Juice quality traits,potassium content,and 1 H NMR derived metabolites of 14 pomegranate cultivars

Abstract

BACKGROUND:

Pomegranate (Punica granatum L.) juice is a highly-valued beverage that has been demonstrated to have large quantities of polyphenolic compounds with powerful antioxidant properties. Currently there is limited information related to other components of the juices of USDA germplasm.

OBJECTIVE:

Preselected pomegranate germplasm was evaluated to identify unique cultivars with consumer-friendly traits and to select candidates that meet or exceed juice quality parameters of ‘Wonderful’ juice.

METHODS

Juices from 14 cultivars and commercial pomegranate juice were analyzed using a variety of methods. Juice quality was determined by measuring pH, titratable acidity, citrate, malate, total soluble solids, maturity index, glucose, fructose, γ-Aminobutyric acid (GABA), antioxidant activity, total phenolics, potassium, glutamate, glutamine, and ethanol.

RESULTS

Candidate cultivars meeting ‘Wonderful’ juice quality specifications and nutritional profile were identified as were candidates with desirable traits not seen in ‘Wonderful’. There were significant differences among cultivars in pH, titratable acidity, citrate, malate, total soluble solids, maturity index, glucose, fructose, GABA, potassium, glutamate, glutamine, and ethanol. There was no evidence for differences in antioxidant activity and total phenolics of juices among cultivars.

CONCLUSIONS:

Pomegranate cultivars other than ‘Wonderful’ could provide a broader palette of flavors for the consumer, but more research is needed to determine commercial potential of USDA germplasm.

Keywords

Amino acids DPPH germplasm lythraceae potassium wonderful pomegranate

1 Introduction

The pomegranate germplasm in the United States is highly diverse and is comprised of hundreds of pomegranate cultivars sourced from domestic and international origins [1]. Most cultivars are in the public domain with the potential to be selected for commercial use by growers or developed into improved commercial cultivars by breeders. Reports have indicated that accessions with softer seeds or lower acidity could increase consumer demand for fresh fruit [1]. Recent sensory panel investigations have identified several cultivars in American germplasm with consumer acceptability [2]. The ‘Wonderful’ cultivar, the industry standard in several countries, is a relatively tart to sweet-tart, bitter, moderately hard-seeded fruit and has astringent juice compared to other cultivars previously analyzed from the collection at the United States Department of Agriculture - Agricultural Research Service (USDA-ARS) National Clonal Germplasm Repository, Davis, CA (NCGR) [3]. The identities of the chemical compounds in ‘Wonderful’ juice that are responsible for this perceived bitterness in sensory panels have not been confirmed. Many of the 14 accessions included in the study have been previously described to have one or more of the following traits: unique color, soft seeds, low acidity, or unique flavor [4 –6]. Most of these accessions have had limited attention in the literature and have yet to be phenotyped for potential use by commercial growers, the food and beverage industries, and breeders. Pomegranate cultivation in the United States is largely a monoculture of ‘Wonderful’. Since 2011, for unknown reasons, major beverage companies in the United States began to only purchase ‘Wonderful’ juice from growers [7]. This preference for ‘Wonderful’ is likely due to industry demand for the red, sweet-tart juice of this cultivar. Additionally, ‘Wonderful’ was the primary cultivar selected for clinical trials in America testing the putative health effects of pomegranate juice on humans [8].

To understand the factors influencing juice quality differences among pomegranate cultivars, analytical chemistry methods are used to evaluate the chemical composition and properties of their juices. The use of multiple methods to evaluate fruit juice allows for more robust quality analyses by overcoming the limitation of one method by employing another method with different strengths [9]. Several methods exist to analyze the components and quality of fruit juices, including gas chromatography [10], liquid chromatography coupled with mass spectrometry [11], acid titration, refractometry [3], and spectrophotometry [12]. While all of these methods are useful, they have different levels of sensitivity and specificity. Additionally, several methods have disadvantages that include destruction of the sample, the inability to determine specific quantities of classes of metabolites, or the requirement of a separation of individual sample components prior to analysis. The ability of NMR (nuclear magnetic resonance) to universally and rapidly detect and quantify organic compounds makes it an efficacious method for evaluating differences in chemical profiles among fruit juices from different cultivars without requiring a separation. NMR is advantageous because it accommodates relatively simple sample preparation and allows for compound identification and quantification without authentic standards [13].

