Saccharomycopsis Fibuligera Alpha Amylases Domain Organization Biology Essay

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.

Amylases are among the most important industrial enzymes covering up to 25% of the market for enzymes used in industries. Among these enzymes, amylase plays the key role in starch conversion producing starch derivatives usable for further processing (1). The enzyme is widely used in the fructose and glucose syrup production, to replace malt in brewing industry, to improve flour in the baking industry, and in production of modified starch for paper industry (2). The enzyme is also used for starch removal in the desizing process in the textile industry and as additive in detergents (3). The application of the enzyme has also been broadened by utilization in the development of renewable energy sources (4), and in the prevention of coronary heart disease and against the development of certain cancers (5). For industrial purposes, namely for gelatinization and liquefaction processes which are conducted at 80-110oC, thermally stable enzyme is required (1). Therefore, amylases have been extensively subjected to research for years with improvement of enzyme specificity and stability as the main goal.

Amylase catalyzes the hydrolysis of 1,4 glycosidic bonds of starch predominantly in random manner resulting in oligosaccharides with reducing glycosidic groups in the configuration. Amylases are metallo-enzymes, which require the presence of calcium ion for both stability and activity. The enzyme is generally stable in a rather wide pH range of 5.5 to 8.0, depending on the origin of the enzyme. The working temperature of the enzyme varies between 25 to 90oC, with an optimum at 50-60oC (6).

The structures and architectures, and possible applications of amylases are well-known and well-explored (2, 7-9). The amylase family consists of starch hydrolases and related enzymes comprising about 20 different enzymes specificities. It is a large enzyme family containing 138 members. This number is still increasing. Despite the availability of crystal structures of the enzyme from bacteria, mammals, and archaea, the structure of amylase from Aspergillus oryzae has for years been the only known fungal one which was elucidated (10). Recently, the structure of amylase from Aspergillus niger, that shares a high sequence homology with the enzyme from A. oryzae was elucidated (13).

A screening at various places in Indonesia resulted in finding the yeast Saccharomycopsis fibuligera strain R64 which demonstrates a high amylolytic activity. This strain was found to secrete two enzymes, identified as and gluco-amylase. Each enzyme was successfully isolated, separated, and independently characterized. The isolated amylase showed an attractive stability upon storage at room temperature up to over three months, indicating its potency as a stable enzyme. The benefit of S. fibuligera as a food borne microorganism and the ability of the S. fibuligera strain R64 to act on raw starch have been an additional attractive value. This has drawn our interest to explore and study the enzyme properties and structure, and further manipulate and modify the enzyme to be useful for industrial and other purposes.

Itoh et al. (11) published the nucleotide sequence of amylase from Saccharomycopsis fibuligera, and described the homology with A. oryzae amylase. Matsui et al. (12) studied the role of amino acid residues near the active site of amylase from Saccharomycopsis fibuligera, using a structural model derived from the X-ray structure of the enzyme from A. oryzae.

This article describes our efforts to modify the enzyme in order to improve the thermostability of the enzyme. Furthermore, making use of protelytic digestion we derived the domain organization, also using the X-ray structure of the enzyme from A. oryzae. (K. Hassan et al., manuscript in preparation). The results provide important information on the enzyme structure and for further manipulation of the enzyme.

Materials and Methods

The yeast S. fibuligera strain R64 was obtained from the culture collection of the Microbiology Laboratory, Inter University Centre for Biotechnology, Bandung Institute of Technology. The strain is kindly maintained and prepared by the Biochemistry Laboratory, Department of Chemistry, Padjadjaran University. All chemicals were from Sigma Aldrich and Merck. DEAE-Toyopearl 650M and butyl-Toyopearl 650M chromatography matrices were purchased from Tosoh (Tosoh Corp. Japan) and Sephadex G-25 gel filtration matrix was from Sigma Aldrich.

