Effects of glucose on insulin fibrillation. Insulin amyloid formation was also investigated by studying the effect of glucose levels that might be seen in diabetic patients on fibrillation. Two glucose concentrations were used, one at 6mM (Figure 3.2.1) and the other at a higher concentration of 12mM (Figure 3.2.2) as might be expected in the normal physiological state or hyperglycaemia, respectively.
The data in Figures 3.2.1 and 3.2.2 show that in the presence of glucose at both 6mM and 12mM when bovine and human insulin are incubated at 37°C and 65°C has a similar trend to the samples with no glucose i.e. pH 2.0 incubations tend to have the highest fluorescence intensity followed by pH 4.0 and then pH 7.4. It is interesting to note that when glucose is added, the ThT signal decreases from 24 hours to 7 days at pH 2.0 which is the opposite to what is seen with protein incubated alone. The trend is the same for both concentrations of glucose at 37°C and 65°C but the Thio T signal is higher at the 12mM concentration than at the 6mM concentration at pH 2.0, 4.0 and 7.4. However at a higher concentration, ThT signal at pH 7.4 increases to a higher level comparable to those seen at lower pH incubation especially at 65°C. This indicates that the presence of glucose has a positive effect on the ability of human and bovine insulin forming fibres, thus diabetic patients might be more likely to suffer from insulin fibril formation.
3.3. Effect of Acetone on Insulin Fibrillation. The presence of ketones in the body is common in people with diabetes and can lead to ketoacidosis if left untreated. The ketone levels rise due to shortage of insulin, in response to the body burning fatty acids resulting in production of acidic ketone bodies (Kitabchi et al., 2006).
The effect of acetone on insulin fiber formation was analysed on both human and bovine insulin at a high concentration of 1.5mM (Gaw, 2004) as would be seen in diabetic cases. The samples were incubated at pH 2.0, pH 4.0 and pH 7.4 at temperatures 37°C and 65°C (Figure 3.3).
At 37°C, the ThT signal is low at pH 2.0 (blue bars) in both human and bovine insulin and increases at pH 4.0 and pH 7.4 (green and red bars) opposite to what is seen in Figure 3.1 when protein is incubated alone at 37°C. However when temperatures are increased to 65°C, ThT signal increases at pH 2.0 to a higher level compared to pH 4.0 and pH 7.4 comparable to what is seen to figure 3.1 at high temperatures. This data shows that acetone suppresses fiber formation only at pH 2.0 at lower temperatures for both human and bovine but promotes fiber formation when temperatures are increased at pH 2.0, 4.0 and 7.4. The ThT signal decreases from 24 hours to 7 days for human insulin opposite to bovine insulin with acetone (A) and protein alone (Figure 3.1) at 65°C. This data suggests that the presence of acetone can promote amyloid formation in bovine and human insulin at pH 2.0, pH 4.0 and at pH 7.4 with increased temperatures which may have an effect on diabetic patients.
3.4. Effect of Agitation on insulin fiber formation. This was tested by incubating both bovine and human insulin at 37°C with no additives for one day and 3 days.
Figure 3.4. Thioflavin T fluorescence of bovine insulin (A) and human insulin (B) agitated at 37°C for 24hrs and 3days.
Figure 3.4 shows an increase in Thioflavin T intensity at pH 2.0, 4.0, and 7.4 at 37°C. This is the opposite of what is seen in figure 3.1 when the protein is incubated at the same temperature with no agitation. The data also shows ThT signal markedly increases at pH 7.4 to a level higher than pH 2.0 and pH 4.0 for both human and bovine insulin compared to figure 3.1 where ThT signal is highest at pH 2.0 and lowest at pH 7.4. Agitation promotes fiber formation more at a high pH than lower pH.
Conditions used in in vitro studies of low pH and high temperatures to form amyloid fibers of insulin have been produced successfully in the results above (shown by the small error bars). Other physiological conditions were used i.e. glucose, acetone important in diabetes and it was found they can promote amyloid formation even at pH 7.4 and low temperatures. Agitation was also used to enhance fibrillation. In conclusion, fibrillation under these physiological conditions shows the relationship between insulin fiber formation and diabetes.
A wide range of pathologies i.e. systemic amyloidoses, type II diabetes and Alzheimer's have been recognised as being caused by mis-folding of proteins following conformational change (Kopito and Ron, 2000). It is also known that protein aggregation and fibril formation involve partially folded intermediate conformations (Fink, 1998). In the case of insulin, amyloid fibers are formed when the domains exposed by unfolding interact with a hydrophobic surface domain that is normally buried, resulting in three insulin dimers forming a hexamer (Brange et al., 1997). These hexamers then go on to form oligomers and eventually amyloid fibers.
