The blue crab (Callinectes sapidus) was captured in Laguna Madre, Tamaulipas, transported, and processed at the facilities of the company “Integradora Pesquera Acuícola” located in San Fernando, Tamaulipas (México) within 24 h after caught. The crabs were cooked using a commercial autoclave and cooled with clean water. The crab meat was removed from shells, introduced into a plastic recipient, and transported on ice to the laboratory for further analysis.

Gelatin (GP 28, 290 bloom, Coloidales Duché, S.A DE CV. Ciudad de México, México) was purchased at a local market. The commercial microbial transglutaminase was Active TG-TI (Ajinomoto USA, Inc., Teaneck, NJ).

Cooked and minced crab meat was washed with water at a 3:1 (water: meat) ratio for 10 min at 4 ºC to dissolve soluble proteins, drained with a cotton fabric filter to remove the excess water containing the soluble compounds and stored in 250 g portions in polyethylene bags until analyzed.

Following the methodology reported by Hernández-Robledo et al. (2017), the meat was mixed with gelatin (0, 0.5, 1, or 3 %) and MTgase (at 0 % as control or 0.5 %) using a food processor (Hamilton Beach Model 72860-MX, Glen Allen, VA, USA) for 3 min and stuffed into stainless steel tubes with cap in both extremes, closed, and immersed in hot water at 90 ºC for 30 min in a water bath with temperature control (Techne Tempette Junior TE-8J). After cooking, the tubes were immersed in ice water (0 ºC) and stored at 4 ºC until analyzed (less than 16 h).

A texturometer (Model TA Plus, Lloyd Instruments, Ametek, UK) was used to perform the uniaxial compression test. Cylindrical gel samples (25 x 30 mm) were compressed to 75 % at 1 mm/s crosshead speed using a 50 mm aluminum probe (P50). The test was carried out at room temperature. Two compression cycles were used for texture profile analysis (TPA). The hardness, cohesiveness, springiness, and chewiness parameters were calculated from the TPA curves. From the first compression cycle, stress-strain plots were obtained, and the Young modulus and maximum stress were calculated. Six replicates were analyzed for each treatment.

The L*, a*, b* CIELab color parameters of the samples were determined using a colorimeter (Bluemetric, model BLUE-HP200, SA de CV, Monterrey, Mexico) in four random points on each gel sample. From the measurements, a*, b*, L*, and the color attributes Hue* and Chroma* were calculated.

A 3 g sample from the crab meat gels was wrapped in 5 paper sheets of 6 cm x 6 cm, placed in 50 mL conical centrifuge tubes, and centrifuged at 15,000 rpm for 15 min (Rotofix 32 A. Hettich Zentrifugen, Tuttlingen, Germany) following the methodology reported by Hernández-Robledo et al. (2017). Subsequently, the final weight of the sample was registered. Four repetitions per treatment were performed. The EW was calculated using Equation 1.

E W   =   ( W i - W f   / W i )   x   100 (Eq. 1)

Where:

EW = amount of extracted water.

Wi = Initial weight.

Wf = Final weight.

The crab meat gels were freeze-dried and ground. A sample was fixed on an aluminum pin using conductive double-sided tape and then observed in an ambient scanning electron microscope (Phenom Pro., Phenom-World BV, Eindhoven, The Netherlands).

Data were analyzed using a factorial ANOVA and the post hoc Tukey test was performed to evaluate the significant effects of the treatments. Results are presented as mean ± standard deviation, and differences were considered statistically significant at a p < 0.05. Also, multiple regression analysis was performed to model the response of the gel properties as a function of the percentage of gelatin and MTgase using a second-order polynomial equation (Design Expert® v10, Stat-Ease, Inc., Minneapolis, MN, USA). Coefficients of the terms in the polynomial equation were calculated and 3D plots were used to describe the behavior of the analyzed properties.

Table 1

Factors	Levels
Gelatin concentration	0, 0.5, 1 and 3 %
MTgase concentration	0 % and 0.5 %

The correlation values obtained from the multiple regression analysis are shown in Table 2. Hardness, cohesiveness, and chewiness showed R² and adjusted R² values higher than 0.9, indicating that the model fits adequately to data and that the model could be appropriate for future data.

