- Email: [email protected]
Digestion study of proteins from cooked meat using an enzymatic microreactor
Meat Science 81 (2009) 405–409
Contents lists available at ScienceDirect
Meat Science journal homepage: www.elsevier.com/locate/meatsci
Digestion study of proteins from cooked meat using an enzymatic microreactor Ph. Gatellier *, V. Santé-Lhoutellier INRA, UR370 QuaPA, Centre de Theix, F-63122 Saint Genès Champanelle, France
a r t i c l e
i n f o
Article history: Received 27 June 2008 Received in revised form 5 September 2008 Accepted 10 September 2008
Keywords: Meat Cooking Protein Digestibility
a b s t r a c t A semi automatic flow procedure with photometric detection was developed for the study of meat protein digestion. This system comprised two independent flow pathways, gathered by two compartments. The gastric compartment was simulated by an ultrafiltration cell fitted with a 10 KDa cut off membrane and the intestinal compartment was simulated by a 1 KDa cut off dialysis membrane. The pathways were filled with solutions simulating digestive conditions. The proposed system was employed in digestion studies of whole protein extracts from raw and cooked (100 °C) meat. A mathematical modelling for the determination of the digestive kinetic constants was established. The results show that meat cooking leads to an important decrease of protein digestibility by proteases of the digestive tract. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Meat and meat products are a good source of proteins for humans. These proteins are well balanced in amino acids and contain all the essential amino acids that humans can not synthesize. Nevertheless, meat processes (storage, cooking, salting, irradiation) can affect the physicochemical state of proteins and thus the bioavailability of amino acids. Indeed, during these processes, proteins are the target of free radical attack. Oxidation of amino acids leads to protein aggregation through the formation of disulfide and dityrosine bridges. Such reactions have been detected during meat cooking (Santé-Lhoutellier, Astruc, Marinova, Grève, & Gatellier, 2008) or during exposure of meat proteins to a chemical oxidative system (Morzel, Gatellier, Sayd, Renerre, & Laville, 2006). Protein aggregation can also be generated by the interaction of free amino groups of lysine with aldehydic products of lipid peroxidation and such reactions have been detected during meat storage (Gatellier et al., 2007; Renerre, Dumont, & Gatellier, 1996). Recently, we have reported that the increase of protein hydrophobicity observed during meat cooking was linked to protein aggregation (Santé-Lhoutellier et al., 2008). Many authors have demonstrated that aggregation of meat proteins can affect their degradation by enzymes of the digestive tract (Kamin-Belsky, Brillon, Arav, & Shaklai, 1996; Liu & Xiong, 2000; Santé-Lhoutellier, Aubry, & Gatellier, 2007; Santé-Lhoutellier et al., 2008). Besides less efficient amino acid assimilation, a reduced protein digestion rate would have a negative impact on human health. It has been demonstrated that fermentation of non hydrolysed proteins by colonic flora increases * Corresponding author. Tel.: +33 473 62 41 98; fax: +33 473 62 42 68. E-mail address: [email protected] (Ph. Gatellier). 0309-1740/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2008.09.002
the risk of colon cancer (Evenepoel et al., 1998; Geypens et al., 1997). Thus estimation of protein digestion rate is of great interest in the evaluation of the nutritional and healthy qualities of meat products and that sensitive tools for the study of this process are required. In digestion studies, animal models such as rats or cannulated mini pigs provide realistic information about the degradation and assimilation of proteins which is closely related to human digestion. Nevertheless these experiments are difficult and need procedures that few laboratories have at their disposal and they are inappropriate for large tests. Therefore it is not surprising that this approach is not commonly used and that in vitro models simulating the conditions existing in vivo have been proposed. Up till now, two in vitro approaches have been exploited for protein digestion rate analysis. Some authors have reported the hydrolysis of proteins, in test tubes, using proteases of the digestive tract (pepsin, trypsin and chymotrypsin) in conditions of pH and temperature which mimic the digestive system (Kamin-Belsky et al., 1996; Liu & Xiong, 2000). Recently we monitored myofibrillar protein digestibility by such a method (Santé-Lhoutellier et al., 2007, 2008). These experiments in test tubes have many advantages. They are usually not expensive or time consuming as they only need basic equipment available in most laboratories. Nevertheless, these tests have a number of drawbacks and merit improvement to mimic the real digestive process. In vivo there is a continuity of protein cut by proteases while this in vitro approach always involves separate use of the gastric and pancreatic proteases. There is no doubt that treatment by pepsin prior to trypsin/chymotrypsin treatment will affect the nature of the latter enzyme products. Moreover, to decrease analysis time and increase sensitivity, high concentrations of
406
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
enzyme are generally used which tends to produce undesirable auto digestion of the enzyme itself. Finally, isolation of hydrolysed amino acids and peptides from proteins always requires trichloroacetic acid treatment and the cut off of this chemical treatment corresponds to compounds of approximately 20 KDa molecular weight which cannot migrate through the intestinal wall. In fact, these kinds of measurements, are more accurately described as in vitro enzymatic hydrolysis of protein than in vitro digestibility of protein. There are computer-controlled systems that allow closer simulation of in vivo dynamic physiological processes occurring within the stomach and small intestine of humans. For example the gastric-small intestine system TIM-1 (TNO, Zeist, The Netherlands) reproduces with high reliability the main parameters of digestion, such as pH, temperature, gastric, biliary and pancreatic secretion, peristaltic mixing and passive absorption of water and small molecules. This system has been extensively used in pharmaceutical studies of drugs sensitive to digestive conditions (Blanquet et al., 2004 and 2005). Even though this system might be closest to an in vivo situation, until now, it has not been applied to meat protein digestibility, perhaps because of the high cost and the difficulty of carrying out such experiments routinely. This report describes efforts to develop an artificial digestive system more closely related to the in vivo process than the test tube method and less expensive than the gastric-small intestine system, involving the establishment of a semi automatic flow procedure that enables both the continuous monitoring of protein hydrolysis by the proteases of the digestive tract and the passive diffusion of amino acids and small peptides through a dialysis membrane. A mathematical model for the determination of the digestive kinetic constants was established and feasibility was ascertained by performing protein digestion assays with raw and cooked meat.
