Evaluation of Morphological, Chemical, and Antioxidant Characteristics, and Phenolic Profile of Three Goji Berry Varieties Cultivated in Southwestern Spain

Abstract

Agricultural diversification represents an important strategy for promoting sustainability and resilience in rural regions. Goji berries (Lycium barbarum) have emerged as a promising alternative crop due to their high nutritional and functional potential. In this sense, the search for new crops to diversify the production in southwestern Spain is of main interest for farmers to adapt their productions to consumers claims and to climate change, having alternatives to the classical crops (olives, grapes for wine, etc.). This study evaluated several quality related parameters of three goji berry varieties cultivated in the southwest of Spain. Texture profile analysis (TPA) and puncture tests revealed varietal differences in firmness, cohesiveness, and springiness, influenced by genotype and harvesting time. Other morphological and quality parameters such as moisture, total soluble solids, titratable acidity and color were also affected. Significant differences in antioxidant capacity (ABTS and DPPH assays) were found among the varieties and harvesting times, with NQ7 exhibiting the highest values. Phenolic compounds were identified and quantified by LC–HRMS/MS, detecting 33 compounds, with most belonging to the hydroxycinnamic acids, flavonols and flavanones families. NQ7 presented the highest total phenolic content (74.787 mg/100 g DW), with rutin, coumaric acid derivatives, and naringenin as major contributors. The correlation analysis confirmed a strong relationship between total phenolic content and antioxidant capacity. Overall, the results indicated that goji berries grown in southwestern Spain exhibited favorable quality and bioactive profiles, supporting their suitability for sustainable production and commercialization, including further applications as functional food ingredients.

1. Introduction

Goji berries (Lycium barbarum) have been consumed for thousands of years in traditional Chinese medicine and are valued for their reputed properties that boost vitality and good health [1]. They were initially introduced into Europe in their dehydrated form, as the long transport times and short shelf life of fresh goji berries limited their availability [2]. However, concerns were later raised by European health authorities about imported berries from China due to the detection of heavy metals (mainly lead and cadmium) and pesticide residues resulting from intensive agricultural practices in the producing regions near the Himalayas. After that crisis, and largely due to the emergence of small farms in latitudes closer to European consumers, such as those located in southern Italy or Greece, goji berries are now part of our daily diet, having been incorporated as a minority fruit in the rich Mediterranean diet [1,3,4,5,6,7,8].

The main drivers of the goji berry market in the West are their taste, their health benefits, and the search for new healthy products to incorporate into a varied diet. Goji berries are a rich source of fiber and nutrients and are low in calories and fat, fitting perfectly with the Mediterranean diet [9,10,11]. Today, goji berry cultivation is widely distributed in arid and semi-arid regions in China, Japan, Korea, Europe, North America and Mediterranean countries. The main plantations are in Ningxia and other provinces in northwest China [12,13,14]. By 2017, the area under goji berry cultivation had increased 360 times compared to 1949, reaching 66,700 hectares, with a total production volume of dried goji of around 180,000 tonnes (>USD 2.1 billion). In 2022, the production of goji berries in China was 300,000 tonnes. In Europe, interest in goji berry cultivation continues to grow, with small farms being established in southern countries such as Italy [15,16,17], Romania [18], Bulgaria [19], Greece [13,20,21,22] and Portugal [23]. Romania currently has the largest cultivation area within the European Union, while France, Germany, and Spain have joined the list of producers in the last decade [18,24].

In terms of demand, European consumers increasingly seek new, healthy and natural products that can be integrated into the Mediterranean diet [25]. This trend includes the goji berry, the consumption of which has risen steadily in recent years. According to the EIN Presswire—Everyone’s Internet News Presswire, the global goji berry market has demonstrated appreciable growth in recent years, due to the healthier lifestyle choices and well-being of consumers in the 21st century, including an increase in vegan and vegetarian trends, celebrity endorsements and influencer marketing. The market value rose from USD 1.47 billion in 2023 to USD 1.54 billion in 2024, with a compound annual growth rate (CAGR) of 5.0%, and is projected to reach USD 1.89 billion by 2028 [26]. Although China continues to dominate global production, the goji berry market has expanded in North America, Europe, Asia-Pacific, Latin America, the Middle East and Africa. Demand is particularly strong in North America and in several EU countries. Regulatory restrictions imposed by the European Union and North America on the import and labeling of goji berries from China are a major drawback for this market and represent an opportunity to develop this crop in our latitudes. Some of these restrictions arise from the limited reliability of China’s organic certification system, which allows the use of certain pesticides and fungicides not authorized in Europe.

In this sense, the current projections indicate an annual growth rate of about 10% between 2024 and 2032 [27]. The global market is expected to reach nearly USD 3 billion by 2032, driven by consumer preferences for healthier lifestyles and the integration of goji berries into multiple sectors, including food, dietary supplements, and cosmetics. The Asia–Pacific region remains the main consumption area, although substantial growth is expected in the North American and European markets, supported by the increasing popularity of “superfoods” and sustainable product labeling trends [1]. Currently, the goji berry market is segmented by application and product type. On the one hand, goji berries are used in food and beverage production, while on the other, they are increasingly incorporated into pharmaceutical, personal care and cosmetic formulations. In the food sector, goji berries are consumed fresh, as juices, or in processed forms such as dehydrated, freeze-dried, powdered, and puréed, or as jams. This pattern reflects a broader European trend toward the consumption of exotic, healthy, natural and locally produced fruits [5,28]. In this sense, it is important to check classical quality parameters (such as texture, color, acidity, morphology, etc.) for fruits to be consumed fresh and how the adaptation to different locations and climates affect them. In this regard, although some researchers have evaluated different quality parameters in fresh goji berries, this information remains scarce because the fruit is highly perishable and its commercialization has mainly focused on the dried form [29,30].

Numerous studies have been carried out on the phenolic composition and antioxidant properties of goji berries. Mocan et al. [18] evaluated four L. barbarum varieties grown in Italy and Romania, showing significant differences in the content of phenolic compounds and flavonoids among varieties and harvesting periods. Similarly, Zhang et al. [29] characterized eight Chinese goji varieties, observing that chlorogenic acid and quercetin derivatives were the predominant phenolic compounds, while total antioxidant capacity was strongly correlated with total phenolic content. More recently, Chen et al. [31] performed a UPLC–MS/MS analysis of four goji varieties at different ripening stages, observing eleven representative phenolic compounds, mainly rutin, isoquercitrin, and chlorogenic acid, the concentrations of which decreased during fruit ripening. Moreover, Fatchurrahman et al. [32] examined the effect of different storage temperatures on the postharvest quality of goji fruits and demonstrated that storage conditions strongly influence the retention of bioactive compounds and their antioxidant capacity. Beyond compositional analyses, numerous in vivo and ex vivo studies have demonstrated that goji berry consumption exerts beneficial effects on human health, including the modulation of lipid profile through increased HDL cholesterol, improvements in glucose metabolism and reductions in fasting glucose levels, as well as the attenuation of oxidative stress, enhancements in quality of life, and anti-fatigue effects [33,34,35,36,37].

In terms of production, the search for new crops to diversify and increase the profitability of farms in the southwest of Spain (Extremadura and Andalusia) has led to research into how this fruit adapts to this region. It is important to assess the viability of this crop as a complementary crop to those already existing in the territory in order for it to contribute to the diversification of farms (environmental and agricultural improvement), and thus to increase their profitability. These measures aim to improve the quality of life in rural areas and contribute to the demographic challenge. In addition, our study highlighted the impact of the cultivation practices (organic) on the phenolic profile of goji berries, an underexplored area, and how the dry Mediterranean climate in this region influences the quality and phytochemical composition of different goji berry varieties. This study aims to evaluate the different morphological and quality parameters of fruits from three different varieties of goji berries cultivated in the southwestern part of Spain. To this end, the phenolic compound profile and antioxidant activity of organically grown goji berry samples were evaluated to verify the effect of variety and harvesting time on phytochemical composition.

