1
MSc Chemistry
Analytical Sciences track
Literature Thesis
Discovery approaches for natural inhibitors
Analytical methods for the discovery and identification of plant-based enzyme
inhibitors with a focus on lipoxygenase.
by
Bob van Dooren
12389331
10-2020
12 ECs
August 3
rdtill September 25
thSupervisor:
Dr. B Hollebrands
Examiner:
Dr. A. Astefanei
Dr. R. Haselberg
2
Abstract
Lipoxygenase enzymes are a group of catalytic enzymes that play a role in the deoxygenation of fatty acids and are present in both plants and animals. Their activity is unwanted in food products due to the formation of compounds that causes off-flavours and odours. The addition of inhibitors can stop this unwanted enzymatic activity, but a universal approach for finding these specific inhibitors is lacking. In this thesis, several analytical methods are introduced that are used in the inhibitor discovery process for both plant and human lipoxygenase. In addition, an overview of industrial processes and their effect on soybean lipoxygenase enzymes is provided. The analytical methods discussed are of spectroscopic, chromatographic, and digital nature. This thesis presents the question: ‘What is currently the best discovery approach for plant-based inhibitors of plant lipoxygenase enzymes?’. It was found that an organic solvent extraction of dried plant material followed by enzyme activity monitoring through the detection of conjugated dienes is the simplest and most often used approach to detect inhibiting compounds in plant extracts. Organic solvent extraction allows for easy separation of the compounds by HPLC and subsequent identification using UV-spectroscopy, mass spectrometry, and NMR. Even though this approach is most often used, it lacks any real high-throughput potential for industrial scale. In addition, the use of in silico testing used for inhibitor discovery of human lipoxygenase enzymes was found not to be beneficial for the plant variant. Lastly, other detection and extraction methods discussed in this thesis such as, oxygen monitoring and supercritical fluid extraction, all have advantages that would not be beneficial for preliminary inhibitor discovery but could be implemented on an industrial scale in the scenario where inhibitors are added to food products as a food processing technique.
3
TABLE OF CONTENT
Abstract ... 2
Chapter 1: Introduction ... 4
Chapter 2: Lipoxygenase mechanics ... 5
Chapter 3: Industrial processes ... 8
3.1 Conventional methods ... 9
3.2 Alternative food processing techniques ... 11
Chapter 4: Plant-specific lipoxygenase inhibitor screening methods ... 15
4.1 Formation of conjugated dienes ... 15
4.2 Oxygen consumption ... 16
4.3 Co-oxidation of colourimetric reagents ... 18
4.4 Formation of secondary products ... 23
4.5 Radiochemical ... 24
Chapter 5: Inhibitor discovery and identification methods ... 26
5.1 Discovery ... 26
5.2 Drying and extraction ... 28
5.3 Separation and identification ... 31
5.4 High-throughput ... 33
5.5 Example of SLOX inhibitors... 33
Chapter 6: Human lipoxygenase inhibitor screening methods ... 35
Chapter 7: Conclusion and discussion ... 37
Acknowledgements ... 39
4
Chapter 1: Introduction
Since the last decade, most western countries have seen a shift in consumer demand towards more sustainable food products. This shift can be witnessed with the growing market for plant-based meat alternatives. For example, the meat substitute industry in the Netherlands has seen an annual growth of around 10% since the early 2010s. Besides, the Netherlands is one of the fastest-growing markets for plant-based meat alternatives globally [1]. The demand for more sustainability within the food industry can also be seen in the wish for less processed food products. Recently, a growing suspicion surrounding the concept of food processing techniques can be seen within Dutch consumers, with specific processes being perceived as risky due to the possibility of synthetic chemical contamination [2]. This scepticism and wish for different food products are changing the food industry, bringing with it new challenges to overcome. Now new plant-based food products need to be developed, which are subjected to the least amount of food processing. However, this creates an obstacle for the food processing industry due to the high amount of catalytic enzymes present in high-protein legumes, such as lipoxygenase enzymes [3].
Lipoxygenases (LOX) are a family of enzymes that are present in both plants and humans. These catalytic enzymes play a role in an organism’s inflammatory response to stress by speeding up the degradation of fatty acids. The presence of active LOX enzymes is unwanted in food products because they are responsible for the formation of off-flavours and odours. Currently, a range of industrial processes is implemented to stop this unwanted enzyme activity, e.g., the application of high heat or the addition of preservatives. Alternatively, LOX enzyme activity can be inhibited by the presence of inhibitory compounds such as abietic acid or 4’,5-dihydroflavone [4][5]. Recently, there has been a growing demand for LOX enzyme inhibiting compounds originating from plant extracts. The addition of such compounds to food products would stop unwanted enzyme activity and allow for less food processing.
Despite the growing demand for plant-LOX inhibitors, only a few dedicated inhibitor discovery approaches exist for this plant variety. Currently, new inhibitors for plant lipoxygenases are discovered by techniques such as in-vitro screening or by activity-based profiling [6]. Per contra, most research about lipoxygenase inhibition focuses on the human variant of the enzyme. Although, most human LOX inhibitors are being tested on plant LOX due to the fragility of the human enzyme variant. Those plant-LOX results are used to model and predict the inhibitory effect on human lipoxygenase [7]. However, this approach could hypothetically be reversed and be applied to plant lipoxygenase as well. High-tech analytical methods exist for the discovery and identification of inhibiting compounds for a wide range of different classes of enzymes. In theory, these methods could be applied for the discovery and identification of plant-based enzyme inhibitors.
This review aims to present an overview of analytical techniques that are applied for the discovery and identification of plant-based enzyme inhibitors with a focus on plant lipoxygenase. Examples of existing plant LOX specific methods will be given and discussed, in addition to discovery methods for other types of enzymes. It will be discussed whether those methods could be applied to plant LOX as well. Furthermore, an overview of industrial processes used for enzyme inactivation is given to aid the comparison of established natural inhibitors.
5
Chapter 2: Lipoxygenase mechanics
Lipoxygenase (EC 1.13.11.-) is a family of catalytic enzymes that play a crucial role in the deoxygenation of polyunsaturated fatty acids. LOXs contain non-heme iron monomeric proteins and are found in most plants, fungi, and animals [8]. Human LOX enzymes are expressed in immune, epithelial, and tumour cells and play a crucial part in inflammation response pathways. Human LOX consists of six enzyme isoforms that are similar in shape but differ in properties, such as substrate specificity. In recent years, their involvement with diseases such as cancer and heart disease have been better understood [9]. Lipoxygenase enzymes are also found in most common fruits and vegetables [10]. Although present in most plants, historically, biochemical characterisation of plant LOX has mostly been performed on the soybean isoforms. Soybean seeds are the richest known source of plant lipoxygenase. Soybean lipoxygenase (SLOX) has been identified to contain four isoforms: 1, 2, 3a, and 3b. However, L-3a and L-3b are highly identical in properties and are, in most cases, referred to together as L-3 [11]. Vick and Zimmerman published a comprehensive overview of plant LOX and showed the catalysed reaction which all four SLOX isoforms take part in (Fig. 1) [12].
Figure 1. The lipoxygenase catalytic reaction. Reproduced from ref. [12].