In this work, two types of insulin (human and bovine) have been analysed under conditions in vitro where insulin is known to form amyloid fibrils (Brange et al., 1997). The amyloid fibers were generated by destabilising the insulin protein using conditions such as temperature (at 37°C and 65°C), pH (acid and neutral), and additives i.e. glucose both high and low concentrations, high concentrations of acetone as would be seen in diabetic ketoacidosis and finally agitation.
4.1. Significance of pH on insulin fibrillation.
Various pH conditions (2.0, 4.0 and 7.4) were used because pH has an influence on insulin association thus has an effect on insulin fibrillation. For instance, in acidic solutions (pH 1.5), insulin maintains its tertiary structure and is monomeric whilst with increasing pH above 2.0 the insulin becomes dimeric (Blundell et al., 1972).
Investigation of the influence of pH on fibrillation in acidic conditions (Figure 3.1 a, b) reveals an increasing tendency to form fibers at pH 2.0 by both human and bovine insulin. These results indicate that the formation of fibrils probably proceeds via monomerisation of insulin (from the hexameric form) before amyloid can form. This also explains the decrease of Thioflavin T signal at pH 4.0 because the insulin molecule is not only monomeric but dimeric as well. However, at pH 7.4, the ThT signal is much lower when compared to pH 2.0 and 4.0. This is because the insulin molecule is hexameric in neutral solutions (Sluzky et al., 1992). Monomers of insulin are less stable and therefore more likely than dimers and hexamers to undergo conformational change. This explains the increase in insulin fibrillation pH 2.0 compared with pH 4.0 and pH 7.4. Since neither human nor bovine insulin form amyloid fibers readily at pH 4.0 and pH 7.4 increasing the temperature and addition of a series of chemicals was used to try to promote fiber formation.
4.2. Effects of temperature.
The effect temperature has on insulin fibrillation at acid and neutral pH is illustrated in Figure 3.1 a, b. The results demonstrate that at 37°C insulin fibrillation in bovine and human is higher in acidic pH compared to insulin fibrillation at neutral pH. This may be because less acidic conditions suppress the nucleation reaction (involves formation of active centers) while allowing the growth reaction to proceed (Lamy and Waugh, 1953). Nucleation reaction is the slowest and requires interaction of three to four insulin molecules. These interactions are mainly between non polar side chains (hydrophobic interactions) and their subunits are suggested to be either the dimer or the monomer of insulin(Burke and Rougvie, 1972). Whereas nucleation seems to require temperatures above normal, the growth of fibrils can proceed at ambient or lower temperatures.
However, when the temperature is increased to 65°C, fiber formation is promoted even at pH 7.4. This might be due to the formation of covalent polymerization products via thiol-catalysed disulfide exchange which becomes significant and increases when there is an increase in temperatures causing fiber formation at neutrality (Costantino et al., 1994).
4.3. Effect of Glucose.
The reference range for glucose in the blood is 4.5mM - 8.0mM (Gaw, 2008). Thus, glucose was used at two concentrations, a physiological concentration of 6mM and a high concentration of 12mM to observe insulin fibrillation in vitro. These two conditions were used because the synthesis of insulin is regulated by glucose levels. Low levels of glucose cause hypoglycaemia a condition common in diabetic cases. High levels of glucose however, causes hyperglycaemia often found in diabetes mellitus. These two conditions occur when the insulin levels are low.
4.3.1. 6mM Concentration.
Figure 3.2.1 and 3.2.2 a, b show that insulin fibrillation is high in acidic solutions compared to neutral solutions in both human and bovine insulin (as seen in Figure 3.1). When glucose is added, the ThT signal decreases between 24hrs and 7 days at both temperatures. This may be because there is fast fibrillation on insulin forming fibrils within a few hours. The protein is partially unfolded forming amyloid fibers then after a while aggregates start forming when the protein is completely unfolded. When the glucose levels are low, the levels of insulin are high thus explaining the rate of insulin fibrillation increases in acid solutions as would be expected with increase in insulin concentration (Waugh, 1957).
4.3.2. 12mM concentration.
