Background information on proteins for meat

Proteins    |    Muscle Structure    |    Muscle Proteins    |    Muscle to Meat    |    Prior Raman work on meat and muscle     |    References

Proteins

Proteins are large naturally occurring heterogeneous polymers built from a small range of monomers called amino acids (referred to as residues when polymerised). There are about 20 commonly found amino acids in nature (generally in the L enantiomer), with a wider range of residues (fully condensed amino acids in a peptide) possible with post-translational modification, one important example is the hydroxylated proline and lysine residues in collagen. The main amino acids are listed below in Table 1, including three and one letter codes used for identification1. Unless otherwise stated Stryer has been used as the source of information.

There are a four of levels of structure in proteins, with primary structure the simplest and quaternary the highest level of organisation. The primary structure is the order in which the amino acids appear (labelled from the nitrogen terminus (unreacted amino group) to the carbon terminus (unreacted carboxyl group)), which determines the forces prevalent at different parts of the protein. The primary structure determines the higher-level structures possible since the forces generated by the residues along the chain hold these structures together. The primary structure is maintained by covalent bonding at the peptide link formed by condensation of the amino group and carboxyl group of adjacent amino acids.protein condensation of peptides

 

Class

Name

Code (3)

Code (1)

Side chain structure

 

Glycine

Gly

G

-H

Aliphatic

Alanine

Ala

A

- CH3

 

Valine

Val

V

-CH(CH3)2

 

Leucine

Leu

L

-CH2CH(CH3)2

 

Isoleucine

Ile

I

-CH(CH3)CH2CH3

 

Proline

Pro

P

-(CH2)3-(aN)

Aromatic

Phenylalanine

Phe

F

-CH2Ph

 

Tyrosine

Tyr

Y

-CH2Ph-OH

 

Tryptophan

Trp

W

-CH2In

Sulphur

Cysteine*

Cys

C

-CH2SH

 

Methionine

Met

M

-CH2CH2SCH3

Acidic

Aspartic Acid

Asp

D

-CH2COOH

 

Glutamic Acid

Glu

E

-CH2 CH2COOH

Basic

Lysine

Lys

K

-(CH2)4NH2

 

Arginine

Arg

R

-(CH2)3NHC(=NH)NH2

 

Histidine

His

H

-CH2Py

Amides

Asparagine

Asn

N

-CH2CONH2

 

Glutamine

Gln

Q

-CH2 CH2CONH2

Hydroxyl

Serine

Ser

S

-CH2OH

 

Threonine

Thr

T

- CH(CH3)OH

Post Translation Modification

Hydroxy Proline

HPro

P

-(CH2)2CHOH-(aN)

Table 1 List of the common amino acids, their identification codes and their structures. * Cysteine becomes known as cystine upon the formation of a disulphide link.

This peptide bond has partial (bond order ~1.4) double bond character, which results in a rigid plane of sixpeptide bond schematic atoms with the possibility of geometric isomers, although the trans isomer is almost exclusively found (only ~1% of proteins contain cis). The bonds adjacent to the peptide bond are free to rotate and it is these rotations that allow the formation of the higher levels of organisation in proteins. A schematic representation of the peptide bond is shown in Figure 1, with the rigid plane of six atoms (CaC(O)N(H)Ca) highlighted by the purple background. The C=O and N-H groups are polarised allowing electrostatic and hydrogen bonding to occur within and between lengths of the peptide chain, these forces creating the local conformations (secondary structure) of the chain. There are a limited number of secondary structures possible and even fewer that occur in significant abundance. The most important two types of secondary structure, in terms of relative abundance, are the a-helix and the b-sheet. However, various turns (such as b-turns) and less systematically organised backbone (generally referred to as ‘random’ or ‘disordered’) are vitally important in determining higher-level structure, though much less abundant than the more prevalent a-helices and b-sheets.

