In vitro protein digestibility of direct-expanded chickpea– sorghum snacks

Funding information International Development Research Centre (IDRC) and Global Affairs Canada (Ottawa, ON) through the Canadian International Food Security Research Fund (CIFSRF), Grant/Award Number: 107984-001/2 Abstract Blending cereals with pulses provides a balanced protein with higher biological value as their amino acid compositions are complementary. Extrusion not only can improve protein digestibility but also may reduce essential amino acid content. This study investigated the effects of extrusion parameters and blend ratio on in vitro protein digestibility (IVPD) and IVPD-corrected amino acid score (IVPDCAAS) of direct-expanded chickpea–sorghum snacks. Chickpea–sorghum blends (50:50, 60:40, and 70:30 chickpea:sorghum, w/w) were extruded at 10 combinations of moisture content (16%, 18%, and 20%) and barrel temperature (120 C, 140 C, and 160 C), and at 169 C and 15% moisture, the conditions identified in a previous study as producing maximal expansion. Chickpea and sorghum flours were extruded at 140 C and 18%moisture for comparison purposes. The IVPD of raw 50:50, 60:40, and 70:30 chickpea–sorghum blends ranged from 76% to 78%; values for raw chickpea and sorghum flours were 79% and 74%, respectively. Extrusion increased IVPD (P < 0.05) of all flours and blends. An increase in extrusion temperature increased the IVPD of extrudates (P < 0.05), whereas an increase in moisture content had the opposite effect (P < 0.05). The IVPDCAAS of raw 50:50, 60:40, and 70:30 chickpea–sorghum blends were 0.64, 0.72, and 0.73, respectively; values for raw chickpea and sorghum flours were 0.74 and 0.27, respectively. Extrusion increased IVPDCAAS (P < 0.05). The 70:30 chickpea–sorghum blend extruded at the maximal expansion exhibited the highest protein quality indicating this to be the optimal condition for snack production.

of 2%-6% on a dry-weight basis (Palavecino et al., 2016). Like other cereals, when compared with human nutritional requirements, sorghum protein is deficient in certain essential amino acids, most importantly lysine; however, it contains sufficient levels of the sulfur amino acids, cysteine, and methionine (Mokrane et al., 2010).
The in vitro digestibility of raw sorghum protein (40%-77%) is reported to be lower compared with other cereals due to the existence of antinutrients such as tannins (Duodu et al., 2002;Elkonin et al., 2013). In order to overcome its limitations in terms of both amino acid composition and protein digestibility, sorghum-based foods require processing as well as blending with a complementary protein source.
Chickpea (Cicer arietinum L.) is the third most widely grown pulse globally, with production of 17 million tons in 2018 (FAOSTAT, 2019).
Chickpea is an important source of protein for human consumption (Liu et al., 2008) and has a protein content of 16%-28% (Chibbar et al., 2010). The most abundant essential amino acids in whole chickpea seed are leucine and lysine, whereas the sulfur-containing amino acids cysteine and methionine are limiting (Wang et al., 2010). The in vitro digestibility of raw chickpea protein has been reported to be 59%-76% (Bhagyawant et al., 2018). Chickpea contains antinutrients such as polyphenols and trypsin and chymotrypsin inhibitors that contribute to lower protein digestibility and that can be destroyed by processing (Bessada et al., 2019).
Combining cereals with pulses provides a balanced protein with high biological value (Arribas et al., 2017). In addition, processing methods such as high-temperature extrusion are reported to improve digestibility of various plant-based protein sources (Nosworthy et al., 2017). However, optimization of extrusion conditions to maximize the protein quality of direct-expanded chickpeasorghum snacks has yet to be reported. Therefore, this study was designed to investigate the effect of extrusion conditions and chickpea:sorghum blend ratio on in vitro protein digestibility (IVPD) and IVPD-corrected amino acid score (IVPDCAAS) of direct-expanded chickpea-sorghum snacks.

| Determination of IVPDCAAS
Amino acid score was determined by comparing the amino acid composition of the target protein to that of the reference protein (FAO/WHO, 1991). The reference amino acid composition was recommended by FAO/WHO (1991) using the amino acid requirements for children 2-5 years of age (amino acid, mg/g protein): histidine, 19; isoleucine, 28; leucine, 66; lysine, 58; methionine + cysteine, 25; phenylalanine + tyrosine, 63; threonine, 34; tryptophan, 11; valine, 35. The FAO/WHO reference pattern for children 2-5 years of age is the amino acid pattern required by the United States FDA for determination of protein content claims (21CFR101.9). This reference pattern was selected for the benefit of product developers and regulatory agents. The lowest amino acid score represents the first limiting essential amino acid.
The IVPD was determined for the 10 treatments according to House et al. (2019). Samples containing 62.5 mg protein were heated to 37 C and adjusted to pH 8.0. The stability of the pH was maintained for 10 min, and then a multienzyme cocktail containing trypsin, chymotrypsin, and protease was added. The pH was recorded for 10 min, and the IVPD was determined from the change in pH over 10 min using the following equation: in vitro protein digestibility % ð Þ= 65:66 + 18:10 × ΔpH 10 min : This pH drop method of analyzing IVPD has been validated in several studies. The IVPDCAAS was calculated as the product of the limiting amino acid score and IVPD . IVPD and IVPDCAAS analyses were done in quadruplicate.
The 70:30 chickpea-sorghum blend was identified as the point at which the IVPDCAAS plateaued and was therefore designated as optimal (data not shown). With the objective of examining the effect of blending chickpea and sorghum on IVPD and IVPDCAAS, chickpea-sorghum blend ratios of 50:50 and 60:40 also were included in the study for comparative purposes.

