Abstract Khai-Nam (Wolffia globosa), a water meal or duckweed, is an aquatic angiosperm commonly dwell in Thailand and neighboring ASEAN countries. With its potential to provide up to 40% (w/w) of protein, Khai-Nam emerges as a promising alternative source of plant-based protein. This research aims to investigate the optimal condition of protein extraction from Khai-Nam by ultrasound-assisted extraction (UAE) method by the response surface methodology (RSM) with analytical Box-Behnken design (BBD). The experiments exhibited that these three factors affected yield of protein extraction. The optimum conditions for extracting protein were a solid-liquid ratio of 0.1:10 g/mL, ultrasound amplitudes of 100% and extraction time of 30 min; after the experiments performed under these conditions, the maximum protein yield was found to be 39.65% w/w, which is similar to the predicted value. The major amino acids in the Khai-Nam protein concentrate (KNPC) were glutamate, aspartate and leucine. Moreover, the water and oil absorption capacity of the KNPC were 6.41 ± 1.89 g water/g protein and 5.68 ± 0.09 g oil/g protein, respectively. In contrast, the maximum foamability, foaming stability, emulsifying activity, and emulsion stability were 292 ± 28.85%, 88.49 ± 0.78%, 100 ± 0.00%, and 90.83 ± 1.18%, respectively. This study suggests that Khai-Nam protein extracted with ultrasound could be a hopeful and viable alternative protein source in food systems, thereby offering a practical solution to the enhancing demand for plant-based proteins.

Keywords: Functional property, Protein extraction, Response surface methodology (RSM), Ultrasound-assisted extraction (UAE), Wolffia globosa

Funding: The authors are grateful for the research funding provided by King Mongkut’s University of Technology North Bangkok. Contract no. KMUTNB-62-DRIVE-07.

Citation:  Phongthai, S., Maitip, J., Promsart, W., Junla, W. and Ungwiwatkul, S. 2025. Optimization of ultrasound-assisted protein extraction from Khai-Nam (Wofflia globose) using response surface methodology. Natural and Life Sciences Communications. 24(2) e2025021.

INTRODUCTION

The continuing rise in the global population, predicted to reach 10 billion by 2050, will intensify the demand for proteins as essential nutrients (Pérez-Vila et al., 2022). Moreover, more customers have recently changed to flexitarian, vegetarian, or vegan diets (Pojić and Tiwari, 2018). New food alternatives must be acquired to ensure food security; hence, research has been conducted on novel sources of vegetable proteins (dos Santos-Silva et al., 2024). Duckweed, Wolffia globosa, is a plant species capable of producing a high protein level. W. globosa, known as Khai-Nam in Thai, is a species of duckweed which classified within the family of Lemnaceae (Yu et al., 2017).  Khai-Nam is the smallest flowering plant with a globe or oval shape, measuring less than a millimeter in diameter. It normally floats in still or slowly flowing water (Romano and Aronne, 2021). Khai-Nam is frequently located in Thailand's freshwater ecosystems and several Southeast Asian nations. This plant serves as a significant nutritional supply, comprising 16–41% protein, 17–35% carbohydrates, 3–9% fat, and having elevated concentrations of minerals and vitamins ranging from 3.5% to 26%, all measured on a dry weight basis. The chemical composition of the plant varied according to the strain and growth conditions (Duangjarus et al., 2022; Xu et al., 2023). W. arrhiza 7678a and W. globosa have been observed to possess 50.89% and 48.20% protein, respectively. They have similar protein contents as eggs (52.7%), beef (40.5%), and pork (27.7%) Ruekaewma et al., 2015; Hu et al., 2022; On-Nom et al., 2023). This research on Khai-Nam's protein extraction potential reassures us about the potential of this plant as a food security solution, suggesting that utilizing Khai-Nam to produce high-protein human food and animal feed may be feasible (Appenroth et al., 2017).

Ultrasound-assisted extraction (UAE), the efficiency extraction technique, has been generally employed for the extraction of notable constituents from biomass, such as proteins, lipids, and bioactive compounds, because of to its numerous advantages, including operation at or near ambient temperature, enhanced efficiency relative to traditional extraction methods, and reduced costs (Xu et al., 2016; Hadiyanto et al., 2018; Sengkhamparn et al., 2019).

