In the creation of meat analogues, it is important to be able to simulate a fibrous texture. Hydrocolloids are customarily added to act as thickeners and gelling agents, but the work of Dinan et al (2018) alerted me to the fact that when used in the right combination, these components can create the texture characteristics we are looking for.
Following the notes from the work of Dina, I give my notes from the work of Dekkers et al (2018) where they reviewed the structuring processes for meat analogues. These are my notes from their work.
A. Hydrocolloid Options for Creating Meat Analogues
The processing methods to investigate are high-temperature shear cell processing or HTSC processing. Raw materials used include:
SPI – soy protein isolate
WG – wheat gluten
PPI – pea protein isolate (low allergenicity and cost; high availability and good nutritional value)
PPI with SPI -> weaker gel than SPI.
Hydrocolloids + PPI = stronger gels.
Dinan (2018) set out to investigate the influence of different hydrocolloids on the textural attributes of PPI-WG sheared in an HTSC. They evaluated
X – xanthan (thickening, gelling, emulsifying, stabilising properties);
CA – iota-carrageenan (thickening, gelling, emulsifying, stabilising properties);
SA – sodium alginate (strong water binding and general gelling agent);
GG – guar gum (water-soluble, non-ionic, very viscous solution at low concentrations + low cost);
CMC – carboxymethyl cellulose (water-soluble, anionic – functioning as a binder and meat extender);
GZ – low acyl gellan gum (used in meat analogues – substantial gelling properties, forms brittle, strong, heat and pH stable gels);
P – methylated pectin ((used in meat analogues);
LBG – locust bean gum (less soluble, lower viscosity than GG used for gelling and as a thickening agent);
The above were evaluated at 1%, 2% and 3%.
CaCl2 of 1% was used for all mixtures.
Protein blends were prepared as follows:
-> 1% CaCl2 into 60% water
-> PPI added (19.5%)
Protein Hydration allowed for 30 minutes
WG mixed in next (19.5 or 18.5% with hydrocolloid at 1%, 2%, or 3%)
Following mixture, processing with HTSC.
The PPI and WG contained at least 78.6 wt% and 72.4 wt% protein, respectively, (with a nitrogen conversion factor of 5.7) on a dry weight basis, according to Dumas measurement. The manufacturer specifies that PPI and WG had average dry matter contents of 93.2 wt% and 92.3 wt%, respectively.
Different hydrocolloids with 1, 2, or 3% concentrations were added to the PPI-WG mixture.
|(%)||Step 1||Step 2||Step 3|
|0||60.0% water + 1.0% CaCl2||19.5% PPI||19.5% WG|
|1||60.0% water + 1.0% CaCl2||19.0% PPI||19.0% WG + 1.0% Hydrocolloid|
|2||60.0% water + 1.0%||18.5%||18.5% WG + 2.0%|
|3||60.0% water + 1.0% CaCl2||18.0% PPI||18.0% WG + 3.0% Hydrocolloid|
The CaCl2 concentration remained the same (1%) for all mixtures. Control (C) did not contain any hydrocolloids (0%). To allow the variation of the hydrocolloid concentrations from 0 to 3% while keeping the total solid concentration at 40 wt%, the concentrations of PPI and WG were reduced in equal proportions (19%, 18.5%, and 18%, respectively).
To prepare different protein blends, 1% CaCl2 was dissolved in the distilled water (60%) before the addition of PPI. Then, PPI was added and properly mixed with a spatula. The beakers with the protein mixture were covered with a parafilm to prevent water evaporation and to allow protein hydration at room temperature for 30 min. After the hydration step, WG and the desired hydrocolloid were first mixed and subsequently stirred into the protein blend with a spatula before processing. After the preparation of different mixtures in this step, they were pro- cessed with the HTSC instrument.
Texturising with Extruders – work by various researchers
- Aganovic, Berger, Palanisamy, and Stefan (2018) investigated the effect of
CA which is iota-carrageenan (thickening, gelling, emulsifying, stabilising properties) (0.75%, 1.5%, 2.25%, and 3%) was investigated in soya meat analogues produced by extrusion processing,
- Nanta, Skolpap, & Kasemwong (2021) investigated the effect of
GG which is guar gum (water-soluble, non-ionic, very viscous solution at low concentrations + low cost)
CA which is iota-carrageenan (thickening, gelling, emulsifying, stabilising properties), and
X xanthan (thickening, gelling, emulsifying, stabilising properties) (1%–7% w/w)
was investigated in SPI-based meat analogues prepared using extrusion.