The a-helix is a right-handed helix conformation, which brings the backbone into maximum van der Waals contact and all the peptide bonds are involved in hydrogen bonding. The hydrogen bonds occur within the length of peptide forming the helix and are parallel to the helical axis while the residue side groups project outwards from the helix. The helix has 3.6 residues per turn and a rise of 0.15 nm per residue. The C=O of residue i hydrogen bonds to the NH of residue i+4 and all the carbonyls point towards the carbon terminal. Proline residues disrupt any secondary structure as they do not have any hydrogen on the nitrogen of the peptide bond, this will cause a break in the helix. The right-handed helix is preferred as it accommodates the b-carbons in the residues, so that a left-handed helix only possible in poly-glycine and peptides made from D-amino acids (e.g. some bacteria)

The other main conformation is the b-sheet, in which the peptide chains are fully extended and aggregate side by side, with the hydrogen bonds occurring between separate lengths of the peptide chain and are perpendicular to the peptide backbone. The lengths of peptide can line up parallel (all going in the same direction, e.g. from N terminus to C terminus) or anti parallel (not going in the same direction). Both arrangements can occur in the same protein and each has significant differences in structure. The parallel type has evenly spaced hydrogen bonds, which are perpendicular to the sheet axis, and the side groups directly above the a-carbon, whereas the anti-parallel type has alternating hydrogen bonds and the R groups have to sit away from the vertical position due to steric hindrance. When a mixed sheet occurs (with both types present) then the angles (torsion and side group etc.) are all compromised. Inclusion of a proline residue causes a bulge in the sheet. 

However, for each of these secondary structures to interact and form tertiary structures the protein must alter its axis and fold over. In order to accomplish this, the various sections of helix and sheet are linked together by ‘random’ sections and various types of turn. This term ‘random’ is misleading as the chain adopts specific rotations about each peptide bonds, dependant on environmental conditions. A more accurate term for such regions would be ‘unsystematic’ as there is no systematic way of describing the angles formed at each peptide without summarising this information for each individual peptide bond. One of the most important turns is the b-turn, which lies in one plane and creates a 180 o turn in the direction of the peptide. The turn is stabilised by a hydrogen bond between the C=O of the ith residue to the N-H of the (i+3)th residue. There are three types of b-turn, I and II differ in which side of the plane the O and H attached to the second peptide link lie, while III is actually a short 310 helix (3 residues per turn and hydrogen bonding occurring between carbonyl oxygen and the amide hydrogen which occurs ten atoms along the peptide backbone) and is non-planar.

The tertiary structure is the overall folding of the peptide chain to form, for example, fibres or globules. The tertiary structure is determined by the secondary structure, which is in turn determined by both the primary structure and environmental factors. A number of forces are employed to control and hold the tertiary structure: electrostatic forces between charged or polar side groups, hydrogen bonding, van der Waals and covalent bonds (disulphide bonds between two cysteine). The folded peptides can then aggregate together (using similar forces) to form larger proteins, this aggregation being referred to as quaternary structure.

Proteins are highly organised macromolecules with high specificity in their chemical compositions, physical structure and function. Slight changes in chemical or physical factor can have a dramatic effect on the protein, completely destroying or changing its functionality by disrupting the overall structure of the protein. This is used extensively in the food industry as much of the food processing alters the proteins of the food in beneficial ways and so it is important to be able to quantify the effect of the various types of processing. Ageing and cooking are used to alter the protein structure of the meat to make it more tender and palatable while pasteurisation destroys the proteins of microbes, preventing the protein from functioning 2-4.

The Structure and Composition of Muscle and Meat

structure of muscles and meatMeat is post mortem animal skeletal muscle tissue, mainly consisting of water, protein, fat and polysaccharide. Post mortem muscle is converted into meat by a series of biochemical pathways that result from the changes in cell environment upon the cessation of blood flow and nerve activity.

In order to understand the Raman spectra of meat, it is necessary to understand muscle structure and biochemistry. The information in this section was taken from three main sources 3, 5, 6. Skeletal muscle is, as the name implies, associated with the skeleton and its function is the movement of the skeleton to affect various actions. Skeletal muscle is distinguished by the striations due to the fibre-based organization in the cell. The tissue consists of connective tissue, intramusclar fat and sarcomeres. Hundreds of cylindrical protein fibres 1 to 40 mm long (sarcomeres), each bound by an elastic membrane called the perimysium, are bound together by connective tissue (epimysium) which is in turn connected to the skeleton. The sarcomeres contain bundles of smaller fibres called myofibrils which are made up of numerous proteins, the most important being actin and myosin which are the proteins responsible for contraction.

 The muscle fibres appear striped under the microscope with a series of bands perpendicular to the myofibrillar axis, as shown in Figure 2. These bands are a consequence of the overlap of the contractile proteins, with darker regions corresponding to regions of high overlap. The light I band consists mainly of myosin, the H zone consists mainly of actin, while the dark regions of the A band consist mainly of actomyosin complex.  Structural proteins that hold the contractile proteins in place appear as dark lines (the Z and M lines).

Structure Of Muscle Proteins.