2. Materials and methods 2.1. Chemicals All reagents and enzymes were purchased from Sigma. Cell and membranes for ultrafiltration were purchased from Millipor. Dialysis membranes were purchased from Spectra/por. 2.2. Animals and samples The experiment was carried out with bovine M. Rectus abdominis. Four animals (Charolais heifers) were killed in the experimental slaughterhouse of INRA Theix Research Centre. Twenty four hours after death, when the ultimate pH was reached, small samples of muscle were cut in the form of parallelepipeds (0.5 0.5 2 cm) in parallel with muscle fibres and cooked. 2.3. Meat cooking Four samples were placed in sealed polypropylene test tubes (inner diameter 10 mm and thickness 1 mm) and heated at 100 °C in a digital temperature-controlled dry bath (Block-heater, Stuart-Scientific) for 0, 2, 10 and 30 min. This treatment reflected meat oven cooking where similar core temperatures can be reached. The core temperature of the samples was measured with a digital thermometer configured with a thin temperature probe. The temperature profile has already been published (Santé-Lhoutellier et al., 2008). After cooking, samples were cooled at room temperature for 15 minutes to 18–20 °C and then frozen at 80 °C until use. Two separate cookings were performed independently on meat from each animal.
2.4. Preparation of meat extract 1 g of frozen meat was homogenised (Polytron PT 2100, from KINEMATICA AG, Switzerland) in 10 ml of a glycine buffer 0.033 M at pH 1.8 and filtered on gauze to eliminate collagen fibers and most of the lipids which could clog up the pores of the ultrafiltration membranes. Protein concentration of the filtrate was estimated by the Biuret method (Gornall, Bardawill, & David, 1949). 2.5. Realisation of the enzymatic microreactor (Fig. 1) 2.5.1. The gastric compartment The gastric phase experiments were performed in an ultrafiltration cell (inner diameter 65 mm) with a polyethersulfone membrane (molecular weight cut off 10 KDa) fitted in the holder. The gastric medium was simulated by a glycine buffer 0.033 M at pH 1.8. The system was used in conjunction with a heating stirring table. The table was magnetically coupled to a stirring bar, which maintained fluid movement during operation. The cell was immersed in a water bath maintained at 37 ± 1 °C. The ultrafiltration cell was connected to a peristaltic pump. All connections were with a polyethylene catheter (inner diameter = 1.14 mm). The solution was pumped out at a constant rate of 1 ml/min and directly transferred into the detector cell fitted with a 280 nm filter. A total void volume of 3.1 ml existed between the ultrafiltration cell exit and the detector cell entry. After optical density measurement the solution was reintroduced in the ultrafiltration cell. The volume was initially made up to 45 ml with the glycine buffer and the baseline was recorded by adjusting electronically the absorbance to zero. 5 ml of meat extract (with a protein concentration of approximately 18 mg/ml) were then added into the ultrafiltration cell to give a final volume of 50 ml. After stabilisation of the optical density, pepsin (25 U/mg of meat protein) was added and the digestion process was monitored by recording the optical density during 120 min. Pepsin digestion was terminated by adding of 250 ll NaOH 12 N which raised the pH to 8. 2.5.2. The intestinal compartment Trypsin (6.55 U/mg of protein) and chymotrypsin (0.33 U/mg of protein) were added to 15 ml of the pepsic digestion solution. This mixture was immediately introduced in a cellulose dialysis membrane (diameter 11.5 mm) with a molecular weight cut off of 1 KDa. The dialysis membrane was immediately immersed in a 25 ml flask containing 10 ml of glycine buffer, 0.033 M at pH 8. The solution was continuously stirred and its temperature was maintained at 37 ± 1 °C. The solution surrounding the dialysis membrane was continuously changed by two peristaltic pumps. The first pump linked a glycine 0.033 M pH 8 stock solution (500 ml) to the top of the flask and the second pump linked the bottom of the flask to the UV detector. The ingoing and outgoing flow rates were the same (1 ml/min) to maintain a constant volume in the flask. The intestinal digestion process was monitored by recording the optical density for 130 minutes.
