1. Introduction

Storage at low temperatures is a method used for food preservation that best preserves the taste, smell, general quality, and nutritional content of foods [1]. The foods have been stored at low temperatures for ages to secure the quality of the resulting product [2]. The storage of food at low temperatures considerably slows down microbiological growth, biochemical reactions, and color and texture deterioration. It also prevents nutrients from being lost and the formation of an off-flavor during storage [3]. One of the critical characteristics influencing the microstructure of the frozen product is ice crystal, which is directly proportional to the freezing rate. Slow freezing causes the production of huge ice crystals, which irreversibly destroy the tissues. A fast freezing results in little ice crystals contribute to the structural quality of the food [4]. Despite its ease, freeze storage methods still have many drawbacks, such as high drip loss, severe shrinkage, diminished nutrients, and color changes [5]. The intensity of the adverse effects of the freezing process on foods relies on the nature of foods as different food materials respond differently to the freezing process [6].

Nelumbo nucifera G, commonly known as lotus, is a perennial aquatic plant belonging to the aquatic genus Nelumbo. It is a good source of carbohydrates, fiber, vitamin C, and minerals such as Ca, P, Fe, and K. All the sections of the N. nucifera and its active ingredients have been employed as a “homology of medicine and food,” serving as both tasty food and potent empirical medicine. [7]. N. nucifera rhizome possesses several phytochemical antioxidants that scavenge reactive oxygen species and break the chain reaction of free radicals [8–10]. At normal temperature, the shelf life of lotus rhizome is around 14–20 days, which may be extended by storage at low temperature. Freeze storage is a preservation technology extensively employed in the food industry to extend the shelf life of lotus roots [11].

The researchers are expressing interest in modifying food material to make it more significant in the industrial and nutritional domains, particularly N. nucifera rhizome starch (NNRS), which was previously modified by acetylation, crosslinking, and oxidation [12]. However, scant information was discovered regarding the impact of storage conditions on NNRS’s phytochemical and antioxidant capacity. Therefore, response surface methodology (RSM) was used to optimize the impact of storage conditions on the phytochemical and antioxidant capacity of NNRS at various storage temperatures and times.

RSM is used to create response surface models to predict changes in response variables with changes in input variables [13]. RSM is also used to get the optimum levels of input variables leading to the desired goals of response variables [14]. In this methodology, the central composite design (CCD) has been used to optimize independent variables and develop a second-order quadratic model for response variables by reducing the number of experimental runs.

2. Materials and Methods

The fresh N. nucifera rhizomes were purchased from the local market, washed in distilled water, cut into small pieces, weighed, packed in airtight polythene bags, and stored under various conditions.

3. Results

The results of dry mass, extract yield, phytochemical composition, and free radical scavenging potential of the fresh and stored rhizome are given in Table 1. The TDM and TEY of fresh rhizomes were found to be 10.33 ± 0.58 and 37.03 ± 0.35%, respectively. The TPC and TFC of fresh rhizome were 21.31 ± 2.03 mg gallic acid eqv./100 g dw and 20.70 ± 1.17 mg catechin eqv./100 g dw, respectively. The TTC, ETC, GTC, and CTC were found to be 22.77 ± 2.38, 19.23 ± 1.75, 20.89 ± 1.99, and 21.99 ± 2.035 mg tannic acid eqv./100 g dw, respectively. The DPPH RSC, ABTS RSC, HRSC, and SORSC of fresh rhizome samples were found to be 94.07 ± 2.40, 79.63 ± 2.69, 97.24 ± 3.09, and 97.58 ± 2.14%, respectively, while those of BHT were 95.37 ± 5.77, 80.36 ± 3.33, 98.89 ± 4.98, and 96.47 ± 2.04%, respectively. The freezing treatment resulted in a significant decrease in TDM, TEY, TPC, TFC, TTC, ETC, GTC, and CTC to 6.71 ± 0.33 (%), 12.35 ± 0.78 (%), and 11.71 ± 0.19 mg gallic acid eqv./100 g dw, 14.05 ± 0.72 mg catechin eqv./100 g dw, 9.91 ± 0.39 mg tannic acid eqv./100 g dw, 8.20 ± 0.37 mg tannic acid eqv./100 g dw, 10.41 ± 1.63 mg tannic acid eqv./100 g dw, and 11.66 ± 1.40 mg tannic acid eqv./100 g dw, respectively. The DPPH RSC, ABTS RSC, HRSA, and SORSC also decreased significantly to 63.40 ± 4.43, 53.92 ± 1.18, 74.56 ± 0.67, and 71.82 ± 1.01%, respectively, under the influence of freeze treatment.