In order to understand and interpret the Raman spectra of meat it is also important to be aware of the detailed structure of meat proteins. The main source of information for this section was Pearson and Young (1989). Meat proteins can be classed as those involved in myofibrillar structure and those in the sarcoplasm. The main proteins in meat are myosin (~50 %) and actin (~20 %), while the next most important proteins in the biochemistry of muscle function are tropomyosin and troponin, which are associated with actin in the thin filament. The sarcoplasmic proteins are mainly composed of enzymes responsible for muscle function.

The myosin molecule, represented in Figure 3, consists of 6 polypeptide units arranged as astructure of meat actin myosin long fibrous tail with two globular heads. The rod allows packing of the myosin into the thick filaments of the myofibril. Its length is ~150 nm with diameters of 8 nm at the head and 2 nm at the tail giving the protein a mass of ~500 kD. The main part of the molecule, the S2 subunit (440 kD), consists of two identical strands which form a coil along ~50% of their length. The remaining 50% of the strands (S1) forms the globular head with which are associated the remaining strands of the molecule (14 and 20 kD strands on each side of the head region). Together the S1 and S2 subunits make up the high molecular mass (HMM) portion of myosin while the small peptides associated with S1 make up the LMM fraction. The secondary structure is 56% a-helical overall (95% for tail, 30% for head).

The head region of the myosin incorporates three active sites: the actin binding site, thprotien structure schematic actin myosine ATP binding/hydrolysing site and the phosphate-binding site. These structural elements are represented in Figure 4 below. The actin binding site is located on the opposite side of the head to the ATP and phosphate binding sites, with a narrow cleft between the lobes. These lobes are connected by two sections of the peptide chain, which cross the cleft. The point believed to generate the movement of the tail lies next to the ATP hydrolysis site and consists of a short loop that directs the peptide into the cleft and then a a-helix with a kink in the middle around a glycine residue. This in turn connects to a longer a-helix that contains a proline/tryptophan kink that alters the angle of the rod approx. 90 o, around which the LMM chains bind to the myosin. This helix then continues into the S2 subunit, which forms the bulk of the thick filament.

In the myofibre, actin occurs in the form of Fibrous (F) actin, which consists of approximately 400 42 kD polypeptide monomers polymerised into a fibrous chain with a mass of 17,000 kD and length of 1.0 mm. Monomeric actin contains less than 30 % a-helical structure. The actin molecule shows a high degree of orientation, with the actin peptides forming arrowhead shapes that point away from the Z-line. Actin forms a helix from two strands, which are bound together at two sites on each of the actin monomers, as shown in Figure 5. The grooves between the two strands of F-actin contain the active sites for binding to the myosin heads and for the regulating proteins.protein structure actin fibres

Tropomyosin composes ~7 % of the total protein found in the myofibril and is closely associated with troponin. Tropomyosin is almost completely a-helical in a double helical strand, contains 284 amino acids and has a molecular mass of 33 kD. Tropomyosin is always closely associated with troponin, a protein that consists of three main sub units, C (18 kD, binds Ca2+), I (23 kD, inhibits actin-myosin interaction) and T (37 kD, binds strongly to tropomyosin). The fibrous T subunit binds along the tropomyosin strand while the I and C subunits form a globular head.

Essentially the tropomyosin is a flexible molecule, which binds to actin at different sites that determine the interaction of actin with myosin. The conformation of tropomyosin is dictated by the conformation of associated protein troponin, which in turn is affected by the concentration of Ca2+. It is these interactions that allow Ca2+ ions to control the process of contraction.

The transition from muscle to meat

 

Following death, muscle attempts to retain its biochemical integrity and maintain the level of ATP6. This is partially achieved through the process of post-mortem glycolysis of glycogen and the production of lactic acid. The reserves of glycogen are used up in the production of lactic acid, reducing the amount of energy available for the replenishing of ATP levels. As glycogen disappears, other energy sources such as creatine phosphate, ADP and AMP are used to maintain cell integrity. When the energy resources are depleted there is no longer any energy available to pump Ca2+ ions out of the sarcoplasm (which catalyses the formation of actomyosin), nor to break the association of actin and myosin. This has the consequence that the myofibrillar proteins form a rigid network of actomyosin. This rigor network is much tougher than the myofibrillar network prior to the formation of actomyosin. When meat is held at refrigeration temperatures, following the formation of rigor, a process called ‘ageing’ or conditioning’ causes the meat to tenderise. This increase in tenderness is due to the breakdown in the structure of the Z-lines which hold the actin and myosin in place6. Toughness, or texture, is considered the most important aspect of meat quality in the western world today 27 and so it is desirable to measure this parameter to predict the quality of the meat. The detailed mechanism for the contraction of muscle is detailed below, starting with in-vivo process followed by how it differs post-mortem.