It has been demonstrated that NMR methodology is useful to the food industry for juice quality control including applications such as the detection of adulterants, quality control during mixture analysis of fruit juice products, and determination of juice authenticity [9 , 14–16]. The juice industry in the United States has experienced high-profile lawsuits and federal investigations involving both the false advertising of commercial juice composition and health benefits to humans through consumption [17]. Therefore, for the pomegranate industry, it is important to be able to detect beverage adulterants, such as sucrose and high fructose corn syrup, and distinguish between the juices of different species and cultivars for economic, quality evaluation and public health reasons. Unlike other techniques, NMR is able to detect adulterants without destroying the sample.

Evaluating the differences in juice quality among cultivars of horticultural commodities is important because cultivar is generally more influential in determining fruit juice composition than 1) site of cultivation, 2) year of harvest, or 3) length of storage [18]. According to [19], citric acid is the chief determinant of sour flavor in pomegranate juice regardless of the sugar concentration. This means that if a juice has a high citric acid concentration, it will have a sour flavor whether or not it has a relatively high sugar concentration. Sweet or “low acid” pomegranates are typically reported to have a citric acid concentration below 0.50% [3 –19]. Standards have been established for titratable acidity (TA) and total soluble solids (TSS) of ‘Wonderful’ pomegranate. Because citric acid is the predominant organic acid in pomegranate juice, TA is expressed in citric acid equivalents. Minimum maturity of ‘Wonderful’ pomegranate fruit has been determined to be 1.85% TA [20] and growers in California typically harvest fruit after reaching at least 15% TSS. Maturity index (MI) is also a measure of maturity in fruit crops and is calculated as the ratio between TSS and TA (sugar to acid ratio) and is used as a tool by growers to determine fruit maturity. The optimum MI for ‘Wonderful’ is greater than 8.1, at which point the fruit is considered ready to harvest. Other components in juice can affect flavor, such as potassium, which is reported to affect juice flavor and consumer perception of acidity [20], malate, and γ-aminobutyric acid (GABA) [21].

The objectives of this research were 1) to determine the nutritional composition and profiles of pomegranate juices of germplasm utilizing advanced analytical chemistry tools (e.g. NMR) among pomegranate cultivars conserved at the NCGR; 2) to compare fruit juice quality traits and metabolites of promising preselected pomegranate cultivars from the NCGR collection with the industry standard, ‘Wonderful’ and commercial ‘Wonderful’ juice; and 3) to identify cultivars that are suitable candidates for the food and beverage industries.

2 Materials and methods

2.1 Plant materials

This research was conducted over two years with fruit sourced from pomegranate accessions preselected for their fruit quality from the USDA-ARS NCGR for Tree Fruit and Nut Crops and Grapes in Davis, CA. The pomegranate cultivars analyzed were preselected based on fruit quality and consumer sensory panel results and included: ‘Al Sirin Nar’, ‘Ambrosia’, ‘Blaze’, ‘Desertnyi’, ‘Eversweet’, ‘Golden Globe’, ‘Green Globe’, ‘Haku Botan’, ‘Loffani’, ‘Parfianka’, ‘Phoenicia’, ‘Purple Heart’, ‘Sakerdze’, and ‘Wonderful’ (Table 1). The accessions in this study that are soft-seeded included ‘Desertnyi’, ‘Eversweet’ and ‘Parfianka’. Low acidity accessions included ‘Ambrosia’, ‘Desertnyi’, ‘Eversweet’, ‘Golden Globe’, ‘Green Globe’, and ‘Loffani’. ‘Wonderful’ fresh fruit and commercial juice were included as a control and as the standard to compare the other accessions in this study.

Table 1
Descriptions of 14 pomegranate cultivars from the USDA-ARS National Clonal Germplasm Repository used in this study