The effects of freezing conditions on dry mass, extract yield, phytochemical composition, and antioxidant potential were optimized by the RSM using a five-level bifactorial CCD. The experimental results of dry mass, extract yield, phytochemical composition, and antioxidant potential of stored rhizome samples at various combinations of S_T and S_t are given in Table 2. The TDM and TEY ranged from 5.30% to 8.20% and 5.75% to 22.02% with mean ± standard deviation of 6.71 ± 0.33 (%) and 12.35 ± 0.78 (%), respectively. The TPC and TFC ranged from 8.4 to 14.8 and 8.266 to 16.854 mg/100 g dw with mean ± standard deviation of 11.71 ± 0.19 and 14.05 ± 0.72 mg/100 g dw, respectively. The TTC, ETC, GTC, and CTC ranged from 7.001 to 13.135, 5.8 to 10.49, 5.9 to 14.5, and 8.2 to 15.3 mg/100 g dw with mean ± standard deviation of 9.91 ± 0.39, 8.20 ± 0.37, 10.41 ± 1.63, and 11.66 ± 1.40 mg/100 g dw, respectively. The DPPH RSC, ABTS RSC, HRSC, and SORSC ranged from 50.87% to 79.117%, 38% to 64.4%, 65.9% to 84.1%, and 63.28% to 81.54% with mean ± standard deviation of 63.40 ± 4.43, 53.92 ± 1.18, 74.56 ± 0.67, and 71.82 ± 1.01%, respectively.

3.1. Response Surface Analysis and Optimization of Results

The experimental data were statistically analyzed using response surface models to find the levels of selected variables to achieve the optimal values of the studied response parameters. Tables 3, 4, and 5 present the statistical terms showing the main, interaction, and quadratic effects obtained by ANOVA. The suggested response surface quadratic model yielded the following polynomial regression equations to show the relationship between the freezing variables and the studied response parameters.

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

$$ ()

Table 3. Parameters of analysis of variance in total phenolic and flavonoid contents of stored N. nucifera rhizome.

Source	TDM (%)			TEY (%)			TPC (g/100 g dw)			TFC (g/100 g dw)
	CE	F value	p value	CE	F value	p value	CE	F value	p value	CE	F value	p value
Model	6.54	50.87	< 0.0001	10.42	268.95	< 0.0001	11.64	841.48	< 0.0001	16.04	91.91	< 0.0001
A—storage temperature (°C)	0.69	150.29	< 0.0001	2.84	483.69	< 0.0001	1.37	1881.12	< 0.0001	1.24	106.24	< 0.0001
B—storage time (week)	−0.56	97.76	< 0.0001	−3.58	773.24	< 0.0001	−1.53	2316.68	< 0.0001	−1.96	265.17	< 0.0001
AB	0.18	3.51	0.07	−0.66	8.61	< 0.0001	0.07	1.76	0.1982	0.27	1.72	0.2030
A2	0.06	1.28	0.27	0.55	23.48	< 0.0001	0.07	5.43	0.0289	−0.62	34.32	< 0.0001
B2	0.08	2.68	0.11	1.01	79.13	< 0.0001	−0.003	0.02	0.9003	−0.98	84.77	< 0.0001
R2	0.92			0.98			0.9946			0.9523
Adjusted R2	0.90			0.98			0.9934			0.9420
Predicted R2	0.87			0.97			0.9913			0.9287
AP (%)	18.85			43.81			70.5448			27.8665
CV (%)	5.02			6.26			1.62			5.15

• Note: R², regression coefficient.
• Abbreviations: AP, adequate precision; TFC, total flavonoid; TPC, total phenolic content.