2. Materials and Methods

2.1. Chemicals

HPLC-grade distilled water, HPLC-grade methanol, HPLC-grade acetonitrile, gallic acid (3, 4, 5-trihydroxybenzoic acid) and potassium hydroxide were sourced from Panreac Applichem ITW Reagents (Darmstadt, Germany). Sodium hydrogen carbonate was acquired from VWR International Eurolab (Barcelona, Spain). HPLC-grade Formic acid was purchased from Fisher Scientific (Madrid, Spain). 2,2′-azinobis-(3-ethylbenzothiazoline-6-sulphonic acid) diammonium salt (ABTS), 2,2-diphenyl-1-picrylhydrazyl (DPPH), 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) and phenolic compound standards were supplied by Sigma-Aldrich (Steinheim, Germany).

2.2. Materials and Sample Collection

Three different varieties of goji berries (L. barbarum) (NQ1, NQ7 and V3) grown under organic systems were collected in 2023 from a private cultivar located in the Jerte Valley (Extremadura, Spain). The plants were 3 years old (2021) and planted in an open field in an experimental block design, with the crop rows in a north–south orientation; we used three replications and five plants of each variety on each replication, with an interval of 2 × 1.5 m in terms of distance. Six-month-old to one-year-old goji berry plants were obtained from commercial nurseries and were planted using an anti-grass mesh with ridges. The distribution of varieties followed the same pattern, with five replicates on each. Drip irrigation lines with a flow rate of 2 L/hour were used.

At the time of planting, goji berry plants were treated with an organic humic amendment containing 37.5% organic matter (“Orgamax E”), using a dose of 600 kg/ha. A second application of the same organic amendment at 37.5% was made in winter 2023, at a dose of 200 kg/ha, with the aim to improve soil properties (infiltration, organic matter, biological activity, slow release of nutrients). In July 2023, a foliar fertilizer (seaweed extract “Bioalgax”—Ascophyllum nodosum) was applied, adding nutrients and biostimulants to improve plant growth, fruit set and fruit enlargement.

During the planting period, no pesticide treatments were carried out to combat pests or diseases (no fungicides, insecticides or acaricides), following an organic production model, in accordance with Regulation (EU) 2018/848 of the European Parliament and of the Council of 30 May 2018 on organic production and labeling of organic products. During the spring of 2023, organic fungicide treatments (wet sulfur) were applied to control fungal diseases (powdery mildew—Oidium).

During the studied period (2021–2024), the average annual temperature was 16.42 °C, with cold, long winters and the warmest months being July and August, reaching a maximum temperature of 41.92 °C in July 2022. Minimum temperatures were recorded in winter, with averages of 11.07 °C and extremes down to −2.74 °C and −2.21 °C in January 2021 and 2023. The average annual rainfall was 1093 mm, though 31 out of 48 months had less than 100 mm/month, with 5 months being completely dry, and the average humidity was 58.08%, peaking in winter and being the lowest in July–August. Actual evapotranspiration (ETo) averaged 1158.92 mm/year, with the highest values in summer (180–199 mm) and the lowest in winter (25–39 mm). Solar radiation averaged 16.14 MJ/m², with the maximum values in April–August (up to 32 MJ/m²) and minimum values in October–January (0.42–0.70 MJ/m²). Prevailing winds were in the NE and SW directions, with average speeds of 1.29 m/s and occasional maxima up to 18 m/s.

The fruits were harvested at commercial ripeness, determined by external color and morphological characteristics. The fruits were collected at four different harvesting times, the first half of October (T1), the second half of October (T2), the first half of November (T3) and the second half of November (T4), giving a total of 12 samples (4 samples per variety during the harvesting time). Fruits were selected randomly by variety, picking them from different plants of each variety to obtain a representative sample of the cultivar.

After harvesting, fruits were randomly distributed, stored under refrigeration (5 ± 2 °C), and transported to the laboratory within 24 h using PET punnets with a lid.

The physicochemical parameters (weight, width and length, moisture content, titratable acidity, pH, TSS, color and texture) of all the samples were analyzed at the time of harvesting, replicating each measurement twice for the weight, three times for pH and acidity and texture, and five for the rest of the parameters. Subsequently, representative subsamples amounting to 50 g were lyophilized and frozen at −80 °C for the following analysis.

2.3. Morphological Measurements

Measurements of three morphometric variables were conducted to evaluate fruit characteristics. The width and length of 20 goji berries were measured using a digital caliper with a precision of 0.01 mm (Comecta SA, Barcelona, Spain). Additionally, 20 fruit samples from each variety were weighed with a precision balance with a sensitivity of 0.01 g (Nimbus, Adam Equipment, Oxford, CT, USA).

2.4. Moisture Content

One gram of each sample was oven-dried at 103 ± 2 °C for 2 h until a constant mass was achieved, following AOAC international methods [38]. The moisture content of the goji berries was determined in triplicate.

2.5. Titratable Acidity (TA), pH, and Total Soluble Solids (TSSs)

Ten grams of goji berries were weighed and homogenized with 15 mL of distilled water in an ultraturrax for 2 min to form a slurry for further analysis. Titratable acidity (TA) was determined following the method described by Sánchez-Parra et al. [39] using a Mettler Toledo T70 automatic titrator (Mettler Toledo, Greifensee, Switzerland). The pH was measured with a combined pH glass electrode (DGi111-SC Mettler Toledo, Greifensee, Switzerland). Total soluble solids (°Brix) were measured using an Atago PAL-1 electronic handheld refractometer (Atago CO Ltd., Tokio, Japan). The device was calibrated with distilled water before each reading.

2.6. Color

The surface color of the skin of the goji berry was measured using a portable Konica Minolta CM–700D instrument (Minolta Corporation Ltd., Osaka, Japan). Measurements were taken at three consecutive equatorial points on the berry’s surface, with D65 as an illuminant and a 10° observer angle. The color was based on the CIELAB space, where L* represents lightness (L* = 0 for black, L* = 100 for white), a* measures the red–green intensity (a* < 0 for green, a* > 0 for red), and b* denotes the blue–yellow intensity (b* > 0 for yellow, b* < 0 for blue). The device was calibrated with a standard white reference before each measurement. Additionally, chroma (C*) and hue angle (h°) were calculated to evaluate the color saturation and the extent of red and yellow hues, respectively.

C* = (a² + b²)^1/2

(1)

h° = arctan (b*/a*)

(2)

2.7. Measurements for the Evaluation of Texture Properties

Three different probes were employed for assays, including the following: (a) the puncture test; (b) texture profile analysis (TPA); and (c) the miniature Kramer Ottawa cell. Texture measurements of samples were conducted using a TA-XT Plus Texture Analyzer (Stable Micro Systems Ltd., Godalming, UK) equipped with a 5 kg load cell for (a) and (b) and a 50 kg load cell for (c). These methods were selected for their effectiveness in capturing different texture attributes and their successful implementation in berry-like fruit crops [40,41]. Exponent software 5.1.1.1 (Stable Micro Systems Ltd., Godalming, UK) was used to acquire and integrate the data. A total of seven textural characteristics were determined, as detailed elsewhere [42].

For the puncture test, a needle probe (P/2N needle probe, Stable Micro Systems Ltd., Godalming, UK) was used to penetrate the goji berry at a speed of 1 mm/s up to a depth of 6 mm. Firmness was calculated as the average of the maximum force recorded in Newtons. TPA was carried out on the goji berry samples using a 5 mm cylindrical probe (P/5 compression platen probe, Stable Micro Systems Ltd., Godalming, UK). The testing parameters included a 1 mm/s test speed, 10 mm/s post-test speed and a 35 mm test distance. Finally, the miniature Kramer Ottawa cell test was performed with a five-blade Kramer chamber, where five goji berries (1 g approximately) were placed in a single layer at the bottom of the chamber, perpendicular to the blades, and subjected to extrusion at a speed of 10 mm/s. All the tests were conducted in triplicate under controlled-temperature conditions (23 ± 2 °C).

