Lablab purpureus (L) bean flour ameliorates plasma proteins and accretion of docosahexaenoic acid (DHA, 22:6, ω‐3) in the plasma, liver, and brain of malnourished rats

Lablab purpureus (L) bean is an undervalued or underutilized orphan crop in many tropical countries, where different forms of malnutrition are associated with stunting growth and cognitive deficiencies. We previously reported that L. purpureus contains ω‐3 α‐linolenic acid (ALA, C18:3, ω‐3), which can act as the precursor of ω‐3 docosahexaenoic acid (DHA, C22:6, ω‐3). Inadequate level of DHA impairs growth, development, and cognitive performance. Therefore, we evaluated if supplementation of L. purpureus seed flour (LPS) affects the nutritional status, in terms of body weight gain, plasma proteins, and DHA levels of malnourished model rats. Three groups of rats, namely, controls, malnourished alone (MN), and 15% LPS‐supplemented malnourished (MN + LPS) rats were fed with LPS for 12 weeks. Afterward, body weight, liver weight, brain weight, plasma proteins, micronutrients, lipid profile, and fatty acid profile of plasma, liver, and brains were determined by standard methods. The levels of liver lipid peroxide (LPO) and proinflammatory TNFα were also measured. The body weight, liver weight, serum total proteins and micronutrients (iron/potassium), and the levels of docosahexaenoic acid significantly increased in the plasma, liver, and brain of MN + LPS rats. Moreover, LPO and TNFα levels reduced significantly in MN + LPS rats. in vitro analysis revealed a significant free radical scavenging and antioxidative potential of L. purpureus seed extract. Thus, L. purpureus not only replenishes protein‐energy malnutrition, but also increases the levels of DHA, an indispensable polyunsaturated fatty acid for brain cognition. Finally, our results suggest that L. purpureus might benefit human malnutrition and related cognitive deficits.

brain, concurrently with reductions of learning-related memory cognitions and mental health (Hashimoto, Hossain, Al, Matsuzaki, & Arai, 2017;Nguyen et al., 2014). Between the LCPUFAs, DHA is crucial for sperm and retinal outer-segment membrane lipid composition (Crawford et al., 1999). DHA may improve attention deficit hyperactivity disorder (ADHD)-a common childhood behavioral disorder (Königs & Kiliaan, 2016). An oral administration of DHA increases the memory of elderly  and Alzheimer's disease model rats (Hashimoto et al., 2002). Though consumption of purified DHA from commercial sources and/or directly from sea-based fish/animal sources is the best way to have an adequate blood/brain levels DHA, it is, however, difficult to afford for the people in poor countries struggling continuously against malnutrition and economic burden. Also, the people living in the landlocked geographical regions without/with limited access to the seafood cannot get sufficient DHA. Given the clear role of DHA in brain functions, much interest has generated whether the levels of blood/brain DHA could be enhanced by dietary means. Despite of having many nutritional components and health-promoting factors, including precursor of DHA (i.e., α-linolenic acid, ALA) and high protein content, L. purpureus has remained a neglected crop. The feeding of low-protein diet is believed to cause protein malnutrition in the experimental rats.
Proteins not only play roles for the structural purposes of the cells, they also perform important roles as enzymes and hormones, and participate in boosting immune systems, regulate cell development and transport functions. When proteins fail to play these roles, an imbalance sets in and protein-energy malnutrition (PEM) may develop in animals. Protein malnutrition may also result in an inadequate intake of micronutrients (Brito et al., 2016;Leite, Jordao, de Andrade, Masson, & Frade, 2011). Therefore, in this investigation, we studied whether the supplementation of LPS improves protein malnutrition and affects the plasma, liver, and brain DHA composition of the experimental malnourished rats. Alongside, we also determined the effects of L. purpureus on the body/liver weight gain, plasma micronutrients, including Na, K, Ca, and Fe levels, lipid profiles, and oxidative stress (OS) in these rats.

| Collection of L. purpureus seeds
Dried L. purpureus seeds were collected from a local market and authenticated by a botanist from the Department of Botany, Jahangirnagar University, Savar, Dhaka. Dried seeds were powdered into fine flour using a mechanical grinder. The diet containing LPS was autoclaved for 20 min at 121 C under 15Ib/in and stored in an airtight container.

