Enhancing the quality and antioxidant capacity of phycocyanin extracted from Spirulina platensis PCC 7345: A quality-by-design approach

1 Introduction

In recent years, the search for natural bioactive compounds with potential health benefits has gained significant momentum in food and pharmaceutical sciences. Spirulina, a blue-green microalga, has garnered considerable attention due to its rich content of bioactive molecules (Paramanya et al., 2023b). Phycobiliproteins are the light-harvesting complex found most commonly in Spirulina. They are further divided into four types based on their absorption wavelengths: phycocyanin (620–640 nm), allophycocyanin (650–660 nm), phycoerythrocyanin (550–575 nm), and phycoerythrin (500–575 nm) (Pagels et al., 2019). Phycocyanin is the most abundant protein found in cyanobacteria and is assumed to be responsible for the antioxidant properties of phycobiliprotein (Paramanya et al., 2023b). Phycocyanin has demonstrated a wide array of potential health-promoting properties, such as antioxidant, anti-inflammatory, and immunomodulatory effects, making it an appealing candidate for applications in nutraceuticals and functional foods (Fernandes et al., 2023). Extensive research has underscored its prowess in scavenging reactive oxygen species (ROS) and mitigating oxidative stress, a fundamental player in the realm of chronic diseases, including diabetes and its multifaceted complications (Martinez-Gonzalez et al., 2017).

Oxidative stress, a hallmark of various chronic diseases, including diabetes, results from the imbalance between harmful reactive oxygen species (ROS) production and the body's repair mechanisms (Pokhrel et al., 2022). Given its robust antioxidant potential, phycocyanin stands out as a promising candidate for mitigating oxidative stress in the context of diabetes, a condition often characterized by elevated oxidative stress levels (Paramanya and Ali, 2019). Phycocyanin exerts control over carbohydrate absorption by inhibiting amylase and glucosidase enzymes, resulting in a gradual increase in post-meal blood glucose levels. This measured response reduces oxidative stress associated with elevated glucose levels and reactive oxygen species production, making phycocyanin a promising candidate for diabetes management. Its multifaceted role in regulating blood sugar and mitigating oxidative stress offers new avenues for improving diabetes care (Poojari et al., 2022; Shrestha et al., 2021).

Phycocyanin's multifaceted attributes, primarily its antioxidant capabilities, position it as a versatile agent in managing diabetes. This versatility suggests ongoing research, product development, and the exploration of phycocyanin's role as a valuable natural compound to enhance health outcomes across different diseases (Poojari et al., 2022; Shrestha et al., 2021). This research aims to extract Phycocyanin from Spirulina and determine its antioxidant potential and capability to inhibit the activity of α-amylase and α-glucosidase enzymes. These findings will further facilitate in evaluating the antiglycating potential of Phycocyanin.

The extraction of phycocyanin can produce variable results attributed to inherent fluctuations in factors including the cell cycle, environmental conditions during organism growth, and methodological nuances (Li et al., 2020). These variations introduce unpredictability in extraction efficiency and impact the quality of phycocyanin (Betz et al., 2011). Quality by Design (QbD) is a systematic approach to product development and manufacturing, aiming to consistently produce high-quality products. It focuses on understanding and controlling critical parameters throughout the product lifecycle, promoting efficiency, risk management, and regulatory compliance (Yu et al., 2014). In the context of phycocyanin extraction, QbD is vital for optimizing processes and ensuring reliable and robust outcomes. The Box-Behnken Design, a response surface methodology, is particularly valuable in this pursuit. Its efficiency in exploring multiple variables simultaneously allows for the identification of optimal conditions, providing statistical rigor and versatility across various scientific fields (López et al., 2020), including the extraction of phycocyanin where it aids in systematically varying factors to enhance the overall efficiency and quality of the extraction process.

The authors observed a lack of research articles utilizing response surface methodology for optimizing phycocyanin extraction, where Silveira et al. (2007) employed factorial design. In contrast, our study takes a Quality by Design (QbD) approach, utilizing the Box-Behnken design to optimize the extraction of high-purity phycocyanin from Spirulina, highlighting a systematic approach to ensure product quality.We assess phycocyanin purity through Fourier-Transform Infrared Spectroscopy (FTIR) and confirm structural integrity using Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE). Additionally, we explore its antioxidant potential through various assays, which are crucial for applications in functional foods and pharmaceuticals. This study reveals that Spirulina-derived phycocyanin inhibits amylase and glucosidase enzymes, offering potential benefits in managing post-meal blood sugar levels for individuals with diabetes. Additionally, its antioxidant properties can help counteract oxidative stress, reducing the risk of diabetes-related complications. These findings suggest the potential use of phycocyanin as a natural intervention for post-meal blood sugar control and diabetes-related complications, opening avenues for further research and product development in this area.