Table 4. Parameters of analysis of variance in total tannin contents of stored N. nucifera rhizome.

Source	TTC (mg/100 g dw)			ETC (mg/100 g dw)			GTC (mg/100 g dw)			CTC (mg/100 g dw)
	CE	F value	p value	CE	F value	p value	CE	F value	p value	CE	F value	p value
Model	9.10	222.86	< 0.0001	7.39	122.70	< 0.0001	7.90	18.62	< 0.0001	9.95	20.45	< 0.0001
A—storage temperature (°C)	1.53	551.85	< 0.0001	1.04	281.44	< 0.0001	1.28	22.28	< 0.0001	1.41	36.55	< 0.0001
B—storage time (week)	−1.48	517.36	< 0.0001	−1.04	281.44	< 0.0001	−1.79	43.49	< 0.0001	−1.65	50.14	< 0.0001
AB	−0.0730	0.4174	0.5246	0.2181	4.09	0.0549	0.0725	0.0238	0.8788	0.1693	0.1763	0.6785
A2	0.3645	40.26	< 0.0001	0.3423	39.00	< 0.0001	1.24	26.78	< 0.0001	0.7609	13.79	0.0011
B2	0.2915	25.75	< 0.0001	0.3057	31.10	< 0.0001	0.7827	10.73	0.0033	0.6162	9.04	0.0063
R2	0.9798			0.9639			0.8019			0.8164
Adjusted R2	0.9754			0.9560			0.7588			0.7765
Predicted R2	0.9699			0.9454			0.7049			0.7273
AP	38.2515			30.1170			11.6557			12.130
CV	0.9798			4.56			15.65			11.98

• Note: R², regression coefficient.
• Abbreviations: AP, adequate precision; CTC, condensed tannin content; ETC, ellagitannin content; GTC, gallotannin content; TTC, total tannin content.

Table 5. Parameters of analysis of variance in free radical scavenging capacity of stored N. nucifera rhizome.

	DPPH RSC (%)			ABTS RSC (%)			HRSC (%)			SORSC (%)
	CE	F value	p value	CE	F value	p value	CE	F value	p value	CE	F value	p value
Model	56.14	29.87	< 0.0001	58.03	340.51	< 0.0001	72.71	621.25	< 0.0001	69.96	238.01	< 0.0001
A—storage temperature (°C)	4.95	44.96	< 0.0001	5.17	696.41	< 0.0001	4.14	1381.81	< 0.0001	3.81	513.94	< 0.0001
B—storage time (week)	−6.5	77.35	< 0.0001	−5.79	873.76	< 0.0001	−4.52	1644.43	< 0.0001	−4.24	637.75	< 0.0001
AB	0.8177	0.4083	0.5291	−0.36	1.14	0.2958	−0.16	0.6914	0.4142	−0.504	3	0.0965
A2	2.92	20.15	0.0002	−1.39	64.98	< 0.0001	0.83	72.11	< 0.0001	0.675	20.84	0.0001
B2	2.93	20.22	0.0002	−1.92	123.91	< 0.0001	0.66	44.73	< 0.0001	0.831	31.57	< 0.0001
R2	0.866			0.9867			0.99			0.981
Adjusted R2	0.8375			0.9838			0.9911			0.977
Predicted R2	0.8			0.9792			0.9888			0.969
AP	15.435			51.015			61.408			38.93
CV	6.99			2.18			0.897			1.4

• Note: R², regression coefficient.
• Abbreviations: ABTS RSC, azino-bis (tetrazolium sulfate) radical scavenging capacity; AP, adequate precision; DPPH RSC, 2, 2-diphenyl-1-picrylhydrazyl radical scavenging capacity; HRSC, hydroxyl radical scavenging capacity; SORSC, superoxide radical scavenging capacity.