| Animals
Protein malnutrition in animals can be induced by variations of composition and duration (weeks to months) of the feeding of restricted diets. The severity of malnutrition depends on the protein contents, namely, diets with low-protein contents may induce severe malnutrition, whereas those with moderate protein contents may induce moderate malnutrition (Lobe, Marica, Bernstein, & C.,, & German, R.Z., 2006;Miñana-Solis et al., 2008;Leite et al., 2011;Brito et al., 2016;Triawanti et al., 2018). In our experiments, the control rats received a normal quantity of protein, whereas the malnutrition model rats were prepared by providing them with low quantity of protein. Briefly, inbred albino Wistar rats (8-WKs, BW 100 g) were divided into three groups: control group (control diet-fed group), malnourished (MN) diet-fed rat group, and 15% L. purpureus-supplemented malnourished rat group (MN + LPS; Figure 1). The MN diet-a low protein diet-contained a lower amount of pure casein (2% by weight, one sixth that of the control diet, Table 1

| Brain tissue preparation
At the end of the experiment, the rats were killed after sodium pentobarbital (65 mg/kg of body weight) anesthesia (i.p., intraperitoneally). F I G U R E 1 In vivo experimental design. Rats were divided into three groups (n = 9): standard pellet diet-fed group (Cont), malnourished diet-fed group (MN), and 15% Lablab purpureussupplemented diet-fed malnourished rat (MN + LPS) group. Diets were fed for 12 weeks, then killed and different parameters were determined After drawing blood, the livers and brains were perfused with ice-cold saline to remove blood. The brain cortex was separated from the whole brain on ice. Tissues were homogenized (10 mg/ml) in phosphate buffer (50 mM, pH 7.4, containing phenylmethylsulfonyl fluoride [PMSF] protease inhibitor), using Dounce glass homogenizer and centrifuged at 500 × g to remove unruptured tissues. The resulting supernatants were assigned as whole homogenates. Whole homogenates were again centrifuged at 10,000 × g for 1 hr to separate the supernatants (cytosolic fractions) for ELISA of tumor necrosis factor alpha (TNFα). All samples were immediately subjected to the assays and/or stored at −80 C.

| Blood biochemistry
Plasma total cholesterol (TC) was determined enzymatically using the cholesterol oxidase method, whereas high-density lipoprotein (HDL)cholesterol (HDL-C) were determined by the same procedure after precipitating low-density lipoprotein (LDL) and very low-density lipoprotein (VLDL), using assay kits for total cholesterol (TC) and HDL-C (total cholesterol E-test and HDL-C test; Wako Pure Chemical Industries, Osaka, Japan). The absorbance of the color complex was measured at 560 nm. Plasma triglyceride (TG) was also measured using a commercially available reagent kit (TG-test; Wako Pure Chemical Industries). In the TG assay protocol, TGs are converted to free fatty acids and glycerol. The absorbance of the color was measured at Serum protein was assayed by protein assay BCA (bicinchoninic acid) reagents. The BCA method quantifies protein in a short time.
The principle of this method is that proteins reduce Cu +2 to Cu +1 in an alkaline solution (the biuret reaction) and result in a purple color formation by bicinchoninic acid. Absorbance was recorded at 560 nm against BSA standard (Wako Pure Chemicals).
Electrolytes were measured by potentiometry (Maas & Sprokholt, 1990). The method used here measured the voltage that develops between the inner and outer surfaces of an ion selective electrode. The electrode (membrane) is made of a material that is selectively permeable to the ion being measured. This potential is measured by comparing it with the potential of a reference electrode. Because the potential of the reference electrode is held constant, the difference in voltage between the two electrodes is attributed to the concentration of ions in the sample. Values for Na and K were expressed as mmol/L, whereas those for Ca and Fe as mg/dl and μg/dl, respectively.