2 Materials and Methods

3 Results

3.1

3.1 Quality by Design approach

3.1.1

3.1.1 Defining the TPPs and determination of CQAs

In our study, we adopted a QbD approach for extracting Phycocyanin from Spirulina platensis PCC 7345. This approach necessitated carefully considering various TPPs based on an extensive literature review. As a critical step in the QbD framework, we established a set of CQAs that directly impact the purity and quality of the extracted Phycocyanin. After comprehensive evaluation, it was determined that Phycocyanin concentration and purity would serve as the ultimate CQAs, as they play a pivotal role in ensuring the quality of the Phycocyanin product. These CQAs were chosen based on their direct influence on the final product's integrity and suitability for various applications (Kumar et al. 2014, Seo et al., 2013).

3.1.2

3.1.2 Risk evaluation

The CQAs chosen for assessment were Phycocyanin concentration and purity. Consequently, it was imperative to assess the potential factors that could influence these primary product qualities. To identify these factors, Ishikawa diagrams (also known as cause and effect diagrams) were created, as illustrated in Fig. 1. The Ishikawa diagram revealed that the concentration of the buffer, the duration of the culture's growth phase, and the number of freeze–thaw cycles were identified as factors with a significant impact on Phycocyanin concentration and purity, as indicated in references (Asayama et al., 2012; Kuddus et al., 2013).

Close

Fig. 1

Ishikawa fishbone diagram representing the factors that influence Phycocyanin Concentration and Purity. [The three factors selected for assessment have been highlighted in yellow.].

Export to PPT

3.1.3

3.1.3 Box Behnken design

The Box-Behnken design was utilized, and Design Expert software was employed to analyze the impact of different factors on the response variables: Phycocyanin concentration and purity. The experimental matrix consisted of 15 distinct criterias, as detailed in Table 1. Within this matrix, Phycocyanin concentration ranged from 0.0121 to 0.2812 mg/ml, while purity ranged from 0.17 to 0.6948. The remaining data points were evenly distributed across this specified range, as referenced in sources (Aoki et al., 2021; Asayama, 2012; Dhoble and Patravale, 2019; Kuddus et al., 2013; Kumar et al., 2014; Seo et al., 2013; Wang et al., 2001).

Table 1 Box Behnken Design for Extraction of Phycocyanin.

		Factor 1	Factor 2	Factor 3	Response 1	Response 2
Std	Run	A:Growth phase	B:Concentration of Buffer	C:Number of cycles	Phycocyanin Concentration	Phycocyanin Purity
		Days	M	Au	mg/ml	Au
9	1	9	0	2	0.0138	0.3112
10	2	9	1	2	0.02	0.259
3	3	3	1	3	0.043	0.202
12	4	9	1	4	0.1207	0.432
4	5	15	1	3	0.2812	0.6948
7	6	3	0.5	4	0.0338	0.17
13	7	9	0.5	3	0.102	0.3567
1	8	3	0	3	0.0608	0.2814
5	9	3	0.5	2	0.0121	0.32
15	10	9	0.5	3	0.1419	0.3306
6	11	15	0.5	2	0.0158	0.3644
14	12	9	0.5	3	0.09698	0.3615
8	13	15	0.5	4	0.1253	0.2833
2	14	15	0	3	0.0332	0.2454
11	15	9	0	4	0.0641	0.5116

3.1.3.1

3.1.3.1 Response 1

In the case of Phycocyanin concentration, the software examined four mathematical models, which included the linear model, 2FI model, cubic model, and quadratic model. Among these options, the quadratic model was recommended for its higher predictive power, whereas the cubic model was identified alliased (as indicated in Table 2). Following the application of ANOVA, the quadratic model was confirmed to be statistically significant, and the lack of fit (p < 0.05) was deemed non-significant.

Table 2 Model Summary Statistics for Response 1.

Source	Std. Dev.	R2	Adjusted R2	Predicted R2	PRESS
Linear	0.0598	0.4513	0.3016	−0.1046	0.0792
2FI	0.0489	0.7332	0.5331	−0.1241	0.0806
Quadratic	0.0359	0.9104	0.7491	−0.2014	0.0862	Suggested
Cubic	0.0246	0.9831	0.8818		*	Aliased

A model diagnostic plot is a standard tool in statistics and data analysis, particularly for evaluating the effectiveness of predictive models like linear regression models. It helps assess how well the model's predictions align with the actual data (Dhoble and Patravale, 2019). In Fig. 2, the data points are distributed close to or along the diagonal line, indicating a good fit of the model. Since there are no noticeable deviations from the 45-degree line, we can infer that the model does not exhibit any bias.