2.8. Phenolic Compounds Extraction

The extraction of phenolic compounds was performed using a mixed solution of methanol:deionized water (80:20, v/v) acidified with 1% formic acid. Thus, 0.2 g of lyophilized sample was extracted with 1 mL of the extraction solvent, and then vortexed for 10 s, sonicated for 10 min and centrifuged at 15,000 rpm for 15 min at 4 °C. The supernatant was collected, and the pellet was re-extracted using the same protocol. The supernatants collected during the extractions were combined into a single sample. The samples were transferred into vials and stored at −80 °C until their analysis.

2.9. Antioxidant Activity

The antioxidant activity (ABTS and DPPH) and total phenolic content were measured using a Synergy HTX Multi-Mode Microplate Reader (Biotek Instruments, Winooski, VT, USA).

2.9.1. ABTS Assay

Free radical scavenging activity was assessed using the ABTS (2,2-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)) discoloration method [43]. Antioxidant activity was expressed as mg of Trolox equivalents per 100 g of dry sample (mg TE/100 g DW). Each value represents the average of three measurements.

2.9.2. DPPH Assay

The DPPH (1,1-diphenyl-2-picryl-hydrazyl) free radical scavenging test was conducted using the method previously described by Ordóñez-Díaz et al. [44]. The antioxidant activity was expressed as mg of Trolox equivalents per 100 g of dry sample (mg TE/g DW). Each value is the average of three measurements.

2.10. Phenolic Compounds Profile

The identification and quantification of the phenolic compounds in the goji berry samples were performed using a micro-flow LC-HRMS/MS mass spectrometer system (Thermo Fischer Scientific, San José, CA, USA) comprising a Vanquish Neo UHPLC system coupled to an Orbitrap Exploris 240 mass spectrometer. Chromatographic separation was performed using a system of Acclaim™ PepMap™ 100 C18 columns (150 mm × 1 mm, 2 μm) (Thermo Fischer Scientific, San José, CA, USA). Mobile phase A was water and mobile phase B was a mixture of acetonitrile and water (75/25, v/v), both acidified with 0.1% of formic acid. The gradient time was 20 min (0–2 min, 0% B; 2–16 min, 0–95% B; 16–20 min, 95% B), with a flow rate of 50 μL/min, and the sample injection volume was 1 μL.

Exploris 240 MS was operated in full-scan, negative-ion mode (m/z 100–1000 Da), with MS/MS being used in tandem, via the data-dependent acquisition method. The resolution was fixed at 60,000 and 15,000 for full-scan and data-dependent mass scanning, respectively. The parameters of Heated-Electrospray Ionization (HESI) were as follows: spray voltage: 2500 V; sheath gas: 35 units; auxiliary gas: 5 units; ion transfer tube temperature: 325 °C; vaporizer temperature: 120 °C; and RF lens: 70%.

The quality control (QC) samples were applied to assess the analytical process and consisted of a pool of goji berries, which were injected regularly throughout the run. Data acquisition and analysis were performed using Trace Finder 5.1 and Xcalibur 4.6 software (Thermo Scientific, San José, CA, USA). Phenolic compounds (Table S1) were identified by comparing the exact mass and the retention time with available standards. When the standards were not available, the phenolic compounds were tentatively identified by comparing the theoretical exact mass and experimental accurate mass of the molecular ion. These were compared against chemical compound databases, such as MzCloud, Phytohub, Phenol Explorer, PubChem, etc., with an error tolerance ≤ 5 ppm.

The phenolic compounds were quantified using standard curves of reference standards, with a linearity ranging from 0.05 to 50 mg/L and a coefficient of determination higher than 0.99 for all standard curves. The limit of detection and limit of quantification ranged from 0.048 to 0.700 mg/L and from 0.16 to 2.333 mg/L, respectively. In their absence, quantification was carried out using a closely related parent compound. The precision of the analytical method was assessed using a quality control as part of routine analysis, with a relative standard deviation (RSD) below 10%.

2.11. Statistical Analysis

Univariate statistical analyses were conducted using the free software R (v.4.3.3; R Core Team, Vienna, Austria). A two-way analysis of variance (ANOVA) was performed, followed by Tukey’s post hoc test for mean comparisons. The significance level was established at p < 0.05. Additionally, Pearson’s correlation test was performed, and a correlation plot was generated using OriginPro 2025 software (trial version; OriginLab Corporation, Northampton, MA, USA). Finally, MetaboAnalyst 6.0 (https://www.metaboanalyst.ca, accessed on 29 October 2025) software was utilized to conduct an ANOVA–Simultaneous Component Analysis (ASCA) to evaluate the individual and combined effects of the experimental factors (Variety and Harvesting Time). ASCA combines the principles of analysis of variance (ANOVA) and principal component analysis (PCA), allowing the decomposition of total variability into components associated with each factor and their interaction. A permutation test (100 iterations) was performed to validate the ASCA results and prevent overfitting [45]. Furthermore, a heatmap was generated to visualize the normalized abundance patterns across varieties and phenolic compound classes.

3. Results and Discussion

3.1. Morphological Parameters

The analysis of different morphological parameters was conducted by evaluating the different varieties and harvesting times. In general, we observed a high influence of the genotype on the morphological characteristics of the fruits.

Table 1. Morphological, physicochemical, color and antioxidant analyses of goji berries.

	Varieties				Harvesting Time
	NQ1	NQ7	V3	p-Value	T1	T2	T3	T4	p-Value
Weight (g/20 berries)	6.69 ± 0.72 b	9.79 ± 2.03 a	9.35 ± 4.92 a	***	7.24 ± 1.58 b	7.98 ± 3.53 b	8.05 ± 1.22 b	11.17 ± 4.69 a	***
Width (mm)	7.59 ± 0.96 b	9.25 ± 1.15 a	7.58 ± 1.60 b	***	7.82 ± 1.27 b	7.95 ± 2.06 ab	8.07 ± 0.97 ab	8.71 ± 1.36 a	*
Length (mm)	13.32 ± 1.55 b	15.01 ± 2.64 a	14.59 ± 3.10 ab	**	11.87 ± 1.73 b	14.73 ± 2.42 a	15.29 ± 1.88 a	15.34 ± 2.61 a	***
Moisture (%)	81.54 ± 2.53 a	79.06 ± 2.60 b	80.08 ± 2.60 ab	*	79.14 ± 1.65	79.10 ± 2.76	81.75 ± 2.06	80.91 ± 3.47	ns
TTA 1 (% citric acid)	1.10 ± 0.06 b	0.94 ± 0.12 c	1.17 ± 0.29 a	***	1.21 ± 0.35 a	0.96 ± 0.12 c	1.07 ± 0.06 b	1.04 ± 0.10 b	***
pH	5.28 ± 0.37 b	5.19 ± 0.55 c	5.56 ± 0.83 a	***	5.37 ± 0.13 b	5.22 ± 0.11 d	5.48 ± 0.62 a	5.31 ± 0.50 c	***
TSS 2 (°Brix)	15.41 ± 5.28 ab	16.65 ± 3.56 a	13.54 ± 2.74 b	**	18.29 ± 4.77 a	16.21 ± 1.70 ab	12.82 ± 3.17 c	13.47 ± 3.99 bc	***
L*	41.97 ± 3.58 b	46.30 ± 2.74 a	42.41 ± 3.36 b	***	43.57 ± 2.49 ab	42.79 ± 3.78 b	46.02 ± 3.42 a	41.85 ± 4.10 b	**
a*	37.37 ± 4.65 b	41.26 ± 2.57 a	35.52 ± 3.21 b	***	38.52 ± 3.50	38.35 ± 3.54	38.96 ± 4.16	36.37 ± 5.52	ns
b*	26.89 ± 5.81 b	35.30 ± 3.29 a	28.17 ± 4.50 b	***	29.70 ± 4.10 b	28.96 ± 5.07 b	33.64 ± 5.87 a	28.18 ± 7.13 b	**
C	46.11 ± 7.01 b	54.34 ± 3.64 a	45.45 ± 4.35 b	***	48.68 ± 5.03 ab	48.14 ± 5.48 ab	51.63 ± 6.06 a	46.10 ± 8.48 b	*
h°	35.30 ± 3.44 c	40.52 ± 2.16 a	38.26 ± 4.27 b	***	37.53 ± 2.31 b	36.83 ± 3.40 b	40.59 ± 4.53 a	37.15 ± 4.42 b	***
ABTS 3	1.01 ± 0.10 b	1.19 ± 0.09 a	0.94 ± 0.10 c	***	1.00 ± 0.12 b	1.09 ± 0.12 a	1.16 ± 0.12 a	0.93 ± 0.12 b	***
DPPH 3	0.88 ± 0.05 b	1.22 ± 0.12 a	0.82 ± 0.04 c	***	0.97 ± 0.20 b	1.01 ± 0.20 ab	1.03 ± 0.24 a	0.88 ± 0.13 c	***