Cultivar Country of Origin Acidity Flavor Peel color Aril color Seed hardness

Al Sirin Nar Turkmenistan Medium Sweet– tart Pinkish red Red Very hard

Ambrosia USA Low Sweet Pink Pink Hard

Blaze USA Medium Sweet– tart Red Red Medium Hard

Desertnyi Turkmenistan Medium Sweet– tart Red Red Soft

Eversweet USA Very Low Sweet Pink and yellow Pink Soft

Golden Globe USA Low Sweet Yellow and pink Pink Hard

Green Globe USA Low Sweet Greenish yellow Pink Hard

Haku Botan Japan Very High Sour White/yellow White Hard

Loffani USA Low Sweet Pink Pink Hard

Parfianka Turkmenistan High Sweet– tart Red Red Soft

Phoenicia USA High Tart Yellow, Green, pink Red Hard

Purple heart USA High Sweet– tart Red Red Medium hard

Sakerdze Turkmenistan Medium Sweet– tart Pinkish red Red Very hard

Wonderful USA Medium– high Sweet– tart Red Red Medium hard

Cultivar	Country of Origin	Acidity	Flavor	Peel color	Aril color	Seed hardness
Al Sirin Nar	Turkmenistan	Medium	Sweet– tart	Pinkish red	Red	Very hard
Ambrosia	USA	Low	Sweet	Pink	Pink	Hard
Blaze	USA	Medium	Sweet– tart	Red	Red	Medium Hard
Desertnyi	Turkmenistan	Medium	Sweet– tart	Red	Red	Soft
Eversweet	USA	Very Low	Sweet	Pink and yellow	Pink	Soft
Golden Globe	USA	Low	Sweet	Yellow and pink	Pink	Hard
Green Globe	USA	Low	Sweet	Greenish yellow	Pink	Hard
Haku Botan	Japan	Very High	Sour	White/yellow	White	Hard
Loffani	USA	Low	Sweet	Pink	Pink	Hard
Parfianka	Turkmenistan	High	Sweet– tart	Red	Red	Soft
Phoenicia	USA	High	Tart	Yellow, Green, pink	Red	Hard
Purple heart	USA	High	Sweet– tart	Red	Red	Medium hard
Sakerdze	Turkmenistan	Medium	Sweet– tart	Pinkish red	Red	Very hard
Wonderful	USA	Medium– high	Sweet– tart	Red	Red	Medium hard

Characteristics known include county of origin, acidity, flavor, peel color, aril color, and seed hardness. All fruits were picked at maturity from the national pomegranate germplasm collection at Wolfskill Experimental Orchard in Winters, CA USA.

Up to twelve fruits of each cultivar were harvested on 14 October 2014, 15 September 2015, and 15 October 2015. After phytosanitary inspection, fruits were wrapped in paper, shipped in shipping tape-sealed cardboard containers by ground and received within two days, and then stored at 6°C and 98% relative humidity until processing. Fresh market quality fruit, as defined by being well-filled, mature, unblemished and greater than 250 g, were chosen for juice analysis from each of the 14 cultivars. Two cultivars, ‘Loffani’ and ‘Green Globe’ were only available for one harvest date (October 2014).

There were two methods of juicing in this study: aril-pressed juicing and peel-pressed juicing. Peel-pressed juice is common in the USA industry, but aril pressed commercial juice exists on a small scale. For aril-pressed juice, fruits were halved and 100 undamaged arils were manually removed and placed in a polyethylene bag. These arils were pressed manually to express the juice directly into a 15 mL test tube; this juice type is called aril-pressed. Peel-pressed juice samples consisted of halved pomegranates that were pressed with peel and septa intact with a juice press exerting approximately 15 MPa. Juice samples from 2014 consisted of aril-pressed juices from a single fruit. Juice samples analyzed in 2015 utilized aril-pressed and peel-pressed juices and both juicing methods consisted of composite samples from multiple fruit per cultivar. Samples derived from both juicing methods were centrifuged at 1000 g for 5 min using a Becton Dickinson DYNAC Centrifuge (Sparks, MD, USA) and aliquots of the supernatants were used for all analyses. For methods utilizing spectrophotometry, supernatants were hundredfold diluted in 6:4 methanol:deionizied water (V/V) as described below.

2.2 Commercial juices

Commercial ‘Wonderful’ juice samples (PomWonderful^TM brand) were purchased at a grocery store in Riverside, CA, USA. According to labeling, the commercial juices contained 100% California pomegranate juice, sourced from concentrate.

2.3 pH measurement

An AB15 pH meter from Fisher Scientific calibrated with aqueous buffers at pH 2.0, 4.0, 7.0, and 10.0 (Fisher Scientific) was used to measure supernatant juice pH following centrifugation. Solution pH values were determined using a glass double junction pH electrode (Oriontrademark 9110DJWP, Thermo Scientific).

2.4 Titratable acidity (TA)

Juice TA was measured with a Hanna HI 84532 fruit juice TA mini-titrator (Woonsocket, RI, USA) for samples prepared by mixing an aliquot of centrifuged juice (5 mL) with deionized water (45 mL). Results are expressed in citric acid equivalents as a percentage acid in the juice sample. Both high and low range titrants were used to titrate high acid and low acid pomegranate juices, respectively. Juice samples were autotitrated with a OH^--based titrant solution to an endpoint pH of 8.1 (Hanna juice titrants for fruit juice, products HI84532–51 and HI84532–50, Woonsocket, RI, USA). One sample was analyzed per juice sample.