Production of amylase

S. fibuligera strain R64 was routinely grown for 48 hours at room temperature (approximately 25oC) on an agar plate containing 6% sucrose and 1.5% bacto agar in an extract of 10% tauge (w/v). For inoculation, the S. fibuligera cells were transferred to the production medium containing 1% sago starch (w/v) and 1% yeast extract (w/v) and were inoculated for 24 hours at room temperature in an Erlenmeyer flask with a constant shaking at 180 rpm on a gyratory shaker (New Brunswick Scientific Co., Edison-NJ, USA). Finally, the enzyme production was carried out for 120 hours at room temperature with a constant shaking at 180 rpm and 10% (v/v) inoculums as the culture starter.

Amylase, which is secreted into the media, was separated from the S. fibuligera cells by centrifugation at 4200 g for 15 minutes. Complete S. fibuligera cell removal was achieved by filtering the media over a filter paper (Whatman Inc., NJ-USA). The filtrate obtained was concentrated in a filter system for cells recycle with a 10 kDa cut-off membrane (Millipore Co., Bedford Massachusetts USA). This concentrated sample is then called crude enzyme extract. Ammonium sulphate was added to the crude enzyme extract to a final concentration of 25% (w/v). The solution was equilibrated overnight with constant stirring and the precipitate formed was removed by centrifugation at 4200 g for 15 minutes. The supernatant was loaded onto a butyl-Toyopearl 650M hydrophobic interaction column equilibrated with 50 mM phosphate-citrate buffer pH 5.8. The column was eluted stepwisely with 25, 20, 15, 10, 5, and 0% ammonium sulphate in 50 mM phosphate-citrate buffer pH 5.8. The remaining hydrophobic contaminants were removed by 30% (v/v) methanol elution. The collected amylase fractions were loaded onto a Sephadex G-25 gel filtration column equilibrated with 50 mM phosphate-citrate buffer pH 5.8. Pure amylase was obtained after chromatography on a DEAE-Toyopearl 650M anion exchanger column using 50 mM phosphate-citrate buffer pH 5.8 with a 0-1 M NaCl gradient as the eluent.

Chemical modification of amylase by CC-PEG 5000

Amylase was modified with PEG-5000 activated by cyanuric chloride (CC-derivative) according to Tsai et al. 14, not correct) Two ml of enzyme was added to 2 ml of 0.1 M sodium borate buffer pH 9.0, containing 0.25 mg activated-PEG to a final mol ratio of 1:2.5, 1:5, and 1:10. The modification was carried out with stirring the mixture for 3 hours at room temperature. The excess of the modifying reagent was removed by dialysis against 20 mM sodium phosphate buffer pH 6.0 for 24 hours.

Enzyme activity assay

The amylase activity was determined according to Fuwa (15). Amylase activity is defined as a 10% decrease in absorbance of starch-iodine complex at 600 nm. In a reaction tube, 250 l of 0.25% (w/v) soluble starch in 0.2 M phosphate citrate buffer pH 5.8 was incubated with 250 l enzyme for 10 minutes at 50oC. To stop the enzymatic starch hydrolysis, 250 l of 1.0 N HCl solution was added. The remaining starch molecules was reacted with I3- by adding 250 l iodine solution (0.2% (w/v) I2 in 2% KI (w/v)) and the reaction mixture was diluted by milli-Q water to a final volume of 5 ml prior to measurement. One unit is? <<< already mentioned: see sentence 2.

The glucoamylase was determined according to Somogyi for reducing sugar (16) with some modifications. The reducing sugar activity was defined as the amount of enzyme that produces reducing sugar equivalent to 1 mol of glucose per minute under the condition described. First, 200 l of 50 mM phosphate-citrate buffer pH 5.8 containing 0.25% soluble starch (w/v) was incubated for 10 minutes at 50oC. To this solution, 20 l enzyme was added to start the reaction and the mixture was incubated at 50oC for another 10 minutes. The enzymatic reaction was stopped by incubation in boiling water for 10 minutes, and finally 5 minutes incubation on ice. The amount of reducing sugar produced was measured by adding 400 l of Nelson reagent (mixture of Nelson A and Nelson B with 4:1 ratio). This mixture was then roughly mixed and boiled for 10 minutes. After 5 minutes of incubation on ice, the colour, indicating the presence of reducing sugar, was developed with the addition of 400 l arsenomolibdate reagent. Finally, 4.9 ml of water was added prior to the colorimetric measurement at 520 nm. The amount of reducing sugar produced was calculated based on the glucose standard curve.