LOX catalyses the incorporation of molecular oxygen into polyunsaturated fatty acids. Usually, oxygen is incorporated at position n-6 or n-10 of the fatty acid, but some exceptions can be found. The product formed is a hydroperoxydiene in which the attacked double bond moves into conjugation with a neighbouring double bond and assumes trans configuration. For this reaction to occur, fatty acids need to contain a cis,cis-1,4-pentadiene structure. The most common plant fatty acid to have this structure is linoleic acid. The isoenzymes SLOX-1, 2, and 3 are categorised by their positional specificity during oxygenation of linoleic acid. They oxygenate at either the C-9 or the C-13 position of linoleic acid, which leads to the formation of two groups of compounds, (9S)‐hydroperoxy (9‐HPOD) and (13S)‐ hydroperoxy (13‐HPOD), respectively [13]. SLOX-1 (94 kDa) has maximal activity at pH 9.0 and converts linoleic acid into 13-HPOD and 9-HPOD at a ratio of 95:5, respectively, in an oxygen-rich environment. SLOX-2 (97 kDa) has maximal activity at pH 6.8 and converts linoleic acid into equal amounts of 13-HPOD and 9-13-HPOD [14]. SLOX-3 (96.5 kDa) has maximal activity at a pH 6.5 and shows a moderate preference for producing 9-HPOD [15].
6 A crucial part of the LOX mechanism is the Fe3+/Fe2+ cycle. This cycle regulates the activity of the LOX
enzyme and is pH dependant. Not all involved parts have been studied or are fully understood, but Kulkarni has presented the current understanding of the Fe3+/Fe2+ cycle (Fig. 2) [16]. This cycle is best
summarised by Chedea et al. [8]. They state that at alkaline pH, the active Fe3+ enzyme form is reduced
to the inactive Fe2+ form. When reduced, the enzyme takes an electron from the substrate, and the
base OH- takes a proton; this produces a free radical from the aforementioned 1,4-diene system.
Molecular oxygen reacts with the radical and forms a peroxyl radical that retakes the electron from the inactive Fe2+ enzyme. This causes the release of the peroxidate anion and regenerates the enzymes
to its active form. Finally, the peroxidate forms hydroperoxide products by reduction with a proton. The Fe3+/Fe2+ cycle is essential for inhibitory studies, with the iron core being the active site of the
enzyme where the inhibitory compounds can bind to or disrupt this cycle in different ways.
Figure 2. Schematic representation of the Fe3+/Fe2+ cycle. (E) Lipoxygenase, (LH) fatty acid, (XH) xenobiotic. Reproduced from
7 It is hypothesised that LOX plays a vital role in pest control due to their involvement with the formation of green leaf volatiles (GLVs). It has been shown that GLVs are formed after the disruption of plant issues by biotic or abiotic stress. These compounds are used for signalling within and between plants, allowing plants and other organisms to recognise and compete with each other. Matsui looked at the hydroperoxide lyase pathway of oxylipin metabolism [17]. They showed parts of the biosynthetic pathway of linolenic acid hydroperoxide, which highlight the many ways reaction products can be formed (Fig 3a). The most abundant reaction products are formed through fatty acid hydroperoxide lyase and are classified as well-known GLVs (Fig. 3b).
Figure 3. (a) Biosynthetic pathway for oxylipins in plants. (b) common GLVs formed by hydroperoxide lyase reaction. Reproduced from ref. [17].
8
Chapter 3: Industrial processes
Food processing has been met with criticism and doubt in the last decade. This criticism is made worse by the lesser understood boundaries of what is and is not considered food processing. Food processing is defined as; ‘any of a variety of operations by which raw foodstuffs are made suitable for consumption, cooking, or storage.’1 Food processing dates back to the prehistoric ages and has seen
rapid developments in the last century. Most processing techniques, either old or modern, apply heat, remove oxygen, or add anti-bacterial additives (Table 1).
Table 1. Chronological development of food processing techniques. Reproduced from ref. [18].
Traditional processing More modern processes (1900 onwards)
Most modern techniques (post-1960)
Canning Extrusion cooking Freeze drying
Fermentation Freezing and chilling Infrared processing
Oven drying Pasteurisation Irradiation
Pickling Sterilisation Magnetic Fields
Salting Ultra-High Temperature (UHT) Microwave processing
Smoking Modified atmosphere packaging
Sun drying Ohmic heating
Pulsed electric fields Spray drying
Ultra-sonification
The processing of food products is linked to a wide range of beneficial properties. The main advantages include the preservation and convenience of food products, the preservation and improvement of nutritional quality, the increase in palatability and sensory experience, and, most importantly, ensuring the safety of food products. Many modern food processing techniques have the dual purpose of preserving food products and their nutritional value [18].
Now that modern consumers are sceptic about conventionally processed food products and food processing, alternatives are being looked for; and an overview of the impact of conventional food processes is needed. The following overview will look at the impact of conventional food processes on plant lipoxygenase activity, mainly soybean lipoxygenase. The processes looked at will be from experimental set-ups because industrial methods and results are not disclosed. However, they will most likely not deviate much from optimised experimental conditions which are linked to high yields and low production costs.
9
3.1 Conventional methods
The processing methods of soybeans and soybean products are comprehensive examples of the effect of industrial processing on plant enzyme activity. Currently, only 2% of soybean protein produced is directly consumed by humans. The other 98% is processed to become livestock feed [19]. One factor in this low consumption is the aversion of western populations of the intense ‘beany’ flavour of unprocessed soybean products [20]. This ‘beany’ flavour is described as bitter and rancid and is caused by the soybean lipoxygenase enzymes breaking down the polyunsaturated fatty acids. This unwanted flavour is linked to the presence of aldehydes, with hexanal being the most abundant [21]. Most industrial food processing methods applied to soybeans are for the specific inactivation of the SLOX enzyme to halter the formation of aldehydes. The most popular processing methods involve the application of heat. Soybeans need to be cooked before they can be consumed; therefore, applying heat is already part of the production procedure.
Chong et al. published an interesting method for soymilk production using superheated steam (SHS) [22]. The SHS method was used for cooking the soybeans and for releasing their protein content in addition to inactivating the SLOX enzyme. The SHS treatment benefited from removing moisture from the product and being an oxygen-free environment that further inactivated SLOX enzymes. The time and temperature of the soybean treatment with SHS were tested to allow optimal protein extraction and lowest SLOX activity. The optimised conditions were found to be a treatment of 9.5 minutes at a temperature of 120°C. This resulted in a product with a crude protein content of 3.0% and a reduced SLOX activity from 0.0236 µmol/(min mg protein) to 0.0081 µmol/(min mg protein). SLOX activity was reduced with 66%, which is considered large, although the initial enzyme activity of the control group was low, to begin with. Fortunately, their experiment included a sensory panel, which showed that the treated soymilk had a significantly lowered ‘beany’ flavour. Similar to this paper, Yuan et al. applied ultrahigh-temperature processing to inactive enzyme activity in soy milk [23]. They looked at the optimal blanching conditions for soybeans to ensure full inactivation of SLOX enzymes. They found that blanching soybeans in water at 80°C for 2 minutes resulted in a soymilk product with no hexanal present. A SLOX activity test was performed, which presented that the SLOX enzymes were fully inactivated. They have shown that an additional blanching step of the beans allows for full deactivation of the enzyme. However, no actual data was given to support this. Similar results were found by Schweiggert et al. who looked at enzyme activities of blanch-processed chilli powders [24]. Blanching conditions at 90°C for 5 minutes were enough to cause LOX inactivation of 100%. They also found supporting evidence that blanching the chilli pepper before grinding was more effective, enzyme inactivation-wise, than in reverse order. These conditions seem to be similar with soybeans and shows the possibility of this method being effective on other types of produce as well.