Close

Fig. 2

Model diagnostic plots for Response 1. (a) Normal residual plot. (b) Predicted vs actual plot. [The predicted values align with the externally scrutinized values(a) and actual observed values (b), resulting in a scatter of data points along or near a 45-degree diagonal line (the line where predicted = actual). Deviations from the 45-degree line reveal the presence of bias in the model.].

Export to PPT

The Box-Cox transformation is a family of power transformations used to stabilize variance and make data conform more closely to a normal distribution (Dhoble and Patravale, 2019). The λ for the given data set = 1, the software suggests the fit for the data should be at λ = −0.26. The suggested transformation for the current data set is inverse square root which will improve the model fitness.

The final equation describing the response1 is Phycocyanin concentration (mg/ml) = −0.395093--0.011435* Growth phase −0.199347* Concentration of Buffer + 0.340080* Number of cycles + 0.022150 (Growth phase * Concentration of Buffer) + 0.003658 (Growth phase * Number of cycles) + 0.025200 (Concentration of Buffer * Number of cycles) −0.000236 Growth phase² −0.002353 Concentration of Buffer² −0.058388 Number of cycles² (Fig. 2, Fig. 3, Fig. 4).

Close

Fig. 3

Box-Cox Model Diagnostic Plot for Phycocyanin Concentration [The X-axis indicates λ = The optimal power transformation while the Y-axis represents the log of residuals.].

Export to PPT

Close

Fig. 4

Contour plots for factors affecting Phycocyanin concentration [Two factors namely A: Growth phase (days) and B: Concentration of Buffer (M) are represented in the X axes, Response: Phycocyanin concentration (mg/ml) is represented in Y axis.].

Export to PPT

3.1.3.2

3.1.3.2 Response 2

Regarding Phycocyanin purity, the software evaluated four mathematical models, specifically the linear model, 2FI model, cubic model, and quadratic model. Among these models, the 2FI model was identified as having the highest predictive power, while the cubic model was observed to be aliased, as indicated in Table 3. Subsequent ANOVA analysis confirmed the significance of the 2FI model, and it was determined that the lack of fit (p < 0.05) was not significant.

Table 3 Model Summary Statistics for Response 2.

Source	Std. Dev.	R2	Adjusted R2	Predicted R2	PRESS
Linear	0.1278	0.2403	0.0331	−0.6239	0.3841
2FI	0.1164	0.5416	0.1979	−1.4278	0.5742	Suggested
Quadratic	0.1333	0.6242	−0.0523	−4.9811	1.41
Cubic	0.0166	0.9977	0.9836		*	Aliased

Fig. 5 displays data points that are closely aligned with or clustered around the diagonal line, suggesting a strong model fit. The absence of significant deviations from the 45-degree line leads us to conclude that the model is devoid of any noticeable bias.

Close

Fig. 5

Model diagnostic plots for Response 2. (a) Normal residual plot. (b) Predicted vs actual plot. [The predicted values align with the externally scrutinized values(a) and actual observed values (b), resulting in a scatter of data points along or near a 45-degree diagonal line (the line where predicted = actual).].

Export to PPT

The Box-Cox Model Diagnotic Plot for Phycocyanin Purity, the λ for the given data set = 1, the software suggests the fit for the data should be at λ = -0.12. The software considers the model as fit and does not recommend any teansformations.

The final equation describing the response 2 is Phycocyanin purity = -0.173547 + 0.074605* Growth phase + 1.74942*Concentration of Buffer + 0.100200* Number of cycles-0.081383 *(Growth phase * Concentration of Buffer)-0.618200* (Concentration of Buffer * Number of cycles) −0.004658* Growth phase² −1.60629* Concentration of Buffer² + 0.006969 (Growth phase² * Concentration of Buffer) + 0.604500 (Concentration of Buffer² * Number of cycles) (Fig. 5, Fig. 6, Fig. 7).

Close

Fig. 6

Box-Cox Model Diagnostic Plot for Phycocyanin Purity [The X-axis indicates λ = The optimal power transformation while the Y-axis represents the log of residuals.].

Export to PPT

Close

Fig. 7

Contour plots for factors affecting Phycocyanin Purity [Two factors namely B: Concentration of Buffer (M) and C: Number of freeze–thaw cycles (arbitrary units) are represented in the X axes, Response: Phycocyanin Purity (au) is represented in Y axis.].