Table 2

Factor	Hardness (N)	Cohesiveness	Springiness	Chewiness (N)	Young modulus (MPa)	Maximum stress (MPa)	EW (%)
Gelatin
Intercept	11.77	0.2169	0.7402	0.6142	78.93	17.9	11.72
A-Gelatin	-8.19	-0.0788	-0.0022	-4.14	36.9	9.06	-9.47
B-MTgase	3.39	-0.0059	0.0144	0.7807	7.93	1.2	1.96
Ab	-2.56	-0.0163	0.0031	-1.11	4.65	2.55	0.2864
A2	26.46	0.1092	-0.0122	8.97	4.52	10.49	10.36
R2	0.9595	0.9658	0.435	0.9595	0.9356	0.9499	0.8933
Adjusted R2	0.9055	0.9203	-0.3183	0.9055	0.8497	0.8832	0.7511

Young modulus maximum stress, and expressible water fitted appropriately to the model (R² > 0.9). The differences between R² and adjusted R² values are lower than 0.2 indicating that the terms included in the polynomial models allow describing the effect of the parameters on these properties.

The model obtained for springiness did not fit well at any of both parameters, indicating that his property was not a good indicator to observe changes in mechanical properties after adding gelatin. Springiness is not the same as elasticity, because this test may involve the fracture of the sample affecting its capacity to return to its original form (Rahman et al., 2021). Young modulus, on the other hand, is associated with elasticity because it considers the compression values (stress and strain) involved in the linear region of the test implying only a small deformation (Lam et al., 2017) that does not fracture (macro or micro) the sample, thus its deformation theoretically is reversible and the sample should return to its original shape.

The analysis of variance (ANOVA) for the fitting of the gelatin and MTgase effect on mechanical and extractable water is shown in Table 3. Models differ significantly (at p < 0.05) for changes induced in hardness, cohesiveness, chewiness, Young modulus, and maximum stress. Gelatin significantly affected these parameter (p < 0.05) except for expressible water. However, MTgase did not affect any parameter, and there was no significant effect on the interaction of gelatin and MTgase on any studied property. The quadratic term (A²) suggests that gelatin had a higher effect on the mechanical properties than in expressible water.

Table 3

Factor	Hardness (N)	Cohesiveness	Springiness	Chewiness (N)	Young modulus (MPa)	Maximum stress (MPa)	EW (%)
Gelatin
Model	0.0199*	0.0155*	0.7018	0.0199*	0.0393*	0.0272*	0.0815
A-Gelatin	0.0206*	0.0037*	0.881	0.0089*	0.0091*	0.0116*	0.0187*
B-MTgase	0.0946	0.4810	0.2551	0.2333	0.1908	0.4133	0.3009
Ab	0.2420	0.1753	0.8265	0.1904	0.4871	0.2045	0.8933
A2	0.0054*	0.0106*	0.6784	0.0071*	0.7356	0.0491*	0.0843

*Indicates statistical significance.

Figure 1 shows the changes in the texture profile analysis parameter. The hardness of gels was not affected by the MTgase level (0 to 0.5 %); however, it was negatively affected by the addition of low levels of gelatin (less than 2 % as calculated for the model). Gelatin added at 3 % improved the hardness of gels, but it was not enough to recover the initial parameters of samples obtained without this hydrocolloid (Figure 1A).

Figure 1

Cohesiveness (Figure 1B) and chewiness (Figure 1D) showed a similar behavior to hardness, with an initial decrease as affected by the addition of 2 % gelatin and a slight increase when added at 3 %. As mentioned previously, when modeling was discussed, an interaction between MTgase and gelatin was not observed. Crab gels showed relatively high values of springiness (72 %), and this parameter was not affected by the addition of MTgase, gelatin, or their combination (Figure 1C).

Small changes in mechanical properties of crab meat gels, when 0.6 % MTgase was added and cooked directly at 90 ºC (by immersion in hot water), were reported previously. However, better mechanical properties were found when gels from crab meat containing MTgase were incubated at 40 ºC for 30 min. This behavior was associated with side-to-side protein aggregations promoted by covalent bondings of adjacent proteins as a result of MTgase activity during incubation at 40 ºC (Martínez et al., 2014).

The disruptive effect of gelatin on the mechanical properties was also observed in gels obtained from Alaska pollock surimi (Hernández-Briones et al., 2009).

Young modulus (Figure 2A) and maximum stress (Figure 2B) showed a different behavior than TPA parameters. These mechanical properties increased when both MTgase and gelatin were added, and showed a combined effect when added at their maximum level (0.5 % and 3 %, respectively). Differences between both mechanical parameters are associated with the basis of each test. TPA simulates the textural perception in the mouth while a meal is being chewed, while Young modulus and maximum stress, as mentioned previously, are fundamental tests (Rahman et al., 2021).