The application of heat has shown to be effective towards enzyme inactivation. Nevertheless, the application of heat in combination with high pressure has shown to be as effective as the latter but with lower temperatures. Heinisch et al. utilised Fourier Transform Infrared Spectroscopy (FT-IR) to study the pressure effects on SLOX structural changes [25]. FT-IR was used to look at the shift of the frequency maximum of the amide I’ band soybean lipoxygenase (Fig 4). At a pressure of 600 MPa, the maximum amide I’ band shifted, which corresponded to an irreversible structural change of the SLOX enzyme leading to a reduction in enzyme activity. Their results showed that full enzyme inactivation was reached at a pressure of 600 MPa at a temperature of 0°C.
10 Similar inactivation results were found for the treatment of soy milk with heat and pressure. Wang et al. looked at the SLOX inactivation rate constants of soy milk and crude soybean extract at a wide range of temperatures and pressures [26]. Their experimental results were used to predict the temperature-pressure combination of certain inactivation rate constants (Fig. 5). This modelled figure is highly similar to the experimental results produced by Heinisch et al. [25], besides some differences that can be explained by the differences in soybean products.
Heat and pressure can be efficiently used on other types of soybean products as well. The effectiveness of pressure was also demonstrated by Manassero et al. who looked at the influence of temperature and pressure on the SLOX activity of calcium-enriched soymilk [27]. They found that for (almost) all the temperatures and pressures analysed (45-65°C & 500-700 MPa) SLOX was inactivated entirely, showing the effectiveness of pressure-based processing for high-end soybean products. However, these conditions seem to be most effective for soybean products alone because similar conditions were tested for LOX activity of lychee in syrup [28]. Treatment of lychee of at 600 MPa at 25°C for 10 minutes led to decreased LOX activity by 86%. However, it must be noted that a canning treatment reached full LOX inactivation at 100°C for 18 minutes.
However, the application of heat treatment on soybeans and soymilk is not always desirable due to some unwanted effects that can occur with overheating of the product. Overheating of soybeans can lead to the destruction of proteins, the loss of vitamins, and flavour and colour changes [29]. Therefore, alternative processing methods that rely less on intense heat have been made.
Figure 4. Effect of pressure on the frequency maximum of the amide I’ band of SLOX. Closed symbols indicate increasing pressure, open symbols decreasing pressure. Reproduced from ref. [25].
Figure 5. Combined predicted pressure-temperature combination of the same inactivation rate constant for SLOX. The upper line indicates a k value of 0.23 min-1, and the lower line indicated a k value of 0.115 min-1. Reproduced from ref. [26].
11
3.2 Alternative food processing techniques
The application of direct heat and pressure is an old conventional food processing technique to inactivate LOX. Many alternative methods exist for the inactivation of LOX, which have unique inactivation mechanisms. Most alternative methods still apply heat but in an indirect way. For example, microwave heating has gained some interest by having some advantages to conventional heating. Microwave heating can generate heat in a shorter period compared to conventional heating, which increases protein solubility. Wang and Toledo analysed the effect of exposure time of a microwave oven on SLOX activity in dry and soaked soybeans [30]. They found that full SLOX inactivation was achieved for soaked soybeans after 210 seconds of microwave radiation when temperatures had reached 100°C. Dry soybeans reached SLOX inactivation of 98% at the same treatment time (Fig. 6). This showed that higher moisture content in soybeans was needed for optimal heat transfer to inactivate enzyme activity completely.
Another example of indirect heat processing is the use of infrared (IR) heating. The use of IR heating has gained popularity in the food industry due to some of its advantages. It is an energy-efficient heating technology with fast heating rates and increased heat penetration. Yalcin and Basman analysed different IR parameters and their effects on LOX activity of dry and soaked soybeans [31]. They found that IR treatment of 1003 W at 10 minutes was enough for complete LOX inactivation of both dried and soaked soybeans. At these conditions, the soybeans reached temperatures between 99-160°C, which are higher compared to conventional methods (Fig. 7). Compared to microwave heating, the effect of IR heating on SLOX activity was less influenced by moisture content. This could be explained with IR reaching higher temperatures. These higher temperatures have been linked to a decrease in food quality, which is why Li et al. looked at the application of temperature-controlled shortwave infrared (SIR) on wheat germ oil [32]. A temperature range of 70-90°C, and a treatment time range of 20-60 minutes was tested. The tested conditions resulted in a LOX enzyme activity range of 10-40%, which showed that full SLOX enzyme inactivation could not be reached with a low-temperature IR treatment.
Figure 7. Effects of IR treatment on LOX-1 activity of soybeans. Reproduced from ref. [31].
Figure 6. SLOX activity as a function of exposure time to microwave.
12 A new combinational treatment was presented by Maetens et al. where germinated soybeans were treated with UV and IR [33]. Treatment time of 24/48/74h and 1.5/2.0/2.5h with varying source distance was tested for UV and IR, respectively, with its effect on LOX activity. It was hypothesised that UV-radiation might cause chemical degradation of the LOX enzyme, which would be enhanced by the thermal IR treatment. This combination would require lower IR temperatures in comparison to a singular IR treatment. The optimal conditions found were UV-exposure of 24h at 17cm distance and IR-exposure of 1.5h at 15cm distance. This resulted in an enzyme activity reduction of 55% and 97% for SLOX-1 and SLOX-3, respectively. However, their paper failed to report on the resulting temperature of the soybeans, so a thorough comparison to a singular IR treatment cannot be made. Nevertheless, the IR exposure time of this method was 1.5h, which is long compared to the conventional IR method. So likely, similar high temperatures were reached with this combinational method as well.
Another versatile method for soybeans is the use of radiofrequency (RF). Thakur and Nelson tested a non-thermal radio frequency approach [34]. A soy flour suspension was kept at 22°C and exposed to 20kHz for 3 hours at different pH’s. It was found that SLOX enzyme activity only dropped in a solution buffered at pH ≤ 5 (Fig. 8). It was thought that ultrasonic vibration promotes acoustic cavitation, which is the formation and collapse of small bubbles. Acoustic cavitation applied at a medium at pH ≤ 5 promotes the formation of H• and OH• radicals, which leads to the formation of hydrogen peroxide.
Hydrogen peroxide is a known LOX inhibitor at low concentrations and at room temperature, which would explain the found results. However, they had reached some enzyme inactivation with some very low pH’s of a soy flour suspension only. This would therefore not work on other soybean products. However, radiofrequency can also be applied in a thermal application.