2.5 TSS

Total soluble solids were quantified with a Vee Gee Scientific PDX– 1 Digital Refractometer (Kirkland, WA, USA), utilizing 0.5 mL of expressed, centrifuged juice sample. Results were reported in percent (%) TSS.

2.6 Maturity index (MI)

The MI was calculated as the ratio between the TSS and the TA.

2.7 Antioxidant activity (AA)

Juice AA, expressed as an inhibition percentage of the stable free radical 2,2– diphenyl– 1– picrylhydrazyl (DPPH), was quantified utilizing methods described in [12] with modifications reported in [6]. Antioxidant activity values were determined by comparing the juice solution absorbance (A_b) with the absorbance (A₀) of a control solution at 517 nm using a Biochrom Ultrospec II spectrophotometer, and calculated using the following equation:

Equation 1. Antioxidant activity

$Antioxidant Activity (inhibition %) = (1 - A_{b} / A_{0}) \times 100$

The control solution was made by adding 2 mL of 6:4 methanol:deionized water (V/V) to 2 mL of 0.075 mM DPPH. For juice solution absorbance, a 20μL aliquot of juice supernatant was added to 1980μL 6:4 methanol:deionizied water (V/V). This 2 mL of hundredfold-diluted juice was added to 2 mL of 0.075 mM DPPH dissolved in ethanol. Juice samples and the control were reacted in the dark for 30 min at room temperature before analysis.

2.8 Total phenolics

Total phenolics (TP) were measured utilizing the Folin-Ciocalteu method as described by [22]. For TP, 3 μL juice supernatant was added to 297μL 6:4 methanol:deionized water (V/V) in a 4 mL cuvette. Then 1.5 mL of 10– fold-diluted Folin– Ciocalteu reagent was added and left to react for 3 min. After this reaction period, 1.2 mL of 7.5% sodium carbonate was added to the cuvette and the sample was left in darkness for 90 min. The cuvette was transferred to a Biochrom Ultrospec II spectrophotometer (Cambridge, England, UK) and the absorbance at 765 nm was measured. Results were expressed as gallic acid equivalents (GAE), which was used as a standard. TP samples were measured in triplicate.

2.9 ¹H NMR experimental parameters and quantification steps

The ¹H spectra were recorded with a Bruker Avance NMR spectrometer operating at 599.58 MHz, using a 5 mm BBI probe. Concentrations of organic acids, sugars, and ethanol were determined using ¹H NMR. Sample preparation and ¹H NMR experiment were performed as described previously [13]. For this analysis, the juice sample (1 mL) was centrifuged again for 10 min at 13,000 g to remove any particulates formed during storage. To 630 μL of supernatant, 57 μL deuterium oxide (D₂O) and 6 μL of the internal standard solution of 4,4– dimethyl– 4– silapentane– 1– sulfonic acid-d6 (50 mM DSS in D₂O) were added. Deuterated formic acid-d2 (CDOOD) or ammonium deuteroxide-d4 (ND₃OD) dissolved in D₂O were used for pH adjustment to 3.5±0.1. After centrifugation (1 min at 2 g), 620 μL of sample was introduced into a 5 mm NMR tube and the ¹H NMR spectrum recorded at 25 °C using WET (Water Suppression through Enhanced T₁ Effects) solvent suppression [23]. Spectra were acquired after a 2 s relaxation delay (d₁) into 32,768 data points and zero filled to 131,072 points, with 1024 scans coadded and 16 dummy scans. Apodization equivalent to 0.5 Hz line broadening was applied prior to Fourier transformation.

Chemical shifts were referenced relative to the internal standard DSS at 0.00 ppm. After phasing and baseline corrections, Chenomx version suite 8.1 (Alberta, Canada) was used to identify the resonances of selected sugars, amino acids and organic acids, and to determine their peak areas (Fig. 1A). The latter were quantified by correcting for incomplete T₁ relaxation in the proton survey experiments using the Inversion-Recovery pulse sequence. As detailed by [24], the T₁ relaxation time of each compound was determined in aril-pressed and peel-pressed pomegranate juice samples using recovery time delays (τ), ranging from 0.05 s to 20 s (Fig. 1C) [25]. After integrating the resonances of the selected compounds, the time delays (τ) were plotted vs peak areas using Wolfram Mathematica software (Oxford, UK), to determine the longitudinal relaxation time T₁ using equation 1, where M_z(τ) corresponds to the intensity of the magnetization along the longitudinal (z) axis, τ is the recovery delay following the inversion pulse, M₀ is the value of the magnetization at thermal equilibrium, and T₁ is the longitudinal relaxation time.