When the glucose concentration is doubled, there is an increase in insulin fibrillation even at pH 7.4 at both temperatures as shown in Figure 3.2.2. Also, for both bovine and human insulin, there is a much higher ThT signal when the temperatures are increased compared to Fig 3.2.1. The trend where ThT signal decreases between 24hrs and 7 days is the same when either 6mM or 12mM glucose is used. Insulin at pH 7 is present as a hexamer thus in neutral solutions insulin exhibits an anomalous aggregation behaviour, being more prone to fibrillation at lower concentrations (Brange and Havelund, 1983). It is possible that when glucose levels are high the glucose might act to stabilise the insulin fibers at neutral pH. Insulin fibrillation is again high in acidic solutions because the insulin is present as monomers, and the extent and rate of aggregation is proportional to the concentration (Brange et al., 1987).
4.4. Effect of Acetone.
Acetone was used as an additive to induce insulin fibril formation at a high concentration as would be seen in diabetic cases. Most people suffering from Type I diabetes and type II diabetes (less common) suffer from a condition known as Diabetic Ketoacidosis (Dunger et al., 2004). This condition is caused by a lack of insulin. Insulin's main function is to lower blood glucose levels reducing lipolysis but if insulin levels drop significantly, the body starts burning fat uncontrollably while blood glucose levels raise leading to the production of acidic ketone bodies (Kitabchi et al., 2006).
Patients with ketoacidosis produce high acetone levels explaining why a high concentration of 1.5mM acetone was used. In figure 3.3 a, b, acetone decreases ThT signal at pH 2.0 at 37°C for both human and bovine opposite to what is seen in figure 3.1. This may be due to the fact that acetone decreases the insulin concentration thus rate of fibrillation decreases in acid solutions. However, ThT fluorescence increases at pH 4.0 and pH 7.4 compared to protein alone at 37°C. Thus it can be concluded that at this two pHs acetone destabalises both human and bovine insulin to promote fiber formation.
When the temperature is increased to 65°C, insulin fibrillation is increased in the three pHs used especially at pH 2.0. This may be because insulin fibrillation is accelerated under acidic conditions and high temperatures due to the dissociation of classic oligomers just as seen when protein alone is incubated at 65°C. However, ThT signal decreases between 24hrs and 7 days for human insulin figure 3.3b compared to figure 3.3a bovine insulin. This might be because human insulin starts forming aggregates at high temperatures when left for a longer period of time.
4.5. Effect of agitation.
This condition was used because it enhances fibrillation by partial unfolding of the monomer. It has also been shown to increase fiber formation in a variety of amyloid forming proteins. A study done by Sluzky et al on fibrillation of insulin increasing with surface area of hydrophobic material also gave evidence that aggregation caused by agitation was due to interfacial and shear forces on the protein although shear forces play a minor role compared with surface interactions (Sluzky et al., 1992). The samples agitated were protein alone with no chemical additives. Figure 3.4 shows the ThT signal at pH 2.0, pH 4.0 and pH 7.0 at 37°C from 24hrs and three days both with and without agitation. With agitation pH 4.0 and pH 7.4 has a high ThT signal compared with incubation of protein without agitation. This shows that agitation provoked fibrillation of insulin in both acidic and neutral solutions mediated by insulin's interaction with hydrophobic surfaces (Sluzky et al., 1991). Shaking allows a continuous creation of new interface and thus provides a large surface area for adsorption of insulin. This study has shown insulin fibrillation increases with increasing rate of agitation in agreement with published literature (Sluzky et al., 1991).
In summary, the two batches of insulin were used because of the small difference between tendency to fibrillation of human and bovine. The tendency of insulin to form fibers increases with increase in temperature, and is influenced by concentration and agitation. The data shown here indicates that fibrillation of insulin requires dissociation of monomer followed by conformational change to form a partially unfolded intermediate for both human and bovine. Under physiological conditions such as acetone, ThT signal decreases between 24hrs and 7 days for human insulin opposite to bovine insulin. This may be caused by difference in amino acid levels. Bovine differs from human insulin in three amino acid residues (Dunn, 2005). The two batches behave similarly in the conditions although bovine insulin seems to be more prone to fibrillation as shown in
Study the effect of other physiological conditions such as using different concentrations of insulin, seeding, acetic acid (influence of ion strength) on the rate of insulin fibril formation. Incubate the bovine and human insulin at 37°C with agitation for longer (one week) and compare the results with those for 24 hrs and 3days. Finally, confirm fiber formation by examining a series of samples of both bovine and human insulin by electron microscopy