Protein concentration determination

The concentration of protein was determined according to Lowry (17). 100 l of enzyme was mixed with 5 ml reagent C (a mixture of 2% (w/v) sodium carbonate in 0.1 N NaOH and 0.5% (w/v) copper sulphate pentahydrate in 1% (w/v) sodium-potassium-tartrate with a 50:1 ratio) and incubated for 10 minutes in room temperature. Generating the blue colour of protein complex, 50 l of Folin-Ciocalteu reagent was added and the mixture was incubated for another 30 minutes. The absorbance of the protein complex was measured at 750 nm and the protein concentration was calculated using BSA as the standard.

Characterization of the modified enzyme

The optimum pH and temperature values, thermal stability, and effect of salts were determined by measuring the enzyme activity under various conditions. The characteristics of the modified enzyme were compared to the native enzyme to investigate the effect of modification on the enzyme activity and stability. The stabilization factor of the modified enzyme was determined by calculating the enzyme residual activity after incubation at 50oC according to Kazan et al. (18).

Proteolytic digestion

Proteolytic digestion was carried out following Gilkes et al. (19) with some modifications. 25 g of freeze-dried amylase was diluted in 400 l of 50 mM Tris-HCl buffer pH 8.0. The variation of conditions during the digestion was made by either adding 50 l of 8 M Urea and 50 l of 80 mM -mercaptoethanol (also reduced!; final urea conc. only 4 M!) in 50 mM Tris-HCl buffer pH 8.0 to create a partially denatured condition; or 50 l of 83 mM DTT and 50 l of 3.3 mM EDTA in 50 mM Tris-HCl buffer pH 8.0 to create a reducing condition with partial unfolding as result of the removal of Ca++ (?); or 50 l of 35 mM SDS in 50 mM Tris-HCl buffer pH 8.0 to create a complete denatured condition. The total volume of the enzyme reaction was adjusted to 500 l by adding an appropriate amount of 50 mM Tris-HCl buffer pH 8.0. This reaction mixture was pre-incubated for 30 minutes at 37oC. Then, into the solution, 50 l of 0.2 M CaCl2 (this restores the Ca-binding site!) and 50 l of 0.02 mg/ml trypsin in 50 mM Tris-HCl buffer pH 8.0 were added to start the proteolytic digestion. Samples were taken during the incubation after 0, 1, 5, 24, 48, and 72 hours for SDS-PAGE analysis to observe the amylase cleavage pattern.

Sequence homology and structure modelling

Search for homologous proteins was done by the program BLAST (20) (provided online at, only proteins with high homology statistics were selected for further homology study. Sequence homology was studied by the program Clustal-W (21). The amino acid sequence amylases of S. fibuligera strain R64 was derived from its cDNA sequence. Sequence homology with the enzymes from A. oryzae, A. niger and S. fibuligera strain HUT7212 was investigated. The model of amylases from S. fibuligera strain R64 was generated by the on-line program FFAS03 (22). The model template chosen was the reference structure with the highest e-value and identity percentage suggested, which was the structure of amylase from A. niger (or oryzae?) (PDB accession code 2GUY). The final model was prepared using the program PyMol (23).