| Extrusion
Extrusion employed a corotating, twin-screw extruder (Model EV-32; Clextral, Firminy, France) according to Bekele et al. (2020). Briefly, chickpea-sorghum blends (50:50, 60:40, and 70:30, chickpeasorghum, w/w) were extruded by setting the barrel temperature of the last three zones at 120 C, 140 C, or 160 C and the feed moisture content at 16%, 18%, or 20%. The blends also were extruded at the conditions where maximum extrudate expansion was observed, 169 C barrel temperature and 15% feed moisture. The conditions for maximum extrudate expansion were determined according to Bekele et al. (2020). For comparison, chickpea (100%) and sorghum (100%) extrudates were produced at a barrel temperature of 140 C and 18% feed moisture. The screw speed and feed rate were maintained at 396 rpm and 26 kg/h, respectively. Extrudates were dried at 105 C for 5 min using a tunnel drier (Chromalox, Pittsburgh, PA, USA). Each sample was produced in duplicate under each processing condition.
Extrudate drying temperature-time was selected because less than 5% available lysine loss was observed in dairy based confectionery heated at 103 C for 5 min (Malec, Llosa, Naranjo, & Vigo, 2005).

| Statistical analysis
The effects of extrusion and blend ratio on IVPD and IVPDCAAS were analyzed using two-way ANOVA and the Fisher post hoc test. The effects of blend ratio, barrel temperature, and moisture content on IVPD were analyzed using three-way ANOVA and the Fisher post hoc test. Differences were considered significant at P < 0.05 (Vik, 2013).
Statgraphics Centurion version 18.1.12 (Statgraphics Technologies, The Plains, VA, USA) was used for analysis.

| RESULTS AND DISCUSSION
3.1 | Amino acid content and amino acid score of chickpea, sorghum, and blends The amino acid compositions of raw chickpea (100:0), raw sorghum (0:100), and raw chickpea-sorghum blends (50:50, 60:40, and 70:30 chickpea:sorghum, w/w) are presented in Table 1. Measurement of the amino acids was performed in duplicate on the raw samples or extruded samples, so no statistical comparison was undertaken as n < 3. Amino acid analysis was performed on all samples listed in  (Li et al., 2018). For the most part, the coefficient of variation for amino acid composition was less than 7, except in the case of phenylalanine for raw and extruded sorghum, and isoleucine, tyrosine, and lysine for extruded sorghum where the coefficient of variation was greater than 7.
The amino acid scores for raw chickpea, raw sorghum, and the raw chickpea-sorghum blends were calculated according to the 1991 FAO reference pattern for children 2-5 years of age (FAO/WHO, 1991) ( Table 2). The first limiting amino acid for raw sorghum (0:100) was lysine, and its amino acid score was 0.37; for raw chickpea and raw 50:50, 60:40, and 70:30 chickpea-sorghum blends, the first limiting amino acid was tryptophan, and their respective amino acid scores were 0.93, 0.94, 0.94, and 0.84. Statistical comparisons were not done on the amino acid score as n < 3.
The amino acid scores for chickpea, sorghum, and chickpeasorghum extrudates also are presented in Table 2 This also could be the reason that tryptophan was found to be the limiting amino acid in this study. In the case of sorghum, a previous T A B L E 1 Amino acid compositions of raw and extruded chickpea and sorghum flours and chickpea-sorghum blends (g/100 g protein)    study reported that lysine was the first limiting amino acid, with an amino score of 0.4 (Mokrane et al., 2010). Guzman-Ortiz et al. (2015) also observed decreases in levels of amino acids after extrusion of a soybean-corn blend at 160 C and 26% moisture.