Data are expressed as mean values ± SD. ¹ Total titratable acidity. ² Total soluble solids. ³ Measurements expressed as mmol/100 g DW. Different letters in the same row indicate significant differences according to Tukey’s test. *, **, ***, ns: significant at t p < 0.05, p < 0.01, and p < 0.001, and not significant, respectively.

shows significant differences between the varieties under study, with the NQ1 berries having the lowest weight, followed by NQ7, and the V3 varieties having the highest average weight. In this sense, we can compare the lower value observed for NQ1 (6.69 g/20 berries FW) under the cultivation conditions of our study with other data found in the literature by Liu et al. [46] (11.30–13.87 g/20 berries FW), in which the average weight was almost double. These differences may be due to different factors affecting both experiments, such as agronomic practices and environment. Moreover, there was a considerable age difference between the two plantations (3 vs. 15 years) that could have greatly affected the factor under study. Regarding the harvesting time, a significant increase in berry weight was observed in T4 compared to the earlier samplings. In terms of berry dimensions, NQ7 exhibited higher values for both length and width compared to NQ1. In this sense, the length and diameter of the fruits reported for the NQ1 variety cultivated in the northwest of Poland (Width: 9.6 mm, Length: 14.2 mm) were slightly higher than those in the southwest of Spain (Width: 7.59 mm, Length: 13.3 mm) [47]. Regarding the harvesting time, it was observed that the goji berry fruits from T1 presented the lowest values for length. This parameter increased with the harvesting time, with values being similar for T2, T3, and T4. A slightly different trend was observed for the width of the fruits, where a gradual increase could be observed, the maximum value being reached for the berries harvested at T4.

3.2. Moisture Content, Total Soluble Solids (TSSs), Titratable Acidity (TA) and pH

The moisture content ranged from 79.06 ± 2.60% to 81.54 ± 2.53%, with NQ1 showing the highest value (Table 1). These results align with those of previous studies [29]. An inverse correlation was observed between moisture and berry size in the NQ7 and NQ1 varieties. In this sense, the samples were grown under the same agronomic conditions, so both moisture and berry size must have a high genetic component. The lowest values for this parameter were found at the beginning of the harvesting times (T1 and T2).

The total soluble solids (TSSs) ranged from 13.54 ± 2.74 to 16.65 ± 3.56 °Brix, with NQ7 having the highest values. These concentrations are in agreement with data previously reported in the literature (15–17.88) [29,46]. On the other hand, some differences were observed in the TSS content of the NQ1 variety grown in northwest Poland, which presented 11.4 °Brix, compared to 15.41 °Brix in the southwest of Spain, this possibly being due to the higher value of solar radiation incidence [47]. The highest values found for TSSs were observed at the beginning of the season (T1).

The values found for the titratable acidity of the goji berries oscillated from 0.94 ± 0.12 to 1.17 ± 0.29%, this being also in agreement with the values observed in the literature by Zhang et al. [29]. The NQ7 variety showed the lowest values for this parameter (0.94 ± 0.12%) and for the pH (5.19 ± 0.55).

3.3. Color

In general, the color of berries is a key quality attribute influencing consumer perception and acceptability. Several studies indicated that fruit color affects purchasing decisions, as consumers often associate vibrant colors with freshness, ripeness, and higher nutritional content [48,49]. The characteristic red color of goji berries depends on the contribution and proportion of pigment compounds such as anthocyanins and/or carotenoids [29].

In the present study, significant differences were observed for all the color parameters among the three goji berry varieties under study, highlighting the potential influence of genetic factors on pigment composition. Among the varieties, NQ7 showed the highest luminosity (L* = 46.30 ± 2.74), making it appear lighter compared to NQ1 and V3 (L* = 42). Additionally, the a* parameter, which quantifies redness, was highest in NQ7 (a* = 41.26 ± 2.57), suggesting that NQ7 has a more intensely red coloration. The b* values, which represent the yellow–blue spectrum, also varied significantly, with NQ7 showing the highest value (b* = 35.30 ± 3.29). These results indicated that NQ7 berries not only appear redder but also exhibit a greater yellow component, likely contributing to a visually appealing orange–red hue (Table 1). Moreover, chroma (C*) and hue angle (h°) parameters further reinforce these findings. The chroma value, which measures color saturation, was the highest for NQ7, suggesting a more intense but slightly duller color compared to the other varieties. The hue angle (h°), a key indicator of color tone, positioned NQ7 closer to the yellow spectrum, while NQ1 (h° = 35.30 ± 3.44) leaned toward orange. Our findings are consistent with those of Jatoi et al. [30].

For the different harvesting times (T1, T2, T3, and T4), significant differences were also observed for all the color parameters except a*. L* exhibited a significant variation, with T3 showing a higher value (L* = 46.02 ± 3.42) than T2 and T4, indicating that the berries at T3 were lighter. The a* component did not present significant differences, suggesting that the red intensity remained stable throughout the harvesting period. However, the yellow–blue component (b*) displayed significant differences, with T3 having the highest value (b* = 33.64 ± 5.87), indicating a more yellow color at this stage. Additionally, C and h° also showed a significant difference, with T3 exhibiting the highest values (C = 51.63 ± 6.06; h° = 40.59 ± 4.53). These results suggest that harvesting time significantly influences the color characteristics of the berries, with T3 exhibiting more vibrant and distinct colors.

3.4. Texture Analyses

The results of the texture properties of the NQ1, NQ7, and V3 goji berry varieties are shown in

Table 2. Textural parameters of goji berries.

	Varieties				Harvesting Time
	NQ1	NQ7	V3	p-Value	T1	T2	T3	T4	p-Value
Texture analysis profile (TPA)
Hardness (N)	3.05 ± 0.95 a	2.06 ± 0.83 b	2.68 ± 0.63 ab	**	2.51 ± 0.89	2.91 ± 1.08	2.22 ± 0.81	2.74 ± 0.76	ns
Springiness (N/mm)	0.72 ± 0.16 b	0.79 ± 0.05 a	0.75 ± 0.08 ab	*	0.78 ± 0.08 ab	0.70 ± 0.13 b	0.72 ± 0.07 b	0.82 ± 0.12 a	**
Cohesiveness (N/mm)	0.35 ± 0.09 b	0.42 ± 0.07 ab	0.44 ± 0.10 a	**	0.48 ± 0.12 a	0.36 ± 0.08 b	0.41 ± 0.05 ab	0.36 ± 0.08 b	**
Chewiness (N)	0.71 ± 0.14 b	0.66 ± 0.23 b	0.89 ± 0.37 a	*	0.92 ± 0.43	0.68 ± 0.16	0.64 ± 0.14	0.77 ± 0.21	ns
2 mm sensor
Fracturability (N)	2.13 ± 0.41 a	1.62 ± 0.42 b	2.25 ± 0.91 a	**	1.58 ± 0.39 b	2.65 ± 0.93 a	1.80 ± 0.23 b	1.97 ± 0.40 b	***
Mini Kramer/Ottawa sensor
Maximum force (N)	19.66 ± 9.05 b	26.95 ± 5.96 a	20.42 ± 9.92 ab	*	18.73 ± 9.40 b	29.98 ± 6.11 a	20.39 ± 10.28 b	20.26 ± 4.89 b	**
Shear force (N/g of the sample)	140.20 ± 57.30 c	170.36 ± 26.77 b	207.51 ± 40.46 a	***	164.03 ± 28.38 b	205.51 ± 24.86 a	165.12 ± 53.39 b	156.09 ± 72.22 b	**

Data are expressed as mean values ± SD. Different letters in the same row indicate significant differences by Tukey’s test. *, **, ***, ns: significant at t p < 0.05, p < 0.01, and p < 0.001, and not significant, respectively.