Results and Discussion

Isolation and purification

Being extracellular, amylase from S. fibuligera strain R64 can be separated from the yeast cells by centrifugation. Although the enzyme can easily be harvested by collecting the medium, the purification of the enzyme is a rather complicated process. The medium has indeed contributed to the major contaminants of the enzyme, but the main impurities of S. fibuligera strain R64 amylase actually originated from the glucoamylase, which possess very similar properties to amylase. Either gel filtration or ion exchanger columns cannot separate the two enzymes. An interesting result was reported by Hostinova et al. (24) where the S. fibuligera strain explored has only secreted glucoamylase. This shows that the type of amylolytic enzyme produced in yeast S. fibuligera is diverging. To discussion or delete; not relevant here. But the and gluco-amylase were successfully separated on a butyl-Toyopearl hydrophobic interaction column. Amylase was eluted earlier than glucoamylase during a stepwise elution with decreasing ammonium sulphate concentration (Fig. 1), which indicates that the amylase is more hydrophilic than glucoamylase. The subsequent gel filtration and anion exchange chromatography successfully removed the remaining contaminants and resulted in pure enzyme as concluded from SDS PAGE analysis (Fig. 2). The enzyme recovery was 32%, which is quite good for the purification of an enzyme from its biological source (Table 1).

Insert figure 1

The purification scheme developed was also very effective as the condition applied in each step was beneficial for the next step in the purification. For example, fractionation of the concentrated crude extract with 25% (w/v) ammonium sulphate was appropriate for the sample requirements for hydrophobic interaction chromatography. Also, the gel filtration step fitted very nicely to remove salts in the sample collected from the hydrophobic interaction column prior to DEAE anion exchanger column. Table 1 also shows that most of the contaminant were removed by hydrophobic interaction column, which shows the effectiveness of this column as main purification step.

Insert figure 2

Insert table 1

Characteristics and stability of the modified S. fibuligera strain R64 amylase

PEG used in our study was activated by cyanuric chloride (CC) resulting in a dichlorotriazine derivative, which will react specifically with nucleophylic groups on the surface of a native protein, generally non-protonated primary amino groups. PEG is a long chain molecule rich in hydroxyl groups, hence it is able to promote the formation of hydrogen bonds (25). PEG has been used as a protein conjugate for pharmaceutical applications, mostly in drug delivery (26). The use of PEG derivatives has shown to give good results in modification studies of liver catalase and serum albumin from bovine and Escherichia coli asparaginase (27, 28). Although there was no significant effect on the enzyme stability, even further destabilized the enzyme, the modified catalase showed resistance to proteolytic digestion with trypsin (29), which interestingly similar to the properties of the modified amylase subjected in this research. (I do not understand this sentence.)

Upon reaction with PEG-5000 the molecular mass of amylase increased by approximately 5 kDa. (Fig. 2). A change in mol ratio of activated-PEG to enzyme from 2.5:1, 5:1, to 10:1 had no significant effect on the increase in molecular mass of the modified enzyme, which suggests that the increase in mol ratio did not alter the amount of PEG molecule bound to the enzyme. This result suggests that most likely only one group on the surface of the molecule reacted with the reagent, and that after that no further reaction occurs. This result differs much from similar experiments on other proteins (27, 28, 29), where extensive modification of amino groups were observed, resulting in highly modified heterogeneous products. The single modified group (or groups) in amylase has not yet been identified. It may be a lysine side-chain with a rather low pK value or the amino group of the still unknown N-terminal residue of the protein. (a referee may suggest that His is also possible) So, we hypothesize that the PEG molecule has bound to the most susceptible nucleophilic group on the surface, then blanketed the enzyme molecule and prevented further binding of other PEG molecules.

Modified amylase from S. fibuligera strain R64 has more than 80% of the enzyme activity of the unmodified enzyme. Although the modified enzyme was active in a broad pH range of 4.5-8.0, similar to that of unmodified one, its precise pH optimum was shifted from 5.0 to 5.5. This change suggests that the modification has resulted in a slightly changed active site environment. The enzyme is active in a broad temperature range with an optimum at 50oC. Modification has shifted the optimum temperature to 55oC. The activities of both the unmodified and modified enzymes were similar below 50oC. But the activity of the unmodified enzyme decreased rapidly above 55oC, and disappeared at 65oC, while the activity of the modified enzyme remained constant until 60oC and was still present at 65oC. The enzyme activity upon incubation at 50oC was measured for 30 minutes in 10 minutes intervals, and showed that the modification had resulted in a twofold more stable enzyme.