Figure 2

The percentage of water extracted was reduced by 20 % by adding up to 1.5 % gelatin (Figure 3). Gelatin is a hydrocolloid that is characterized by the ability to trap water, which can help crab meat gel to improve its water-holding capacity and thus, decrease the amount of water it can lose. However, previous studies (Martinez et al., 2014) with blue crab meat gels, the addition of transglutaminase (6 g/kg), incubated at 40 °C and cooked at 90 °C, showed lower percentages of water extracted than gels without enzyme. In this study, adding MTgase did not affect the amount of extracted water alone or combined with gelatin, even though gelatin also acts as a substrate for the enzyme (Haug et al., 2004). Studies on adding MTgase and gelatin to surimi gels reported a decrease in water loss (Hernández-Briones et al., 2009; Kaewudom et al., 2013).

Figure 3

The effect of gelatin addition on color attributes of crab meat gels is shown in Table 4. No significant differences (p ≤ 0.05) were observed in these parameters as a function of the level of gelatin or MTgase, indicating that, regardless of the concentration of gelatin or the enzyme, the gels had similar color attributes. Gelatin is a translucent food additive (Siccha and Lock, 1992) and therefore, adding this protein does not affect the color of gels.

Table 5

MTgase (%)	Gelatin (%)	L*	C*	H*
Control
0	0	73.73 ± 1.18a	8.40 ± 0.88a	105.27 ± 1.43a
0.5	0	71.16 ± 0.98a	8.18 ± 0.62a	103.77 ± 2.12a
Gelatin
0	0.5	73.21 ± 0.66a	8.57 ± 0.50a	105.01 ± 1.66a
	1	72.57 ± 0.78a	8.83 ± 0.48a	104.47 ± 1.64a
	3	74.00 ± 1.81a	10.213 ± 1.09a	101.39 ± 1.12a
0.5	0.5	72.00 ± 0.72a	8.76 ± 0.68a	104.36 ± 1.95a
	1	72.362 ± 1.17a	7.51 ± 2.80a	102.77 ± 2.32a
	3	70.862 ± 2.02a	8.28 ± 0.50a	102.6 ± 1.17a

^a,b Different letters indicate statistical difference (P < 0.05) among treatments (MTgase level for control or % gelatin for each level of MTgase).

Microstructure SEM images of crab meat proteins containing different levels of gelatin or MTgase are shown in Figure 4. Control gels (without MTgase) containing 1 % of gelatin (Figure 1CC) were more fibrous, less compact and denser than gels containing 0.5 % or 0.3 % of gelatin (Figure 1A and 1E). These images agree with the results obtained for hardness, cohesiveness, and chewiness, all of which were negatively affected by adding 1 % MTgase but increased slightly by adding 3 % gelatin and had a lower amount of extracted water (higher water holding capacity).

Figure 4

Gels containing MTgase and gelatin seemed denser than control gels, suggesting a better interaction among adjacent proteins by the enzyme effect. Gels containing only 0.5 % gelatin and 0.5 % MTgase seemed more fibrous (Figure 1b) than gels containing 1 % gelatin, which showed lower TPA attributes (except springiness) (Figure 1). Gels containing 3 % gelatin and 0.5 % MTgase were denser and more fibrous than gels with 1 % gelatin, and showed better TPA parameters (Figure 1).

It is important to consider that gelatin is a recognized substrate for MTgase (Haug et al., 2004), and it has been reported to negatively affect the mechanical properties of surimi gels (Hernández-Briones et al., 2009; Kaewudom et al., 2013), especially at levels equal or higher than 1.5 %. However, in this study, adding gelatin improved the texture fundamental properties (Young modulus and maximum stress). Additionally, TPA parameters (except springiness) were improved by adding 3 % of gelatin, even after the negative effect caused by adding 1 % (Figure 1).

Thermally denatured muscle proteins form an opaque and not reversible gel. Gelatin proteins are different as they interact during cooling to form a reversible translucid gel after being thermally denatured. Both of them are recognized as a substrate for MTgase. Meat crab proteins have been reported as a unique and different kind of proteins, which, after being denatured and aggregated by thermal processing, keep the ability of gelling again (Martínez-Robledo et al., 2014), suggesting a partially reversible gel. SEM images indicate that gelatin, a protein that forms reversible gels, was able to interact with muscle crab proteins, and both were an appropriate substrate for MTgase, forming a structural network where crab meat proteins-gelatin interactions took place.