Numerous research studies utilize UAE to extract proteins from both plant and animal sources. Ochoa-Rivas et al. (2017) evaluated alkali extraction, microwave-assisted extraction (MAE), and UAE for protein extraction from peanut flour. The optimal protein yield was achieved using UAE, which also enhanced the in vitro protein digestibility and the functional properties (water absorption, foaming, and emulsifying activities) of the isolated proteins. According to Li et al. (2021a), using UAE at 250 W for 20 minutes under a 60% duty cycle significantly increased the protein yield of brewers' spent grains to 86.16%, compared to the traditional extraction method without ultrasound, which yields a protein yield of 45.71%. Moreover, ultrasound-assisted extraction of chicken liver proteins achieved a yield of 67.6%, in contrast to the 43.5% yield from traditional extraction, and this method can improve the functional properties of extracted proteins (Zou et al., 2017). Therefore, UAE method was chosen for protein extraction in this investigation due to its high efficiency, brief operational duration, and well-known advantage as a green technology.

There are a number of characteristics that can have a significant impact on the extraction efficiency of the conventional extraction process. These parameters include the extraction duration, temperature, and the ratio of solid-liquid. Normally, the standard method optimization is achieved by one factor of time, a lengthy approach that may overlook the interactions between variables. It is possible to overcome these constraints by utilizing the response surface methodology (RSM), which incorporates the possibility of interaction effects among variables (Banik and Pandey, 2008). In recent years, RSM has become known as a significant optimization method, proficiently modelling and enhancing biochemical and biotechnological processes (Sarringkarin and Laokuldilok, 2017; Marliani et al., 2022; Phothitontimongkol and Prasertboonyai, 2022; Fortuna et al., 2024). This makes it easier to evaluate the effects of a wide variety of process parameters and the interactions between those factors and response variables by using a limited number of experimental sets (Tang et al., 2010).

Through the utilization of RSM, the aim of this investigation was to determine the most suitable factors for the ultrasound-assisted extraction (UAE) of protein from Khai-Nam biomass. Three independent parameters, the solid-liquid ratio, the amplitude of the sonication, and the extraction time were chosen as the for this investigation. Khai-Nam protein concentrate was also evaluated for its potential as a source of alternating proteins, and its amino acid compositions and functional properties were estimated.

MATERIAL AND METHODS

Materials

Khai-Nam (KN), naturally grown, was collected from the pond located at the front of King Mongkut's University of Technology North Bangkok (KMUTNB), Rayong province, Thailand (12°49'27.1"N 101°12'55.3"E). KN was rinsed with tap water and dried at 60°C for 24 hours, the grinder was used to grind into a fine powder. The KN powder was screened through 60-mesh sieves and kept in a desiccator until use. All compounds employed in this work were of analytical grade.

Chemical composition

The KN biomass was investigated for moisture, ash, fat, and protein content following the protocols established by the Association of Official Analytical Chemists (AOAC, 2000). The content of carbohydrate was assessed utilizing the Compendium of Methods for Food Analysis, Thailand, First Edition, 2003. The concentrated protein content from KN was determined using a factor of nitrogen conversion by 6.25.

Khai-Nam preparation and protein extraction

KN was defatted followed method of Phongthai et al. (2017). The KN powder was combined with 95% ethanol in a 1:5 weight-to-volume ratio and agitated at ambient temperature for one hour. The precipitate pellet was gathered and then re-extracted for two times. The final KN percent was dehydrated overnight in an oven at 30°C. Finally, the defatted KN (DFKN) was stored in aluminum container within a zip-lock plastic bag at -20°C prior to utilization in subsequent studies.

DFKN was solubilized in distilled water (1:10, w/v) for protein extraction, thereafter adjusting the pH to 10 with 3 M sodium hydroxide. The resulting mixtures were extracted utilizing an VCX130PB Vibracell ultra-sonicator (Sonic and Materials, Inc., USA) in accordance with the experimental design parameters. After centrifugation at 10,000 g, and 4°C for 30 minutes, the supernatant solution was collected, and its pH was corrected to 3.0 by 3 M HCL, after which it was centrifuged under the aforementioned conditions. The precipitate pellet was modified to a neutral pH of 7.0, and subsequently lyophilized. KN protein concentrate (KNPC) was determined using the gravimetric technique and stored in the refrigerator at 4°C until further experiments.

Experimental design and statistical analysis

The protein extraction process from DFKN was statistical optimized using response surface methodology (RSM) and the trial version of Design Expert software version 11.0 (Stat-Ease, Inc., USA). The statistical Box-Behnken design (BBD) was select as a model design with three extraction parameters (solid-liquid ratios, amplitudes, and extraction time) at three different levels: -1, 0, +1, as shown in Table 1. The BBD comprises 15 experimental runs including 12 factorial and 3 central points (Table 2). Equation (1) represents the quadratic model used to predict the optimal point,

where Y represents the predicted response, while β₀, βᵢ, βᵢᵢ, and βᵢⱼ are the constant regression coefficient values of the quadratic model, with Xᵢ and Xⱼ as the independent variables.