| Lipid peroxide (LPO) assay
LPO levels, as indicator of OS, were assessed as previously described by Hossain et al., (2019). Briefly, 200 μl of 8.1% sodium dodecyl sulfate, 3.0 ml of 0.4% thiobarbituric acid in 20% acetic acid (pH 3.5), and 700 μl distilled water were added to 100 μl of homogenate; the mixture was then incubated at 95 C for 1 hr. After cooling, 1.0 ml water and 4.0 ml n-butanol-pyridine (15:1, v/v) were added and the mixture was vigorously agitated for 20 min. After centrifugation at 1800 × g for 10 min, the absorbance of the upper organic layer was determined at 553 nm.
The wells were then washed, blocked with BSA (1%) in tris-buffered saline (TBS) for 4 hr. After wash, anti-rabbit TNFα 1 antibody (at 1:1000 dilutions) was added and incubated at 4 C for 12 hr. On the next day, the wells were washed, incubated with HRP-coupled 2 anti-rabbit antibody for 2 hr at room temperature. After wash, tetramethylbenzidine was used to develop color; the color reaction was stopped by adding 0.1 N HCl. The absorbance of the plate was determined at 450 nm. All absorbance values were normalized to protein concentration.

| Fatty acid composition
The fatty acid profiles of plasma, liver, and brain tissues were determined by direct transmethylation (Lepage et al., 1986), as described previously . Briefly, to 100 μl of plasma or brain/liver tissue homogenates, 2.0 ml methanol-n-octane (4:1, v/v) containing 10 μg tricosanoic acid as an internal standard and 200 μl acetyl chloride were added. The mixture was incubated at 100 C for 60 min and cooled, then neutralized with 0.5 N aqueous NaOH solution containing 10% sodium chloride.
The neutralized mixture was shaken for 10 min at room temperature and centrifuged at 1800 × g for 5 min. The octane phase with the fatty acid methyl esters was directly subjected to gas chromatography. The gas chromatographic separation was carried out on a Model CA, USA) was initially maintained at 100 C for 1 min, raised to 180 C at 20 C/min, then raised to 240 C at 2 C/min, further raised to 260 C at 4 C/min and maintained for 5 min. The identities of the peaks were established by comparing with those of reference compounds and, in part, by gas chromatography-mass spectrometry.

| Preparation of methanolic extract of LPS
A part of LPS was soaked in methanol for 24 hr with continuous shaking (SI-300/300R/600/600R, Jeio Tech, South Korea). Afterwards, the filtrate was concentrated under vacuum rotary evaporator (J.P. Selecta, s.a, Spain) and stored for the in vitro tests of antioxidative poentials ( Figure 2). The extract was used to determine total polyphenol contents, total flavonoid contents, DPPH-free radical scavenging, and anti-LPO activity.

| Antioxidative potential assay
Total polyphenol content (TPC) of L. purpureus flour extract was determined spectrophotometry by using the Folin-Ciocalteu method (Hossain et al., 2012). Briefly, 0.50 ml of the diluted methanolic extract was transferred in triplicate to separate tubes and allowed to dry. Then, 5.0 ml of a 1/10 dilution of Folin-Ciocalteu's reagent in water was added to each tube and vortexed. Polyphenols present in the extract reduce phosphotungustomolybdic acid of Folin-Ciocalteu reagent in alkaline solution to produce a highly blue colored solution.
Afterwards, 4.0 ml of a sodium carbonate solution (7.5% w/v) was added. The tubes were then allowed to stand at room temperature for 60 min. The color complex was measured at 765 nm. The concentration of total phenol compounds in the extract was determined against gallic acid as standard (μg of gallic acid equivalent [GAE]/gm of dry flour). The total flavonoid content of the extract was measured by the aluminum chloride colorimetric assay against catechin standard (Akter, Haque, Islam, Rahaman, & Bhowmick, 2015) and its concentration in the extract was expressed as catechin equivalent (mg of catechin/mg of extract).
In vitro antioxidant effect of L. purpureus flour extract was performed by determining: (a) the DPPH-free radical scavenging ability, and (b) anti-LPO ability, as described previously from this laboratory .
(a) The DPPH (2, 2 diphenyl-1-picrylhydrazyl) free radical scavenging activity of L. purpureus extract was evaluated as described previously (Hossain et al., 2012). The free radical scavenging ability was evaluated through recording the change of absorbance produced by the reduction of 0.2 mM of DPPH. DPPH scavenging activity was calculated as (%): where, A DPPH = absorbance of the DPPH (0.2 mM)-alone; A DPPH + scavenger = absorbance of the DPPH in the presence of scavenger (here, the L. purpureus extract). Protein concentrations were measured by BCA method.

