Introduction

Meat analogues or plant-based meat alternatives are sustainable alternatives to meat [1]. However, current meat analogues would benefit from improvement in texture and fibrousness, as the texture and appearance of meat analogues should resemble meat [2,3,4]. It is known that the addition of hydrocolloids to the formulation may help create this structure. Carbohydrate-based hydrocolloids are water-soluble polysaccharides that generally form a phase separate from proteins, and are often applied for their thickening and gelling properties [5,6,7]. These properties are employed to improve the texture by acting as crosslinkers or solidification agents, and binding protein filaments together in order to enhance the strength [6, 8]. The addition of hydrocolloids to plant proteins may also improve the fibrousness of the products [9], probably through the same mechanism.

In our previous paper [10], we found that low acyl gellan gum gave considerably higher strength and tensile stress values in meat analogues than other hydrocolloids (xanthan, iota-carrageenan, sodium alginate, guar gum, carboxymethyl cellulose, low methylated pectin, and locust bean gum). Low acyl gellan gum is a thickening and gelling agent [11] with applications in confectionary, bakery and dairy products [12]. Native gellan gum, produced by the microorganism Sphingomonas elodea [13], has a high amount of substitution of acyl groups on its backbone. This is reduced on low acyl gellan gum through an alkaline treatment at high temperatures. The reduction in the number of acyl groups makes gels from the gellan gums firmer. Moreover, low acyl gellan gum has high gelation efficiency and good thermal stability. Therefore, it can tolerate high temperatures of processing in the high-temperature shear cell (HTSC), and then, it can produce firm gels during cooling at gelation temperatures between 40 and 70 °C [14]. Upon gelation, gellan gum forms a stacked double helix type of junction zones [15, 16]. The first step in the gelation of gellan gum is the transition from the discorded coil state to a double helix form [16]. Thereafter, aggregates of double helices are created by cations such as calcium ion bridges between [13] carboxyl group pairs on neighboring gellan helices producing a very rigid or brittle structure, which induces gellan gelation [17]. In fact, the firmness of a gel can be enhanced by incorporating appropriate quantities of calcium ions [18, 19]. Thus, the gelling ability of low acyl gellan gum explains its application in protein-based gels, generally in combination with calcium to increase gel strength [17]. However, the gels produced using only low acyl gellan gum and calcium ions lack the desired elasticity for our purposes in the recent study. In this study, our objective was not solely focused on gel formation but rather on improving the overall strength and fibrous appearance of the meat analogues. Based on our previous study [10], where we examined various concentrations of gellan gum (0%, 1%, 2%, and 3%) in blends containing pea protein isolate (PPI) and wheat gluten (WG), we observed a significant increase in the strength of the meat analogues as the concentration of gellan gum increased. It is important to note that all four concentrations displayed noticeable differences, providing compelling evidence for the positive impact of gellan gum, particularly at higher concentrations, on enhancing the desired attributes. The reason for the need of relatively higher gellan gum concentrations in meat analogues could be that the available water content in these blends is lower compared to a dispersion with water and gellan gum only. Given these findings, we made a deliberate decision to utilize an intermediate concentration of 2% gellan gum in this current study. This concentration struck a balance between the desired strengthening effects and the potential drawbacks associated with excessive gel formation or compromised texture.

It has been reported before that gelling or thickening hydrocolloids such as pectin, can be instrumental in the creation of internal structure during heating [20, 21]. The proposed mechanism was the fact that the pectin exists as a separate, dispersed phase in the protein matrix. In addition, the hydrocolloid phase may accumulate between different domains of proteins (e.g., PPI and WG) and act as a lubricant or compatibilizer between these domains, facilitating the creation of a fibrous morphology. Low acyl gellan gum shows gradual gelation with temperature that may be altered by addition of divalent ions such as Ca2+, which makes it an interesting hydrocolloid to assess in combination with a typical blend of proteins, that is used for creation of meat analogues.