Throughout the amylases family, calcium ion has been recognized to play a crucial role in maintaining enzyme activity by keeping the integrity of the enzyme structure (7). Therefore, addition of a metal chelator like EDTA inactivates the enzyme due to extraction of the calcium ion. The activity of S. fibuligera strain R64 amylase also disappeared during EDTA treatment, but could be recovered by titrating back a sufficient amount of calcium, magnesium, or zinc (data not shown). The ability of magnesium or zinc to replace calcium in the activation of the inactivated enzyme is quite common in the amylase family. Interestingly, we found that the modified enzyme was surprisingly more resistant to the addition of EDTA. The modified amylase remained active upon EDTA titration, which indicated inability of EDTA to extract calcium from the enzyme. This result suggests that the modification by CC-PEG has caused better protection of calcium ion in the enzyme structure. (see also Fig. 4).

An increase in thermal stability of a protein can be caused by either the introduction of a new disulphide bridge, increase of the internal hydrophobic packing, increase in the number of hydrogen bonds, or introduction of salt bridges (30, 31). In this case, the large PEG molecule most likely covers the protein surface resulting in a lower local ionic strength and strengthening of hydrogen bonds and salt bridges.

Proteolysis of amylase under various conditions

Structural analysis of S. fibuligera strain R64 amylase was conducted by proteolytic digestion using TPCK treated-trypsin. Hence the cleavage was directed to only exposed lysine or arginine residues.

After 24 hours proteolytic digestion under native condition, two fragments with relative molecular masses of 39 and 10 kDa (designated as p39 and p10) appeared (Fig. 3A). Further digestion increased the amount of these fragments as the enzyme band at 54 kDa (p54) disappeared. These two fragments were stable and suffered no further degradation. This result suggests that the enzyme consists of two separate domains, each tightly packed and highly protected against further trypsin digestion. Most likely, trypsin has cleaved a lysine or arginine residue, which is located in loop between these two domains.

When the enzyme was treated with EDTA to remove calcium and reduced by DTT, followed by proteolysis after returning to native conditions, the cleavage pattern observed was similar to that of the untreated enzyme. However, the time needed for cleavage was shorter. As shown in figure 4A, p39 and p10 appeared within the first hour indicating a faster digestion of the enzyme. Surprisingly, a smaller fragment with a relative molecular mass of 35 kDa (p35) appeared after 48 hours followed by a decrease of the intensity of p39. This suggests that p35 most likely originated from further digestion of p39. We hypothesize that the reducing condition has lowered the compactness of p39 due to reduction of one of more disulphide bridges, resulting in a shorter digestion time and formation of p35. Most likely, the reducing condition has exposed a lysine residue in the 39 kDa frafment resulting in producing p35 upon proteolytic digestion.

Figure 3

Figure 4

Proteolytic digestion of the partially denatured enzyme in 4 M urea in the presence of

β-mercaptoethanol also resulted in formation of the p39 and p10 fragments, similar to that of the native condition. Like proteolytic digestion under reducing condition, the fragments formed again rapidly within less than an hour (Fig. 5A). However, there was no p35 formed during the digestion. This suggests that disulphide bridges play an important role in keeping the integrity of p39.

Finally, it was no surprise that full degradation was observed in the digestion of completely denatured enzyme (Fig. 6A). The enzyme structure collapsed totally, allowing trypsin to cleave all bonds next to lysine and arginine residues, resulting in small fragments undetectable in SDS PAGE analysis. These results show that a stepwise proteolytic digestion can be used to reveal the domain organisation of an enzyme.