| Statistical analysis
Results are reported as mean ± SEM values. Data were subjected to Student's t test and/or one-way ANOVA followed by Fisher's PLSD test for post hoc comparisons. A level of p < .05 was considered statistically significant.

| Data availability statement
All data generated or analyzed during this study are included in this published article.

| Effects of L. purpureus supplementation on body, brain and liver weights
The body weight (BW) of MN rats decreased by 16%, when compared with that of the control. The BW of MN + LPS rats rose to that of the control rats ( Figure 3a). The brain weight of the MN rats decreased slightly (by~5%) when compared with that of the control rats; however, the values did not reach significance.
The brain weight of MN + LPS rats, though increased by~8% as compared with that of the MN rats (g), the difference was not statistically significant ( Figure 3b). However, the changes in the body weight and brain weight gave rise to a significant increase in the brain:body weight ratio in MN rats when compared with that of the control rats. L. purpureus-supplementation decreased the brain:body weight ratio to that of the control rats in the MN + LPS rats ( Figure 3c).
Liver weight of the MN rats decreased significantly when compared with that of the control rats. However, the supplementation with L. purpureus increased significantly the liver weight when compared with that of the MN rats ( Figure 3D). Food intake was not statistically different among the rat groups (data not shown).

| Effect of L. purpureus supplementation on plasma total protein and micronutrients
Malnutrition decreased significantly the level of plasma total protein in the MN rats, when compared with that of the control rats. Total protein level, however, increased significantly in MN + LPS rats, as compared with those of the control and MN rats ( Table 2). The levels of Na, K, and Ca decreased in the MN rats, but not significantly. Fe level decreased significantly in the MN rats. L. purpureus supplementation, however, increased significantly the levels of K and Fe in the plasma of these rats. Blood glucose levels were not altered in either of the rat groups (Table 2).
F I G U R E 3 Effects of Lablab purpureus supplementation on the body weight (a), brain weight (b), brain:body weight ratio (c), and liver weight (d) of the in vivo experimental rats. Data were subjected to one-way ANOVA, followed by

| Effect of L. purpureus supplementation on liver fatty acid profile
The effects of L. purpureus supplementation on liver fatty acid profile of the MN rats are shown in Table 4

| Effect of L. purpureus supplementation on brain fatty acid profile
The levels of the saturated PLA and STA and monounsaturated POA and OLA increased significantly in brain tissues of the MN rats (Table 5). Levels of PUFAs such as LLA, AA, DPA were also high in the MN rats, when compared with those in brains of the control rats.
However, feeding of MN rats with L. purpureus increased AA and DHA levels, concurrently with increases in the molar ratios of DHA/AA in their brains. The degree of USI was also significantly higher in the (MN + LPS) rats.

| Effect of L. purpureus supplementation on OS and TNFα in liver tissues of malnourished rats
Malnutrition augmented the degree of OS, as indicated by significant increases in the levels of LPO in the liver tissues of MN rats, as compared with those in the control rats. L. purpureus supplementation, however, decreased significanly LPO levels in the hepatic tissues of these rats (Figure 4a). Levels of proinflammatory TNFα also rose significantly in the liver tissues of MN rats. However, supplementation of L. purpureus to MN rats decreased significantly the TNFα levels in these rats when compared with those of the malnourished (MN) rats ( Figure 4b).

| In vitro results
Total polyphenol and total flavonoid contents of L. purpureus were 17.15 mg gallic acid equivalent (GAE)/gm dry flour and 9.55 mg catechin equivalent (CTE)/gm dry flour, respectively.  [Correction added on 27 March 2020, after first online publication: In Table 3    [Correction added on 27 March 2020, after first online publication: In Table 4  antioxidative effect of the L. purpureus extract is shown in Figure 5.