Table 1. Levels of the three factors employed in the experimental design.

Table 2. Box-Behnken design with three parameters across 15 runs; experimental and predicted protein yields are compared.

Model adequacy, regression coefficients, and statistical significance were assessed using analysis of variance (ANOVA). To visualize the relationships between the responses and independent variables, response surface plots based on the fitted polynomial regression equations were created using the trial version of Design Expert 11.0 software. The data was analyzed statistically at a level of significance of P = 0.05. The adequacy of the model was additionally confirmed using model analysis, the coefficient of determination (R²), and a lack-of-fit test. A mathematical model was developed to elucidate the impacts of different process parameters and their interactions on each assessed response.

Finally, the optimized conditions for maximum protein extraction yield were determined utilizing the response optimizer function from Minitab 18 software (trial version). The experiment was conducted under optimal UAE conditions to validate the predicted results.

Analysis of amino acid profiles

According to the AOAC technique (AOAC, 2000), amino acid profiles of KNPC were analyzed, and identified using a 6890N GC-MS (Agilent Technologies, USA) equipped with a Zebron ZB-AAA GC column. The column size was 10 mm x 0.25 mm with 0.25 m film thickness.

Functional properties

Water absorption capacity

The water absorption capacity was evaluated using the methodology established from Stone et al. (2015). KNPC 0.5 g was deposited in a 25 ml centrifuge tube. Subsequently, 10 ml of distilled water was added, mixed completely, and then separated the pellet by centrifuging at 5,000 × g for 20 min at 25°C. Then, the supernatant was removed, and the pellet was weighed with the centrifuge tube. The capacity of water absorption of KNPC was quantified to water absorbed/protein by weight.

Foaming activity and foam stability

Foaming activity (FA) and foam stability (FS) were evaluated following the procedure of Cao et al. (2009). The KNPC and distilled water mixture with 1% (W/V) concentration was prepared and adjusted to pH 7.0. Twenty milliliters of the mixed solution were homogenized at 10,000 revolutions per minute for one minute. The total volume of the post-homogenization (0 min) mixture was instantly measured and subsequently at thirty minutes. The FA and FS were determined using equations (2) and (3),

FA (%) = (A – B)/B × 100                                        (2)

FS (%) = (A_30min – B)/(A_0min – B) × 100                 (3)

where A represents the volume after whipping (mL), and B denotes the volume before whipping (mL).

Emulsifying activity and emulsifying stability

The emulsifying properties, emulsifying activity (EA), and emulsifying stability (ES) were assessed following the method described by Cao et al. (2009). The KNPC solutions (5%, w/v) were 10 mL at pH 7.0 and then homogenized the mixture by using a homogenizer at 10,000 rpm for 1 min. Then, 10 mL of peanut oil was added and homogenized at 20,000 rpm for 1 min. Centrifuging the mixture at 1,500 × g for 5 min. The KNPC's emulsifying properties were computed using equations (4) and (5),

EA (%) = (A/B) × 100                                                             (4)

ES (%) = (Aincubate/B) × 100                                                 (5)

where A represents the volume of the emulsifying layer (mL) post-centrifugation, B refers to the total volume (mL), and Aincubate denotes to the emulsifying volume after incubation at 80°C for 10 minutes, then followed by centrifugation.

RESULTS

Characterization of KN biomass

The proximate composition of KN biomass was analyzed, and the results are shown in Table 3. Analysis of KN biomass composition revealed KN's dry weight compositions, including 39.18% protein, 34.14% carbohydrates, 16.57% ash, 5.79% fat, and 4.32% moisture.

Table 3. The compositions of KN biomass.

Effects of the process variables on KN protein yield and fitting the model

The results derived from the 15-combination of the independent variables are presented in Table 2. The extraction yield varied between 15.98% and 39.45%. The maximum yield value of 39.45% was achieved with a solid-liquid ratio of 0.1:10 g/mL, ultrasonic amplitude of 100%, and an extraction time of 20 min. Table 4 presents the statistical analysis of the experimental data. The generated model demonstrated statistical significance (P < 0.05) in forecasting protein yield.

Table 4. Variance analysis and regression coefficients of models for extraction yield of KN.

Note: DF – degree of freedom; SS – sum of square; MS – mean square; F – F-statistics test to determine significance; P – probability value

A response surface regression analysis was performed to create mathematical models for the experimental data, aiming to determine the ideal region for protein yield. The corresponding equation (6) is presented as follows:

Y = 22.763- 8.996 X₁ + 2.005 X₂ +3.138 X₃ + 5.431 X₁X₁ - 0.811 X₂X₂ - 0.819 X₃X₃ - 0.532 X₁X₂ + 0.276 X₁X₃ - 0.365 X₂X₃                     (6)

The impacts of the three parameters on the protein yields may be predicted due to the regression model Eq. (6). The 3D response surface plots, shown in Figure 1, demonstrated the independent and dependent variable relationship. These plots illustrate two variables, with the third variable maintained at a constant value. It was found that decreasing solid-liquid ratio (X₁), and increasing sonication amplitude (X₂) and time (X₃) increases the extraction yield of protein from KN.