In-vivo muscle contraction

The cross bridges between myosin and actin seem to be the site of the only major movement during muscle contraction. X-ray diffraction studies show that in order for the cross bridge to form the myosin heads have to reorient from a helical pattern about the myofibrillar axis to a less regular pattern28. A maximum of only 20-30 % of the myosin heads can possibly orientate in such a way as to bind to actin leaving the other 80 % orientated closer to the thick filament. The radial movement of the myosin heads during contraction causes a reduction in the helical periodicity about the myofibrillar axis. The myosin head rotations are believed to be unsynchronised which is confirmed by the fact that force generation is known to be steady rather than stepped. However there is a 10-15 ms delay between the cross bridge movement and the development, which suggests that the contraction is caused by structural reorganisation after the development of cross bridges.

Current research has shown that the actual conformational change in myosin inducing the rotation is confined to a few small regions of the myosin heads 29, 30. The head is comprised of two domains linked by a number of interlinking portions of the polypeptide 31. These two domains are then linked to the tail portion of the polypeptide by a long section of peptide containing a kink32, 33. Upon hydrolysis of ATP these domains rotate relative to each other, causing the connecting section to change angle, thus rotating the head relative to the fibre axis 34-37.

In the thin filament of relaxed muscle, troponin covers the carboxy-terminal end of tropomyosin, thereby keeping the troponin bound to the edge of the actin groove. Under these circumstances the thin filament is rigid and motion of actin and tropomyosin are both restricted. When Ca2+ ions are present they bind to the troponin, changing the conformation of the T subunit, pulling the troponin from the carboxy-terminal of the tropomyosin. The tropomyosin then moves into the actin groove where it is bound by seven weak interactions to the actin. As the interactions are weak compared to the interactions with the edge of the groove, the rigidity of the tropomyosin and actin is reduced. Since the tropomyosin is bound inside the groove, the actin and myosin are able to bind. Initial binding of myosin is slight but it activates myosin ATPase, possibly by constraining the interaction of tropomyosin and actin. The actin is now very flexible and the binding sites become more accessible to the myosin binding heads.

schematic flow chart contraction of muscle and development of rigor mortisThe sequence of events in the contractile cycle can be summarised as shown in Figure 6 and as described in the text below. The association of ATP with the S1 portion of the myosin molecule causes the dissociation of myosin from actin (see stage 1 in Figure 6). Upon dissociation the myosin head undergoes a rotation from ~45 o to nearly 90 o relative to the thick filament axis (2). The bound ATP is then hydrolysed to ADP and Pi (inorganic phosphate) with another conformational change that continues to store the energy from the ATP hydrolysis and the myosin is unable to bind with the actin at this stage (this is known as the refractive stage). This transient state then undergoes a restructuring allowing the myosin to bind with the actin (3). It is believed the latter stage is the rate-limiting step when the cytoplasmic levels of Ca2+ are elevated. It is at the next step (4) that the control step can be switched off when low levels of Ca2+ are present. The myosin will bind with the actin as long as there is no tropomyosin blocking access to the actin binding sites.  The myosin-actin cross bridge now dissipates the stored energy in order to effect a cross bridge movement from 90 o to ~45 o (5). The Pi is then released (6) followed by a final conformational change (7) before the ADP is released to allow an ATP molecule to bind to the myosin (8).

Post-Mortem muscle contraction.

The following section used a number of sources, 3, 6, 38.

Upon death the blood supply ceases to flow round the muscle and so oxygen and glucose are no longer delivered to the muscle cells. The muscle cells continue metabolism in an effort to maintain the cell as it was before death, but this uses up the remaining oxygen and so the metabolism rapidly becomes anaerobic. The most important difference, in this context, between aerobic and anaerobic respiration is that the latter does not completely oxidise glucose molecules to CO2 and H2O, but rather produces lactic acid as the end product. As metabolism continues the glucose/glycogen supply and the pH both drop, which significantly alters the biochemistry of the cell as homeostatic conditions are destroyed.