Export to PPT

3.2

3.2 FTIR analysis of extracted phycocyanin

The FTIR spectrum of the extracted phycocyanin displayed transmittance maxima at 1649 cm⁻¹, as shown in Fig. 8. This is consistent with the findings of Al-Malki (2020), wherein the maximum transmittance of Phycocyanin was 1652 cm⁻¹.

Close

Fig. 8

FTIR spectrum for phycocyanin extract shows a transmittance maxima at 1649 cm⁻¹ [Graph of Transmittance (%) vs wavenumber (cm⁻¹) shows maximum transmittance at 1649 cm⁻¹. This corresponds to the signal emitted by phycocyanin at 1651 cm⁻¹. ].

Export to PPT

3.3

3.3 SDS PAGE of extracted Phycocyanin

The two distinct protein bands with molecular weights between > 15 and < 25 kD likely correspond to pure Phycocyanin (Fig. 9). Both α and β subunits form the core structure of the protein and are present in equal numbers (Cai et al., 2001).

Close

Fig. 9

SDS-PAGE displaying subunits of Phycocyanin [Lane A: Protein Ladder, Lane B: Extracted and Purified Phycocyanin].

Export to PPT

3.4

3.4 GCMS analysis of Phycocyanin

The GC–MS analysis of Phycocyanin revealed significant peaks at 21.10 min and 26.56 min (Fig. 10). To confirm the presence of phycocyanin, a comparison with a standard and referencing Al-Malki's findings in 2020 was conducted, leading to a conclusive identification of phycocyanin in the extract.

Close

Fig. 10

GC Spectrum of Phycocyanin [Graph of Intensity VS time in mins].

Export to PPT

Additionally, the chromatogram's five peaks were cross-referenced with library databases, and the results have been compiled into a cumulative table for further analysis [Table 4].

Table 4 Result of GCMS.