. Several firmness parameters could be calculated via texture profile analysis (TPA), including hardness, cohesiveness, and chewiness, which are key descriptive metrics with which to quantify the energy required to cut and break a semi-solid food until it is ready to be swallowed. Additionally, the springiness was also calculated, defined as the rate at which a deformed sample regains its shape, and is also considered a key factor [50].

The hardness value showed statistically significant differences among the goji berry varieties, NQ1 being harder (3.05 ± 0.95 N) than NQ7 (2.06 ± 0.83 N). Regarding springiness, significant differences were observed among the varieties, the NQ7 berries presenting a higher value (0.79 N/mm) than those obtained from NQ1 (0.72 ± 0.16 N/mm), indicating a higher ability to recover their shape after deformation. On the other hand, cohesiveness showed significant variability, with the berry samples from the V3 variety showing a higher value (0.44 ± 0.10 N/mm) than the samples from NQ1 (0.35 ± 0.09 N/mm), suggesting higher internal structural integrity. Regarding chewiness, the highest values were observed for V3 (0.89 ± 0.37 N).

Regarding the harvesting time (T1, T2, T3, and T4), distinct patterns in the textural parameters of the berry fruit samples were observed, reflecting the effect of harvesting time on their mechanical properties. Statistically significant differences were found for springiness and cohesiveness but not for hardness and chewiness. Regarding springiness, the samples obtained at T4 showed the highest value (0.82 ± 0.12 N/mm, p < 0.01), indicating that the samples exhibited the best elasticity. On the other hand, for cohesiveness, the berry fruit samples obtained at the beginning of the season (T1) showed higher values (0.48 ± 0.12 N/mm, p < 0.01) than the samples harvested at T2 and T4, suggesting a stronger internal structure (Table 2).

The puncture test, linked to fracturability, showed significant differences among the goji berry varieties. Specifically, V3 and NQ1 showed the highest values (2.25 ± 0.91 N and 2.13 ± 0.41 N, respectively), suggesting that a higher force for penetration is needed compared to NQ7 (1.62 ± 0.42 N). Puncture tests have been widely reported in the literature to measure the firmness of horticultural products [51,52]. The goji berry skin break force, measured by the puncture test, has been recommended as a maturity indicator in some studies [53].

On the other hand, the harvesting time influenced the values obtained for this parameter, the samples harvested at T2 presenting the highest value (2.65 ± 0.93 N), indicating that the force required to penetrate the skin was the highest at this stage.

Regarding the measurement of the maximum force using the miniature Kramer Ottawa cell, the results showed significant differences. The force applied to the NQ7 samples (26.95 ± 5.96 N) was higher than that applied to NQ1 and V3 (19.66 ± 9.05 and 20.42 ± 9.92 N, respectively), constituting evidence of higher overall resistance in these samples. Concerning the shear force, V3 showed the highest value (207.51 ± 40.46 N/g, p < 0.001), followed by NQ7 (170.36 ± 26.77 N/g) and NQ1 (140.20 ± 57.30 N/g). This suggests that V3 requires more energy to cut, this likely being related to parameters such as cohesiveness and hardness (Table 2).

Furthermore, the harvesting time also influenced those parameters, the maximum force required to compress the sample being found at T2 (29.98 ± 6.11 N). In addition, the shear force reached its maximum value at this stage (205.51 ± 24.86 N/g), indicating that T2 samples showed greater firmness and required the most energy for their deformation.

The parameters used in this study can be considered suitable for objectively monitoring goji berry texture to identify cultivar differences and predict their behavior during their harvesting time. In this sense, firmness or hardness is much more sensitive than a simple peak force metric, as demonstrated in the study by Giongo et al. [54], where the textural effect of external (skin) and internal (mesocarp and cavity) tissues was defined by the different parameters set in the various analyses performed on raspberries. In summary, the NQ1 samples were characterized by greater hardness and fracturability, while the NQ7 samples had higher values for springiness and maximum force parameters. Meanwhile, V3 presented higher values for cohesiveness, chewiness, fracturability and shear force. Regarding harvesting time, the results showed a significant impact on firmness parameters, including springiness, cohesiveness, fracturability, maximum force and shear force. It is of note that the first sampling times, especially T2, showed higher values for fracturability, maximum force and shear force, while the later sampling time (T4) showed higher springiness. These results emphasize the importance of studying both variety and harvesting time, and their influence on the mechanical and sensory properties of goji berries, which could have implications for their processing and consumer acceptance.

3.5. Antioxidant Capacity

The antioxidant capacity of the samples was evaluated using two different assays, ABTS and DPPH, which are widely used in the literature to evaluate free radical scavenging activity. The results of both assays showed significant differences among the varieties, with the berry samples obtained from the NQ7 variety presenting the highest values (Table 1). Moreover, significant differences among different harvesting times were observed, suggesting that external factors play an important in modulating the antioxidant response.

In the case of the ABTS assay, the highest antioxidant capacity values (1.09 ± 0.12 and 1.16 ± 0.12 mmol TE/100 g DW, respectively) were observed for the berry fruit samples harvested at T2 and T3. Similarly, the DPPH assay indicated that the samples harvested at T3 had the highest antioxidant capacity values (1.03 ± 0.24 mmol TE/100 g DW), followed by T2 (1.01 ± 0.20 mmol TE/100 g DW), while T4 presented the lowest values (0.88 ± 0.13 mmol/100 g DW). In this context, goji berries exhibit high antioxidant capacity values, comparable to or even surpassing those of fruits typically considered rich in antioxidants, such as strawberries and blueberries [55].

3.6. Total Polyphenols and Phenolic Compound Profile

A total of 33 phenolic compounds were tentatively identified and quantified in the goji berries samples (

Table 3. Concentration (mg/100 g DW) of phenolic compounds identified in goji berry samples of different varieties and harvesting times.