There are in total 30 lysine and arginine residues in the amino acid sequence of the enzyme (including the signal peptide, since we have not yet successful to conclude the signal peptide cleavage site) and it is difficult to point which residue will suffer the cleavage because of the unavailability of the enzyme structure. However, we still can explore the domain organisation of the enzyme by digesting the enzyme stepwisely. Therefore, the enzyme was subjected to a set of proteolytic digestion performed under native, reducing, partially and fully denatured conditions. Under the native condition, the enzyme retains all the covalent and non-covalent bonds necessary for the enzyme architecture. Under reducing condition, the disulphide bridges, if present, will be reduced thus results in a less compact and intact structure. Next, under partial denaturation, the presence of denaturing agent will further deteriorate the enzyme structure and finally under denatured condition, the enzyme structure is completely denatured.

Effect of modification on the proteolytic digestion

At first, we should consider that modification of the enzyme resulted in a molecular mass about 5 kDa higher than that of the unmodified enzyme.

As shown in the figure 3A modified amylase was surprisingly not digested by trypsin. The reason for this may be that the lysine, susceptible to tryptic cleavage in the unmodified enzyme is identical to the residue modified by PEG. Another possibility is that as a general effect the modification by PEG resulted in protection against tryptic cleavage elsewhere in the molecule.

Figure 5

Figure 6

With the following paragraph I have problems: Fig 4A (I prefer Fig 4B): the produced band is really at 40 kDa (or 35+5) and not 44 kDa (39+5)? The figure is not convincing. My interpretation is that a 44 kDa band is produced. This would mean that the same Lys or Arg bond is cleaved as in the unmodified enzyme. Another problem that it is impossible to make a hypothesis about the location of your 35+5 kDa fragment?

Taking the assumption that the modification has occurred on the domain A/B, the proteolytic digestion of the modified enzyme under reducing condition has produced the p35 (appear as p40) and p10, but interestingly not p39 (appear as p45, which is absent), close to the digestion of the unmodified enzyme. The p35 fragment, however, was produced with a shorter digestion time than the unmodified enzyme under similar condition (Fig. 4a and 4b) but the protein band of the modified enzyme fragment was less intense than that of the unmodified enzyme. Since the similar experiment on the unmodified enzyme (Fig. 4a) has shown that the disruption of disulphide bridges has most likely decrease the compactness of the A/B domain, this result suggested that the modification was most likely occurred on a lysine in the A/B domain that was part of the p35 fragment upon proteolytic digestion.

The proteolytic digestion of the modified enzyme under partially denatured condition (Fig. 5B) showed a similar outcome to that of the reducing condition except that a longer digestion time was needed. This result confirms the hypothesis that modification has occurred on a lysine residue that was part of p35 fragment upon proteolytic digestion (???).

The effect of modification can clearly be observed when the modified enzyme was subjected to proteolytic digestion under complete denatured condition. While the native enzyme was completely digested (Fig. 6A), the modified enzyme surprisingly survived the proteolytic digestion. As shown in figure 6B, the p54 and p35 fragments were still present, even after 72 hours of digestion, while the p10, similar to that of the unmodified enzyme, was degraded into small non-detectable fragments soon after it was formed. The results from the proteolytic digestion of the modified enzyme also confirmed that the modification with PEG has most likely occurred in the 39 kDa catalytic domain..

Prediction of domain organization of S. fibuligera strain R64 amylase

The structures of all (?) amylase from fungi (including yeasts) consist of three domains namely A, B, and C. The structure of the A/B domain is a 8(/)-TIM barrel: with eight anti-parallel -sheets surrounded by eight helices. The B-domain contains the calcium-binding site. The A and B domains form the catalytic domain, while the function of the C domain is less defined. The A/B domain is connected to the C domain by a flexible loop. The X-ray structure of a fungal amylase (from Aspergillus oryzae) is shown in Fig. 7, while Fig. 8 presents the alignment of the amino acid sequences of the enzymes from A. oryzae and S. fibuligera strain R64 These two sequences are identical at …% of the amino acid positions, with a higher level of homology in the A/B domains than in the C-domain. But the homology in the latter domain is still significant.