Figure 1. Response surface plots of independent variables on KN protein yield (%), solid-liquid ratio and amplitude (A), solid-liquid ratio and extraction time (B), amplitude and extraction time (C).

Optimization and validation

The optimal condition after the statistical prediction for protein extraction of KN biomass consisted of solid-liquid ratio of 0.1:10 g/mL, ultrasound amplitudes of 100% and extraction time of 30 min (Figure 2). The experiment was conducted under the optimum UAE conditions to confirm the predicted results. Table 5 compares the predicted and experimental values under optimal conditions. At optimal conditions, the experimental values indicated that the models correlated well with the predicted values, showing a percentage error of less than 5%. The mathematical models can accurately predict the protein extraction process from KN for any solid-to-liquid ratio, ultrasound amplitudes, and extraction times in a range of experiments.

Figure 2. Response optimization plot for KN protein extraction process.

Table 5. Predicted Experimental values for the optimized extraction factors.

Amino acid analysis

The amino acid composition of KNPC was analyzed and reported in Table 6. KNPC contained 228.69 mg/g of total amino acids, of which 42.84% were essential. In addition, the major essential amino acids in the KNPC were leucine, valine, and phenylalanine, whereas glutamate, aspartate, and arginine were the dominant non-essential amino acids. The data indicates that KNPC possesses significant potential as an alternative protein source, as its essential amino acids fulfill the requirements for adults as recommended by FAO/WHO/UNU (Table 7).

Table 6. The amino acid profile of KNPC.

Table 7. Indispensable amino acid of KNPC and the FAO/WHO/UNU suggested requirements for adult.

Functional properties of KNPC

Table 8 displays the functional properties of the KNPC, including the water and oil absorption capacities, foaming activity, foam stability, emulsifying activity, and emulsifying stability.

Table 8. Some functional properties of the KNPC.

DISCUSSION

According to proximate analysis, protein and carbohydrates contents of KN biomass were 39.18% and 34.14% of dry weight, which were similar to the results of Duangjarus et al. (2022) and Sirirustananun (2018). These findings indicate that KN has the potential to be a dominant candidate for the role of a plant-based protein supply.

The optimal conditions for protein extraction from KN were successfully predicted using RSM and BBD. The solid-liquid ratio, amplitude, and extraction duration exhibited a positive and substantial linear correlation with protein yield. The solid-liquid ratio was observed to have the highest linear effect (X₁, P < 0.0000), followed by the extraction time (X₃, P = 0.0002) and amplitude (X₂, P = 0.0014). The quadratic of solid-liquid ratio (X₁X₁) significantly influenced protein extraction. The interaction terms between solid-liquid ratio, sonication amplitude, and extraction time were not statistically significant (P > 0.05). The coefficient of determination (R²) of the model is 0.9863, indicating that 98.63% of the variations are accounted for by the fitted model. Moreover, it was found that the lack of fit value (0.652) at P > 0.05 was not significant, suggesting that the created model accurately represents the actual relationship between the selected factors.

The response surface plot representing the interaction between solid-liquid ratio and amplitude (Figure1A), demonstrating that a reduced solid-liquid ratio resulted in an increased protein yield with increased ultrasonic amplitude. The protein yield enhanced with increased extraction time at a low solid-liquid ratio (Figure 1B).

A greater protein yield was noted in a lower solid-liquid system compared to a higher solid-liquid system. The extraction attained equilibrium due to the comparatively elevated protein concentration in the solution with a greater liquid volume than in a solid-liquid system with more solids (Tang et al., 2010). The amplitude and extraction duration significantly affected the protein yield, as shown in the response surface plot in Figure 1C. Wang et al. (2008) indicated that sonication duration was the most critical factor influencing UAE. As sonication duration increases, ultrasonic waves effectively disrupt cell walls, hence improving the bulk transfer of intracellular products into the solvent; at this point excessive sonication time diminishes solvent permeability into the cell walls due to the accumulation of suspended contaminants. The findings indicate that prolonged extraction time and increased amplitude improve protein extractability in a low solid-liquid system.