To the best of our knowledge, the influence of low acyl gellan gum on textural attributes of meat analogues produced has not been reported yet. Therefore, the aim of this study is to assess the effects of the addition of low acyl gellan gum and the influence of CaCl2 on the textural attributes and the micro-and macro-structure of PPI-WG products sheared in the HTSC.

We suggest that if the mechanism of lubrication between the different protein domains is correct, then the addition of low acyl gellan gum to a material consisting of WG and PPI will result in a more pronounced formation of fibers and tensile strength [6,7,8, 22]. Since Ca+ 2 ions modify the mechanism of gelation of low acyl gellan gum in type and strength, the addition of CaCl2 to the blend is an important parameter as well, all the more because it also influences the behavior of a PPI-WG blend itself [23,24,25,26]. Finally, it was reported that longer hydration of the protein blend before the addition of hydrocolloids, enhances the formation of fibrousness [21]. Therefore, we expect that also with low acyl gellan gum, the sequence of addition of the gum, the calcium chloride, and the hydration of the proteins will be of influence.

Materials and methods

Materials

Pea protein isolate (NATURALYS© S85F) and vital wheat gluten (VITENS© CWS) were obtained from Roquete Frères S.A. (St. Louis, Missouri, USA). PPI and WG had an average dry matter content of 93.2 wt% and 92.3 wt%, respectively. The low acyl gellan gum (GelzanTM CM) and CaCl2 were bought from Sigma-Aldrich co., LLC (Zwijndrecht, the Netherlands).

Methods

Mixing procedures and formulations

PW sample

The description and the formulations investigated in this study are shown in Tables 1 and 2, respectively. All samples were prepared at a constant dried ingredient concentration of 40 wt%. For the basic preparation method of the PW sample containing PPI and WG, first PPI (20%) and deionized water (60%) were mixed with a spatula in a beaker until a uniform mixture was obtained. Then, the protein blend was covered with a parafilm to prevent water evaporation and hydrated for 30 min at room temperature. After hydration, WG (20%) was mixed with a spatula into the hydrated PPI mixture. This treatment (PW) was defined as the control in this study. Cognizant of previous research findings, our choice of a 50/50 ratio for PPI/WG, accompanied by 1% calcium, was motivated by the observed benefits to the fibrous structure of the products. Notably, studies such as Dekkers et al. (2018) [27] have reported that gels with a minimum of 50% soy protein isolate (50/50 ratio for SPI/WG) exhibited a continuous SPI with a dispersed WG phase with acceptable fibrous structure. Building upon this knowledge, our previous works [29, 30] have consistently employed the 50/50 PPI and WG ratio. These studies reinforce the robustness and reliability of this formulation selection based on our own research endeavors.

Table 1 The description of the names of different materials. In this table, the abbreviations of P, W, C, G, BH, and AH mean pea protein isolate (PPI), wheat gluten (WG), CaCl2, low acyl gellan gel, before PPI hydration and after PPI hydration, respectively
Full size table
Table 2 Composition of six different formulations investigated in this study. In this table, the abbreviations of P, BH, and AH mean before and after pea protein isolate hydration, respectively
Full size table

PWC sample

For the preparation of the PWC sample containing PPI, WG, and CaCl2, and investigation of the effect of CaCl2 in the PPI-WG mixture, 1% CaCl2 was dissolved in the deionized water (60%) before the addition of PPI. To have the constant concentration of 40 wt% in this treatment, both PPI and WG concentrations were reduced to 19.5% (Table 2). The rest of the procedure remained similar as described in Sect. 2.3.1.

PWC-G-BH sample

For the preparation of the PWC-G-BH sample containing PPI, WG, CaCl2, and low acyl gellan gum (G) with the addition of low acyl gellan gum before hydration (BH) of PPI, 2% low acyl gellan gum was mixed with deionized water (60%). Subsequently, 1% CaCl2 was dissolved and mixed with a spatula in the gel. After that, PPI (18.5%) was mixed in the gel mixture. After hydration for 30 min, WG (18.5%) was mixed with a spatula into the hydrated PPI mixture.