The sequence of Sfamy64 in Fig. 8 includes the signal peptide. We do not yet know how the proteolytic processing to the mature form of the enzyme proceeds. Matsui et al. (12) expressed the Sfamy gene in Saccharomyces cerevisiae and could demonstrate that the N-terminal sequence of the mature protein starts with glutamic acid at position 27 of the translated cDNA sequence. These authors explain this finding by cleavage by a signal peptidase of the Ala-Gln (18-19) bond, followed by further posttranslational processing by a calpain-like endopeptidase encoded by the KEX2 gene of Saccharomyces cerevisiae with cleavage after the Lys-Arg sequence at positions 25-26. It may be that this processing also occurs after translation of the Sfamy gene in Saccharymycopsis fibuligera itself. But this is still unknown. It is interesting that three glucoamylase genes of Saccharymycopsis fibuligera, which do not show homology with the amylase gene, have similar sequences of the signal peptides, including predicted cleavage sites by signal peptidase (give litt. reference), followed by the sequence LXKR after 4-7 residues (Fig. ?…). Hustinova et al. (Archives 411) also assume that the mature glucoamylase sequences start after the Lys-Arg sequence, but do not present evidence for this (MY HOBBY; you may postpone this fragmant to a later paper, Khomaini??, and change it if we really know the N-terminus of mature Sfamy. This week I received an e-mail from Hjalmar Permentier that the proteolytic digestion experiment gagal; also of the positive control. He will try again.).

The availability of X-ray structures of homologous fungal amylases and the sizes of the proteolytic fragments as revealed by SDS PAGE, allow to make hypothesis about the likely positions of the tryptic cleavage positions in primary and tertiary structure. It is very likely that the primary cleavage with formation of the fragment of 39kDa occurred in the flexible loop between the A/B- and C-domains.

So far fine. But where did the cleavage occur? Itoh et al. indicate in their Fig. 3 the transition of the catalytic to the C-domain. But in this region there are no Lys or Arg residues. More in the direction of the N-terminus there is an arginine (position 396 in your alignment), but located in an alpha-helix (position z in your fig. 8?). Not likely a cleavage site. More to the N-terminus there are two more lysines in this helix and Lys-381 outside this helix (y in your Fig. 8?). Is it possible that this helix may be part of the p10 fragment and also of the flexible linker between the A/B and C-domains? Where is your residue x located? The further digestion to the p35 fragment is also mysterious. (35 AA shorter). May also be from the N-terminus? A special feature of the amylase structure that all disulfide bonds are local ones (neighbouring Cys residues).

Therefore I do not continue with suggestions for your paper.

Only at the end still some comments for the future

A. oryzae aw starch binding domain (SBD) (8). However, the domain C function remains inconclusive. The domain C of amylase from Lactobacillus amylovorus, for example, was found to responsible for thermostability and optimum working temperature (32). The C domain was also reported to stabilize the catalytic domain (33). The catalytic site is located in the cleft between A and B domains and the integrity of this catalytic domain is supported by the presence of calcium ion installed in the B domain. Additional domains such as N or D domains are indeed present in the several amylase family members but the function of those is not yet clearly understood. The core and C-domains is separated by a tunnel and interconnected by a loop. The structure of A. oryzae amylase (PDB accession code 7TAA) shows that the nature of the domain interface is flat surfaces constructed by hydrophobic residues interconnected by weak hydrophobic interactions and salt bridges.