3. Results and discussion 3.1. Kinetics measurement in the gastric compartment The record for the pepsic digestion of raw meat is presented in Fig. 2. The rise observed before pepsin injection corresponded to small proteins (<10 KDa) pre existing in the meat extract. After addition of pepsin a lag phase, with no change in the absorbance, was observed which corresponded to the void volume of the
407
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
a Peristaltic pump IN OUT
IN UV detector OUT Cell Recorder Thermometer Water bath Stirring bar Membrane
Heating and stirring table
b IN
Pump 1 OUT
Pump 2 IN
OUT
Glycine pH 8 IN
Stirring table
UV detector Dialysis membrane
OUT
Flask Water bath
Recorder
Stirring bar Heating and stirring table Fig. 1. Schematic diagram of the enzymatic microreactor. (a) The gastric compartment and (b) the intestinal compartment.
3.1.1. Mathematical modelling In order to characterize the gastric digestive process we determined two parameters: The maximum absorbance (Amax), which was a good index of the level of proteolysis, and the half life of the reaction (t1/2) corresponding to Amax/2, which measured the digestion speed. These two parameters can not be extracted directly from the record because absorbance had not reached its maximal value after 120 min. These parameters were obtained by plotting the inverse of absorbance (1/A) against the inverse of incubation time (1/t) and fitting an exponential curve with the Eq. (1):
Pepsin 0.8 1/A = 0.97 Exp 13.0 (1/t) R2 = 0.996
15
0.6 1/OD
Absorbance (OD units)
1
0.4
10 5 0 0
0.2
0.05
0.1
0.15
0.2
1/t
0 0
20
40
60
80
100
120
140
1=A ¼ 1=Amax ebð1=tÞ
ð1Þ
Time (min.) Fig. 2. Ultrafiltration profile of peptides produced by pepsic digestion of raw meat extract. In inset: Exponential curve of the inverse of absorbance versus the inverse of time and its equation.
where b is a coefficient describing the shape of the curve. Extrapolation to 1/t = 0 gives Amax and t1/2 calculated from Eq. (1) is given by the formula: t1/2 = b/Ln 2. An example of this mathematical treatment is given in the inset of the Fig. 2.
system and the delay required for the formation of molecular weight products less than 10 KDa. A rapid increase of the absorbance was then seen. Finally the absorbance increased slowly to reach its maximum value. The speed and amplitude of this second rise depended on cooking time.
3.1.2. Application of the model to raw and cooked meat Fig. 3 shows the effect of cooking time on the two kinetic parameters. An important and significant decrease in Amax was observed with increased cooking time, which indicated a negative effect of cooking on proteolysis intensity. The maximum decrease
408
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
a
b
1.2
30
a 25
0.8
t 1/2 (min.)
A max (OD units)
1
b 0.6 0.4
20
a b
a
b
a b
15 10
c d
0.2
5
0
0 0
2
10
30
0
2
10
30
Cooking time (min.)
Cooking time (min.)
Fig. 3. Effect of the cooking time on the maximum absorbance (a) and the half life (b) measured from the gastric digestion curve. All values are reported as the mean ± SEM of four determinations. Values not bearing common superscripts differ significantly (p < 0.05).
measured between raw and 30 min. cooked meat was 88%. In parallel an increase in t1/2 indicated a slowing down of the digestive process but this effect of cooking was only significant after 30 min. heating. At this time a 33% increase in t1/2 was seen. This negative effect of cooking on protein digestibility has been observed in a test tube experiment where myofibrillar proteins were exposed to pepsin (Santé-Lhoutellier et al., 2008), with a 50% decrease in pepsin activity after 30 min. cooking at 100 °C.
Absorbance (OD units)
0.12
3.2. Kinetics measurement in the intestinal compartment The record of pancreatic protease digestion of raw meat is presented in Fig. 4. This shows a rapid increase in absorbance followed by a slow decrease. The initial increase could be attributed to a large amount of 1 KDa peptides being formed during the prior digestion by pepsin and to the beginning of the proteolysis by pancreatic proteases. During the second stage the 1 KDa peptides were eliminated more rapidly by the buffer circulating system than they were formed by the pancreatic protease hydrolysis. This phenomenon led to impoverishment of the dialysis solution. Even if the dialysis membrane used in this study did not have the physiological properties of the intestinal membrane this system mimicked
0.04
0.06 0.04 0.02
0
20
40
60
80
100
120
140
Time (min.) Fig. 4. Dialysis profile of peptides produced by trypsic and chymotrypsic digestion of raw meat extract. The hatched zone corresponds to the area under the logarithmic curve.
well the continuous displacement of amino acid and small peptides from the lumen of the intestine to the circulating blood flux.
b
a
Abs = -0.0247Ln(t) + 0.141 R2=0.964
0.08
0
5 a
4
0.03
0.02
b bc c
0.01
AUC (OD units)
[a] (OD units / Ln (min.))
a
0.1
3 b
2 c c
1
0
0 0
2
10
Cooking time (min.)
30
0
2
10
30
Cooking time (min.)
Fig. 5. Effect of the cooking time on the director coefficient (a) and the area under the curve (b) measured from the intestinal digestion curve. All values are reported as the mean ± SEM of four determinations. Values not bearing common superscripts differ significantly (p < 0.05).