Thermal RF heating generates heat within the food product by ionic depolarisation and dipole rotation, which can rapidly deliver energy to the food material. It also has lower frequencies than microwave and IR heating, which results in higher penetration depth and lower heating times. Jiang et al. applied a frequency of 27.12 MHz (a 103 multitude increase compared to non-thermal RF) on soybeans, which
allowed for rising temperatures comparable to conventional heating methods [34]. The use of RF is seen as more effective and has faster heating rates compared to conventional heating methods (Fig. 9). The highest SLOX inactivation of 95.2% was gained with a treatment time of 270s at an average Figure 8. Inactivation of LOX activity in whole soy flour by RF in dependence of pH and exposure time. Reproduced from ref. [34].
13 temperature of 113°C. These conditions, especially treatment time, are concise and optimal, making it seem a way better treatment method than conventional methods. However, these experiments are all performed on a small scale and could hypothetically lose their advantages when production is scaled upwards to an industrial setting.
Figure 9. Temperature rise curve of soybeans subjected to RF treatment and conventional thermal treatment. Reproduced from ref. [34].
A new and highly experimental food processing technique is the application of non-thermal plasma (NTP) or ‘cold plasma.’ Plasma is an ionised gas that is conventionally created with pressures higher than 105 Pa and with substantially high power, but non-thermal plasma is created at near ambient pressure and with less power. It was used to treat olive oil and improve its shelf life by Amanpour et al. [35]. They used a flowing set-up that could hypothetically be used to treat higher sample volumes of food products (Fig. 10). A treatment of NTP with argon gas at 2L/min at 7kV for 135 seconds was tested on olive oil. Olive oil LOX enzyme activity was lowered to 57.1%. Similar results were found for wheat germs by Tolouie et al. [36]. They analysed the parameters of time and voltage of a more straightforward but similar set-up on wheat germs. They found the lowest LOX enzyme activity of 50% with NTP plasma at 20 kV for 25min. Additionally, a storage test showed that after 30 days, the treated samples had regained LOX activity close to the starting values. This showed that NTP could not be used to destroy LOX enzymes permanently but could be a useful tool to lengthen the shelf life of the product. These examples show the possibility that this set-up could gain similar results for soybean products like soy milk. This method would be challenging to implement within the cooking operation but could be used on the finished product to extent the shelf life as well. Although no examples were found were soybean products were treated with NTP.
Figure 10. NTP-treatment of olive oil: left) NTP-configuration; right) NTP-treatment of oil sample (detail). Reproduced from ref. [35].
14 Gamma radiation treatment is a food processing method that is mainly used to sterilise packaged food products. However, this technique can have some advantages when used for enzyme inactivation. Barros et al. looked at the nutritional content and LOX activity of dried soybeans after different doses, 2.5/5/10 kGy, of gamma radiation [37]. The different doses of radiation only had a small but significant influence on the nutritional content of soybeans. The highest SLOX reduction was found at a gamma radiation dose of 10 kGy resulting in an average SLOX reduction of 80%. These results showed that gamma radiation could be used to inactivate SLOX enzymes before it has the chance to react during conventional soaking and cooking steps.
It can be concluded that most methods that apply direct heat of 100°C or higher can fully inactivate the LOX enzyme. High temperature is needed to prepare soybeans for consumption and has shown to be a useful enzyme inactivator. Therefore, it is logical that direct heat is the conventional food processing technique in the food industry. However, questions have been raised about the possible degradation of nutrients at these high temperatures, which have resulted in the development of alternative food processing methods. Methods that applied indirect heat sources that could reach temperatures of the conventional methods were most successful in full or almost full LOX enzyme inactivation. Non- or low thermal methods shown had reached LOX enzyme inactivation in a range of 50-80%.
15
Chapter 4: Plant-specific lipoxygenase inhibitor screening methods
This chapter will discuss plant-specific lipoxygenase inhibitor screening methods. Enzyme inhibitors are molecules that bind to enzymes and lower their enzymatic activity. When an inhibitor binds to the active site of an enzyme, it stops the enzyme from being able to bind to its substrate, thereby stopping enzyme activity. Two types of enzyme inhibitors exist, reversible and irreversible. Reversible inhibitors bind to the enzyme with non-covalent interactions such as hydrogen or ionic bonds. They can inhibit enzyme activity in different ways. For example, inhibitor binding to the active site blocks the enzyme from reaching the substrate. Alternatively, binding to other parts of the enzyme, which alters the properties of the enzyme, which diminishes the active site of the enzyme. Reversible inhibitors can be removed or inactivated by dilution. Irreversible inhibitors covalently bind to and modify the enzyme, which can lower or completely stop enzyme activity. Irreversible inhibitors usually forgo a chemical change when binding to the active site of an enzyme.
The effectiveness of an enzyme inhibitor is commonly tested by analysing enzyme activity before and after the addition of said inhibitor. For soybean lipoxygenase, enzyme activity can be measured by different methods, all having their respective benefits and drawbacks. This chapter will discuss the different SLOX activity measurement methods and how they could be used for screening methods.
4.1 Formation of conjugated dienes
The spectroscopic measurement of the formation of conjugated dienes is currently one of the most popular methods for enzyme activity analysis of soybean lipoxygenase. When lipoxygenase catalyses the hydro peroxidation of linoleic acid, an increase in absorption at 234nm can be observed due to the conjugated bonds formed during the reaction. This absorption wavelength is specific for hydroperoxide with a high extinction coefficient of ε = 25000 M−1 cm−1 [38]. Most current
spectroscopic methods are a copy or a slightly modified version of the method published by Axelrod et al. [11]. They published three procedures for all three SLOX isozymes, which were optimised for enzyme activity and detection. The enzyme reaction is carried out at room temperature in a cuvette during analysis. For SLOX-1, the enzyme is added to a modified borate buffer at pH 9.0 with a linoleic acid substrate mixture. The mixture is mixed, and the absorption at 234nm is measured. The enzyme reaction rate is determined from the slope of the straight-line portion of the curve. For SLOX-2, the enzyme is added to a modified phosphate buffer at pH 6.1 with an arachidonic acid mixture. Arachidonic acid is chosen because SLOX-2 is far more active against it than linoleic acid. The hydroperoxide products are measured at 238nm. For SLOX-3, the enzyme is added to a modified phosphate buffer at pH 6.5 with a linoleic acid mixture. The absorption is measured at 280nm. SLOX-3 is similar to SLOX-1. However, the primary hydroperoxide product of SLOX-3 is partially transformed into keto-diene products, which have an absorption maximum at 280nm. This spectroscopic technique is highly sensitive but has problems with turbidity and UV interferences. Because of these limitations, this assay is only used on purified enzymes.
16
4.2 Oxygen consumption
Measurement of oxygen consumption has been one of the oldest methods to analyse lipoxygenase activity. Considering that LOX incorporates molecular oxygen into its substrate, oxygen depletion within a medium correlates to enzyme activity. The oldest technique is a manometric approach, where the decrease in oxygen volume is measured using a Warburg apparatus (Fig. 11) [39]. The enzyme and substrate were buffered to the desired pH and placed in separate compartments. After calibration, the two were mixed, and oxygen consumption was measured every 5 minutes for 30 minutes. The method was deemed effective but was plagued by multiple drawbacks. This complicated analogue method had a long calibration time and total assay period, which allowed for secondary reactions to become more significant. Besides, shaking of the set-up, which was needed as a mixing step, tended to inactivate purified enzyme extracts [40].