Figure 7

Mostly, amylase catalytic domain is constructed by 8(/)-TIM barrel, eight anti-parallel -sheets surrounded by eight helices. This motif is typical for and highly conserved within the amylases family and related to their activity (7). We have found that the sequence of S. fibuligera strain R64 is almost identical to that of S. fibuligera strain HUT7212. There are only three amino acid residues in the S. fibuligera strain R64 differ from that of reported by Itoh (11). Hence, it is amenable that amino acid sequence of the S. fibuligera strain R64 amylase shares also high homology and identity with that of A, niger and A. oryzae (Fig. 7). Moreover, those residues crucial for activity and construction of Aspergillus amylase structures are conserved and located at the similar position. Therefore, it is tempting to employ the structure of A. oryzae amylase to explain the phenomena observed in our works, which was suggested by Matsuura suggestion (10) decades ago. However, the FFAS03 program and our homology study suggested that we should employed the newly elucidated A. niger amylase structure as the template to put up the S. fibuligera strain R64 amylase structure model.

Based on the enzyme model built and the proteolytic digestion studies, we were trying to predict the position of amino acid that has most likely suffered proteolytic cleavage and chemical modification. They were two plausible positions; the first is the lysine residue located at the loop connecting the two A/B and C domains while the second is most likely located in the surface of the A domain. During the cleavage of the unmodified enzyme under native condition, a lysine residue at the loop was most likely cleaved by trypsin resulted in p39 and p10 fragments, suggesting the p39 and p10 were actually the canonical A/B domains and C domain. When proteolytic digestion was done under reducing condition, whereas disulphide bridges are reduced and the A/B domains become less compact, the cleavage may be occur to the next lysine residue located ahead in the sequence. This lysine residue should resulted p35 upon proteolytic digestion and is located on the surface of the A domain. Adapting the structure of A. niger amylase, it is clear that the cleavage has occurred on the loop region.

Modification of the enzyme with CC-PEG should be occurred at the exposed lysine residues in the enzyme surface. Taking into consideration that the modified enzyme digestions under reducing and partially denatured conditions have resulted in p35 and p10, a lysine on the surface of the A domain was the most likely candidate for the modified lysine residue. The model indicates that modification of this lysine residue with a large molecule of PEG has plausibly occurred at the back of the A/B domains, the other site of the active site cleft (Fig. 8). The big PEG 5000 molecule most likely wraps the enzyme molecule through a random hydrogen bond with amino acid residues on the enzyme surface. Hence, we speculate that the modification has "locked" the enzyme resulted in a less flexibility of these domains. This hypothesis may also explain why the modified enzyme was more resistant to inactivation due to the calcium ion extraction, has a slightly lower activity, and more over, and was more resistant to proteolytic digestion.

Figure 8

The results presented are indeed interesting. Referring to the proteolytic digestion of the unmodified enzyme, the cleavage has most likely occurred to the lysine residue on the loop region. However, the proteolytic digestion of the modified enzyme suggested that the modification should be occurred on a lysine residue in the A/B domain, which was not cleaved upon proteolytic digestion. Since modification should only be occurred on the susceptible and exposed lysine residues, this result indicated that these lysine residues are available in the A/B domain therefore PEGylation has occurred. It remains unclear since these lysine residues should also be plausible cleavage site for trypsin. We proposed that the domain A/B is highly compact thus although such cleavage has occurred to some lysine residues on the surface of this domain, the domain A/B has not undergone any disintegration. The proteolysis of the unmodified enzyme under reducing condition supported this speculation with an indication that disulphide bridges play important role in supporting the integrity of the A/B domain


We have successfully determined the amino acid sequence (no not here, this article) and recently crystallized the enzyme (Zeily et al.?) reported in this publication. We are also preparing a publication concerning the study on the domain C function and currently pursuing an extensive study of the enzyme fragmentation analysis by means of mass spectrometry. Hopefully, we will able to elucidate the enzyme structure in the near future and confirm the reported results. The chemical modification and proteolytic digestion studies we reported here has provided a shed of light of the domain organisation, beneficial for the determination of the enzyme structure.

What to be published about the future. Khomaini et al. is in preparation. X-ray and mol. Biology studies in progress? The works should be repeated to have a more convincing evidence, the results of Khomeini should be confirmed.