The drop in energy sources for ATP means that the production of ATP from ADP using glucose is eventually reduced, causing the cell to use ADP, AMP and creatine phosphate (CP) as emergency energy supplies to replenish the levels of ATP. However, as the energy reserves are not being replenished these backup supplies quickly disappear and the cell is unable to metabolise. One of the many consequences of this is that ATP is no longer available to allow dissociation of the actomyosin complex, causing an ‘irreversible’ binding (it is chemically reversible by adding ATP, but biochemically irreversible due to lack of ATP). At stage one there is no ATP available to fill the empty nucleotide-binding pocket and so the complex cannot continue the cycle any more.

Another consequence of the diminishing energy supply is that soon the cell membranes can no longer obtain the energy required to pump excess Ca2+ ions out of the sarcoplasm. The release of Ca2+ is the stimulus for contraction of the muscles during life, and relaxation is effected by pumping the ions back out of the sarcoplasm and so preventing the reformation of the bonds between actin and myosin. However, in death, the energy supplies are continuing to diminish which means that the energy required to manufacture ATP is not being replenished and without ATP the contraction and binding of the actomyosin becomes irreversible. This is the state that is known as rigor mortis, with the contracted and tightly bound actomyosin complex responsible for the increased toughness.

It is not surprising that the decreasing extensibility of the muscle is directly related to the decreasing levels of ATP and the fibres contract and bind without relaxation - at 50% original ATP level the muscle elasticity has dropped to 50% of the original level 6. The extensibility of the muscle changes dramatically on the onset of rigor. Pre-rigor muscle will stretch 20% under a 50g / cm2 load compared to 1% for rigor tissue. It is possible to prevent shortening (and concomitant increase in toughness) by restraining the ends of the muscle so that the actin and myosin are unable to generate enough force to slide past each other. This method (known as the tender-stretch method) is used to reduce rigor toughness in a number of muscles 6. However, this is not feasible for the majority of the muscles, as this would require considerably more resources to dissect each muscle and then hang under tension. It is important to note here that it is not the amount of overlap of actin/myosin that determines toughness, but the number of interactions between actin and myosin. Hence, post-rigor meat will not toughen upon cooking (no new actomyosin bridges forming) despite shortening, while cooking prerigor muscle causes an increase in toughness due to the formation of new actomyosin bridges.

As the glucose is anaerobically metabolised the pH continues to drop as the lactic acid builds up in the sarcoplasm. The rate of pH fall is determined by various factors including the species/breed of the animal (e.g. the pH in pig muscles drops quicker than for beef), type of muscle, pre-slaughter stress and activity. The final pH depends on the amount of glycogen present upon death, with a drop of one pH unit for every 46.5 mmol lactic acid/g muscle and with two lactic acid molecules produced per glucose molecule. In living muscle the myofibrillar proteins acted as buffers against increases in pH, and now in death they also buffer against a sharp decline in pH due to their amphoteric properties. In pigs the resting pH is typically 7.18- 7.3, while for beef it is typically 7.08 and these values rapidly drop upon death and eventually reach a minimum (5.4-5.7 and 5.5-5.6 for pork and beef, respectively) by 48 hours 6. 

Tension development begins at different pH values depending on the load place on the muscle, though all muscles will contract to some degree at pH 5.75. Despite the considerable decrease in extensibility, the actual tension development is a maximum of only 5 % of the maximum in the living tissue.

For most cuts the best that can be done is to reduce rigor toughness by controlling the environmental conditions of the post-mortem muscle. Temperature affects different species in different ways e.g. beef will be toughest stored at 1-2 oC and most tender at 16-25 oC, while pork will be toughest at 37 oC and more tender at all lower temperatures. Storage of beef at temperatures below 14 oC causes a condition known as cold shortening, where the muscle shortens significantly, whereas at 14-19 oC there is minimal shortening.  Shortening of 35-40 % gives maximum toughness, with maximum actin-myosin overlap. Shortening of 50-60 % will cause tearing of the Z-disks, which destroys the anchoring of the actomyosin complex, increasing tenderness. This means that the muscle must be allowed to condition and age for a longer period to overcome the effects of cold shortening. Rapid freezing before the onset of rigor causes a condition known as thaw rigor upon defrosting (see below). Extended ageing will not completely negate the toughening effect of cold shortening.  If rapid freezing is desired, electrical stimulation must be employed to prevent cold shortening.

In freshly slaughtered muscle the conditions have not had time to change significantly from in-vivo conditions. The exact state of the muscle depends on a considerable range of factors such as lairage, handling, slaughter technique, animal stress etc., but this is not of primary importance for this project. These factors affect not only the state of the fresh muscle but will also heavily influence the biochemistry of the post-mortem tissue, thereby affecting the end point of the ageing process. It will be assumed that the freshly slaughtered muscle is relaxed and effects of pre-slaughter handling and different ageing conditions will not be considered.