After the experiments performed under the optimal conditions (solid-liquid ratio of 0.1:10 g/mL, ultrasound amplitudes of 100% and extraction time of 30 min), the maximum protein yield was found to be 39.65% w/w, which is similar to the predicted value. This indicates that utilizing UAE has the potential to enhance the yield of extracted protein. Negi et al. (2024) reported that utilizing UAE not only accelerates the protein extraction process but also has the potential to change the final protein product's yield, structural, techno-functional, and nutritional aspects. The studies conducted by Nitiwuttithorn et al. (2024) revealed that the use of UAE combined with alkaline extraction led to a 1.83-fold increase in the protein yield of duckweed, W. arrhiza, compared to the use of only alkaline extraction. In addition, it has been suggested that the use of ultrasound has the potential to reduce the time required for protein extraction from sacha inchi (Plukenetia volubilis) down from 60 minutes (conventional extraction) to only 19 minutes (Chirinos et al., 2024).

The amino acid profile analysis showed that the dominant amino acids in the KNPC were glutamate, aspartate, and leucine (28.39, 23.89, and 21.85 mg/g, respectively). The result is comparable to Duangjarus et al. (2022). They found that the predominant in concentrated-protein hydrolysate from Wolffia globosa were glutamate, aspartate and leucine (28.89, 22.85, and 19.56 mg/g, respectively).

It has been reported that glutamate and aspartate residues in the anionic peptides can stimulate the binding of metal ions, which is vital for their antimicrobial activity (Li et al., 2021b). According to Table 7, methionine + cysteine, and histidine were identified as the most limiting amino acids, followed by isoleucine and lysine, in relation to the essential amino acid requirements for adults as established by the FAO/WHO (1985), whereas leucine, phenylalanine + tyrosine, threonine, and valine in KNPC are present in enough quantities as required by standard references.

Protein isolated from Khai-Nam also improves the commercial value by expanding food applications (Kaplan et al., 2019). One of the most important factors that determines the functional characteristics of proteins is their interaction with oil, water, and gas, which are the three crucial elements of food systems. Protein functioning depends on their ability to absorb water into water-oil or oil-water interfaces, their capacity to form films, and their capacity to inhibit coalescence. Proteins play an important role as surface active agents. The solubility, viscosity, and gelling properties of a protein are primarily influenced by its interactions with water. However, its interactions with oil and gas determine its emulsifying and foaming capabilities (Bashir et al., 2016).

There were several factors affected the capacity of a protein to hold onto water, including the composition of amino acids, protein's structure, and the extent of the protein's surface exposed to water or oil (Wani et al., 2011). According to Table 7, the KNPC used in this study absorbed 6.41 ± 1.89 grams of water per gram of protein. This is correlated to the previous report that studied on the jackfruit seed protein isolate. It found that 6.46 grams of water per gram of protein reported for jackfruit seed protein isolate that was processed using ultrasound (Ulloa et al., 2017) and higher than the values of 3.22, 3.57, and 3.13 g water/g protein observed for the protein isolate of physic nut (Saetae et al., 2011). The results indicate that KNPC can effectively hold onto water, probably because of the attraction between the protein's water-binding parts and water molecules. One of the most important characteristics of proteins in the viscous-rich foods, including baked products, custards, doughs, and soups, is their capacity to absorb water. This is because these foods must possess the capacity to retain water without causing the breaking down of proteins, which is a factor that contributes to the thickening and viscosity of the substance (Seena et al., 2005).

The nonpolar amino acid side chains and the lipid hydrocarbon chains from hydrophobic interactions affect proteins' oil-binding capacity, a crucial functioning property in the food industry. In this study, the oil absorption capacity of KNPC was found to be 5.68 ± 0.09 g oil/g protein, which higher than the 0.59 ± 0.16 g oil/g protein of Spirulina platensis (Bashir et al., 2016) and the 3.8 ± 0.2 g oil/g protein that found in pea protein isolates through the process of alkali extraction-isoelectric precipitation (Stone et al., 2015). This result indicated that KNPC contains a significant amount of nonpolar amino acids capable of forming bonds with lipid hydrocarbon chains. These findings are significant for formulating ground meat and substitutes and extenders to create tremendous products, such as baked goods, doughnuts, frankfurters, sausages, and soups (Kaur and Singh, 2007).

Foaming activity, a protein's capacity to generate foam under specific parameters, and foaming stability, the protein's efficacy in maintaining foam volume over a designated duration, are key areas of interest in this study. The values of foaming activity and foaming stability of KNPC were observed at 296.26 ± 28.85% and 88.49 ± 0.78%, respectively. The foam capacity plays a crucial part in various food items including cream-related desserts, and certain confectionery products (Singh et al., 2008). Overall, KNPC has the potential to be considered a highly effective foaming agent in a wide variety of food products.