| Effect of extrusion on IVPD
The IVPDs of raw and extruded blends are presented in Table 3. Differences (P < 0.05) between extruded and raw samples of the same blend ratio as well as differences (P < 0.05) between blend ratios, but within raw or similar extrusion conditions, are indicated. The IVPD of raw sorghum (74%) was lower (P < 0.05) than those of raw chickpea (79%) and the raw chickpea-sorghum blends (76%-78%). The IVPD of raw sorghum determined in this study falls within the ranges reported by Elkonin et al. (2013), 40%-76%, and by Bhagyawant et al. (2018), 59-76%.
Studies have reported that one of the explanations for lower digestibility of sorghum is that the prolamin protein (kafirin) forms oligomers or polymers of high molecular weight that are linked together by disulfide bonds and that are resistant to hydrolysis by proteases (Duodu et al., 2002;Nunes et al., 2005). Extrusion increased (P < 0.05) the IVPD of all samples, but the increase in IVPD was less for sorghum.

| Effect of extrusion conditions on IVPD
Extrusion temperature, moisture content and blend ratio had significant (P < 0.05) effects on IVPD (Figure 1). However, the interaction effects were not significant (P > 0.05). IVPD was higher (P < 0.05) for higher extrusion temperatures. In contrast, an increase in moisture content or the proportion of sorghum in the blend resulted in lower (P < 0.05) IVPD. Previous work showed that an increase in extrusion temperature resulted in a concomitant rise in IVPD of a sorghummaize blend and a flaxseed-maize blend (Licata et al., 2014;Min et al., 2015). Multiple reasons for this phenomenon exist, including the alteration of noncovalent interactions resulting in "opening" of the protein, as well as inactivation of protease inhibitors and other antinutritional factors. Ainsworth et al. (1999) reported that IVPD increased with an increase in extrusion temperature to a point, after which IVPD decreased. The explanation for this was that at higher extrusion temperatures, the extrudate had undergone thermal crosslinking during nonenzymatic browning reactions, resulting in lower IVPD. Similarly, others have reported reductions in IVPD at higher extrusion moisture contents (Ghumman et al., 2016;Palanisamy et al., 2019). This might be due to the decrease in shear in the extruder barrel associated with the increase in moisture content.
In line with this study, Licata et al. (2014) reported a decrease in IVPD of a sorghum-maize extrudate with an increase in the proportion of sorghum (range of 15%-60%) in the extrudate. This might be due to crosslinking of high molecular weight sorghum proteins. The sorghum-maize blend was extruded at 120 C and 150 C barrel temperature and 21% and 26% moisture content.
T A B L E 3 Effect of extrusion and chickpea-sorghum blend ratio on in vitro protein digestibility (IVPD) and in vitro protein digestibility corrected amino acid score (IVPDCAAS) of chickpea-sorghum blends Note. IVPD and IVPDCAAS data were analyzed using two way-ANOVA with the Fisher post hoc test (n = 4). Data presented as mean ± standard deviation. Significant differences between extruded and raw samples for the same blend ratio are designated by the symbol *, P < 0.05. Significant differences between blend ratios, but within raw or similar extrusion conditions, are designated by different letters, P < 0.05. The coefficient of variation was ≤4 for IVPD and IVPDCAAS data.

| Extrusion and IVPDCAAS
IVPDCAAS calculates protein quality. IVPDCAAS of raw and extruded chickpea and sorghum flours and chickpea-sorghum blends is presented in Table 3 and 24% moisture) were different than in the current study, which might explain the difference in results between the studies. One of the limitations of the current study was that the available lysine content of the extruded snacks was not determined. Knowledge of available lysine content may have strengthened the interpretation of the IVPDCAAS results.

| CONCLUSIONS
The limiting amino acid was lysine for raw sorghum and tryptophan for raw chickpea and raw chickpea-sorghum blends. Extrusion shifted the limiting amino acids of raw 50:50 and 60:60 chickpea-sorghum blends to lysine. Extrusion increased IVPD of sorghum, chickpea, and chickpea-sorghum blends. Increasing the proportion of chickpea in the chickpea-sorghum blend and the extrusion temperature increased IVPD of the chickpea-sorghum blend, whereas increasing feed moisture content decreased IVPD. The protein quality of chickpeasorghum extrudates was affected significantly by blend ratio, extrusion, and extrusion conditions. Extrusion improved the protein quality of chickpea-sorghum extrudates but not that of the sorghum extrudate. The study illustrated that blending sorghum with chickpea F I G U R E 1 Effect of blend ratio, barrel temperature, and feed moisture content on in vitro protein digestibility of chickpea-sorghum extrudates. Blend ratios are (a) 50:50, (b) 60:40, and (c) 70:30 chickpea-sorghum. Data were analyzed via three-way ANOVA with the Fisher post hoc test (n = 4). Interaction effects were not significant (P > 0.05). Significant differences (P < 0.05) between blend ratios are labeled with different English letters; significant differences (P < 0.05) between temperatures are labeled with different Greek letters; significant differences (P < 0.05) between moisture contents are labeled with different numbers of stars. The coefficient of variation was ≤4 was advantageous from a protein quality point of view. Snacks pre-