 

 References

 

1.         Stryer, L., Biochemistry. 4 ed.; W. H. Freeman and Company: New York, 2000.

2.         Lawrie, R. A., Meat Science. 5 ed.; Pergamon Press: Oxford, 1991.

3.         Ledward, D. A., et al., The Chemistry of Muscle-based Foods. The Royal Society of Chemistry: 1992.

4.         Wong, D. W. S., Mechanism and Theory in Food Chemistry. 1 ed.; Van Nostrand Reinhold: New York, 1989; p 428.

5.         Bourne, G. H., Structure Part 1. 2 ed.; Academic Press: London, 1972; Vol. 1, p 576.

6.         Pearson, A. M.; Young, R. B., Muscle and Meat Biochemistry. Academic Press, Inc.: London, 1989; p 457.

7.         Barron, L. D., et al., Applied Spectroscopy 1996, 50, (5), 619-629.

8.         Bouraoui, M., et al., Food Research International 1997, 30, (1), 65-72.

9.         Carey, P. R., Journal of Biological Chemistry 1999, 274, (38), 26625-26628.

10.       Lippert, J. L., et al., Journal of the American Chemical Society 1976, 98, (22), 7075-7080.

11.       Overman, S. A.; Thomas, G. J., Biochemistry 1999, 38, (13), 4018-4027.

12.       Thomas, G. J., Biopolymers 2002, 67, (4-5), 214-225.

13.       Yu, N., CRC Critical Reviews in Biochemistry 1977, 229-280.

14.       Spiro, T. G., Biological Applications of Raman Spectroscopy. John Wiley and Sons: New York, 1986; Vol. 1.

15.       Tu, A. T., Raman Spectroscopy in Biology: Principles and Applications. John Wiley and Sons: 1982.

16.       Asher, I. M., et al., Laser Raman Spectroscopy: a New Probe of the Molecular Conformations of Intact Muscle and its Components. In Physiology of Smooth Muscle, Bulbring, E.; Shuba, M. F., Eds. Raven: New York, 1976; pp 229-238.

17.       Caillé, J. P., et al., Canadian Journal of Physiology and Pharmacology 1987, 65, 1416-1420.

18.       Caillé, J. P., et al., Biophysical Journal 1982, 37, 120a.

19.       Caillé, J. P., et al., Federation Proceedings 1983, 42, (4), 997.

20.       Caillé, J. P., et al., Biophysical Journal 1984, 45, 382a.

21.       Pézolet, M., et al. VIIth International Conference  of Raman Spectroscopy, North Holland, 1980; Murphy, W. F., Ed. North Holland, 1980; pp 600-601.

22.       Pézolet, M., et al., Biophysical Journal 1988, 53, 319-325.

23.       Caillé, J. P., et al., Biophysical Journal 1987, 51, 279a.

24.       Pézolet, M., et al., Biochimica et Biophysica Acta. 1978, 533, 263.

25.       Pézolet, M., et al., Biochimica et Biophysica Acta. 1978, 544, 394.

26.       Aubin, M., et al., Biochimica et Biophysica Acta 1980, 631, 90-.

27.       Huffman, K. L., et al., Journal of Animal Science 1996, 74, (1), 91-97.

28.       Padrón, R., et al., Journal of Molecular Biology 1998, 275, (1), 35-41.

29.       Kinose, F., et al., The Journal of Cell Biology 1996, 134, (4), 895-909.

30.       Rayment, I., et al., Science 1993, 261, (5117), 50-58.

31.       Nyitrai, M., et al., European Journal of Biochemistry 2000, 267, (14), 4334-4338.

32.       Lörinczy, D.; Belagyi, J., European Journal of Biochemistry 2001, 268, (22), 5970-5976.

33.       Houdusse, A., et al., Proceedings of the National Academy of Sciences of the United States of America 2000, 97, (21), 11238-11243.

34.       Burghardt, T. P., et al., Biochemistry 2001, 40, (15), 4821-4833.

35.       Burghardt, T. P., et al., Biochemistry 2001, 40, (15), 4834-4843.

36.       Nath, S.; Khurana, D., Current Science 2001, 81, (1), 78-82.

37.       Holmes, K. C.; Geeves, M. A., Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 2000, 355, (1396), 419-431.

38.       Bourne, G. H., Structure Part 2. 2 ed.; Academic Press: London, 1972; Vol. 2, p 651.