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
3.2.1. Mathematical modelling The intestinal digestion process is more difficult to analyse because several processes contribute to the absorbance. The previous proteolysis by pepsin interferes with proteolysis by pancreatic proteases. Moreover the continuous elimination of small peptides by dialysis interferes with their continuous formation. In order to characterize the intestinal digestive process the decrease in absorbance was fitted by a logarithmic function with the Eq. (2):
A ¼ aLnðtÞ þ b
ð2Þ
where a is the coefficient of the logarithmic curve describing the speed of the decrease. In addition the area under the curve (AUC), from the beginning of the decreasing phase to the end of measurement, was calculated and used as an index of the intensity of proteolysis. 3.2.2. Application of the model to raw and cooked meat Fig. 5 shows the effect of cooking time on the two kinetic parameters (a and AUC). An important and significant decrease in a value and AUC was observed with increasing cooking time, which indicated a negative effect of cooking on digestion efficiency. The maximum decrease of these two parameters, measured between 0 and 30 min. of cooking, was around 80%. We have previously described a biphasic effect of cooking time when myofibrillar proteins of cooked meat were exposed to pancreatic proteases in test tubes (Santé-Lhoutellier et al., 2008). The difference between the test tube and microreactor results could be explained by differences in experimental conditions, especially the size of the peptide cut off used in each study (approximately 20 KDa in the test tube conditions, with trichloro-acetic acid treatment, and only 1 KDa using the dialysis membrane). 4. Conclusions The system developed was able to semi automatically carry out in vitro digestion of meat protein with continuous monitoring of the proteolysis process. The results obtained with raw and cooked meat were more precise than those usually achieved by the test tube procedure and results were recorded in real time, supporting the value of the system to carry out in vitro assays. This system can be used in studies concerning both the kinetics of protein digestion and diffusion through artificial membranes of
409
the peptides produced. In particular this system could be useful in pre-screening experiments prior to in vivo studies. Further studies will focus on the identification and amino acid composition of the peptides produced. A completely automated, computer-controlled system is also in preparation.
References Blanquet, S., Garrait, G., Beyssac, E., Perrier, C., Denis, S., Hébrard, G., et al. (2005). Effects of cryoprotectants on the viability and activity of freeze dried recombinant yeasts as novel drug delivery systems assessed by an artificial digestive system. European Journal of Pharmaceutics and Biopharmaceutics, 61, 32–39. Blanquet, S., Zeijdner, E., Beyssac, E., Meunier, J. P., Denis, R., Havenaar, M., et al. (2004). A dynamic artificial gastrointestinal system for studying the behaviour of orally administered drug dosage forms under various physiological conditions. Pharmaceutical Research, 21, 585–591. Evenepoel, P., Claus, D., Geypens, B., Maes, B., Hiele, M., Rutgeerts, P., et al. (1998). Evidence for impaired assimilation and increased colonic fermentation of protein related to gastric acid suppression therapy. Alimentary Pharmacology and Therapeutics4, 12, 10–11. Gatellier, Ph., Gomez, S., Gigaud, V., Berri, C., Le Bihan-Duval, E., & Santé-Lhoutellier, V. (2007). Use of a fluorescent front face technique for measurement of lipid oxidation during refrigerated storage of chicken meat. Meat Science, 76, 543–547. Geypens, B., Claus, D., Evenepoel, P., Hiele, M., Maes, B., Peeters, M., et al. (1997). Influence of dietary protein supplements on the formation of bacterial metabolites in the colon. Gut, 41, 70–76. Gornall, A. G., Bardawill, C. J., & David, M. M. (1949). Determination of serum proteins by means of biuret reaction. Journal of Biological Chemistry, 177, 751–766. Kamin-Belsky, N., Brillon, A. A., Arav, R., & Shaklai, N. (1996). Degradation of myosin by enzymes of the digestive system: comparison between native and oxidatively cross-linked protein. Journal of Agricultural and Food Chemistry, 44, 1641–1646. Liu, G., & Xiong, Y. L. (2000). Electrophoretic pattern, thermal denaturation, and in vitro digestibility of oxidized myosin. Journal of Agricultural and Food Chemistry, 48, 624–630. Morzel, M., Gatellier, Ph., Sayd, T., Renerre, M., & Laville, E. (2006). Chemical oxidation decreases proteolytic susceptibility of skeletal muscle myofibrillar proteins. Meat Science, 73, 536–543. Renerre, M., Dumont, F., & Gatellier, P. (1996). Antioxidant enzyme activities in beef in relation to oxidation of lipid and myoglobin. Meat Science, 43, 111–121. Santé-Lhoutellier, V., Astruc, T., Marinova, P., Grève, E., & Gatellier, P. (2008). Effect of meat cooking on physicochemical state and in vitro digestibility of myofibrillar proteins. Journal of Agricultural and Food Chemistry, 56, 1488–1494. Santé-Lhoutellier, V., Aubry, L., & Gatellier, P. (2007). Effect of oxidation on in-vitro digestibility of skeletal muscle myofibrillar proteins. Journal of Agricultural and Food Chemistry, 55, 5343–5348.