- J. Zhang et al. (2020) investigated the effect of
CA which is iota-carrageenan (thickening, gelling, emulsifying, stabilising properties) and
SA which is sodium alginate (strong water binding and general gelling agent)
in peanut protein dispersion to produce fibrous products via extrusion.
Different studies showed that different hydrocolloids resulted in the improvement of meat analogues in only some aspects. Therefore, Dinan et al (2018) broadened the comparison of a more extensive range of hydrocolloids in meat analogues. To the best of their knowledge, the relationship between the addition of different hydrocolloids at different concentrations to PPI-WG meat analogues containing CaCl2 processed with a HTSC has not yet been quantified.
It is expected that the different hydrocolloids will allow achieving the best possible match of the second phase and thus improved properties of products regarding their fibre formation, tensile strength and WHC can be achieved upon the addition of appropriate hydrocolloids (Boison et al., 1983; Palanisamy et al., 2018; J. Zhang et al., 2020).
Finally, the hydrocolloid concentration in PPI-WG blends gives us a parameter to adapt the level of fibrousness, and other textural properties (Nanta et al., 2021).
The HTSC Instrument
A HTSC instrument, designed and manufactured at Wageningen University, was used to structure different protein blends (Wageningen, The Netherlands). The HTSC consists of a rotating bottom cone and a stationary top cone, which generates a simple shear flow. The temperature in the cones and rotation speed were controlled by circulating oil (Thermal H10, JULABO, Germany) and a Haake drive (Haake Polylab
QC, Germany), respectively. The sheared protein materials were pro- cessed in the pre-heated shear cell at 120 ◦C for 15 min at a constant shearing rate of 30 rpm. After shearing, the system was cooled down to 25 ◦C within 10 min without any shearing (0 rpm). The products were taken out after cooling and left at room temperature in a closed Ziplock bag for at least 1 h before further analyses. After cooling, colour and tensile test measurement tests were performed. The remaining samples were frozen at 18 ◦C for more analysis. Each sample was prepared and analyzed in triplicate unless mentioned otherwise.
The addition of the following transformed the PPI-WG products into fibrous materials:
X – xanthan (thickening, gelling, emulsifying, stabilising properties),
GG – guar gum (water-soluble, non-ionic, very viscous solution at low concentrations + low cost),
CMC – carboxymethyl cellulose (water-soluble, anionic – functioning as a binder and meat extender), and
GZ – ow acyl gellan gum (used in meat analogues – substantial gelling properties, forms brittle, strong, heat and pH stable gels)
What did not work?
The following did not transform the PPI-WG products into fibrous materials
SA – sodium alginate (strong water binding and general gelling agent), and
P – methylated pectin ((used in meat analogues)
Which ones contributed to a stronger structure (more tensile strength)?
X – xanthan (thickening, gelling, emulsifying, stabilising properties);
CA – iota-carrageenan (thickening, gelling, emulsifying, stabilising properties);
GG guar gum (water-soluble, non-ionic, very viscous solution at low concentrations + low cost)
GZ – low acyl gellan gum (used in meat analogues – substantial gelling properties, forms brittle, strong, heat and pH stable gels) – especially GZ.
LBG – locust bean gum (less soluble, lower viscosity than GG used for gelling and as a thickening agent)
added to PPI-WG products.
B. Structuring processes for meat analogues
What are the different options to structure meat? Dekkers et al (2018) did an excellent review of the current methods. I list them as they appear in their work, in an abbreviated manner.
Tissue-engineering techniques can be used to in vitro culture animal muscle cells, which may then be transformed into meat (Langelaan et al., 2010; Post, 2012). To culture muscle fibres, first myoblast cells should be harvested from the skeletal muscle from an animal. Subsequently, these myoblasts are replicated by a standard cell culture methodology using serum-supplemented medium with all the necessary nutrients, including amino acids, lipids, vitamins and salts, for cells to grow. When a sufficient number of cells is obtained, they are placed onto a scaffold with anchor points to connect and align the cells, yielding a multicellular tissue. Electric fields or other means are often proposed to ensure alignment and ‘normal’ development of muscle fibres. After approximately three weeks, muscle fibres have matured and can be harvested. These fibres are 2–3 cm long and less than 1mm thick. For upscaling this technique, a large-scale reactor is required in which the myoblasts can be cultured and exposed to the right environmental stimuli. The sensitivity of mammalian cells requires a high control over the growing conditions in the reactor and contamination should be avoided. Currently, these muscle fibres have been used to make a single hamburger as a proof of concept (Post, 2014).