The above polynomial regression equation describes the coefficient for intercept, main effects, interaction terms, and quadratic effects. The effect of S_T and S_t on the phytochemical and antioxidant activity of lotus rhizome is shown by the sign and magnitude of the main effect. The statistical results showed a significant (p < 0.05) positive main effect of the selected storage variables on each studied parameter with relatively higher F values and lower p values. S_T showed significant (p < 0.0001) linear positive effects on TDM, TEY, TPC, and TFC, an interaction positive effect on TEY, a quadratic positive effect on TPC, and a quadratic negative effect on TFC. However, S_T showed a nonsignificant quadratic and interception effect on TDM. S_t showed a significant (p < 0.0001) linear negative effect on TDM and TEY and a quadratic positive effect on TEY. However, no significant effect of S_t was observed on TPC and TFC.

S_T showed a significant (p < 0.01) linear positive effect on TTC, ETC, GTC, and CTC and a significant quadratic positive effect on TTC, ETC, GTC, and CTC. S_t showed a significant (p < 0.0001) linear negative effect on TTC, ETC, GTC, and CTC and a significant (p < 0.01) quadratic positive effect on TTC, ETC, and GTC.

S_T showed a significant (p < 0.0001) linear positive effect on DPPH RSC, ABTS RSC, HRSC, and SORSC and a quadratic positive effect on DPPH RSC, HRSC, and SORSC. S_t showed a significant (p < 0.0001) linear negative effect on DPPH RSC, ABTS RSC, HRSC, and SORSC, a quadratic negative effect on ABTS RSC, and a quadratic positive effect on DPPH RSC, HRSC, and SORSC.

Three-dimensional (3D) response surface plots were drawn to show the main, linear, quadratic, and interaction effects of S_T and S_t on TDM and TEY (Figures 1(a) and 1(b)); TPC and TFC (Figures 2(a) and 2(b)); TTC, ETC, GTC, and CTC (Figures 3(a), 3(b), 3(c), and 3(d)); and DPPH RSC, ABTS RSC, HRSC, and SORSC (Figures 4(a), 4(b), 4(c), and 4(d)) of lotus rhizome.

The applicability of the proposed response surface model was evaluated by computing the predicted values of the analyzed responses using the quadratic model’s polynomial regression equations. The linear regression plots comparing predicted response values against experimental values revealed a relatively proficient agreement between the experimental and predicted values of the analyzed responses (Figures 5(a), 5(b), 5(c), 5(d), 5(e), 5(f), 5(g), 5(h), 5(i), 5(j), 5(k), and 5(l)). The observed values of correlation coefficients (R² = 0.8019–0.9926), coefficient of variance (CV), and adequate precision (AP) suggested the suitability, reliability, and accuracy of the proposed model to investigate the links between freeze conditions and the phytochemical and antioxidant potential of lotus rhizome.

The optimization criteria and results of graphical and numerical optimization of the S_T and S_t to achieve the optimal TDM, TEY, TPC, TFC, TTC, GTC, ETC, CTC, DPPH RSP, ABTS RSC, HRSP, and SORSC with maximum desirability are presented in Figures 6(a), 6(b), 6(c), 6(d), 6(e), 6(f), 6(g), 6(h), 6(i), 6(j), 6(k), 6(l), 6(m), and 6(n). The coded optimal levels (COLs) of S_T and S_t at the maximum response of the studied parameters were 1.69654°C and −0.9687 weeks, respectively. The following generalized equation was used to calculate the actual optimum levels (AOLs) of the selected variables from the coded ones suggested by the design [22].

$$ ()

where AOL is the specific AOL of a particular variable, COL is the coded optimum level of the factors, I_AL is the interval between two actual levels, and CP_AL is the central point of actual levels of that variable. The AOLs of S_T and S_t calculated values at the maximum response of the studied parameters were 8.48°C and 2.0313 weeks, respectively.