Retention Time	Hit	Compound name	Molecular Formula	Match Factor (MF)	Reverse Match Factor (RMF)	ID in mainlib database
10.58	1	5H-Cyclopropa(3,4)benz(1,2-e)azulen-5-one, 1,1a-α,1b-β,4,4a,7a-α,7b,8,9,9a-decahydro-7b-α,9-β,9a-α-trihydroxy	C41H66O8	495	663	21221
	2	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-1,1,6,8-tetramethyl-5-oxo-3-[[(1-oxodecyl)oxy]methyl]-9	C41H66O8	458	542	5809
	3	1′,1′-Dicarboethoxy-1β,2β-dihydro-3′H-cycloprop[1,2]cholesta-1,4,6-trien-3-one	C34H50O5	456	698	5215
	4	2β,4a-Epoxymethylphenanthrene-7-methanol, 1,1-dimethyl-2-methoxy-8-(1,3-dithiin-2-ylidene)methyl-1,2,3,4,4a,4b,5,6,7,8,8	C27H38O4S2	453	765	190470
	5	Carda-16,20(22)-dienolide, 3-[(6-deoxy-3,4-O-methylenehexopyranos-2-ulos-1-yl)oxy]-7,8-epoxy-11,14-dihydroxy-12-oxo-, (3.beta.,5.beta.,7.beta.,11.alpha.)-	C30H36O11	451	794	19022
	6	9-Octadecenoic acid (Z)-, 3-[(1-oxohexadecyl)oxy]-2-[(1-oxooctadecyl)oxy]propyl ester	C55H104O6	429	564	191098
	7	3′H-Cycloprop(1,2)-5α-cholest-1-en-3-one, 1′,1′-dicarboethoxy-1β,2β-dihydro-	C34H54O5	426	629	186161
	8	Octadecanoic acid, 1-[[(1-oxohexadecyl)oxy]methyl]-1,2-ethanediyl ester	C55H106O6	425	554	22164
20.04	1	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-3-(hydroxymethyl)-1,1,6,8-tetramethyl-5-oxo-9aH-cyclopropa[3,4]benz[1,2-e]azulene-9,9a-diyl ester, [1aR-(1aalpha,1bbeta,4abeta,7aalpha,7balpha,8alpha,9beta,9aalpha)]-	C50H82O9	375	436	5809
	2	4a-Phorbol 12,13-didecanoate	C40H64O8	328	496	27185
	3	Hexadecanoic acid, 1a,2,5,5a,6,9,10,10a-octahydro-5,5a-dihydroxy-4-(hydroxymethyl)-1,1,7,9-tetramethyl-11-oxo-1H-2,8a-methanocyclopenta[a]cyclopropa[e]cyclodecen-6-yl ester, [1aR-(1aalpha,2alpha,5beta,5abeta,6beta,8aalpha,9alpha,10aalpha)]-	C36H58O6	302	434	169962
	4	Cholestan-3-ol, 5-chloro-6-nitro-, acetate (ester), (3β,5α,6β)-	C29H48ClNO4	301	561	13491
	5	3-Pyridinecarboxylic acid, 2,7,10-tris(acetyloxy)-1,1a,2,3,4,6,7,10,11,11a-decahydro-1,1,3,6,9-pentamethyl −4-oxo-4a,7a-epoxy-5H-cyclopenta[a]cyclopropa[f]cycloundecen-11-yl ester, [1aR-(1aR*,2R*,3S*,4aR*,6S*,7S*,7aS*,8E,10R*,11R*,11aS*) ]-	C32H39NO10	295	447	6247
	6	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-3-(hydroxymethyl)-1,1,6,8-tetramethyl −5-oxo-9aH-cyclopropa[3,4]benz[1,2-e]azulene-9,9a-diyl ester, [1aR-(1aα,1bβ,4aβ,7aα,7bα,8α,9β,9aα)]-	C40H64O8	294	510	6718
	6	Spiro[cyclohexane-1,6′-[2]oxabicyclo[3.2.0]heptan]-3′-one, 4′-cyclohexylidene-7′-(Z)-(4-hexadecanoyloxybenzoylmethylene)-1′- (4-hexadecanoyloxyphenyl)	C63H92O7	289	440	84646
	7	Pentacyclo[19.3.1.1(3,7)0.1(9,13)0.1(15,19)]octacosa-1(25),3,5,7(28),9,11,13(27),15,17,19(26),21,23-dodecaene-25,26,27,28-tetrol, 5,11,17,23-tetrakis (1,1-dimethylethyl)	C44H56O4	284	516	5967
	8	Docosanoic acid, 1,2,3-propanetriyl ester	C69H134O6	274	558	2816
	9	8-Acetoxy-5b-methylperhydrocyclobuta[n]phenanthrene-11-ol-2,3-dione-2a-propanoic acid lactone	C22H28O6	264	624	13412
21.10	1	5H-Cyclopropa(3,4)benz(1,2-e)azulen-5-one, 1,1a-α,1b-β,4,4a,7a-α,7b,8,9,9a-decahydro-7b-α,9-β,9a-α-trihydroxy-3-hydroxymethyl-1,1,6,8-α-tetramethyl-4a-methoxy-, 9,9a-didecanoate	C41H66O8	624	726	21221
	2	4a-Phorbol 12,13-didecanoate	C40H64O8	551	664	27185
	3	Octadecanoic acid, 1-[[(1-oxohexadecyl)oxy]methyl]-1,2-ethanediyl ester	C55H106O6	544	628	22164
	4	Octadecane, 3-ethyl-5-(2-ethylbutyl)-	C26H54	541	753	7340