	Varieties				Harvesting Time
	NQ1	NQ7	V3	p-Value	T1	T2	T3	T4	p-Value
Caffeic acid	0.014 ± 0.013 b	n.d.	0.020 ± 0.018 a	***	0.006 ± 0.005 c	0.024 ± 0.018 a	0.015 ± 0.017 b	0.001 ± 0.001 d	***
Caffeoylglucoside	0.410 ± 0.064 c	0.625 ± 0.060 a	0.473 ± 0.077 b	***	0.486 ± 0.078 bc	0.555 ± 0.109 a	0.542 ± 0.127 ab	0.429 ± 0.109 c	***
Caffeoylquinic acid	0.269 ± 0.220 a	0.053 ± 0.022 b	0.286 ± 0.262 a	***	0.178 ± 0.226 bc	0.398 ± 0.255 a	0.182 ± 0.169 b	0.053 ± 0.020 c	***
4,5-Dicaffeoylquinic acid	0.002 ± 0.002 b	0.003 ± 0.003 a	0.002 ± 0.002 c	***	0.002 ± 0.002 bc	0.005 ± 0.002 a	0.001 ± 0.001 c	0.001 ± 0.001 c	***
Chlorogenic acid	0.455 ± 0.340 b	0.079 ± 0.047 c	0.601 ± 0.575 a	***	0.257 ± 0.320 bc	0.811 ± 0.582 a	0.319 ± 0.302 b	0.126 ± 0.057 c	***
Ferulic acid	0.293 ± 0.319 a	0.335 ± 0.251 a	0.211 ± 0.222 b	***	0.020 ± 0.006 c	0.081 ± 0.112 c	0.403 ± 0.086 b	0.614 ± 0.104 a	***
Feruloylglucose	5.221 ± 1.220 a	3.727 ± 0.882 b	5.147 ± 1.676 a	***	6.607 ± 1.335 a	4.492 ± 0.647 b	3.820 ± 0.793 b	3.875 ± 0.619 b	***
Dihydroferulic acid 4-glucuronide	0.056 ± 0.033 b	3.069 ± 1.152 a	0.092 ± 0.044 b	***	0.776 ± 1.085 c	1.047 ± 1.470 b	0.790 ± 1.168 c	1.676 ± 2.467 a	***
Dihydroisoferulic acid	0.005 ± 0.001 a	0.004 ± 0.001 b	0.005 ± 0.001 a	***	0.004 ± 0.001 bc	0.006 ± 0.002 a	0.004 ± 0.001 b	0.005 ± 0.001 a	***
Dihydroferulic acid	0.019 ± 0.004 c	0.040 ± 0.007 a	0.035 ± 0.008 b	***	0.039 ± 0.012 a	0.029 ± 0.010 bc	0.031 ± 0.012 b	0.026 ± 0.009 c	***
p-coumaric acid	0.045 ± 0.021 b	2.034 ± 0.660 a	0.112 ± 0.047 b	***	0.636 ± 0.903 bc	0.454 ± 0.626 c	0.770 ± 1.062 b	1.060 ±1.458 a	***
Coumaroylglucoside	0.150 ± 0.026 b	3.913 ± 1.477 a	0.234 ± 0.040 b	***	2.146 ± 3.035 a	0.920 ± 1.075 d	1.214 ± 1.615 c	1.450 ± 1.966 b	***
Coumaroyl dihexoside	0.045 ± 0.015 b	0.809 ± 0.067 a	0.070 ± 0.034 b	***	0.311 ± 0.388	0.287 ± 0.396	0.298 ± 0.349	0.336 ± 0.424	ns
Sinapic acid	0.009 ± 0.011 c	0.015 ± 0.010 b	0.018 ± 0.016 a	***	0.001 ± 0.001 d	0.032 ± 0.008 a	0.014 ± 0.006 b	0.009 ± 0.005 c	***
Sinapoyl hexoside	1.608 ± 0.324 a	1.153 ± 0.137 b	0.869 ± 0.057 c	***	1.233 ± 0.533 a	1.046 ± 0.113 b	1.284 ± 0.395 ab	1.278 ± 0.354 a	**
Total hydroxicinnamic acids	8.603 ± 1.531 b	15.858 ± 2.414 a	8.175 ± 1.336 b	***	12.701 ± 3.916 a	10.185 ± 2.509 bc	9.687 ± 3.382 c	10.940 ± 5.853 b	***
Rutin	0.129 ± 0.037 b	30.947 ± 2.636 a	0.147 ± 0.117 b	***	11.618 ± 17.882 a	10.647 ± 16.130 b	9.812 ± 14.980 c	9.552 ± 14.654 c	***
Quercetin rutinoside (Rutin isomer)	0.375 ± 0.050 a	n.d.	0.365 ± 0.187 a	***	0.286 ± 0.222 a	0.329 ± 0.277 a	0.219 ± 0.179 b	0.154 ± 0.138 c	***
Rutin hexoside	0.060 ± 0.018 b	0.003 ± 0.001 c	0.066 ± 0.035 a	***	0.034 ± 0.025 c	0.058 ± 0.052 a	0.048 ± 0.038 b	0.031 ± 0.024 c	***
Quercetin 3-glucoside (Isoquercitrin)	0.006 ± 0.004 b	0.354 ± 0.065 a	0.015 ± 0.008 b	***	0.109 ± 0.158 ab	0.108 ± 0.139 b	0.141 ± 0.204 ab	0.142 ± 0.212 a	*
Kaempferol rutinoside	0.162 ± 0.099 b	2.077 ± 0.374 a	0.066 ± 0.034 b	***	0.682 ± 0.975 bc	0.617 ± 0.811 c	0.952 ± 1.248 ab	0.822 ± 1.042 ab	***
Isorhamnetin rutinoside	0.044 ± 0.036 b	2.732 ± 0.653 a	0.024 ± 0.014 b	***	0.870 ± 1.312 b	0.799 ± 1.196 b	1.252 ± 1.912 ab	0.812 ± 1.182 b	**
Isorhamnetin glucoside	0.001 ± 0.001 b	0.021 ± 0.004 a	0.002 ± 0.002 b	***	0.007 ± 0.009 b	0.008 ± 0.008 ab	0.010 ± 0.012 ab	0.008 ± 0.011 ab	**
Taxifolin	0.034 ± 0.020 a	0.004 ± 0.001 c	0.021 ± 0.018 b	***	0.006 ± 0.003 c	0.021 ± 0.013 b	0.034 ± 0.024 ab	0.018 ± 0.022 b	***
Total flavonols	0.810 ± 0.143 b	36.152 ± 2.462 a	0.707 ± 0.383 b	***	13.611 ± 20.083 a	12.595 ± 17.980 ab	12.473 ± 18.127 bc	11.546 ± 16.945 c	***
Apigenin	0.001 ± 0.001 b	0.002 ± 0.002 a	n.d.	***	0.0005 ± 0.0006 c	0.0019 ± 0.0024 a	0.0012 ± 0.0011 b	0.0008 ± 0.0004 c	***
Apigenin rutinoside	n.d.	0.001 ± 0.001 b	0.001 ± 0.001 a	***	0.0005 ± 0.0004 b	0.0005 ± 0.0002 b	0.0011 ± 0.0006 a	0.0006 ± 0.0002 b	***
Diosmetin	0.004 ± 0.001 b	0.002 ± 0.001 c	0.014 ± 0.007 a	***	0.004 ± 0.002 b	0.005 ± 0.003 b	0.009 ± 0.009 a	0.009 ± 0.009 a	***
Total flavones	0.005 ± 0.001 b	0.005 ± 0.003 b	0.015 ± 0.007 a	***	0.005 ± 0.003 c	0.008 ± 0.002 b	0.011 ± 0.009 a	0.010 ± 0.009 a	***
Naringenin	3.007 ± 1.502 b	10.639 ± 5.123 a	2.198 ± 1.768 c	***	2.979 ± 1.388 c	7.106 ± 6.214 a	7.308 ± 6.68 a	3.733 ± 3.210 b	***
Naringenin 6-glucoside	1.663 ± 0.554 b	2.422 ± 1.190 a	1.243 ± 0.774 c	***	0.957 ± 0.159 d	2.088 ± 0.445 b	2.766 ± 1.092 a	1.293 ± 0.776 c	***
Narirutin	0.039 ± 0.032 b	7.047 ± 1.352 a	0.214 ± 0.071 b	***	2.764 ± 4.106 a	2.319 ±2.542 b	2.582 ± 3.959 a	2.495 ± 3.827 a	**
Narirutin glucoside	4.278 ± 1.227 a	2.637 ± 0.639 b	1.691 ± 0.651 c	***	3.104 ± 1.285 b	1.773 ± 0.262 c	3.693 ±1.495 a	2.358 ± 1.801 c	***
Eriodyctiol	0.047 ± 0.026 a	0.027 ± 0.017 b	0.030 ± 0.022 b	***	0.020 ± 0.013 c	0.039 ± 0.018 b	0.046 ± 0.029 a	0.035 ± 0.026 b	***
Total flavanones	9.034 ± 2.520 b	22.772 ± 5.980 a	5.286 ± 2.781 c	***	9.824 ± 5.498 c	13.324 ± 8.968 b	16.395 ± 11.595 a	9.914 ± 7.769 c	***
Phloretin	0.003 ± 0.001 a	n.d.	0.002 ± 0.002 b	***	0.001 ± 0.001 b	0.003 ± 0.002 a	0.003 ± 0.002 a	0.001 ± 0.001 b	***
Phloridzin	0.002 ± 0.001 a	n.d.	0.001 ± 0.001 b	***	0.001 ± 0.001 b	0.001 ± 0.001 a	0.001 ± 0.001 a	0.001 ± 0.001 c	***
Total dihydrochalcones	0.005 ± 0.001 a	n.d.	0.004 ± 0.003 b	***	0.002 ± 0.001 b	0.004 ± 0.003 a	0.004 ± 0.003 a	0.002 ± 0.002 b	***
Total phenolic compounds	18.458 ± 2.630 b	74.787 ± 4.339 a	14.186 ± 4.120 c	***	36.143 ± 29.382 a	36.115 ± 29.366 a	38.570 ± 32.885 a	32.412 ± 29.804 b	***