Contents lists available at ScienceDirect
Meat Science journal homepage: www.elsevier.com/locate/meatsci
Digestion study of proteins from cooked meat using an enzymatic microreactor Ph. Gatellier *, V. Santé-Lhoutellier INRA, UR370 QuaPA, Centre de Theix, F-63122 Saint Genès Champanelle, France
a r t i c l e
i n f o
Article history: Received 27 June 2008 Received in revised form 5 September 2008 Accepted 10 September 2008
Keywords: Meat Cooking Protein Digestibility
a b s t r a c t A semi automatic flow procedure with photometric detection was developed for the study of meat protein digestion. This system comprised two independent flow pathways, gathered by two compartments. The gastric compartment was simulated by an ultrafiltration cell fitted with a 10 KDa cut off membrane and the intestinal compartment was simulated by a 1 KDa cut off dialysis membrane. The pathways were filled with solutions simulating digestive conditions. The proposed system was employed in digestion studies of whole protein extracts from raw and cooked (100 °C) meat. A mathematical modelling for the determination of the digestive kinetic constants was established. The results show that meat cooking leads to an important decrease of protein digestibility by proteases of the digestive tract. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Meat and meat products are a good source of proteins for humans. These proteins are well balanced in amino acids and contain all the essential amino acids that humans can not synthesize. Nevertheless, meat processes (storage, cooking, salting, irradiation) can affect the physicochemical state of proteins and thus the bioavailability of amino acids. Indeed, during these processes, proteins are the target of free radical attack. Oxidation of amino acids leads to protein aggregation through the formation of disulfide and dityrosine bridges. Such reactions have been detected during meat cooking (Santé-Lhoutellier, Astruc, Marinova, Grève, & Gatellier, 2008) or during exposure of meat proteins to a chemical oxidative system (Morzel, Gatellier, Sayd, Renerre, & Laville, 2006). Protein aggregation can also be generated by the interaction of free amino groups of lysine with aldehydic products of lipid peroxidation and such reactions have been detected during meat storage (Gatellier et al., 2007; Renerre, Dumont, & Gatellier, 1996). Recently, we have reported that the increase of protein hydrophobicity observed during meat cooking was linked to protein aggregation (Santé-Lhoutellier et al., 2008). Many authors have demonstrated that aggregation of meat proteins can affect their degradation by enzymes of the digestive tract (Kamin-Belsky, Brillon, Arav, & Shaklai, 1996; Liu & Xiong, 2000; Santé-Lhoutellier, Aubry, & Gatellier, 2007; Santé-Lhoutellier et al., 2008). Besides less efficient amino acid assimilation, a reduced protein digestion rate would have a negative impact on human health. It has been demonstrated that fermentation of non hydrolysed proteins by colonic flora increases * Corresponding author. Tel.: +33 473 62 41 98; fax: +33 473 62 42 68. E-mail address: [email protected] (Ph. Gatellier). 0309-1740/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2008.09.002
the risk of colon cancer (Evenepoel et al., 1998; Geypens et al., 1997). Thus estimation of protein digestion rate is of great interest in the evaluation of the nutritional and healthy qualities of meat products and that sensitive tools for the study of this process are required. In digestion studies, animal models such as rats or cannulated mini pigs provide realistic information about the degradation and assimilation of proteins which is closely related to human digestion. Nevertheless these experiments are difficult and need procedures that few laboratories have at their disposal and they are inappropriate for large tests. Therefore it is not surprising that this approach is not commonly used and that in vitro models simulating the conditions existing in vivo have been proposed. Up till now, two in vitro approaches have been exploited for protein digestion rate analysis. Some authors have reported the hydrolysis of proteins, in test tubes, using proteases of the digestive tract (pepsin, trypsin and chymotrypsin) in conditions of pH and temperature which mimic the digestive system (Kamin-Belsky et al., 1996; Liu & Xiong, 2000). Recently we monitored myofibrillar protein digestibility by such a method (Santé-Lhoutellier et al., 2007, 2008). These experiments in test tubes have many advantages. They are usually not expensive or time consuming as they only need basic equipment available in most laboratories. Nevertheless, these tests have a number of drawbacks and merit improvement to mimic the real digestive process. In vivo there is a continuity of protein cut by proteases while this in vitro approach always involves separate use of the gastric and pancreatic proteases. There is no doubt that treatment by pepsin prior to trypsin/chymotrypsin treatment will affect the nature of the latter enzyme products. Moreover, to decrease analysis time and increase sensitivity, high concentrations of
406
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
enzyme are generally used which tends to produce undesirable auto digestion of the enzyme itself. Finally, isolation of hydrolysed amino acids and peptides from proteins always requires trichloroacetic acid treatment and the cut off of this chemical treatment corresponds to compounds of approximately 20 KDa molecular weight which cannot migrate through the intestinal wall. In fact, these kinds of measurements, are more accurately described as in vitro enzymatic hydrolysis of protein than in vitro digestibility of protein. There are computer-controlled systems that allow closer simulation of in vivo dynamic physiological processes occurring within the stomach and small intestine of humans. For example the gastric-small intestine system TIM-1 (TNO, Zeist, The Netherlands) reproduces with high reliability the main parameters of digestion, such as pH, temperature, gastric, biliary and pancreatic secretion, peristaltic mixing and passive absorption of water and small molecules. This system has been extensively used in pharmaceutical studies of drugs sensitive to digestive conditions (Blanquet et al., 2004 and 2005). Even though this system might be closest to an in vivo situation, until now, it has not been applied to meat protein digestibility, perhaps because of the high cost and the difficulty of carrying out such experiments routinely. This report describes efforts to develop an artificial digestive system more closely related to the in vivo process than the test tube method and less expensive than the gastric-small intestine system, involving the establishment of a semi automatic flow procedure that enables both the continuous monitoring of protein hydrolysis by the proteases of the digestive tract and the passive diffusion of amino acids and small peptides through a dialysis membrane. A mathematical model for the determination of the digestive kinetic constants was established and feasibility was ascertained by performing protein digestion assays with raw and cooked meat.