A more modern approach that overcomes these problems is the polarographic method. This method generally uses a Clark oxygen electrode, with platinum and silver electrodes at 0.60 volts, to detect oxygen concentrations. Most methods use a set-up similar to (Fig. 12), which consists of a buffered enzyme and substrate solution added to a sealed reaction cell with the oxygen electrode. Often, oxygen concentration and temperature are measured and recorded electronically [41]. Enzyme activity is calculated from the initial rate of oxygen absorption, which is often measured for 5 minutes maximum [40]. It has been proven that the polarographic method of oxygen detection shows similar results compared to the chemical determination of oxygen concentrations by Mn(II) oxidation, even in complex media [42].
Figure 12. Schematic representation of a polarographic benchtop device. Reproduced from ref. [41].
The main advantage of polarographic oxygen measuring is the short analysis time compared to the manometric approach, which results in low denaturing effects. Also, the polarographic approach does not require an optically transparent substrate solution, which is heavily needed with a spectrophotometric approach. Where the spectrophotometric approach cannot measure vegetable homogenates, the polarographic approach can. Pinksy et al. used this approach to measure the LOX activity of 50 common fruits and vegetable homogenates [10]. Plant tissues were homogenised in a 1% Triton, phosphate buffer (pH 7.0), and only a few samples needed additional centrifuging and
Figure 11. The Warburg apparatus. Reproduced from ref. [39].
17 filtration. On the other hand, the polarographic method has the disadvantages that oxygen uptake is not specific to the LOX enzyme only. Secondary forms of oxygen uptake are caused by other systems but are lessened by the short assay period. However, it is recommended that two enzyme activity approaches are used when the identity of the analysed enzyme has not been established.
The polarographic method shows to have potential in use for inhibitor discovery. Recently, an improved method was published by Reyes-De-Corcuera et al. [41]. They had shortened the assay period to two minutes and incapsulated the linoleic acid substrate with β-cyclodextrin, which stabilised the substrate solution at a pH below 8.0. Their rapid and quantitative method of LOX activity analysis can be used in an industrial setting. However, their use of a stabilising compound could be challenging to scale up. Furthermore, the polarographic method has already been used for some inhibitor testing. Salas et al. used the method for enzyme activity analysis of homogenates of genetically modified potatoes [43]. Moreover, Schubert et al. used this method to successfully analyse the inhibitory effect of pomegranate seeds on SLOX (Fig. 13) [44]. These examples highlight that the polarographic method is best used for vegetable homogenate samples and that this approach could be used to test potentially found inhibitors on more ‘food product-like’ sample types. This method would be too complicated for initial inhibitor discovery but would be more useful for quality control in a scenario where the addition of enzyme inhibitors in food products has become common.
Figure 13. SLOX inhibition by pomegranate fermented juice (pfj) and pomegranate cold pressed seed oil (pcpso) extract. Positive control , BHA. Reproduced from ref. [44].
18
4.3 Co-oxidation of colourimetric reagents
An interesting approach to SLOX activity testing is the detection of co-oxidised colourimetric reagents, commonly called pigments. A range of methods uses the co-oxidation of pigments by lipid hydroperoxides (LOOHs) created by the LOX reaction as an indirect indication of enzyme activity. The use of pigments adds an extra level of modification within the detection procedure that can help solve some of the problems faced with vegetable homogenate analysis. This is one of the reasons why many methods and modification are still being published. Many types of pigments have been tested with varying degree of success. One of the more successful and well-known pigments is benzoyl-leuco methylene blue (BLMB). Auerbach et al. had published the method of LOX enzyme detection by the conversion of BLMB to methylene blue (MB) [45]. MB is an intensely coloured product which is light and temperature stable. In this method BLMB, a colourless compound is oxidised by lipid hydroperoxides to form MB, a colourant which strongly absorbs at 660nm. This reaction is catalysed by haemoglobin and results in an alcohol by-product (reaction 1).
Reaction 1. The colourimetric reaction of BLMB to MB. Hydroperoxide (ROOH) is converted to alcohol (OH) Reproduced from ref. [45].
The linear range of this method is 1.6-32 nmol (0.5-10 µg) of 13-HPODE, which is considered very low. However, this method can only be applied to LOX enzyme extracts. Even so, LOX substrates and products are not very water-soluble and cause the sample to be turbid, which lowers the sensitivity of the method. This problem has been solved by the voltammetry detection of MB, published by Pérusse and Leech [46]. The previous spectrophotometric assay was easily adapted to a voltammetric format as MB is an electroactive product. MB was detected by a flow-injection analysis (FIA) system potentiostat. Results of voltammetric detection were similar to spectrophotometric detection (Fig. 14), but this method has the benefit of being able to analyse turbid samples like vegetable homogenates.
Figure 14. FIA monitoring of MB by UV-VIS absorbance at 600 nm (triangles) and by SWV peak currents (crosses). Reproduced from ref. [46].
19 Despite the overall benefits, this method has a lower reproducibility (Fig. 15). Assay components adsorbed onto the electrode surface, which had to be polished between assay runs. The application of rapid scanning techniques and appropriate controls could hypothetically solve these adsorption problems. However, this would add extra analysis time to a method which is being sold as a highly rapid technique. Nevertheless, this extra time could be excepted when vegetable homogenates are being analysed effectively.
Figure 15. Variation of the MB peak current with inhibitor quercetin concentration. Reproduced from ref. [46]. Additionally, MB has one more property that could potentially stop itself from being used for inhibitor discovery. Any compound that disrupts the enzymatic reaction can be detected using MB. This includes LOX inhibitors but also non-specific antioxidants. Antioxidants can disrupt the LOX enzymatic reaction in addition to disrupting the MB reaction. This could be useful to test the antioxidative properties of inhibitors by applying the method on pure lipid hydroperoxides, which was done by Pérusse and Leech [46] (Table 2). However, when food products are being analysed for their LOX enzyme activity only, the potential presence of antioxidants could influence results drastically.
Table 2. LOX inhibition and antioxidant activity of selected compounds. Reproduced from ref. [47]
Inhibitor IC50 (nM) Concentration (nM) added to
antioxidant assay % anti-oxidant Dithizone 180 330 100 Curcumin 20 81 97 Caffeic acid 2 10 100 Quercetin 4 20 100 Phenidone 7 16 100 Esculine 2 3.2 90 NDGA 1 2.4 90
One of the oldest, yet effective pigments used, is iodine. Lipid hydroperoxides are reduced to alcohols by iodide, which in turn converses iodide to the triiodide anion (reaction 2). This conversion can be measured spectrophotometrically at 290 or 360nm [47](Fig. 16).
Reaction 2, the colourimetric reaction of iodine. Hydroperoxide (ROOH) is reduced to alcohol (ROH) by iodine (I-). Reproduced
20 Figure 16. The relationship between the amount of hydroperoxide reacted and I3- produced. Measured at 290 nm (triangles)
and 360 nm (circles). Reproduced from ref. [47]
The use of iodide is nonetheless commonly plagued with two disadvantages. This approach suffers from low sensitivity. Iodine content is difficult to estimate, even after modification of the method. Moreover, this approach is highly affected by interferences from oxygen [48]. Oxygen interference is somewhat less of a problem with modern technology, where samples and all other solvents can be deoxygenated by purging with nitrogen gas. DeLong et al. performed the iodometric assay, including deoxygenation steps, on a variety of fruits and vegetables [49]. They found that the iodometric assay underestimated LOOH concentrations and could not detect LOOH concentrations below 11 µm. In general, the iodine method provides no additional advantages that are needed for LOX enzyme activity testing in modern laboratories.