Data are expressed as mean values ± SD. Different letters in the same row indicate significant differences by Tukey’s test. *, **, ***, ns: significant at t p < 0.05, p < 0.01, and p < 0.001, and not significant, respectively.

). Among them, the hydroxycinnamic acids were the family with the higher number of compounds (n = 15), followed by flavonols (n = 8), flavanones (n = 5), flavones (n = 3) and dihydrochalcones (n = 2). Concerning hydroxycinnamic acids in goji berries, our data are in agreement with those previously reported by other authors [56], where several coumaric acid and caffeic acid derivatives were identified. In the same vein, Milinčić et al. [57] identified 15 hydroxycinnamic compounds out of a total of 28 phenolic compounds. Regarding flavonoids, it is of note that both profiles and quantification are highly variable in the literature [5,23,24,57]. In this sense, Ma et al. [58] observed, in a transcriptomic and metabolomic study of “Ningqi No. 1”-variety samples cultivated in different areas, that the flavonoid biosynthesis pathway was highly influenced by environmental factors. Table 3 shows the content of phenolic compounds for the three goji berry varieties and the different harvesting times. Table S1 shows the information for the identification of phenolic compounds, including retention time, the instrumental and experimental accurate mass, and the error between the exact accurate mass and the mass found of the detected compounds (ppm), while Figure S1 shows the HPLC-HRMS/MS chromatograms of phenolic compounds. In general, the phenolic compound compositions of the goji berries differed among the varieties under study. The samples from the NQ7 variety showed the highest concentration of hydroxycinnamic acids, flavanols and flavanones, in agreement with results obtained by other researchers regarding the content of phenolic compounds in NQ7 samples cultivated in China [59]. The samples from the NQ1 and V3 varieties had a more similar profile, showing significant differences in the total content of flavones, flavanones and dihydrochalcones (Table 3). These differences in the phenolic family profiles can be observed in the heatmap, which highlights the distinct accumulation patterns across the three goji berry varieties (Figure S2).

In general, the total phenolic content remained stable during the first three harvesting times but decreased significantly in T4 (Table 3). In contrast, the results obtained showed significant differences between the varieties, with NQ7 having the highest phenolic content (74.787 ± 4.339 mg/100 g DW). These results are in agreement with those found for the antioxidant capacity, since total polyphenols are one of the main contributors to antioxidant capacity. The cultivation of goji berries under the climatic conditions of the southwest of Spain may promote the accumulation of phenolic compounds by enhancing the antioxidant defense mechanisms of plants. This effect was previously reported by Breniere et al. [8], who observed an increase in total phenolic content of around 15.5%, on a dry mass basis, in goji berries grown under water stress.

As stated above, NQ7 showed the highest concentration of hydroxycinnamic acids (Table 3). These differences were mainly due to the content of p-coumaric acid and its hexoside derivatives (coumaroylglucoside and coumaroyl dihexoside), dihydroferulic acid 4-glucuronide, and caffeoylglucoside, among others (Table 3). In this context, Zhao and Shi [59] observed a higher concentration of hydroxycinnamic acids in NQ7, especially ferulic and p-coumaric acids. On the other hand, feruloyl glucose, chlorogenic acid and caffeoylquinic acid were significantly higher in NQ1 and V3. These findings are in agreement with previous research, where several authors observed that ripe goji berries of the NQ1 variety had a higher content of chlorogenic acid [59]. Nevertheless, although the content of hydroxycinnamic acid compounds was mostly similar between NQ1 and V3, some differences were observed. The samples from NQ1 showed higher sinapoyl hexoside content, whilst V3 showed a higher concentration of caffeic acid, chlorogenic acid and sinapic acid.

Regarding harvesting time, it was noted that the content of hydroxycinnamic acids was higher for the samples harvested at T1, with lower values found for this family of compounds during the T2, T3 and T4 harvesting times. The main contributions in the first sampling were feruloylglucose, coumaroylglucoside and sinapoyl hexoside. Although these remained as the main compounds in the goji berry samples harvested during the T2, T3 and T4 periods, the concentration of most of them decreased (except for the case of sinapoyl hexoside, the concentration of which decreased at T2 and then remained constant). On the other hand, other groups of compounds such as dihydroferulic acid 4-glucuronide, p-coumaric acid or ferulic acid increased in concentration with harvesting time, reaching values 2.1-, 1.4- and 1.5-fold higher from T3 to T4, respectively (Table 3).

The difference in the flavonol content of NQ7 compared with that of the other varieties was remarkable, mainly due to the contribution of rutin, reaching 41.3% in this variety, but only 0.7 and 1.1% in NQ1 and V3, respectively. This compound has been identified as the main phenolic compound in previous research [23,57]. It was observed that other compounds, such as isoquercitrin, kaempferol 3-rutinoside, isorhamnetin 3-rutinoside and isorhamnetin 3-glucoside, were significantly higher in the NQ7 berry samples. In this regard, flavonol glucosides have been quantified at high concentrations in goji samples [23,56]. An isomer of rutin (quercetin rutinoside) was identified and quantified for samples from the NQ1 and V3 varieties, but it was not detected in the NQ7 berry samples (Table 3). In agreement with these findings, several isomers of quercetin rutinoside have been identified in the literature for goji berry samples [29].

Given that the three goji berry varieties were cultivated under identical growing conditions, the observed variability in phenolic compound profiles is likely attributable to genotype-dependent biosynthetic regulation. The differential expression or activity of key enzymes within the phenylpropanoid–flavonoid pathway, including phenylalanine ammonia-lyase (PAL), chalcone synthase (CHS), flavone synthase (FNS) and flavonol synthase (FLS), may influence the accumulation of specific phenolic compounds during biosynthesis [60].

In relation to harvesting time, a gradual decrease was observed for the total concentration of flavonols. This was largely due to the decrease in the major contributor for this family of compounds, namely rutin (Table 3). On the other hand, some minor compounds increased their concentration over the harvesting time, such as kaempferol 3-rutinoside and isorhamnetin 3-rutinoside.

The total content of flavones was significantly higher in the samples obtained from the V3 variety, diosmetin being the main compound found. It was observed that flavones showed a slight increase with the harvesting time (Table 3).

The goji berry samples from the NQ7 variety exhibited significantly higher flavanone concentrations. Regarding the individual compounds of this family, it is worth noting that the NQ7 samples showed the highest content for naringenin, naringenin 6-glucoside and narirutin, representing 26.8% of the total, whilst the samples from the NQ1 variety showed the highest content of nariturin glucoside and eriodyctiol. Of note is the great difference in the concentration of narirutin, from 7.047 mg/100 g DW in the NQ7 samples to 0.124 and 0.039 mg/100 g DW in V3 and NQ1, respectively. In this sense, it is interesting to highlight the likely involvement of narirutin glucoside in the biosynthesis pathway, considering the different accumulation mechanisms observed in the different varieties under study. On the other hand, an increase in flavanone content was observed during harvesting time, reaching the highest values at T3.