| DISCUSSION
The results of this study suggest that L. purpureus (L) flour supplementation improves the body weight and increases the levels of protein, potassium, and iron in plasma. Moreover, levels of DHA increases significantly in plasma, liver, and brain tissues, with concurrent ameliorations of hepatic oxidative and proinflammatory stress, and plasma lipid profile in MN rats. Usually, the loss of body weight is used as a marker for the first sign of malnutrition. In this regard, Fletcher and Carey (2011) reported that a loss of 10% of body weight reflects a profound effect of malnutrition. Thus, the decrease in the body weight by 16% authenticates that protein-malnutrition was generated in our experimental MN model rats. Malnutrition can evidently be replenished by the supplementation of adequate diet. Consequently, the increase (p < .05) in the body weight of MN + LPS rats suggests that L. purpureus, at least partially, was able to replenish the malnutrition of the MN rats in our experimental paradigm. Malnutrition, however, caused only a 5% decrease in the brain weight (Figure 3b). This indicates that malnutrition could not sufficiently affect the brain weight of our MN model rats. Here, we speculate that the brain was somehow resistant to malnutrition at the adult age of the rats.
L. purpureus supplementation increased the brain weight by only 8% of the MN + LPS rats; however, the effect on the brain weight was not statistically significant. In contrast, the liver weight of the MN rats decreased significantly (by~17%), as compared with that of the control rats (Figure 3d). The mechanism as to why malnutrition decreased the weight of liver remains to be clarified. We speculate that proteinenergy malnutrition is accompanied by metabolic changes, whereby the animals attempt to guarantee the energy supply to organs of utmost importance at the expense of structural fat and protein.
Our assumption was consistent with the previous reports (Cahill, 1970). Opleta et al. (1998) also reported that malnutrition decreases both cell number and cell size in the hepatic tissues and hence decreases the liver weight. Thus, the increased turnover of the metabolites from the liver and/or worsening of liver cells might be the reasons for which the liver weight of the low protein-supplemented MN model rats decreased. L. purpureus supplementation significantly increased the plasma proteins of the MN + LPS rats, indicating that proteins contained in L. purpureus flour might have contributed to increase the plasma proteins as well as liver weight in these rats.
LLA (C18:2, ω-6) and ALA (18:3, ω-3) are two parent EFAs, which must be supplied in the diet of animals. The balance between ω-3 and T A B L E 5 Effect of Lablab purpureus supplementation on the brain (μg/mg protein) fatty acid profile of the rats [Correction added on 27 March 2020, after first online publication: In Table 5  vious studies have shown that protein-energy malnutrition is associated with reduced levels of PUFAs in the plasma and RBC membranes (Holman et al., 1981;Wolff et al., 1948;Koletzk, Abiodun, Laryea, & Bremer, 1986). The results of decreased levels of LLA and ALA in the MN rats of our investigation are also consistent with the reports of Smita, Muskiet, and Boersma (2004). The LLA and ALA, respectively, represent the major ω-6 and ω-3 PUFAs in LPS [9]. Therefore, the  (Table 4) brought about a significant decrease (p < .05) in the DHA/AA molar ratio and degree of USI.
However, the supplementation of L. purpureus to MN rats significantly decreased the levels of AA and increased DHA levels, DHA/AA molar ratios and the USI of the liver tissues.
Levels of ω-6 LLA and AA increased, whereas those of DHA decreased in the brain tissues of MN rats. Interestingly, the supplementation of L. purpureus increased DHA levels (by approximate ly twofolds), concurrently with an increase in the DHA/AA molar ratio.