2. Materials and methods 2.1. Chemicals All reagents and enzymes were purchased from Sigma. Cell and membranes for ultrafiltration were purchased from Millipor. Dialysis membranes were purchased from Spectra/por. 2.2. Animals and samples The experiment was carried out with bovine M. Rectus abdominis. Four animals (Charolais heifers) were killed in the experimental slaughterhouse of INRA Theix Research Centre. Twenty four hours after death, when the ultimate pH was reached, small samples of muscle were cut in the form of parallelepipeds (0.5 0.5 2 cm) in parallel with muscle fibres and cooked. 2.3. Meat cooking Four samples were placed in sealed polypropylene test tubes (inner diameter 10 mm and thickness 1 mm) and heated at 100 °C in a digital temperature-controlled dry bath (Block-heater, Stuart-Scientific) for 0, 2, 10 and 30 min. This treatment reflected meat oven cooking where similar core temperatures can be reached. The core temperature of the samples was measured with a digital thermometer configured with a thin temperature probe. The temperature profile has already been published (Santé-Lhoutellier et al., 2008). After cooking, samples were cooled at room temperature for 15 minutes to 18–20 °C and then frozen at 80 °C until use. Two separate cookings were performed independently on meat from each animal.
2.4. Preparation of meat extract 1 g of frozen meat was homogenised (Polytron PT 2100, from KINEMATICA AG, Switzerland) in 10 ml of a glycine buffer 0.033 M at pH 1.8 and filtered on gauze to eliminate collagen fibers and most of the lipids which could clog up the pores of the ultrafiltration membranes. Protein concentration of the filtrate was estimated by the Biuret method (Gornall, Bardawill, & David, 1949). 2.5. Realisation of the enzymatic microreactor (Fig. 1) 2.5.1. The gastric compartment The gastric phase experiments were performed in an ultrafiltration cell (inner diameter 65 mm) with a polyethersulfone membrane (molecular weight cut off 10 KDa) fitted in the holder. The gastric medium was simulated by a glycine buffer 0.033 M at pH 1.8. The system was used in conjunction with a heating stirring table. The table was magnetically coupled to a stirring bar, which maintained fluid movement during operation. The cell was immersed in a water bath maintained at 37 ± 1 °C. The ultrafiltration cell was connected to a peristaltic pump. All connections were with a polyethylene catheter (inner diameter = 1.14 mm). The solution was pumped out at a constant rate of 1 ml/min and directly transferred into the detector cell fitted with a 280 nm filter. A total void volume of 3.1 ml existed between the ultrafiltration cell exit and the detector cell entry. After optical density measurement the solution was reintroduced in the ultrafiltration cell. The volume was initially made up to 45 ml with the glycine buffer and the baseline was recorded by adjusting electronically the absorbance to zero. 5 ml of meat extract (with a protein concentration of approximately 18 mg/ml) were then added into the ultrafiltration cell to give a final volume of 50 ml. After stabilisation of the optical density, pepsin (25 U/mg of meat protein) was added and the digestion process was monitored by recording the optical density during 120 min. Pepsin digestion was terminated by adding of 250 ll NaOH 12 N which raised the pH to 8. 2.5.2. The intestinal compartment Trypsin (6.55 U/mg of protein) and chymotrypsin (0.33 U/mg of protein) were added to 15 ml of the pepsic digestion solution. This mixture was immediately introduced in a cellulose dialysis membrane (diameter 11.5 mm) with a molecular weight cut off of 1 KDa. The dialysis membrane was immediately immersed in a 25 ml flask containing 10 ml of glycine buffer, 0.033 M at pH 8. The solution was continuously stirred and its temperature was maintained at 37 ± 1 °C. The solution surrounding the dialysis membrane was continuously changed by two peristaltic pumps. The first pump linked a glycine 0.033 M pH 8 stock solution (500 ml) to the top of the flask and the second pump linked the bottom of the flask to the UV detector. The ingoing and outgoing flow rates were the same (1 ml/min) to maintain a constant volume in the flask. The intestinal digestion process was monitored by recording the optical density for 130 minutes.