Moreover, two other phenolic compounds belonging to the dihydrochalcone family were identified and quantified at low concentrations in the NQ1 and V3 samples. These compounds were not detected in the goji berry samples obtained from the NQ7 variety (Table 3).

In general, and based on the results of the previous literature, some phenolics compounds are related to sensory characteristics such as bitterness, pungency, fruitness/harmony, etc. In this case, TPCs were correlated with sensory descriptors for bitter taste sensation in olive oil [61] and, more specifically, phenolic compounds such as 3,4-DHPEA-EA. Moreover, tannins have been described to play an important role in the perception of astringency in foods, depending on their molecular weight and their interaction with salivary proteins [62,63]. Those compounds are not so important in goji berries, and no information has been found in the literature regarding goji berries and sensory perception.

3.7. ANOVA–Simultaneous Component Analysis (ASCA)

The ANOVA–Simultaneous Component Analysis (ASCA) applied to the dataset revealed clear and statistically significant effects of variety, time, and their interaction on the multivariate structure of the data. The permutation tests (100 iterations) confirmed the significance of all three effects, with p < 0.01 for variety, p = 0.03 for time (Figure 1), and p < 0.01 for the variety × time interaction (Figure S3). This demonstrates that both genetic and temporal factors had a strong influence on the agronomic traits evaluated, and that the varieties responded differently throughout the experimental period.

Moreover, the scree plots showed that the first principal component explained most of the variation within each effect, suggesting that each factor was dominated by a single major trend. For variety (model a), the first component accounted for about 80% of the explained variance, indicating well-defined genotype differentiation (Figure 2), while for time (model b) (Figure 2) and interaction (model ab) (Figure S5), the first two components captured approximately 30–35% of the total variance, highlighting consistent temporal dynamics and interactive patterns

The major pattern associated with variety showed a clear separation among the three genotypes (NQ1, NQ7, and V3), with NQ7 exhibiting an opposite trend to those of NQ1 and V3, reflecting distinct genetic behaviors and potential differences in physiological or metabolic responses. In the case of time, the first component (48.8% of explained variance) displayed a clear oscillatory trend across sampling points (T1–T4), suggesting progressive physiological or developmental changes over time (Figure 3). The interaction patterns revealed crossing trajectories among varieties, confirming that the genotypes did not follow parallel temporal responses and thus exhibited significant variety × time interaction effects (Figure S5). This indicates that each genotype underwent specific dynamic adjustments during the experimental period, an important feature for understanding genotype-dependent adaptation and optimizing agronomic management strategies.

3.8. Pearson Correlation

The Pearson correlation matrix revealed several correlations between the studied variables (Figure 4 and Figure 5). The color gradient, ranging from red (negative correlation) to blue (positive correlation), helps visualize the strength and direction of these relationships.

As expected, the antioxidant activity assays, ABTS and DPPH, exhibited strong positive correlations with TPC (r = 0.88 and r = 0.93, respectively), confirming the key role of phenolic compounds in the antioxidant capacity of goji berries. Additionally, a positive correlation between ABTS and DPPH (r = 0.86) confirmed that both assays evaluated a similar group of compounds linked to the antioxidant activity of the samples, beyond the different methodological procedures.

The color parameters (L, a* and b*) also showed positive correlations with the antioxidant capacity tests and TPC (Figure 1). This supports the idea that goji berry samples with more intense colors, particularly those with a higher red component (a*), tend to have a higher concentration of phenolic compounds, and, consequently, higher antioxidant capacity.

In general, it was observed that redder fruits (higher a*) showed higher antioxidant activity (0.51 and 0.5 with ABTS and DPPH, respectively) and phenolic content (0.56), indicating that color can be a good visual indicator of high-quality fruit. In addition, strong positive correlations were observed among TSS, TPC, and antioxidant capacity (Figure 4), suggesting that sweeter cultivars, specially NQ7, also contain higher levels of health-promoting compounds. These relationships can guide cultivar selection and help determine the optimal harvest stage for maximum fruit quality.

The correlation analysis of textural properties is shown in Figure 2, providing an insight into the mechanical behavior of goji berries. Hardness is negatively correlated with springiness (r = −0.45) and cohesiveness (r = −0.61), indicating that harder berries tend to be less elastic and cohesive. This finding suggests that as goji berries become firmer, their ability to return to their original shape and maintain their internal structure diminishes. Additionally, chewiness showed a moderate positive correlation with hardness (r = 0.41), meaning that firmer berries require more effort to chew. On the other hand, shear force correlated positively with maximum force (r = 0.45), as could be expected, since both parameters reflected resistance to deformation. Interestingly, fracturability exhibited weak correlations with other textural parameters, suggesting that it may reflect different structural characteristics of the berries, complementing the information obtained from the other measurements used.

These results highlight the complex interactions between the three methods, as each one focused on the different aspects and mechanical principles of texture. The puncture test measures resistance to penetration, offering an indication of surface hardness. Texture profile analysis, on the other hand, evaluates multiple parameters, such as hardness, cohesiveness, and elasticity, providing a more comprehensive view of the mastication process. The miniature Kramer Ottawa cell assesses a food’s resistance to deformation under cutting and compression conditions [42,64]. As a result, the methods are not directly comparable, but they can complement each other to provide a more complete understanding of the textural properties of samples.

In summary, the texture evaluation showed strong positive correlations among fracturability, maximum force, shear force, and hardness, reflecting the firmness and structural resistance of goji berries. On the other hand, springiness showed a negative correlation with cohesiveness and hardness, showing that more elastic fruits tend to be less firm. The high correlation between cohesiveness and chewiness suggested that internal tissue cohesion influences masticatory sensory.

Additionally, to evaluate the relative contribution of phenolic compounds to antioxidant activity, an additional Pearson correlation analysis was performed. In general, the correlation between families and individual compounds showed a similar trend, although with slight differences (Figure 6). Specifically, the families showing the highest correlation with ABTS data were flavanones, mainly isorhamnetin 3-rutinoside (0.76), isorhamnetin 3-glucoside (0.74), and kaempferol 3-rutinoside, and flavonols, notably naringenin (0.79) and naringenin 6-glucoside (0.78). Regarding DPPH, although the flavanone (0.93) and flavanol (0.93) families were also the main contributors, rutin (hydroxycinnamic acid) predominated, with a correlation of 0.92.

4. Conclusions

This study included the physicochemical and morphological characterization of three varieties of goji berries (Lycium barbarum) cultivated in southwestern Spain. Significant differences were observed among the varieties and harvesting periods in terms of their physicochemical, morphological, and textural attributes, confirming the strong influence of both genotype and the harvesting period. We should emphasize that this is a preliminary study and that the results obtained are conditioned by the regional conditions of this trial. The NQ7 goji variety exhibited the highest total phenolic content (74.787 mg/100 g DW), unlike the V3 (14.186 mg/100 g DW) and NQ1 (18.458 mg/100 g DW) varieties. In addition, NQ7 showed the best results for antioxidant capacity (ABTS and DPPH), as well as a distinctive color, SST and texture parameters that enhanced its potential for fresh consumption and processing. LC–MS/MS analysis identified 33 phenolic compounds, predominantly hydroxycinnamic acids, flavonols, and flavanones, highlighting rutin, coumaric acid derivatives, and naringenin as key contributors to the antioxidant potential of goji varieties. The correlations observed between the total phenolic content and antioxidant activity reinforce the importance of phenolic composition as a quality indicator.

Finally, the results suggested that the cultivation conditions used in this location showed that goji berries can be successfully introduced as an alternative cultivation in southwestern Spain, representing a promising crop for agricultural diversification, high-value fruit production, and the development of functional foods within Mediterranean agroecosystems.