Ferrous oxidation-xylenol orange can be seen as the successor of iodide and BLMB for LOOHs detection. Like BLMB, this assay is based on the oxidation of ferrous (Fe2+) to ferric (Fe3+) ions in an acid
environment by LOOHs. The ferric ions subsequently bind to the xylenol orange dye. The dye changes colour from bright orange to a blue colour which has high absorbance between 500 and 650nm [50]. The FOX assay has been tested multiple times on LOX of plant material in addition to some inhibitor testing. Similar results compared to the 234nm spectrophotometric method were found in both cases (Fig. 17 & 18). It has been found that the linearity of the absorbance is lost at LOOH concentrations above 25 µm [51].
Figure 17. Time course of hydroperoxides generation by tomato (circles) or potato (squares) LOX, measured by FOX (open symbols) or absorbance at 234 nm (filled symbols). Reproduced from ref. [51].
Figure 18. Effect of LOX inhibitors on enzymatic activity measured by FOX and absorbance at 234 nm. Reproduced from ref. [51].
21 The FOX assay also has a few advantages that make it very suitable for a high-throughput approach. The assay is relatively simple, produces results rapidly, is not sensitive to ambient oxygen or light, and the FOX-complex is stable for more than 24 hours. Waslidge and Hayes already optimised their method to work with a 96-well plate, which showed its potential as a high throughput method [50]. Timabad et al. has shown more of this potential by adapting the FOX assay to be performed in a reaction volume of 0.25 ml [52]. Also, their method allowed the detection of LOOH originating from rice LOX in the range of 0.1-1.5 µM. This was lower than the limit of detection of 5.0 µM for the non-high throughput method of Delong et al. [49]. Considering the limiting but successful inhibitor testing and the high throughput format of the FOX assay, this assay has high potential as enzyme activity testing at an industrial setting. The FOX-complex being stable for several hours will be most useful for quality control between different batches and will allow for most reproducible results.
A more specialised method is the use of luminol as chemiluminescence detection. Kondo et al. published an improved method that detected the chemiluminescence of luminol and cytochrome-c [53]. LOOHs formed by SLOX reacts with cytochrome-c to produce oxygen radicals. These radicals oxidise luminol to the excited state. Light generated from the luminol is detected by a CL analyser and correlates to SLOX activity. The addition of cytochrome-c resulted in a 50 times increase in sensitivity. Therefore, the detection limit of this assay is 20 units of SLOX, which is defined as the amount of enzyme increasing n the absorbance at 234nm of 0.001/min by the spectroscopic method (Fig. 19).
Figure 19. Proportionality between reaction velocity and enzyme concentration by the Chemiluminescence Assay Method. Reproduced from ref. [53].
The assay format has been used by Kondo et al. to look at soybean seedlings, which allowed the useful comparison of relative enzyme activity [54]. It was also used by Liu and Pan to look at the inhibitory effect of green tea extract on fish LOX (Fig. 20) [55]. This method allowed for rapid analysis with incubation times being only 3 minutes long. Also, this method showed to be highly effective for tissues containing low amounts of LOX enzymes. Be that as it may, this assay format is mostly used for relative enzyme activity, and the high level of sensitivity is not needed when used for inhibitor discovery of plant material with high levels of naturally occurring LOX enzymes.
22 Figure 20. Inhibitory effect of Green Tea Extract (GTE) on α-tocopherol acetate (AA) and fish gill LOX indicated by
chemiluminescence intensity. Reproduced from ref. [55].
Fluorescence detection is a method which has not been widely investigated for the use of LOX enzyme activity detection of plant material. Only recently has this possibility been tested by Whent et al. [56]. They tested the parameters of a high throughput 96-well plate fluorescence method. Soybeans were ground with water and centrifuged; the supernatant was added without additional filtering in a pH 9.0 borate buffer with fluorescein. The mixture was placed on a 96-well plate without incubation time needed. The decrease in fluorescence was measured every 10 seconds for 6 minutes. The area-under-the-curve (AUC) was used for SLOX-1 quantification. The excitation and emission wavelength were 485 and 515nm, respectively. Quantification, linearity, range, reproducibility (Fig. 21), accuracy and precision were all tested; all factors were found to be excellent.
Figure 21. Reproducibility of the high-throughput LOX-1 fluorescein assay. Columns with the same letter indicate no statistical difference at P = 0.05. Reproduced from ref. [56].
A high throughput fluorescence assay was also used for inhibitor testing on human 5-LOX [57]. In this case, H2SCFDA was used instead of fluorescein. They were able to reliably identify small 5-LOX inhibitors and found similar results in comparison to the spectroscopic assay. These examples show that a fluorescence assay has the potential to be applied as a high throughput method but on enzyme extracts only. This method has not been tested on vegetable homogenates, but if so, it would most likely be plagued by the turbidity of the sample. However, compared to the spectroscopic assay, this method is better suited for high-throughput.
23
4.4 Formation of secondary products
The detection of secondary products from the LOX enzymatic reaction is another enzyme activity detection approach that has some benefits. As discussed in previous chapters, the LOX enzymatic reaction is quickly followed by the formation of secondary products, which for soybean products is mainly hexanal formation. Hexanal concentration can easily be measured with gas-chromatography (GC) in a variety of vegetable products. Bergman et al. published a procedure to analyse hexanal concentration in a variety of rice species [58]. Their method was able to detect hexanal within 6 minutes using a GC-FID and an internal standard (TMP) (Fig. 22-I). Their method also found a linear response between sample weight and hexanal concentration (Fig. 22-II). This showed consistency between hexanal concentration and sample species.
Figure 22. I, left) Chromatogram of typical rice extract. Peak a, hexanal and d, TMP. II, right) Linear response of hexanal of aromatic rice. Reproduced from ref. [58].
Moreover, Yuan and Chang showed a consistent correlation between hexanal concentrations and LOX activity in soymilk (Table 3) [59]. Hexanal concentration was also measured by GC-FID with an internal standard within 7 minutes, and LOX activity was determined spectrophotometrically. The usefulness of GC-FID for LOX activity detection is further demonstrated by Yu et al. [60]. They used a headspace GC-FID to detect hexanal concentrations of a range of differently processed soymilk. This method was used to analyse the ways different processing method affected SLOX and the subsequent hexanal formation. They found significant differences in hexanal concentrations between processing methods which showed that this could also be used to analyse the effect of theoretically added SLOX inhibitors. However, it needs to be noted that their method needed 30 minutes of equilibration within the headspace oven, which could pose a threat for the hypothetical inhibitory testing. Nevertheless, this approach could be advantageous in scenarios where exact enzymatic activity does not need to be known, but low hexanal concentrations need to be confirmed. The correlation between enzymatic activity and hexanal concentrations is strong enough for that to be effective.
Table 3, Significant Correlations (p ≤ 0.005) among protein, fatty acids, LOX activity, and soymilk odour composition. Reproduced from ref. [59].