The DHA/AA ratio is considered an antioxidative indicator in the brain tissues Hashimoto et al., 2002). Recently, Serini and Calviello (2015) also reported that ω-3, in particular DHA, reduces OS in neurodegenerative diseases. Reductions in the levels of DHA and/or DHA/AA molar ratios are associated with an impairment in learning and memory (Hashimoto et al., 2002(Hashimoto et al., , 2017. Neurodegenerative diseases such as Alzheimer's disease is also accompanied with a decrease in brain DHA levels (Hashimoto et al., 2002). The results of increased levels of DHA or DHA/AA ratio are, therefore, suggestive of the beneficial effects of L. purpureus flour in malnutrition-induced impairments of learning and memory. Furthermore, the USI increased significantly in the plasma, liver, and brain tissues of MN + LPS rats. The larger USI indicates a greater membrane fluidity, which plays a crucial role in many cellular and physiological functions. For example, an increase in membrane fluidity ameliorates hypertension , platelet aggregation Hashimoto et al., 2006), hepatic secretory functions (Hashimoto et al., 2001), exo/endocytosis or neurotransmitter release from the synaptic plasma membranes/brain cognitions (Hashimoto et al., 2017), and cellular glucose transport (Weijers, 2012). These results thus demonstrate that LPS could be used not only in replenishment of proteinenergy malnutrition, it may also replenish essential DHA levels in brain and, consequently, may contribute positively to brain cognition impairments. In contrast, malnutrition increased significantly the levels of saturated fatty acids such as PLA, STA, and TCA, and monounsaturated fatty acids such as POA, OLA, and NVA in brain tissues. It, therefore, remains to be clarified as to whether the increase in the levels of saturated fatty acids was an adaptive response to the EFAdefiency during malnutrion in our MN rats (Table 5).
LPS contains a considerable amount of polyphenols and flavonoids: it exhibited in vitro DPPH-free radical scavenging knack as well as anti-LPO effects ( Figure 5). We, therefore, infer that the antioxi- is severely compromised in malnutrition (Aly, 2014). It is, therefore, conceivable that the hepatic rise in LPO levels, induced by dietary protein depletion, may be one of the important contributing factors in the activation of proinflammatory response. The speculation is well consistent with the increase (p < .05) in hepatic TNFα levels in the MN rats than those in L. purpureus-supplemented rats.
The beneficial effects of L. purpureus supplementation should, nevertheless, be claimed with precautions. Some previous reports claim that the antinutrients such as phytate reduces the absorption of essential trace elements and minerals during gastrointestinal passage, and hence leads to calcium, iron, and zinc deficiencies (Zhou & Erdman, 1995;Jenab &Thompson, 2002;Gemede, 2014). In contrast, the beneficial properties of phytate, for example its antioxidant (Graf, 1987;Minihane & Rimbach, 2002) and anticancer activities (Shamsuddin, 1995(Shamsuddin, , 2002, have also been reported by some other invstigators. In the present investigation, the levels of Na, K, and Ca decreased, though not significantly; Fe levels, however, decreased significantly in the MN rats when compared with those of the control rats. We have previously reported that L. purpureus contains 1% phytate (Hossain et al., 2016). If the phytate present in L. purpureus could reduce the absorption of microelements/minerals, the levels of these elements would decrease further in MN + LPS rats. Instead, the levels of plasma Fe and K increased in MN + LPS rats. As malnutrition is usually accompanied with electrolyte imbalance, it is essential that the imbalance be corrected. L. purpureus contains considerable amounts of micronutrients and minerals (Shahuu et al., 2014).
Therefore, the levels of Fe and K increased and contributed to the electrolyte and mineral pools in the malnourished rats (MN + LPS).
Plasma levels of TG, TC, HDL-C, LDL-C, and VLDL-C were not noticeably altered in the MN rats when compared with those of the control rats. In contrast, Oyagbemi and Odetola (2013) reported an increased level of both plasma TC and TG in the malnourished rats. In our investigation, L. purpureus supplementation to the malnourished rats rather decreased (p < .05) the levels of all these parameters of the lipid profile. The mechanism(s) of action as to why the feeding of Plasma glucose levels were not significantly altered either in the MN or in the MN + LPS rats when compared with those of the controls.

| CONCLUSION
The results of the present study clearly suggest that the supplementation of L. purpureus bean flour increases the body weight as well as levels of plasma protein, potasium, iron, and essential fatty acids (LLA and ALA) in the plasma, liver, and brain. All these results suggest that