3. Results and discussion 3.1. Kinetics measurement in the gastric compartment The record for the pepsic digestion of raw meat is presented in Fig. 2. The rise observed before pepsin injection corresponded to small proteins (<10 KDa) pre existing in the meat extract. After addition of pepsin a lag phase, with no change in the absorbance, was observed which corresponded to the void volume of the
407
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
a Peristaltic pump IN OUT
IN UV detector OUT Cell Recorder Thermometer Water bath Stirring bar Membrane
Heating and stirring table
b IN
Pump 1 OUT
Pump 2 IN
OUT
Glycine pH 8 IN
Stirring table
UV detector Dialysis membrane
OUT
Flask Water bath
Recorder
Stirring bar Heating and stirring table Fig. 1. Schematic diagram of the enzymatic microreactor. (a) The gastric compartment and (b) the intestinal compartment.
3.1.1. Mathematical modelling In order to characterize the gastric digestive process we determined two parameters: The maximum absorbance (Amax), which was a good index of the level of proteolysis, and the half life of the reaction (t1/2) corresponding to Amax/2, which measured the digestion speed. These two parameters can not be extracted directly from the record because absorbance had not reached its maximal value after 120 min. These parameters were obtained by plotting the inverse of absorbance (1/A) against the inverse of incubation time (1/t) and fitting an exponential curve with the Eq. (1):
Pepsin 0.8 1/A = 0.97 Exp 13.0 (1/t) R2 = 0.996
15
0.6 1/OD
Absorbance (OD units)
1
0.4
10 5 0 0
0.2
0.05
0.1
0.15
0.2
1/t
0 0
20
40
60
80
100
120
140
1=A ¼ 1=Amax ebð1=tÞ
ð1Þ
Time (min.) Fig. 2. Ultrafiltration profile of peptides produced by pepsic digestion of raw meat extract. In inset: Exponential curve of the inverse of absorbance versus the inverse of time and its equation.
where b is a coefficient describing the shape of the curve. Extrapolation to 1/t = 0 gives Amax and t1/2 calculated from Eq. (1) is given by the formula: t1/2 = b/Ln 2. An example of this mathematical treatment is given in the inset of the Fig. 2.
system and the delay required for the formation of molecular weight products less than 10 KDa. A rapid increase of the absorbance was then seen. Finally the absorbance increased slowly to reach its maximum value. The speed and amplitude of this second rise depended on cooking time.
3.1.2. Application of the model to raw and cooked meat Fig. 3 shows the effect of cooking time on the two kinetic parameters. An important and significant decrease in Amax was observed with increased cooking time, which indicated a negative effect of cooking on proteolysis intensity. The maximum decrease
408
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
a
b
1.2
30
a 25
0.8
t 1/2 (min.)
A max (OD units)
1
b 0.6 0.4
20
a b
a
b
a b
15 10
c d
0.2
5
0
0 0
2
10
30
0
2
10
30
Cooking time (min.)
Cooking time (min.)
Fig. 3. Effect of the cooking time on the maximum absorbance (a) and the half life (b) measured from the gastric digestion curve. All values are reported as the mean ± SEM of four determinations. Values not bearing common superscripts differ significantly (p < 0.05).
measured between raw and 30 min. cooked meat was 88%. In parallel an increase in t1/2 indicated a slowing down of the digestive process but this effect of cooking was only significant after 30 min. heating. At this time a 33% increase in t1/2 was seen. This negative effect of cooking on protein digestibility has been observed in a test tube experiment where myofibrillar proteins were exposed to pepsin (Santé-Lhoutellier et al., 2008), with a 50% decrease in pepsin activity after 30 min. cooking at 100 °C.
Absorbance (OD units)
0.12
3.2. Kinetics measurement in the intestinal compartment The record of pancreatic protease digestion of raw meat is presented in Fig. 4. This shows a rapid increase in absorbance followed by a slow decrease. The initial increase could be attributed to a large amount of 1 KDa peptides being formed during the prior digestion by pepsin and to the beginning of the proteolysis by pancreatic proteases. During the second stage the 1 KDa peptides were eliminated more rapidly by the buffer circulating system than they were formed by the pancreatic protease hydrolysis. This phenomenon led to impoverishment of the dialysis solution. Even if the dialysis membrane used in this study did not have the physiological properties of the intestinal membrane this system mimicked
0.04
0.06 0.04 0.02
0
20
40
60
80
100
120
140
Time (min.) Fig. 4. Dialysis profile of peptides produced by trypsic and chymotrypsic digestion of raw meat extract. The hatched zone corresponds to the area under the logarithmic curve.
well the continuous displacement of amino acid and small peptides from the lumen of the intestine to the circulating blood flux.
b
a
Abs = -0.0247Ln(t) + 0.141 R2=0.964
0.08
0
5 a
4
0.03
0.02
b bc c
0.01
AUC (OD units)
[a] (OD units / Ln (min.))
a
0.1
3 b
2 c c
1
0
0 0
2
10
Cooking time (min.)
30
0
2
10
30
Cooking time (min.)
Fig. 5. Effect of the cooking time on the director coefficient (a) and the area under the curve (b) measured from the intestinal digestion curve. All values are reported as the mean ± SEM of four determinations. Values not bearing common superscripts differ significantly (p < 0.05).