A wide variety of amino acids, including charged, nonpolar, and uncharged polar, are typically considered to be the main components of protein molecules. Individual amino acids possess both hydrophilic and hydrophobic characteristics. This enables proteins to engage with both aqueous and lipid molecules, rendering them efficient emulsifiers. From Table 7, the emulsifying activity and emulsion stability of KNPC were 100 ± 0.0% and 90.83 ± 1.18%, respectively. According to Stone et al. (2015), commercial protein isolate products produced from egg, whey, soy, and pea displayed emulsifying activity and emulsion stability that ranged from 106.2 ± 0.0% to 210.4 ± 14.4% and 94.7 ± 2.3% to 100.0 ± 0.0%, respectively.  Consequently, the KNPC demonstrated emulsifying capacity and stability similar to the other protein isolate products. Thus, KNPC could be advantageous for inclusion in emulsion-based food products, such as sausage, salad dressing, or butter.

Nevertheless, the functionalities of protein in food formulation are also influenced by a variety of factors, including the concentration used, processing parameters (e.g., temperature, shear force, pH), as well as the co-ingredients used. It is possible that the over- or under-use of KNPC is the result of undesirable characteristics in food products. Therefore, it is necessary to optimize the level of addition prior to its implementation in a realistic food formulation.

CONCLUSION

Protein extraction from Khai-Nam biomass was optimized using RSM. The optimum extraction conditions were KN biomass consisting of a solid-liquid ratio of 0.1:10 g/mL, ultrasound amplitudes of 100%, and an extraction time of 30 min. Under the optimum conditions, the actual yield from the experiment corresponded with the predicted yield. Furthermore, the KNPC exhibited significant functional attributes in water and oil absorption capacity, foaming, and emulsification. Consequently, KNPC may serve as an alternative protein source useful in food systems. However, in protein extraction using UAE method, it is crucial to evaluate the energy consumption, economic feasibility, and industrial scalability to assess the potential of this method for application in protein extraction from Khai-Nam.

ACKNOWLEDGEMENTS

This research was funded by King Mongkut’s University of Technology North Bangkok. Contract no. KMUTNB-62-DRIVE-07. The authors are greatly grateful for the support from Faculty of Science, Energy and Environment, King Mongkut’s University of Technology North Bangkok, Rayong Campus. This study was partially supported by Chiang Mai University. The authors greatly thank the Faculty of Agro-Industry, Chiang Mai University, for supporting all research facilities.

AUTHOR CONTRIBUTIONS

Suphat Phongthai: conceptualization, investigation, writing—reviewing and editing, writing an original draft; Jakkrawut Maitip: conceptualization, data curation, writing—reviewing and editing; Woranika Promsart: research design, investigation, data analysis, writing an original draft; Wararak Junla: investigation; Sunisa Ungwiwatkul: conceptualization, research design, data curation, investigation, data analysis, writing an original draft, writing—reviewing and editing, funding acquisition. All authors have read and approved of the final manuscript.

CONFLICT OF INTEREST

The authors declare that they hold no competing interests.

REFERENCES

AOAC. 2000. Official methods of the association of official analytical chemists. Association of Official Analytical Chemists, Washington D.C.

Appenroth, K. J., Sree, K. S., Böhm, V., Hammann, S., Vetter, W., Leiterer, M., and Jahreis, G. 2017. Nutritional value of duckweeds (Lemnaceae) as human food. Food Chemistry. 217: 266-273.

Banik, R. M. and Pandey, D. K. 2008. Optimizing conditions for oleanolic acid extraction from Lantana camara roots using response surface methodology. Industrial Crops and Products. 27(3): 241-248.

Bashir, S., Sharif, M .K., Butt, M. S., and Shahid, M. 2016. Functional properties and amino acid profile of Spirulina platensis protein isolates. Pakistan Journal of Scientific and Industrial Research Series B: Biological Sciences. 59(1): 12-19.

Cao, X., Wen, H., Li, C., and Gu, Z. 2009. Differences in functional properties and biochemical characteristics of congenetic rice proteins. Journal of Cereal Science. 50(2): 184-189.

Chirinos, R., Scharff-Salinas, R., Rodriguez-Diaz, J., Figueroa-Merma, A., Aguilar-Galvez, A., Guzmán, F., and Campos, D. 2024. Conventional and ultrasound-assisted extractions of protein from sacha inchi (Plukenetia volubilis) and their impact on the physicochemical and structural characteristics. Applied Food Research. 4(2): 100545.

dos Santos-Silva, A. C., Saraiva, B. R., Lazzari, A., Dos Santos, H., de Oliveira, É. L., Sato, F. and Matumoto-Pintro, P.T. 2024. Optimization and characterization of protein extraction from asparagus leafy by-products. Foods. 13(6): 894.