	5	Octadecanoic acid, 3-[(1-oxohexadecyl)oxy]-2-[(1-oxotetradecyl)oxy]propyl ester	C51H98O6	538	691	21877
	6	Tetracosane, 12-decyl-12-nonyl-	C43H88	521	746	21916
	7	Tetratetracontane	C44H90	524	757	5519
	8	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-3-(hydroxymethyl)-1,1,6,8-tetramethyl-5-oxo-9aH-cyclopropa[3,4]benz[1,2-e]azulene-9,9a-diyl ester, [1aR-(1aα,1bβ,4aβ,7aα,7bα,8α,9β,9aα)]-	C40H64O8	524	663	6718
	9	Heptadecane, 9-hexyl-	C23H48	517	778	7302
	10	Octadecanoic acid, 2-[(1-oxotetradecyl)oxy]-1,3-propanediyl ester	C53H102O6	515	679	21858
26.56	1	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-1,1,6,8-tetramethyl-5-oxo-3-[[(1-oxodecyl)oxy]methyl]-9aH-cyclopropa[3,4]benz[1,2-e]azulene-9,9a-diyl ester, [1aR-(1aα,1bβ,4aβ,7aα,7bα,8α,9β,9aα) ]-	C50H82O9	528	627	5809
	2	4a-Phorbol 12,13-didecanoate	C40H64O8	505	730	27185
	3	Docosanoic acid, 1,2,3-propanetriyl ester	C69H134O6	459	812	2816
	4	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-3-(hydroxymethyl)-1,1,6,8-tetramethyl-5-oxo-9aH-cyclopropa[3,4]benz[1,2-e]azulene-9,9a-diyl ester, [1aR-(1aα,1bβ,4aβ,7aα,7bα,8α,9β,9aα)]-	C40H64O8	458	757	6718
	5	Hexadecanoic acid, 1a,2,5,5a,6,9,10,10a-octahydro-5,5a-dihydroxy-4-(hydroxymethyl)-1,1,7,9-tetramethyl-11-oxo-1H-2,8a-methanocyclopenta[a]cyclopropa[e]cyclodecen-6-yl ester, [1aR-(1aα,2α,5β,5aβ,6β,8aα,9α,10aα)]-	C36H58O6	430	604	169962
	6	Acetic acid, 17-(4-hydroxy-5-methoxy-1,5-dimethylhexyl)-4,4,10,13,14-pentamethyl-2,3,4,5,6,7,10,11,12,13,14,15,16,17-tetradecahydrocyclopenta[a]phenanthryl ester	C33H56O4	429	713	34865
	7	Octadecanoic acid, 2-[(1-oxohexadecyl)oxy]-1-[[(1-oxohexadecyl)oxy]methyl]ethyl ester	C53H102O6	422	608	21985
	8	Octadecanoic acid, 2-[(1-oxotetradecyl)oxy]-1,3-propanediyl ester	C53H102O6	414	645	21858
	9	1,3-Dipalmitin trimethylsilyl ether	C38H76O5Si	403	672	28050
	10	Octadecanoic acid, 2-[(1-oxododecyl)oxy]-1,3-propanediyl ester	C51H98O6	397	613	21974
34.44	1	9-Octadecenoic acid (Z)-, 3-[(1-oxohexadecyl)oxy]-2-[(1-oxooctadecyl)oxy]propyl ester	C55H104O6	462	644	191098
	2	Octadecanoic acid, 1-[[(1-oxohexadecyl)oxy]methyl]-1,2-ethanediyl ester	C55H106O6	453	623	22164
	3	5H-Cyclopropa(3,4)benz(1,2-e)azulen-5-one, 1,1a-α,1b-β,4,4a,7a-α,7b,8,9,9a-decahydro-7b-α,9-β,9a-α-trihydroxy-3-hydroxymethyl-1,1,6,8-α-tetramethyl-4a-methoxy-, 9,9a-didecanoate	C41H66O8	448	647	21221
	4	Decanoic acid, 1,1a,1b,4,4a,5,7a,7b,8,9-decahydro-4a,7b-dihydroxy-1,1,6,8-tetramethyl-5-oxo-3-[[(1-oxodecyl)oxy]methyl]-9aH-cyclopropa[3,4]benz[1,2-e]azulene-9,9a-diyl ester, [1aR-(1aα,1bβ,4aβ,7aα,7bα,8α,9β,9aα)]-	C50H82O9	444	556	5809
	5	2β,4a-Epoxymethylphenanthrene-7-methanol, 1,1-dimethyl-2-methoxy-8-(1,3-dithiin-2-ylidene)methyl-1,2,3,4,4a,4b,5,6,7,8,8a,9-dodecahydro-, acetate	C27H38O4S2	431	792	190470
	6	1′,1′-Dicarboethoxy-1β,2β-dihydro-3′H-cycloprop[1,2]cholesta-1,4,6-trien-3-one	C34H50O5	424	707	5215
	7	3-Phorbinepropanoic acid, 9-acetyl-14-ethyl-13,14-dihydro-21-(methoxycarbonyl)-4,8,13,18-tetramethyl-20-oxo-, 3,7,11,15-tetramethyl-2-hexadecenyl ester	C55H76N4O6	414	566	24447
	8	3′H-Cycloprop(1,2)-5α-cholest-1-en-3-one, 1′,1′-dicarboethoxy-1β,2β-dihydro-	C34H54O5	410	661	186161
	9	Octadecanoic acid, 2-[(1-oxohexadecyl)oxy]-1-[[(1-oxohexadecyl)oxy]methyl]ethyl ester	C53H102O6	389	617	5581
	10	Docosanoic acid, 1,2,3-propanetriyl ester	C69H134O6	388	801	2816