Ph. Gatellier, V. Santé-Lhoutellier / Meat Science 81 (2009) 405–409
3.2.1. Mathematical modelling The intestinal digestion process is more difficult to analyse because several processes contribute to the absorbance. The previous proteolysis by pepsin interferes with proteolysis by pancreatic proteases. Moreover the continuous elimination of small peptides by dialysis interferes with their continuous formation. In order to characterize the intestinal digestive process the decrease in absorbance was fitted by a logarithmic function with the Eq. (2):
A ¼ aLnðtÞ þ b
ð2Þ
where a is the coefficient of the logarithmic curve describing the speed of the decrease. In addition the area under the curve (AUC), from the beginning of the decreasing phase to the end of measurement, was calculated and used as an index of the intensity of proteolysis. 3.2.2. Application of the model to raw and cooked meat Fig. 5 shows the effect of cooking time on the two kinetic parameters (a and AUC). An important and significant decrease in a value and AUC was observed with increasing cooking time, which indicated a negative effect of cooking on digestion efficiency. The maximum decrease of these two parameters, measured between 0 and 30 min. of cooking, was around 80%. We have previously described a biphasic effect of cooking time when myofibrillar proteins of cooked meat were exposed to pancreatic proteases in test tubes (Santé-Lhoutellier et al., 2008). The difference between the test tube and microreactor results could be explained by differences in experimental conditions, especially the size of the peptide cut off used in each study (approximately 20 KDa in the test tube conditions, with trichloro-acetic acid treatment, and only 1 KDa using the dialysis membrane). 4. Conclusions The system developed was able to semi automatically carry out in vitro digestion of meat protein with continuous monitoring of the proteolysis process. The results obtained with raw and cooked meat were more precise than those usually achieved by the test tube procedure and results were recorded in real time, supporting the value of the system to carry out in vitro assays. This system can be used in studies concerning both the kinetics of protein digestion and diffusion through artificial membranes of
409
the peptides produced. In particular this system could be useful in pre-screening experiments prior to in vivo studies. Further studies will focus on the identification and amino acid composition of the peptides produced. A completely automated, computer-controlled system is also in preparation.
References Blanquet, S., Garrait, G., Beyssac, E., Perrier, C., Denis, S., Hébrard, G., et al. (2005). Effects of cryoprotectants on the viability and activity of freeze dried recombinant yeasts as novel drug delivery systems assessed by an artificial digestive system. European Journal of Pharmaceutics and Biopharmaceutics, 61, 32–39. Blanquet, S., Zeijdner, E., Beyssac, E., Meunier, J. P., Denis, R., Havenaar, M., et al. (2004). A dynamic artificial gastrointestinal system for studying the behaviour of orally administered drug dosage forms under various physiological conditions. Pharmaceutical Research, 21, 585–591. Evenepoel, P., Claus, D., Geypens, B., Maes, B., Hiele, M., Rutgeerts, P., et al. (1998). Evidence for impaired assimilation and increased colonic fermentation of protein related to gastric acid suppression therapy. Alimentary Pharmacology and Therapeutics4, 12, 10–11. Gatellier, Ph., Gomez, S., Gigaud, V., Berri, C., Le Bihan-Duval, E., & Santé-Lhoutellier, V. (2007). Use of a fluorescent front face technique for measurement of lipid oxidation during refrigerated storage of chicken meat. Meat Science, 76, 543–547. Geypens, B., Claus, D., Evenepoel, P., Hiele, M., Maes, B., Peeters, M., et al. (1997). Influence of dietary protein supplements on the formation of bacterial metabolites in the colon. Gut, 41, 70–76. Gornall, A. G., Bardawill, C. J., & David, M. M. (1949). Determination of serum proteins by means of biuret reaction. Journal of Biological Chemistry, 177, 751–766. Kamin-Belsky, N., Brillon, A. A., Arav, R., & Shaklai, N. (1996). Degradation of myosin by enzymes of the digestive system: comparison between native and oxidatively cross-linked protein. Journal of Agricultural and Food Chemistry, 44, 1641–1646. Liu, G., & Xiong, Y. L. (2000). Electrophoretic pattern, thermal denaturation, and in vitro digestibility of oxidized myosin. Journal of Agricultural and Food Chemistry, 48, 624–630. Morzel, M., Gatellier, Ph., Sayd, T., Renerre, M., & Laville, E. (2006). Chemical oxidation decreases proteolytic susceptibility of skeletal muscle myofibrillar proteins. Meat Science, 73, 536–543. Renerre, M., Dumont, F., & Gatellier, P. (1996). Antioxidant enzyme activities in beef in relation to oxidation of lipid and myoglobin. Meat Science, 43, 111–121. Santé-Lhoutellier, V., Astruc, T., Marinova, P., Grève, E., & Gatellier, P. (2008). Effect of meat cooking on physicochemical state and in vitro digestibility of myofibrillar proteins. Journal of Agricultural and Food Chemistry, 56, 1488–1494. Santé-Lhoutellier, V., Aubry, L., & Gatellier, P. (2007). Effect of oxidation on in-vitro digestibility of skeletal muscle myofibrillar proteins. Journal of Agricultural and Food Chemistry, 55, 5343–5348.