Duangjarus, N., Chaiworapuek, W., Rachtanapun, C., Ritthiruangdej, P., and Charoensiddhi, S. 2022. Antimicrobial and functional properties of duckweed (Wolffia globosa) protein and peptide extracts prepared by ultrasound-assisted extraction. Foods. 11(15): 2348.

FAO/WHO. 1985. Energy and protein requirements. Report of Joint FAO/WHO/UNU Expert Consultation Technical Report. Series No. 724. FAO/WHO and United Nations University, Geneva.

Fortuna, D., Darmawati, E., Sunarti, T. C., Mardjan, S. S., Widayanti, S. M., Yuliana, N. D., and Purwanti, N. 2024. Response surface methodology for optimization of mucilage extraction from Dioscorea alata tuber. Natural and Life Sciences Communications. 23(2): e2024019.

Hadiyanto, H. and Adetya, N. P. 2018. Response surface optimization of lipid and protein extractions from Spirulina platensis using ultrasound assisted osmotic shock method. Food Science and Biotechnology. 27(5): 1361-1368.

Hu, Z., Fang, Y., Yi, Z., Tian, X., Li, J., Jin, Y., and Zhao, H. 2022. Determining the nutritional value and antioxidant capacity of duckweed (Wolffia arrhiza) under artificial conditions. LWT - Food Science and Technology. 153: 112477.

Kaplan, A., Zelicha, H., Tsaban, G., Meir, A. Y., Rinott, E., Kovsan, J., Novack, L., Thiery, J., Ceglarek, U., and Burkhardt, R. 2019. Protein bioavailability of Wolffia globosa duckweed, a novel aquatic plant–A randomized controlled trial. Clinical Nutrition. 38(6): 2576-2582.

Kaur, M. and Singh, N. 2007. Characterization of protein isolates from different Indian chickpea (Cicer arietinum L.) cultivars. Food Chemistry. 102 (1): 366-374.

Li, J., Hu, S., Jian, W., Xie, C., and Yang, X. 2021b.  Plant antimicrobial peptides: Structures, functions, and applications. Botanical Studies. 62(1): 1-15.

Li, W., Yang, H., Coldea, T.E., and Zhao, H. 2021a. Modification of structural and functional characteristics of brewer's spent grain protein by ultrasound assisted extraction. Lwt. 139: 110582.

Marliani, N., Artika, I. M., and Nurcholis, W. 2022. Optimization extraction for total phenolic, flavonoid contents, and antioxidant activity with different solvents and UPLC-MS/MS metabolite profiling of Justicia gendarussa Burm f. Chiang Mai University Journal of Natural Sciences. 21(3): e2022046.

Negi, P., Kalsi, R., Bhasin, J. K., Kashyap, P., Thakur, A., and Goksen, G. 2024. Ultrasound-driven advancements in food waste protein extraction: Assessing yield, nutritional impacts, techno-functionality, and structural modifications. Sustainable Chemistry and Pharmacy. 42: 101767.

Nitiwuttithorn, C., Wongsasulak, S., Vongsawasdi, P., and Yongsawatdigul, J. 2024. Effects of alkaline and ultrasonication on duckweed (Wolffia arrhiza) protein extracts’ physicochemical and techno-functional properties. Frontiers in Sustainable Food Systems. 8: 1343615.

Ochoa-Rivas, A., Nava-Valdez, Y., Serna-Saldívar, S. O., and Chuck-Hernández, C. 2017. Microwave and ultrasound to enhance protein extraction from peanut flour under alkaline conditions: Effects in yield and functional properties of protein isolates. Food and Bioprocess Technology. 10: 543-555.

On-Nom, N., Promdang, P., Inthachat, W., Kanoongon, P., Sahasakul, Y., Chupeerach, C., and Temviriyanukul, P. 2023. Wolffia globosa-based nutritious snack formulation with high protein and dietary fiber contents. Foods. 12(14): 2647.

Pérez-Vila, S., Fenelon, M. A., O'Mahony, J. A., and Gómez-Mascaraque, L.G. 2022. Extraction of plant protein from green leaves: Biomass composition and processing considerations. Food Hydrocolloids. 133: 107902.

Phongthai, S., Lim, S. T., and Rawdkuen, S. 2017. Ultrasonic‐assisted extraction of rice bran protein using response surface methodology. Journal of Food Biochemistry. 41: e12314.

Phothitontimongkol, T. and Prasertboonyai, K. 2022. Process optimization and modeling of low-cost activated carbon preparation and its application in chromium (VI) removal. Chiang Mai Journal of Science. 49(6): 1559-1575.