[MF = comparison of the unknown’s mass spectrum’s peaks to those of the peaks in the library’s spectra. RMF = match factor when the peaks in the unknown’s spectrum that are not in the library’s known reference spectrum are ignored. Interpretation: >900 = Excellent, 800–900 = Good, 700–800 = Fair, and < 600 = Poor match.].

3.5

3.5 Antioxidant potential of Phycocyanin

Spirulina is recognized for its potent antioxidant properties, and these attributes are mainly due to the presence of phycobiliproteins within it. Among the phycobiliproteins, phycocyanin is the predominant component in Spirulina (Martelli et al., 2014). In our study, we conducted a series of diverse antioxidant assays to evaluate and quantify the antioxidant potential of Phycocyanin extracted from Spirulina platensis PCC 7345, as illustrated in Fig. 11.

Close

Fig. 11

Antioxidant Potential of Phycocyanin [Results expressed in the percentage of inhibition; mean (n = 3), p < 0.05.].

Export to PPT

From the Graph (Fig. 11), the following data has been derived (Table 5).

Table 5 Determining IC50 value for antioxidant assays.

	DPPH		ABTS		NO
	Phycocyanin	Standard	Phycocyanin	Standard	Phycocyanin	Standard
Equation from Graph	y = 0.7118x + 21.028	y = 0.3942x + 19.811	y = 0.5217x + 37.87	y = 0.6088x + 10.998	y = 0.443x + 18.216	y = 0.4622x + 32.138
R2	0.9667	0.9858	0.9368	0.9596	0.9239	0.9743
IC50 (µg/ml)	40.7024	76.58	23.25091	64.063	71.7471	38.65

[Standard: Ascorbic acid].

The IC50 values represent the concentration of phycocyanin at which it achieves 50% inhibition in various antioxidant assays. Specifically, in the DPPH, ABTS, and NO Radical scavenging assays, the IC50 values for phycocyanin were determined to be 40.70 µg/ml, 23.25 µg/ml, and 17.74 µg/ml, respectively.

3.6

3.6 Enzyme inhibitory assays

3.6.1

3.6.1 Inhibition of α-Amylase enzyme

Both α-Amylase and α-Glucosidase enzymes are pivotal in carbohydrate metabolism, and elevated levels of these enzymes are commonly observed in hyperglycemic patients. Inhibiting these enzymes is a fundamental strategy for preventing glycation and its associated complications, such as diabetes and obesity [Gong et al., 2020; Payan, 2004].

Fig. 12a presents the findings of the α-Amylase inhibitory assay, demonstrating the potent inhibitory effect of phycocyanin on α-amylase across various concentrations. The highest level of inhibition (61.47%) was achieved at a concentration of 100 µg/ml, with an approximate IC50 value of 72.24 µg/ml (Table 5). This underscores the concentration-dependent inhibitory capacity of phycocyanin on α-amylase.

Close

Fig. 12

Enzyme Inhibitory assays (A) α-amylase percent inhibition and (B) α-glucosidase percent inhibition [Results expressed in the percentage of inhibition; mean (n = 3), p < 0.05.].

Export to PPT

3.6.2

3.6.2 Inhibition of α-Glucosidase enzyme

Fig. 12b presents the results of the α-Glucosidase inhibition assay, demonstrating the potent inhibitory effect of phycocyanin on α-Glucosidase across various concentrations. The highest level of inhibition (76.54%) was achieved at a concentration of 100 µg/ml, with an approximate IC50 value of 82.45 µg/ml (Table 6). This underscores the concentration-dependent inhibitory capacity of phycocyanin on α-Glucosidase.

Table 6 Determining IC50 value for enzyme inhibition assays.

	α-Amylase	α-Glucosidase
Equation from Graph	y = 0.6332x + 4.2546	y = 0.6202x − 1.1349
R2	0.917	0.9616
IC50 (µg/ml)	72.24	82.45

4 Discussion

The adoption of a Quality by Design (QbD) framework in this study involved a meticulous review of the literature to define Target Product Profiles (TPPs) and Critical Quality Attributes (CQAs) (Dhoble and Patravale, 2019). The chosen CQAs, Phycocyanin concentration, and purity were selected due to their direct influence on the quality and efficacy of the final Phycocyanin product. The risk assessment phase unveiled factors with notable impacts on Phycocyanin concentration and purity. Ishikawa diagrams (cause and effect diagrams) were instrumental in identifying influential variables (Aoki et al., 2021; Asayama, 2012; Kuddus et al., 2013; Kumar et al., 2014; Seo et al., 2013; Wang et al., 2001). Notably, the concentration of the extraction buffer, the growth phase of the culture, and the number of freeze–thaw cycles emerged as key contributors to these critical attributes. The Box Behnken design, employed for its systematic evaluation of factors, facilitated the exploration of Phycocyanin concentration and purity across a broad spectrum. The results exhibited variations within the matrix of 15 formulations, ranging from 0.0121 to 0.2812 mg/ml for concentration and 0.17 to 0.6978 for purity.

The FTIR analysis corroborated the quality and consistency of the extracted Phycocyanin. Notably, a transmittance maximum at 1649 cm⁻¹ was consistent with previous findings, reinforcing the reliability of the Phycocyanin product. Al-Malki (2020) extracted phycocyanin from Ulva lactuca algae and reported the maximum transmittance of phycocyanin as 1652 cm⁻¹. A similar FTIR spectrum was observed by Husain et al. (2022), where phycocyanin was extracted from Plectonema sp. The FTIR spectrum of the analyzed sample, as depicted in Fig. 8, reveals four prominent peaks, each associated with specific functional groups or molecular vibrations inherent to phycocyanin. The signal observed at approximately 1649 cm⁻¹ is indicative of the amide I band, a characteristic feature of proteins like phycocyanin. This band primarily arises from the stretching vibrations of the carbonyl (C = O) group within the peptide bonds of the protein structure (Sadat and Joye, 2020).