PWC-G-AH

For the preparation of the PWC-G-AH sample containing PPI, WG, CaCl2, and low acyl gellan gum (with the addition of low acyl gellan gum after hydration (AH) of PPI), 1% CaCl2 was dissolved in deionized water first. Then, PPI (18.5%) was added and properly mixed with a spatula. The beaker with the protein mixture was covered with a parafilm and hydrated for 30 min. After the hydration step, WG (18.5%) and low acyl gellan gum (2%) powders (Table 2) were first mixed together and subsequently stirred into the protein blend with a spatula before further processing in the HTSC.

PW-G-BH sample

For the preparation of the PW-G-BH sample containing PPI, WG, and low acyl gellan gum with the addition of low acyl gellan gum before hydration (BH) of PPI, 2% low acyl gellan gum was mixed with deionized water. Next, PPI (19%) was mixed in the mixture. After hydration for 30 min, WG (19%) was mixed with a spatula into the hydrated PPI mixture.

PW-G-AH sample

For the preparation of the PW-G-BH sample containing PPI, WG, and low acyl gellan gum with the addition of low acyl gellan gum after hydration (AH) of PPI, deionized water (60%), and PPI (19%) were added into a beaker and properly mixed with a spatula. The beaker with the protein mixture was covered with a parafilm and hydrated for 30 min. After the hydration step, WG (19%) and low acyl gellan gum (2%) were first mixed and subsequently stirred into the protein blend with a spatula before processing.

High-temperature shear cell (HTSC) structuring of protein blends

A high-temperature shear cell (HTSC), designed at Wageningen University, was used to structure different protein blends. In the HTSC, the shearing zone was formed between two cones: a rotating bottom cone and a stationary top cone. The top cone had a diameter of 135 mm and a height of 510 mm. These cones were closely spaced, with a small 2 mm gap between them, creating a high-pressure environment of 6.5 bar. This configuration allowed for controlled and precise shearing within the HTSC, facilitating accurate experimentation and analysis [29]. The protein blends were processed in the pre-heated shear cell at 120 °C for 15 min at a constant shearing rate of 30 rpm [31]. After shearing, the HTSC was cooled down to 25 °C within 10 min without any shearing (0 rpm). After cooling, the products, measuring 80 mm in diameter, were taken out of the shear cell and were rested left at room temperature in a closed Ziplock bag for at least 1 h before the tensile test was performed. The remaining material was frozen at -18 °C for SEM analysis. Each sample was prepared and analyzed in independent triplicates.

Scanning electron microscopy (SEM)

SEM was used to gain information on the microstructure (~ 5–300 μm at magnifications of 250 X up to 10,000 X) of different products according to the procedure described by Taghian Dinani, et al. [29, 30] and Schreuders et al. [32]. In more detail, rectangular pieces, measuring 13 mm × 5 mm, were carefully cut from the samples that underwent processing in the HTSC. These pieces were then immersed in a glutaraldehyde solution (2.5% v/v) and gently shaken for a duration of 8 h using a Mini Rocker-Shaker MR1 from Riga, Latvia. Subsequently, the glutaraldehyde solution was replaced with deionized water, and the samples were left to soak overnight. To facilitate further processing, the samples underwent a series of ethanol solutions with varying concentrations (10%, 30%, 50%, 70%, 96%, and 100% v/v), with each immersion lasting 1 h. The samples were then subjected to critical point drying using a CPD 300 instrument from Leica in Vienna, Austria. After drying, the samples were manually fractured in a direction parallel to the shear and affixed to stubs using carbon cement glue. A sputter coating of 12 nm of tungsten was applied using an SCD 500 instrument from Leica. Finally, the surfaces of the samples were analyzed using a field emission scanning electron microscope (Magellan 400, FEI, Eindhoven, the Netherlands) with secondary electron detection at 2.00 kV and a current of 13 pA.