The filamentous fungus Fusarium venenatum is used since the mid- 1980s as a basis for meat analogues that are marketed under the brand name Quorn. The fungus is produced in a continuous fermentation process in bioreactors. The conditions in the bioreactor are critical for the production of the fungus, for example temperature and pH should be monitored and strictly controlled. After fermentation, the RNA has to be degraded into monomers by a heat treatment, so it can diffuse out of the cells. The residual biomass is heated and centrifuged to obtain a paste-like product with 20 wt.% solids (Wiebe, 2002, 2004). The filamentous fungus is unordered after this centrifugation step, and therefore further process steps, such as forming, steaming, chilling, and texturizing, are required to obtain fibrous products. Minced-type products, such as chunks, sausages, and burgers, are commercially available from this material (Wiebe, 2002). While this route has been a commercial success for decades, the process is relatively intensive in its use of resources, and high in energy usage in the process and ingredient production (Smetana, Mathys, Knoch, & Heinz, 2015).
Boyer patented wet spinning of proteins for the application of meat analogues in 1954 (Boyer, 1954). A solution containing protein is extruded through a spinneret, and subsequently immersed into a bath containing a non-solvent for the protein. Exchange of solvent and nonsolvent results in precipitation and solidification of the extruded protein phase, forming stretched filaments with a thickness in the order of 20 μm (Boyer, 1954; Rampon, Robert, Nicolas, & Dufour, 1999; Tolstoguzov, 1988). Several studies showed the food-grade production of fibres with plant-based materials such as soy, pea and fababean (Gallant, Bouchet, & Culioli, 1984; Rampon et al., 1999). The type of structure that is formed depends on the solidification mechanism: fibres are obtained when the dispersed phase solidifies and the continuous phase can be washed away; capillary filled gels are obtained when the continuous phase is solidified and the dispersed phase stays liquid, and fibre filled gels are obtained when both the dispersed and the continuous are solidified (Tolstoguzov, 1988). Wet spinning is mostly used for the creation of individual fibres, and is one of the standard techniques for the production of membranes for industrial separation purposes (Ho & Sirkar, 2012).
In electrospinning, a biopolymer solution is pushed through a hollow needle or spinneret that has an electric potential relative to a ground electrode. Accumulation of charge on the surface of the droplet that emerges from the spinneret causes surface instabilities that ultimately grow into very thin fibres (≈100 nm), which are attracted to the ground electrode (Schiffman & Schauer, 2008). Food-grade electrospinning is mostly presented for applications in which nanofibers were used as carriers or delivery systems for bioactives components, such as polyphenols and probiotics (Librán, Castro, & Lagaron, 2017), but electrospinning can also be used to produce fibres for the application of meat analogues (Nieuwland et al., 2014). Electrospinning of proteins has been reported for several animal-based proteins, such as whey, collagen, egg and gelatin, but only sparingly for plant proteins (Anu Bhushani & Anandharamakrishnan, 2014; Ghorani & Tucker, 2015). Zein, when dissolved in a 80 wt.% ethanol solution, was spun into fibres with a thickness in the nanoscale (Miyoshi, Toyohara, & Minematsu, 2005). Nieuwland et al. showed the possibility to use zein as a carrier for other proteins, such as soy protein (Nieuwland et al., 2014). The proteins are required to be highly soluble, and behave like a random coil instead of globulins. Those requirements are generally not met by plant proteins, since they are in their native state globular and upon denaturation, insoluble aggregates are formed.
Extrusion is the most commonly used commercial technique to transform plant-based materials into fibrous products. We distinguish two classes of structuring with extrusion: low-moisture and highmoisture. In low-moisture extrusion, flours or concentrates are mechanically processed into texturized vegetable proteins (TVP), which are dry, slightly expanded products that are moisturized afterwards. In high moisture extrusion, fibrous products are produced with moisture contents above 50 wt.%. The proteins are plasticized/molten inside the barrel by a combination of heating, hydration and mechanical deformation. When this protein-‘melt’ flows into the die, it gets aligned by the (inhomogeneous) laminar flow and is cooled to prevent expansion. High moisture extrusion processes were extensively studied in the eighties and nineties (Arêas, 1992; Cheftel, Kitagawa, & Queguiner, 1992; Mitchell & Areas, 1992). Mitchell and Areâs proposed the ‘suspension model’ to mechanistically explain extrusion texturization (Mitchell & Areas, 1992). According to this model, the biopolymer melt forms two phases; a homogenous continuous phase and a dispersed, insoluble phase. The insoluble, dispersed phase is either formed during processing in the barrel at high temperatures, or was already present in the raw materials prior to processing. Whether materials/ingredients can be extruded depends on the ratio of soluble and insoluble components; too many insoluble components disturb protein cross-linking and result in incoherent products. Although extrusion processing has been studied extensively for many years, the control over the process and the design of extruded products is still mostly based on empirical knowledge (Emin & Schuchmann, 2017). Although extrusion is relatively energy intensive, it is the most widely applied technology for the production of meat analogues.