Furthermore, the signal at 1587 cm⁻¹ aligns with the amide II band and/or aromatic rings present in proteins, emphasizing the proteinaceous nature of phycocyanin (Hernández et al., 2010). Symmetric bending vibrations of methyl groups is associated with the signal around 1398 cm⁻¹ (Hannah and Mayo, 2004). Additionally, the FTIR signal at 1115 cm⁻¹ is consistent with C-O stretching vibrations in ethers or esters, providing insights into specific chemical bonds within the phycocyanin molecular structure. This signal may also correspond to C–C stretching vibrations in alkanes, adding to the comprehensive characterization of phycocyanin's chemical composition (Liu and Kim, 2017).

The SDS-PAGE analysis revealed two distinct protein bands with molecular weights between > 15 and < 25 kD, indicative of pure Phycocyanin. This observation aligns with the findings reported by Santiago-Santos et al. (2004), where similar SDS-PAGE bands of Phycocyanin extracted from Calothrix sp. were characterized with molecular weights of approximately 17 kD and 21 kD, respectively. This alignment with established literature further validated the structural integrity of the protein core composed of both α and β subunits.

The GC spectrum displayed noticeable peaks at 21.10 min and 26.56 min. Through meticulous scrutiny and comparison with the standard, and in concordance with the findings of Al-Malki's research in 2020, it was evident that analogous peaks were observed. Notably, Qian and colleagues (2023) reported similar peak observations with commercial phycocyanin. This comprehensive analysis confirmed the presence of phycocyanin in the extract.

An array of diverse antioxidant assays underscored the robust antioxidant potential of Phycocyanin. The calculated IC50 values, representing the concentration at which 50% inhibition occurs in these assays, provided empirical evidence of Phycocyanin's superior antioxidant capacity compared to Spirulina platensis PCC 7345 aqueous extract. Comparing these findings to our prior research on the antioxidant potential of Spirulina platensis PCC 7345 aqueous extract (as documented in Paramanya et al., 2023a), it becomes evident that phycocyanin exhibits significantly higher antioxidant potential than the extract itself. This observation suggests that phycocyanin may play a pivotal role as one of the key factors contributing to Spirulina's overall antioxidant properties. A consistent rise in antioxidant potential in a concentration-dependent manner was observed in both ABTS and DPPH assays, aligning with the findings reported by Wu et al. 2016.

Our findings suggest that phycocyanin has the potential to regulate α-amylase and α-glucosidase activity, a significant enzyme in carbohydrate digestion. Similar concentration-dependent inhibition of these enzymes by phycocyanin has been reported by Prabakaran et al. (2020), where phycocyanin showed a dose-dependent and maximum inhibition effect at 250 μg/ml. This inhibition may have implications for conditions like obesity and diabetes, where carbohydrate metabolism plays a crucial role. Balyan et al. (2023) have reported a direct correlation between the inhibition of these enzymes and the antiglycation potential of their samples. Subsequent phases of our research will focus on elucidating the antiglycation potential of the extracted Phycocyanin.

5 Conclusions

Quality by Design (QbD) serves as a systematic framework for optimizing the phycocyanin extraction process, ensuring heightened control and predictability concerning concentration and purity outcomes. Employing Ishikawa diagrams and Box-Behnken design, the investigation discerned crucial factors influencing phycocyanin extraction, facilitating risk mitigation and comprehensive exploration. Robust analytical techniques confirmed the consistency and structural integrity of the extracted phycocyanin. The discerned antioxidant properties and concentration-dependent inhibition of carbohydrate-metabolism enzymes position phycocyanin as a promising candidate for diverse applications. This research contributes a refined extraction methodology and paves the way for further scientific inquiries into the health, wellness, and industrial utility of phycocyanin.

Funding

Researchers Supporting Project Number (RSPD2023R710) from King Saud University, Riyadh, Saudi Arabia.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

CRediT authorship contribution statement

Additiya Paramanya: Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Visualization. Abeeb Oyesiji Abiodun: Conceptualization, Methodology, Validation, Resources, Writing – review & editing, Funding acquisition. Mohammad Shamsul Ola: Validation, Resources, Writing – review & editing, Funding acquisition. Ahmad Ali: Validation, Resources, Data curation, Visualization, Supervision, Project administration.