Correlation variables r Soybeans
Linoleic acid vs hexanal in raw soymilk 0.93 Three normal varieties
LOX activity vs hexanal in boiled soymilk 0.94 Five varieties
LOX activity vs hexanal in raw soymilk 0.63 Five varieties
LOX activity vs hexanal in 0 min boiled soymilk 0.92 Three normal varieties
Protein content vs hexanal in raw soymilk 0.81 Three normal varieties
Protein content vs total hexanal during cooking 0.85 Three normal varieties
24 This approach should be met with some form of caution because it has been found that this method is prone to interferences. Lei and Boatright found that hexanal could be formed from isolated soybean protein without LOX present, through the presence of reducing agents like DTT (Fig. 23) [61]. Their study also showed that iron could act as a catalyst for hexanal production. This shows that the conditions of the soybean products should be well understood before GC analysis. Boatright and Lu found similar results for other types of reducing agents [62]. However, they used radiolabelled linoleic acid to track the hexanal production origination from SLOX activity and found that the influence of reducing agents was relatively small in comparison to the amount of hexanal formed by SLOX.
Figure 23. Formation of hexanal in defatted soy flour and ISP influenced by various amount of DTT or erythorbate. (A) in DF soy flour as influenced by DTT; (B) in ISP as influenced by various amount of DTT or erythorbate. Reproduced from ref. [61].
4.5 Radiochemical
Lastly, another method for enzyme activity is based on the use of radiolabeled substrates. This approach looks very similar to that of the spectroscopic method but uses radiochemical substrates and measured with radio chromatography or mass spectroscopy. This technique has mostly been used to study and characterise animal enzymes when their enzymatic reaction was still unclear. It has been used for animal LOX enzymes to study the formation of LOX products and its secondary by-products [63] [64]. This method has also been performed on some vegetables such as tomatoes [65]. For the analysis of vegetables, 14C-labeled linoleic acid is used as a substrate, whilst other solvents are kept
similar to the spectroscopic method. After the enzymatic reaction analysis, solvent extraction is often performed. This extract can be analysed by HPLC-MS, which is done in more modern techniques. However, in most cases where vegetables were analysed, the solvent extract was run on a TLC plate, then afterwards, dried spots were scraped off and analysed with a scintillation counter [66].
This technique is not often used for enzyme activity monitoring anymore. This can be explained by the method of not being fast or straightforward. Besides, expensive substrates and specialised equipment are needed. Furthermore, the scraping of the TLC plates allowed for the rise of individuals errors, making the overall method relatively ineffective.
25 In the table below, a small summary of the previously discussed enzyme activity detection methods is given. This table highlights the main advantages and disadvantages of each given method. In short, the spectroscopic method is most favourable when fast and straightforward enzymatic activity analysis is needed for plant extracts. However, this method relies on purified LOX enzymes and can, therefore only be performed on a laboratory scale. When analysis of vegetable homogenates is needed, all other methods can be used to a certain extent. Radiolabelling is most useful in preliminary enzyme kinetic studies and offers the least advantages in search of plant-based inhibitors. Oxygen consumption or pigment detection offers advantages which are favourable on an industrial setting. Monitoring of oxygen consumption can be easiest scaled up to an industrial setting to monitor enzyme activity. However, the detection of secondary products could hypothetically also be highly favourable for the food processing industry. In most cases, exact enzymatic activity is not needed to be known. The formation of hexanal is known to cause most off-flavours in food products; therefore, its detection can be enough in some scenarios.
Table 4, a summary of LOX enzymatic activity detection methods with their respective advantages and disadvantages
Technique Advantages Disadvantages Spectroscopic method – the
formation of conjugated dienes can be monitored with UV-spec at 234nm. Increase of intensity is related to enzyme activity.
• High extinction coefficient • Fast procedure
• Requires no special equipment
• Can be used with high throughput
• Can only be performed on pure enzyme
• Prone to UV interferences
Oxygen method – determining the rate of deoxygenation of the LOX enzyme by monitoring oxygen consumption. Either manometric or polarographic
• Requires little sample preparation
• Can analyse vegetable homogenates
• Little interferences
• Implementable at industrial scale
• The method needs special equipment
• Requires analysis in an oxygen-free environment
Pigment method – the detection of pigments co-oxidised by LOX reaction products as an indication of enzyme activity
• Wide range of useful pigments
• Appliable on turbid samples • Some can measure
antioxidative properties as well
• Most formed complexes are stable which allow for a high-throughput approach
• Can lead to false positives • Not all pigments are highly
specific
• Requires incubation time
Secondary products method – the detection of secondary products of the LOX reaction like hexanal as an indication of enzyme activity
• High linearity
• Can analyse complex food samples
• Useful for the food industry (hexanal being the most unwanted compound)
• Prone to interferences
• Least indicative of total enzyme activity
Radiolabelling method – detecting radiolabelled products originating from the radiolabelled substrate to determine enzyme activity.
• Most useful to study enzyme kinetics and enzyme specificity • The complete enzymatic
reaction can be tracked • Can be performed on
complex mixtures
• Requires expensive chemicals • Requires specialised equipment • The method is not fast nor simple
26
Chapter 5: Inhibitor discovery and identification methods
5.1 Discovery
The demand and subsequent search for inhibitory compounds requires well-thought-out discovery approaches. However, the term ‘discovery’ means finding something that was previously unknown. In the field of inhibitor research, true ‘discovery’ approaches are scarce. When looking for plant-based inhibitory compounds, most (plant) material researched originates from medicinal plants which are historically known to have bioactive compounds present. In this case, research is about identification but not discovery. An example of a proper LOX inhibitor discovery approach is a skin test. Heinemann et al. used a skin test to look at the anti-inflammatory properties of a plant extract [67]. Sodium lauryl sulphate (SLS)-induced irritant contact dermatitis (ICD) was invoked by human volunteers and treated with a plant extract. Afterwards, the anti-inflammatory potency was determined by visual score and moisture content. Fig. 24 shows their found results which showed improved visual score for the treated samples. This method could hypothetically be used to discover the anti-inflammatory (LOX inhibition) of plant extracts and any new substance, without regard to the many problems this method faces. The acetone solvent in which the active plant compounds would be dissolved caused skin irritation by itself, which compromised the experimental set-up. And if this approach were to be used as a discovery method on a bigger scale, it would call for the use of animal testing which is deemed cruel and not necessary when other approaches are available.
Figure 24. Increase of the visual score during SLS irritation and parallel application of the test substances. The boxplots show the differences between scores on day five and day 1 for Isatis extracts (e1, e2, e3), tryptanthrin (try), acetone (ac), and untreated control (control). Reproduced from ref. [67].
Other discovery approaches can be found within the pharmaceutical industry, like virtual screening. Examples of these will be given in the next chapter because these approaches are ordinarily not applied for the discovery of plant-based inhibiting compounds. Generally, plant-based inhibiting compounds are identified and analysed from medicinal plants, as previously stated. This will likely continue since recent evidence has shown that chemically diverse compounds originating from higher plants have anti-inflammatory characteristics, including some LOX inhibitory properties [68]. This approach to developing medication from medicinal plants was described by Calixto et al. and is visualised in Fig. 25 [69]. Their approach highlighted the following steps. First, correct botanical identification, collection,
27 and drying of the plant material are needed. Secondly, extraction of the plant material is needed, which needs to be compatible with other analysis methods. This crude plant extract is screened using the desired bioassay. This is followed by several steps of chromatographic separation, where each fraction is submitted to the same bioassay. Subsequent identification of the pure compounds is performed. When compounds are identified, toxicology testing is needed if the compounds will be used for human consumption. Finally, when compounds are a part of medicine development, structure synthesis and modification is often performed, though this is not needed for LOX inhibitors which will be added to food products.