Tensile strength analysis

The tensile strength of the sheared protein blends was measured with a texture analyzer (TA.XT Plus Connect, United Kingdom) using a static load of 50 N with a constant deformation of 1 mm/s. Three perpendicular and parallel tensile bars to the shear flow direction (Fig. 1) were cut out in a bone shape to determine the degree of anisotropy (AI) [30, 33]. The two ends of the tensile bars were placed and secured into two clamps in the texture analyzer. The extensional force needed to break the samples was recorded by software (Exponent Connect, Stable Micro systems Ltd.). Then, the tensile stress (σ) [N/m2] for both parallel and perpendicular directions to the shear flow was calculated using Eq. (1):

Fig. 1
figure 1

B. Indication of the place and direction of the specimens that were cut out of the product made in the HTSC. These tensile bars were consequently used to perform the tensile test

Full size image
$$\sigma \left( t \right) = \frac{{F\left( t \right) \times H\left( t \right)}}{{H\left( 0 \right) \times A\left( 0 \right)}}$$
(1)

In this equation, F(t) is the force [N] needed to break the tensile bar and A(0) is the initial contact surface of the tensile bar [m2] calculated by multiplying the measured length (l) and width (w) of the tensile bar [mm] according to Fig. 1. If h(0) specifies the length between the clamps at the starting point (15.5 mm), h(t) is the gap [mm] that arises upon tearing the tensile bar at time t concerning the starting point. The tensile strain (ε) for both parallel and perpendicular directions to the shear flow was calculated with Eq. (2):

$$\varepsilon = {\rm{ln}}\frac{{{\rm{h}}\left( {\rm{t}} \right)}}{{{\rm{h}}\left( 0 \right)}}$$
(2)

Young’s modulus was calculated from the linear part of the curve of the tensile stress versus the tensile strain. The AI, indicating the fibrousness of a sample, is defined as the ratio between the tensile strain (AIstrain) or the tensile stress (AIstress) in parallel and perpendicular directions according to Eqs. (3) and (4).

$${\rm{A}}{{\rm{I}}_{{\rm{stress}}}} = \frac{{{\sigma _{{\rm{parallel}}}}}}{{{\sigma _{{\rm{perpendicular}}}}}}$$
(3)
$${\rm{A}}{{\rm{I}}_{{\rm{strain}}}} = \frac{{{\sigma _{{\rm{parallel}}}}}}{{{\sigma _{{\rm{perpendicular}}}}}}$$
(4)

It is important to highlight that we made every effort to maintain consistency by initiating the tensile tests precisely one hour after sample production in the HTSC. However, it is necessary to acknowledge that slight variation in the tensile testing time occurred due to the nature of conducting the test on three perpendicular and three parallel specimens. Despite this variation, we conducted the tests efficiently, ensuring that there were no significant differences between the three perpendicular and three parallel specimens conducted on different samples. Furthermore, the tests were performed at a consistent time, and under similar conditions for all treatments and formulations.

Statistics

Complete Randomized Design in the SPSS statistical software (Version 28.0, IBM, Armonk, NY) was used to study the effects of different sample formulations and preparations on the investigated responses. A descriptive Duncan’s test at a significance level of 95% (P ≤ 0.05) was applied to investigate the statistical significance of different results. In this study, all reported results are shown as mean ± the standard deviation (SD).

Results and discussion

Macrostructure

The macrostructure and fiber formation of the products with different formulations and preparation procedures (PW, PWC, PWC-G-BH, PWC-G-AH, PW-G-BH, and PW-G-AH, explained in Sect. 2.3 and summarized in Table 1) were visually inspected by folding the products produced in the HTSC parallel to the shear flow direction. Figure 2 shows that different structures were produced upon the addition of low acyl gellan gum to the PPI-WG blend before and after hydration and with or without the addition of CaCl2. The product without CaCl2 and low acyl gellan gum hydrocolloid (PW as the control sample) did not have a fibrous structure and was basically a brittle gel.