4. Discussion

The antioxidant activity of plant foods is recognized as an indicator of the potential health benefits associated with their consumption. Because of their high perishability and seasonality, plant foods are mostly consumed or used as processed products, and freeze storage is one of the technologies used to produce high-quality foods [23]. However, as previously noted, plant foods can be harmed during freezing and frozen storage, resulting in reduced final product quality and functional qualities, particularly after thawing, compared to similar fresh products. Cell breakages can cause the compartmentalization of antioxidants such as anthocyanins and other phenolic compounds and their destruction due to interaction with oxidative enzymes. If frozen foods are handled and processed correctly, their useful characteristics can be preserved after freezing [24].

The observed experimental results showed a significant linear positive effect of S_T and a linear negative effect of St on the lotus rhizome’s phytochemical content and antiradical potential. The results indicated that the phytochemical content and antiradical potential of lotus rhizome are decreased with a decrease in the storage temperature and an increase in the storage time. Studies on the effect of frozen storage conditions (−23°C and −70°C for 6 months) on the TPC of cherries have showed a 25% degradation after 3 months of freezing and a 50% degradation after 6 months of freezing at the selected temperature range [25]. The previous studies reported a decrease in the TFC and antioxidant potential of green and purple leaves of sweet potatoes and outer leaves of onions after storage at 0°C [26]. Similarly, flavonoids and tannins have been reported to be decreased with a decrease in temperature and an increase in freezing duration. Plant tannins have also been reported to be destructed and modified at low temperatures and cause the ripening of fruits and vegetables [27]. The linear negative impact of Ft implies that extended exposure to freezing temperatures is thought to progressively deteriorate or lessen the antioxidants’ potency. The longer the compound is frozen, the lower its antioxidant capacity.

Generally, ice crystals arise in the extracellular environment because of the lower solute concentration and higher freezing point. Ice crystals typically form apoplastic gaps at −2°C resulting in supercooling of the plant cells [28, 29]. If the cell wall is intact, it acts as an effective barrier, preventing ice from penetrating the cell. However, the creation of ice in the extracellular compartment will result in a differential in osmotic pressure, which will suck the liquid water out of the supercooled cells, resulting in tissue dehydration. This shrinks the protoplasm and collapses the cell wall [28, 30]. The loss of water from the intracellular compartment affects turgidity, increases intracellular solute concentration, and, as a result, affects the ionic strength, pH, and viscosity of the fruit’s unfrozen portions, making it susceptible to chemical and enzymatic modifications that can interfere with product quality. Excessive softness of the fruit tissues increases their vulnerability to microbiological deterioration and physical damage, lowering the product’s shelf life [31].

5. Conclusions

A strong association was observed between storage conditions and phytochemical components, particularly total phenolics and flavonoids, and the free radical scavenging potential of N. nucifera rhizome against DPPH, ABTS, hydroxyl, and superoxide radicals. The phytochemical content and free radical scavenging potential of N. nucifera rhizome were found to be a linear function of storage temperature and a linear negative function of storage time. The lotus rhizome showed optimal values of phytochemical constituents and antioxidant activity when stored at a low temperature of 8.48°C for 2 weeks. Both phytochemical content and free radical scavenging potential were decreased with a further decrease in temperature. High R² values indicate that the model is adequate and applicable. The current experimental findings, the developed response model, and the optimization results would be a significant addition to the literature. It would serve as a manual for scientists and business owners seeking the best storage conditions to preserve the bioactive phytochemical antioxidants in food for nutritional, pharmaceutical, and clinical applications.

Conflicts of Interest

The authors declare no conflicts of interest.

Author Contributions

Haq Nawaz created and designed the study, and Farwa Mumtaz contributed to data acquisition and manuscript writing. Rimsha Aosaf and Muhammad Yousaf contributed to interpreting the results. Vania Amin and Mubashir Nawaz contributed to data analysis and wrote up the original draft, and Nafeesa Tahir contributed to data collection and performed statistical analysis.

Funding

No funding was received for this study.