Pojić, M., Mišan, A., and Tiwari, B. 2018. Eco-innovative technologies for extraction of proteins for human consumption from renewable protein sources of plant origin. Trends in Food Science & Technology. 75: 93-104.

Romano, L. E., and Aronne, G. 2021. The world smallest plants (Wolffia sp.) as potential species for bioregenerative life support systems in space. Plants.10(9): 1896.

Ruekaewma, N., Piyatiratitivorakul, S., and Powtongsook, S. 2015. Culture system for Wolffia globosa L. (Lemnaceae) for hygiene human food. Songklanakarin Journal of Science and Technology. 37(5): 575-580.

Saetae, D., Kleekayai, T., Jayasena, V., and Suntornsuk, W. 2011. Functional properties of protein isolate obtained from physic nut (Jatropha curcas L.) seed cake. Food Science and Biotechnology. 20(1): 29-37.

Sarringkarin, W. and Laokuldilok, T. 2017. Optimization of the production conditions of glutinous rice bran protein hydrolysate with antioxidative properties. Chiang Mai University Journal of Natural Sciences. 16(1): 1-18.

Seena, S. and Sridhar, K. R. 2005. Physicochemical, functional and cooking properties of under explored legumes, Canavalia of the southwest coast of India. Food Research International. 38(7): 803-814.

Sengkhamparn, N., Lasunon, P., and Tettawong, P. 2019. Effect of ultrasound assisted extraction and acid type extractant on pectin from industrial tomato waste. Chiang Mai University Journal of Natural Sciences. 18(2): 214-225.

Singh, P., Kumar, R., Sabapathy, S.N., and Bawa, A.S. 2008. Functional and edible uses of soy protein products. Comprehensive Reviews in Food Science and Food Safety. 7(1): 14-28.

Sirirustananun, N. 2018. Appropriate proportion of water meal (Wolffia arrhiza (L.)) and commercial diet in combined feeding for tilapia fingerlings rearing. International Journal of Agricultural Technology. 14(2): 249-258.

Stone, A. K., Karalash, A., Tyler, R. T., Warkentin, T. D., and Nickerson, M. T. 2015. Functional attributes of pea protein isolates prepared using different extraction methods and cultivars. Food Research International. 76: 31-38.

Tang, D. S., Tian, Y. J., He, Y. Z., Li, L., Hu, S. Q., and Li, B. 2010. Optimisation of ultrasonic-assisted protein extraction from brewer’s spent grain. Czech Journal of Food Sciences. 28(1): 9-17.

Ulloa, J. A., Villalobos Barbosa, M. C., Resendiz Vazquez, J. A., Rosas Ulloa, P., Ramírez Ramírez, J. C., Silva Carrillo, Y., and González Torres, L. 2017. Production, physico-chemical and functional characterization of a protein isolate from jackfruit (Artocarpus heterophyllus) seeds. CyTA - Journal of Food. 15(4): 497–507.

Wang, J., Sun, B., Cao, Y., Tian, Y., and Li, X. 2008. Optimisation of ultrasound-assisted extraction of phenolic compounds from wheat bran. Food Chemistry. 106(2): 804-810.

Wani, A.A., Sogi, D.S., Singh, P., Wani, I.A., and Shivhare, U.S. 2011. Characterisation and functional properties of watermelon (Citrullus lanatus) seed proteins. Journal of the Science of Food and Agriculture. 91(1): 113-121.

Xu, G., Liang, C., Huang, P., Liu, Q., Xu, Y., Ding, C., and Li, T. 2016. Optimization of rice lipid production from ultrasound-assisted extraction by response surface methodology. Journal of Cereal Science. 70: 23-28.

Xu, J., Shen, Y., Zheng, Y., Smith, G., Sun, X. S., Wang, D., Zhao, Y., Zhang, W., and Li, Y. 2021. Duckweed (Lemnaceae) for potentially nutritious human food: A review. Food Reviews International. 39(7): 3620-3634.

Yu, C., Zhao, X., Qi, G., Bai, Z., Wang, Y., Wang, S., Ma, Y., Liu, Q., Hu, R., and Zhou, G. 2017. Integrated analysis of transcriptome and metabolites reveals an essential role of metabolic flux in starch accumulation under nitrogen starvation in duckweed. Biotechnology for Biofuels. 10:1-14.

Zou, Y., Li, P. P., Zhang, K., Wang, L., Zhang, M. H., Sun, Z. L., and Wang, D. Y. 2017. Effects of ultrasound-assisted alkaline extraction on the physiochemical and functional characteristics of chicken liver protein isolate. Poultry Science. 96(8): 2975-2985.

OPEN access freely available online

Natural and Life Sciences Communications

Chiang Mai University, Thailand. https://cmuj.cmu.ac.th