Fig. 2
figure 2

Visual pictures of the macrostructure of products with PPI and WG (PW), PPI, WG and CaCl2 (PWC), PWC with the addition of low acyl gellan gum before (PWC-G-BH) and after PPI hydration (PWC-G-AH), PW with the addition of low acyl gellan gum before (PW-G-BH) and after PPI hydration (PW-G-AH). In this figure, the red-underlined letters indicate the types of visual structures formed, where (A) indicates the formation of large and little fibrils, (B) indicates the formation of only large fibrils, (C) indicates a layered structure, (D) indicated curd-like structure without obvious fibers and (E) indicates the formation of a brittle gel

Full size image

However, after the addition of CaCl2 to the PPI-WG mixture (PWC), a layered structure was observed orientated along the shear direction. No small or large fibers were observed in this product. Previous research by shearing PPI (19.5 wt%) and WG (19.5 wt%) with sodium chloride (1 wt% NaCl) in the HTSC at 120 °C and 30 rpm for 15 min showed the formation of a fibrous structure [31]. We expect that CaCl2 induces stronger solidification of the proteins than NaCl. Indeed, it was reported that gelation of proteins is affected by substituting NaCl with CaCl2 [34, 35]. PPI without salt has a decreased solubility and therefore gel formation is affected [34], but more importantly WG, being a continuous phase and probably responsible for the creation of a fibrous morphology, may not be able to form its network with CaCl2 instead of NaCl. Therefore, a brittle gel has formed [36].

The addition of low acyl gellan gum before PPI hydration to the PPI-WG blends with and without CaCl2 (PWC-G-BH and PW-G-BH, respectively) yields rather similar results. Only thick fibers could be observed in both samples (Fig. 2). The absence of finer fibers may be related to low acyl gellan gum being an efficient (water) binder. When low acyl gellan gum absorbs too much water, the consistencies of both protein phases may not be optimal anymore [13], which is essential for the deformation of the domains into a fibrous morphology. The addition of low acyl gellan gum to a PPI-WG blend after hydration but without CaCl2 (PW-G-AH) gives a curd-like consistency in which no fibers were observed at all. However, the addition of low acyl gellan gum as a powder along with WG powder, to the hydrated mixture of PPI and CaCl2 (PWC-G-AH) yielded a quite pronounced fibrous structure compared to the other materials in Fig. 2. The hydration of PPI in a CaCl2 solution, therefore, seems important for the solubility of PPI, and, thus, for structure formation in a PPI-WG blend. The initial absorption of the CaCl2 into the PPI phase may give the WG the time to develop and create the gluten network that is needed to form a fibrous structure [31].

Microstructure study by scanning electron microscopy (SEM)

SEM images were made at different magnifications (250 X, 2500 X, and 10,000 X times) in the direction of the shear flow to give an understanding of the network formation on the microscale. Figure 3 shows that PW and PWC microstructures at 250 X magnification were similar in different aspects. No orientation in the shear flow direction could be observed; however, some small threads (yellow arrows in Fig. 3) could be detected in the structure of PW. PWC on the other hand showed larger filaments and globules with some cracks in between the filaments (red arrows indicating the cracks in Fig. 3). Zhang et al. (2022) [37] found that a potato protein isolate (30%) – WG blend (18%, /v) without CaCl2 did not give an obvious fibrous structure. We see the same for PW blends at a magnification 250 times. In the same work the authors reported that cracks developed after the addition of 0.004 g/ml CaCl2, resulting in gaps in the gel. One can compare this to the cracks in the PWC blends in Fig. 3. On a smaller length scale (2500 and 10,000 X in Fig. 3), the PW and PWC structures do show an opener morphology, which could be described as small globular domains, denoted as a ‘cauliflower’ morphology by [38].