Figure 25. Procedure for obtaining the active principles from medicinal plants. Reproduced from ref. [69].
All of these steps need to be executed correctly and be compatible with each other to be effective. Many different methods are developed for each of these steps to accommodate the challenges that emerge from the wide range of plant materials analysed. Naturally, these methods all have their advantages and disadvantages.
28
5.2 Drying and extraction
Inhibitor testing of medicinal plants starts with the correct collection and drying of plant material. Not one drying method is universally effective for all types of plants and plant material. Plant research will require the optimisation of the plant drying method considering the strong influence it can have on the final concentration of the extract. Oberthür et al. looked at how temperature affected the pigment concentration in dried leaves [70]. They found that both low and high temperatures influenced the final concentration of the pigments significantly yet different for every component (Fig. 26). Sultana et al., on the other hand, looked at the effects of extraction solvents and techniques on antioxidant levels of a group of plant materials [71]. They found that aqueous solvent (80% organic solvent) extracts prepared by both shaker and reflux technique yielded higher antioxidant levels for all plants analysed. However, differences in highest antioxidant levels were found with different organic solvent at individual plant species. These examples show that the most effective solvent/technique and temperature is plant species dependant.
Figure 26. Indican concentrations in leaves of selected accessions of Isatis tinctorial L. (TW, JW, KW, FW, SW), after different drying procedures. Reproduced from ref. [70].
Nevertheless, the use of organic solvent extraction is losing popularity with other techniques. Conventional solvent extractions are known to be high in solvent consumption and longer reaction times. Besides, many other methods have been developed that offer some other benefits. For example, a slightly modified solvent extraction already offers many benefits. Benthin et al. published a pressurised liquid extraction method for medicinal plants [72]. Their results showed that all plants needed individual optimisation, but overall their method used half to 1/5th the volume of solvent needed compared to the non-pressurised method. Their method also had reaction times on average of 15 minutes, which they argued allowed their method to be used in high-throughput screening programmes. Interestingly, the use of organic solvents could be almost entirely scrapped in the pressurised hot water extraction method (PHWE). This method is known to be an effective extraction method for medicinal plants. In most cases, only low amounts of organic modifiers like ethanol are used to increase efficiency. Alternatively, surfactants can also be added to increase the extraction power of hydrophobic species at lower temperatures. However, the application of this method on plant material is somewhat limited because more elaborate optimisation is needed per individual species [73].
The previous method can be further modified and turned into a subcritical water extraction (SWE). SWE belongs to the class of subcritical fluid extraction (SFE), which is part of the group ‘green extractions’. The extraction methods are developed to have reduced energy requirements and
non-29 hazardous solvents whilst ensuring the quality of the extract. SFE methods operate at the supercritical conditions of the solvents used. At these conditions, the low density and high diffusivity allow for extremely efficient extraction rates. Many solvents can be used, and most of these are deemed as safe (Table 5). Often carbon dioxide is used for three reasons: the gas is harmless for humans and the environment (with the quantities worked with), it is suitable for temperature-sensitive compounds, and it causes an oxygen-free environment [74].
Table 5. Critical properties of some solvents used in SFE. Reproduced from ref. [74]
Solvent Critical Property
Temperature (°C) Pressure (atm) Density ƿSCF (g/mL) Solubility σSFC (cal-1/2 cm -3/2) Carbon Dioxide 31.2 72.9 0.470 7.5 Ethane 32.4 48.2 0.200 5.8 Ethene 10.1 50.5 0.200 5.8 Methanol -34.4 79.9 0.272 8.9 Nitrous Oxide 36.7 71.7 0.460 7.2 n-Butene -139.9 36.0 0.221 5.2 n-Pentane -76.5 33.3 0.237 5.1 Sulfur hexafluoride 45.8 37.7 0.730 5.5 Water 101.1 217.6 0.322 13.5
One added benefit of subcritical gas usage as an extraction solvent is that after the extraction is finished and subcritical conditions are lost, gas can be easily removed and a liquid, solvent-free extract is left [75]. Also, the use of a gas like carbon dioxide allows for the addition of polar modifiers which help increase extraction yields of higher polarity compounds [76]. An example of a CO2 SFE set-up is
given in figure 27. It can be said that the SFE set-up is of medium to high complexity in comparison to other more conventional extraction techniques.
Figure 27. Schematic presentation of a supercritical carbon dioxide extraction system coupled with cosolvents. Reproduced from ref. [76].
Although, in the field of inhibitor discovery, often a water extraction is favourable. And with this, SWE offers higher efficiencies. Luo et al. compared the efficiencies of SWE and hot water extraction (HWE) on antioxidants from bran [77]. They found that SWE offered a higher yield of polyphenolic compounds in comparison to HWE (Table 6).
30
Table 6, The comparison of SWE with HWE. SWE: subcritical water extraction; HWE: hot water extraction; TPC: total polyphenolic contents; GAE: gallic acid equivalents; dw: dry weight. Reproduced from ref. [77].
Method TPC (mg GAE/g dw) TPC (mg GAE/g dw extract)
SWE 42.453 ± 0.275 578.33 ± 15.46
HWE 31.813 ± 0.786 320.15 ± 5.15
Per contra, SWE can be a technique that is too extreme. Singh and Saldana showed that the SWE of potato peels operating at temperatures higher than 180°C lead to the degradation of the phenolic compounds extracted [78]. Granting that operating conditions below 180°C still resulted in superior results compared to conventional methods. Generally speaking, SWE is a highly efficient green extraction technique which grants many advantages on an industrial scale. SWE is not often used with initial inhibitory discovery research due to the high cost of the set-up. On the other hand, the set-up can be customised to perform, for example, continuous extraction [79]. Customisation and low solvent usage are preferable for the industry. This industrial application was further backed by results from Ahmadian-Kouchaksaire et al., who showed the successful SWE of phenolic antioxidants from crocus petals in a large scale setting [80].
Another green extraction method which can be applied on an experimental scale is microwave-assisted extraction (MAE). Molecules align themselves with the direction of an oscillating microwave electromagnetic field. When this field is removed, the molecules return to a random orientation. This reorientation produces a displacement inside the material, which generates heat (Fig. 28). This unique heating principle gives MAE several advantages. Because heat transfer is coming from inside-out, MAE has high heat transfer and depth. Heating does depend on moisture content, but this generally causes no problem with plant material. MAE also has short heating times and consumes less energy. MAE does not particularly need special hazardous solvents and is seen as eco-friendly. However, MAE is prone to local overheating when the distribution of the microwaves are not uniform [74].
Figure 28. (A) Experimental set-up for conventional extraction of high-added-value molecules from plant matrices at laboratory scale. (B) Ultrasound-assisted extraction principle and cavitational phenomenon. (C) Microwave-assisted extraction equipment used at laboratory scale showing the molecular rotation mechanism. Reproduced from ref. [74].