Mixing of proteins and hydrocolloids
Fibrous products can be obtained by mixing protein with hydrocolloids that precipitate with multivalent cations (Kweldam, 2011). After mixing, the fibrous products are washed and the excess water is removed by pressing, yielding dry matter contents between 40 and 60 wt.%. Despite the initial ordering in the shear direction, the subsequent steps destroy this large range ordering, limiting the use to minced meat products, such as burgers and schnitzels. In this process, various combinations of proteins, hydrocolloids and multivalent cations can be used. For example, a product based on casein and alginate was introduced in 2005 under the brand name Valess. Plant proteins such as soy, rice, maize and lupine can be employed in a similar way. This process is well scalable, yields products with some degree of structure, but still is relatively intensive in its use of resources.
In freeze structuring or freeze alignment, an aqueous solution or slurry of proteins is frozen to generate structure. Heat removal from a well-mixed slurry leads to an isotropic structure, but when heat is removed unidirectionally without mixing, then alignment of ice crystal needles yields anisotropic structures. Directional freezing has been studied for structuring of meat, fish, and plant proteins (Consolacion & Jelen, 1986; Lugay & Kim, 1978; Middendorf, Waggle, & Cornell, 1975), but is also employed for the creation of porous metal and ceramic materials. The size of the ice crystal needles can be tailored with the freezing temperature and rate (Lugay & Kim, 1978). Subsequently, the frozen product is dried without melting the ice crystals, for example by freeze drying, to obtain a porous microstructure with sheetlike parallel orientation of the proteins. These sheets are connected forming a cohesive fibrous product (Consolacion & Jelen, 1986). To obtain distinct fibrous products, the proteins should have relatively good solubility prior to freezing, and during the freezing process, these proteins become insoluble (Lugay & Kim, 1978).
Shear Cell Technology
Based on the recognition that extrusion is an effective, but not a well-defined process, a technology based on well-defined shear flow deformation was introduced a decade ago to produce fibrous products (Manski, van der Goot, & Boom, 2007b). Shearing devices inspired on the design of rheometers (Manski, van der Goot, & Boom, 2007a), socalled shear cells, were developed in which intensive shear can be applied in a cone-in-cone or in a couette geometry (Krintiras, Göbel, Bouwman, van der Goot, & Stefanidis, 2014; Van den Einde et al., 2004). The final structure obtained with this technique depends on the ingredients and on the processing conditions. Fibrous products are obtained with calcium caseinate and several plant protein blends, such as soy protein concentrate, soy protein isolate (SPI) – wheat gluten (WG), and SPI – pectin (Dekkers, Nikiforidis, & Goot, 2016; Grabowska, Tekidou, Boom, & van der Goot, 2014; Grabowska et al., 2016; Manski et al., 2007b). The structures prepared with calcium caseinate showed anisotropy on a nanoscale, while for the plant-based material, anisotropy was observed up to the micrometre-scale. The technology was successful up to pilot scale (Krintiras, Gadea Diaz, van der Goot, Stankiewicz, & Stefanidis, 2016).
Somayeh Taghian Dinani, Nicole Louise Broekema, Remko Boom, Atze Jan van der Goot, Investigation potential of hydrocolloids in meat analogue preparation, Food Hydrocolloids, Volume 135, 2023, 108199, ISSN 0268-005X, https://doi.org/10.1016/j.foodhyd.2022.108199. (https://www.sciencedirect.com/science/article/pii/S0268005X22007196)
Birgit L. Dekkers, Remko M. Boom, Atze Jan van der Goot (2018) Review: Structuring processes for meat analogues. https://doi.org/10.1016/j.tifs.2018.08.011; Received 7 May 2018; Received in revised form 17 August 2018; Accepted 17 August 2018; Trends in Food Science & Technology 81 (2018) 25–36; Available online 20 August 2018; 0924-2244/ © 2018 The Authors. Published by Elsevier Ltd. This is an open-access article under the CC BY-NC-ND license; (http://creativecommons.org/licenses/BY-NC-ND/4.0/).