Fig. 3
figure 3

Overview of the SEM pictures at different magnifications (top to bottom: 250-, 2500- and 10,000-times magnitude, respectively) for the microstructure of products with PPI and WG (PW), PPI, WG and CaCl2 (PWC), PWC with the addition of low acyl gellan gum before (PWC-G-BH) and after PPI hydration (PWC-G-AH), PW with the addition of low acyl gellan gum before (PW-G-BH) and after PPI hydration (PW-G-AH). In this figure, the red arrows indicate the cracks in the structure, yellow arrows indicate threads in the products at 250 times magnitude, blue arrows indicate packed globular domains, and orange arrows indicate elongated packed globular domains

Full size image

The addition of low acyl gellan gum in the case of PWC-G-BH gives a comparable structure as with PW and PWC. There is no pronounced orientation in the shear flow direction, but there are some small threads (yellow arrows) and some cracks between the filaments (red arrows). PWC-G-BH shows a web-like structure with large smooth surfaces in between (at the length scale of 10,000 times magnitude in Fig. 3).

The addition of the low acyl gellan gum to PWC in PWC-G-AH gave more orientation in the shear flow direction. This is in line with earlier observations after addition of pectin (2.2 wt%) to SPI [21]. We here see filaments (at 250 X) that are elongated and oriented in the direction of the shear flow (yellow arrows). Cracks in-between filaments occur, just as in the PWC structure (red arrows). On a smaller length scale (10,000 X), the PWC-G-AH material exhibits string-like connected domains, resembling a ‘spider web’ morphology [38]. While the morphology is rather open, there are also some packed globular domains present (blue arrows in Fig. 3). Thus, the ‘cauliflower’ and ‘spider web’ morphologies coexist in the same structure for the PWC-G-AH product. The coexistence of both morphologies is a characteristic of the gel formation of globular proteins, and thus, PWC-G-AH behaves like a gel of globular proteins [39]. Grabowska et al. (2014) [38] showed that sheared SPI (30 wt%) blends with NaCl2 gave similar morphologies.

The addition of low acyl gellan gum to PW blends gives comparable structures for both PW-G-BH and PW-G-AH. PW-G-BH and PW-G-AH show no cracks in between filaments at 250 X, and, thus, these structures are more comparable with PW than PWC at this magnitude. These blends formed denser structures of small filaments oriented in the shear flow direction. PW-G-BH features oval globules (blue arrows) at 2500 X magnification whereas PW-G-AH shows elongated globules and larger elongated filaments (orange arrows in Fig. 3). At a magnification of 10,000 X for PW-G-AH and PW-G-BH blends, no web-like morphology as with PWC-G-AH and PWC-G-BH is seen, nor a cauliflower morphology as by PW and PWC. The morphology is rather dense and does not contain fiber-like domains or globules but rather dense plates. Wang et al. (2021) [40] observed that a composite gel made of rice glutelin (9% wv) and sugar beet pectin (SBP) (4% w/v) yielded a uniform, interconnected gel, which was destroyed at high calcium concentrations (400 mM). Similarly, Fig. 3 shows that PW-G-BH and PW-G-AH are denser, without any gaps, while addition of CaCl2 gave cracks, similarly as found by Wang.

The differences between PW-G-BH and PW-G-AH products and PWC-G-BH and PWC-G-AH products in morphologies suggest an effect of CaCl2 on the action of low acyl gellan gum. Ca2+ ions change and promote the gelation of low acyl gellan gum creating a network due to intermolecular associations between Ca2+ and low acyl gellan gum [41]. In addition, CaCl2 screens the charge interactions of PPI, which could result in more aggregation [42]. We expect that the combined effect of the intermolecular associations and formation of PPI aggregates being enclosed by a porous WG phase explains the morphology obtained for PWC-G-BH and PWC-G-AH blends.

Shear stress and strain

Figure 4A and B show tensile stress and strain values, respectively, of the different materials. The Young’s modulus for the different materials can be found in the Appendix section, Figure S1. PWC-G-AH is significantly stronger in both parallel (P\(\le\)0.001) and perpendicular (P\(\le\)0.001) directions (Fig. 4A) and in tensile strain in the parallel direction (P\(\le\)0.001) to the shear flow (Fig. 4B). However, its tensile strain differs significantly from that of PWC and PW-G-AH in the perpendicular direction (P\(\le\)0.001). PWC has the lowest tensile stress and strain values in the parallel (stress 15.67 ± 6.45 kPa in Fig. 4A; strain 0.14 ± 0.03 in Fig. 4B) and perpendicular (stress 18.89 ± 4.90 kPa in Fig. 4A; strain 0.15 ± 0.02 in Fig. 4B) directions to the shear flow. However, this was not significantly different from PW (P\(>\)0.05).

The addition of low acyl gellan gum before and after hydration to PWC (PWC-G-BH and PWC-G-AH) results in significant differences in tensile stress in both parallel and perpendicular directions (Fig. 4A) and for the tensile strain in the parallel direction (P\(\le\)0.001) (Fig. 4B). Moreover, both samples (PWC-G-BH and PWC-G-AH) had significantly higher tensile stress values in parallel and perpendicular directions and significantly higher strain values in the parallel direction than PWC (P\(\le\)0.001). Therefore, not only the addition of low acyl gellan gum to PWC blends (in PWC-G-BH and PWC-G-AH) is effective, but also the order of addition is important. One can imagine that the sequence is important for the transient consistency of the different phases, with one sequence making sure that during shearing and heating, different phases have a similar consistency, which allows deformation. A different sequence may mean that one phase has already softened while another is still dehydrated and stiff, impeding deformation of domains into an oriented structure. This effect is possibly larger for PWC-G-AH sample than PWC-G-BH because initially no gel structure between low acyl gellan gum, water, and CaCl2 can occur in PWC-G-AH because low acyl gellan gum was added after PPI hydration in CaCl2 solution. However, the addition of WG after PPI hydration in the PPI-CaCl2-G blend could interfere with this network resulting in disruption of the network in PWC-G-BH.

Fig. 4
figure 4

A) Tensile stress (kPa) and B) tensile strain (-) of shear-induced structured products with PPI and WG (PW), PPI, WG and CaCl2 (PWC), PWC with the addition of low acyl gellan gum before (PWC-G-BH) and after PPI hydration (PWC-G-AH), PW with the addition of low acyl gellan gum before (PW-G-BH) and after PPI hydration (PW-G-AH) in parallel and perpendicular direction to the shear flow. In each figure, various English letters display statistically significant difference of results (p ≤ 0.001). Upper letters indicate significant differences for parallel direction and lower letters for perpendicular direction to the shear flow

Full size image

Both samples of PW-G-BH and PW-G-AH blends show insignificant differences in shear stress and strain values (Fig. 4A and B, respectively) for both parallel and perpendicular directions to the shear flow (P\(>\)0.05). Therefore, the addition of low acyl gellan gum before or after PPI hydration to PPI-WG blend without CaCl2 has no significant effect on the strength of the samples. The increase in stress and strain values upon low acyl gellan gum addition to a PPI-WG blend (in PW-G-BH and PW-G-BH) without CaCl2 and independent of the order of addition, may be related to better solidification of PPI-WG blends in presence of low acyl gellan gum. PPI and WG have different viscoelastic behavior during heating at 120 °C [31]. The viscosity of WG upon heating is lower than PPI which could have an influence on the deformation and alignment of protein domains in the shear flow direction [27, 31]. However, the addition of low acyl gellan gum could alter the viscosity of WG, and, thus, the difference between viscosities of WG and PPI could become smaller [43], which will lead to more deformation of the individual domains, possibly leading to long fibers or bi-continuous structures [31].

The addition of low acyl gellan gum to PWC blends (PWC-G-BH and PWC-G-AH) in presence of CaCl2 showed significantly higher tensile stress values in the parallel direction compared to the addition of low acyl gellan gum to PW blends (PW-G-BH and PW-G-AH), and, thus, stronger structures were made when adding the hydrocolloid to PPI-WG blends with CaCl2. The addition of CaCl2 to a PPI-WG blend with low acyl gellan gum could have two possible effects. As was mentioned before the first effect is that CaCl2 could induce gelation due to complexation of Ca2+ and the hydrocolloid [41]. The second effect is that the divalent Ca ions could act as an ionic bridge between the proteins and low acyl gellan gum, which increases the adhesion between the different components [44, 45].

Anisotropy index (AI)

In Fig. 5, the anisotropy indices in mechanical strength, based on shear stress and strain, for different blends are shown. The materials without low acyl gellan gum (PW and PWC) had indices close to 1 (AI stress and strain ≤ 1), indicating no anisotropy. PW and PWC have similar mechanical strength (P > 0.05) in both parallel and perpendicular directions [38]. Figure 2 shows no fibrous morphology for these two blends. Upon addition of low acyl gellan gum, mechanical anisotropy (AI ˃ 1) was observed in all the other materials (Fig. 5). Thus, addition of low acyl gellan gum improved the structure’s anisotropy, independent of the addition order and presence of CaCl2.

Fig. 5
figure 5

The anisotropic index (AI) based on stress and strain of shear-induced structuring of products with PPI and WG (PW), PPI, WG and CaCl2 (PWC), PWC with the addition of low acyl gellan gum before (PWC-G-BH) and after PPI hydration (PWC-G-AH), PW with the addition of low acyl gellan gum before (PW-G-BH) and after PPI hydration (PW-G-AH). In this figure, various English letters display statistically significant difference of results (p ≤ 0.001). Upper letters indicate significant differences for AI stress and lower letters for AI strain

Full size image

The macrostructural differences between PW and PWC blends without low acyl gellan gum and PWC-G-BH, PWC-G-AH, PW-G-BH, and PW-G-BH blends with low acyl gellan gum in Fig. 2 could be attributed to the differences in mechanical strength (in Figs. 4 and 5) between their continuous and dispersed phases. The low acyl gellan gum may either form a separate, weaker dispersed phase, or may accumulate between the different protein phases (PPI and WG) and reduce the adhesion between these phases. Both effects would lead to more anisotropy if the different domains deform during heating and shearing [21, 31, 38, 46].

Even though the mechanical anisotropy obtained with different orders of addition of low acyl gellan gum to PW and PWC blends was comparable in Fig. 5, different macrostructures were observed in Fig. 2. The AI for PW-G-AH is larger than 1 but was not accompanied by a fibrous morphology in Figs. 2 and 3. It was previously reported that a fibrous morphology was not always related to mechanical anisotropy, which is indeed the case for PW-G-AH [32]. However, PWC-G-BH and especially PWC-G-AH showed pronounced fibers and thus their fibrous morphology. These materials have larger tensile stress and strain values in the parallel direction, which could indicate the presence of more elongated domains after shearing.

Conclusion

In this study, the effect of the addition of low acyl gellan gum to a blend of PPI and WG was investigated. In addition, the effect of the addition of CaCl2 was assessed, as it changes the solidification profile of the low acyl gellan gum, but also has an influence on the behavior of the proteins. The order of addition of the low acyl gellan gum and CaCl2 in a PPI-WG blend was found to be important because a change in the order of addition resulted in structural and mechanical differences. The addition of CaCl2 and low acyl gellan gum to the protein blend, especially when added after hydration of the proteins improved the macro- and microstructure properties in terms of fiber formation (investigated by SEM pictures and visual pictures) and the tensile strength (investigated by tensile stress, tensile strain, and AI parameters). We expect that similar behavior will be found with other similar hydrocolloids. This insight into the behavior of low acyl gellan gum may be used as a starting point for further research with other hydrocolloids. Especially with charged hydrocolloids, we expect that the interaction with electrolytes will be key.