Fig.1

¹H NMR spectra. Expansion of the ¹H NMR spectra of ‘Purple Heart’ pomegranate juice expressed from whole fruit (peel-pressed juice), highlighting Chenomx integrations for fructose (in orange), glucose (in green), and citrate (in black) (Panel A). Expansion of the ¹H NMR spectrum of ‘Haku Botan,’ ‘Wonderful,’ ‘Purple Heart,’ and ‘Eversweet’ pomegranate juice expressed from whole fruit, highlighting the different concentrations for acids and sugars (Panel B). Inversion-Recovery ¹H NMR spectra of ‘Wonderful’ pomegranate juice using relaxation delay times ranging from 0.05 to 20 s with scatter plot of the intensity of the citrate resonance as a function of the relaxation delay and the equation used to calculate the T₁ relaxation time (Panel C).

2.2.9.1 Equation 2. Equation for T₁ relaxation time measurements using the inversion-recovery method

$M_{z} (τ) = M_{0} - 2 M_{0} e^{(- \frac{τ}{T_{1}})}$

The peak areas of DSS and juice compounds in the ¹H survey spectra were corrected for incomplete T₁ relaxation and normalized to the number of protons giving rise to each resonance (Fig. 1). The concentrations of compounds in each sample were determined using the Chenomx resonance area of each metabolite relative to the DSS internal standard integral, Equation 3.

2.2.9.2 Equation 3. Calculation of the compound concentration, based on the DSS internal standard concentration, and the T1 corrected area per proton for DSS and the compound

$Compound Concentration = Compound T_{1} Corrected Area \times \frac{DSS Concentration}{DSS T_{1} Corrected Area}$

NMR was chosen for quantifying, within a single analysis, a large range of organic compounds, without the necessity of the corresponding standards and without the need for expensive labeled isotope standards. Furthermore, ¹H-NMR is a very reproducible technique. However, ¹H-NMR has low sensitivity, which can be improved by acquiring a large number of scans to improve the signal to noise ratio. In this investigation, we used a high number of scans (1024 scans) to enhance compound detection i.e. the phenolic compounds present in pomegranate juice (Fig. 1). Even with 1024 scans, phenolic compound signals, such as punicalagin, were not abundant enough in the samples to be quantified.

Due to the high number of scans, the analysis time was long (90 min per sample) even with a d₁ of 2 s applied before each scan. To obtain complete relaxation, the d₁ is typically 5 times longer than the T₁ of the selected signal [23, 24]. To reach the full relaxation of all selected metabolites, the d₁ is conventionally set at more than 13 s (5 times DSS T₁, Table S1), which would have drastically increased the analysis time of the juice samples. With a d₁ set at 2 s, a T₁ correction was necessary for the compounds with a T₁ longer than 0.4 s (d₁ divided by 5).

The WET (Water Suppression through Enhanced T₁ Effects) solvent suppression was used to suppress the most abundant signal, due to the water naturally present in the juice samples [23]. The pulse width was similar to previous reports [13]. Metabolites were identified and quantified using Chenomx software, which contains a library of metabolite spectra. To determine concentrations, after phasing and baseline correction, the predicted spectra from the library were fit to the metabolite signals to measure metabolite areas (Fig. 1), and concentrations were calculated using Chenomx software.

2.10 Ion chromatography

Potassium concentration is high in pomegranate juice and fruit and knowing if there are differences between cultivars is important for the beverage industry in terms of health claims and nutrition labels [9]. In the present work, potassium was determined by ion chromatography using a Dionex ICS– 4000 capillary HPIC system (Thermo Fisher, Fair Lawn, NJ). Ions were separated using a 500 μL/min isocratic flow of methanesulfonic acid (20 mM), on an IonPac TM CS– 12A 5μm RFIC 3×150 mm column. To limit column contamination, a guard column (IonPac TM CG– 12A 5μm RFIC 3×30 mm) was placed in front of the analytical column. The injection volume was 25 μL and the cell temperature was maintained at 30°C. A 4 mm converter-cation self-regenerating suppressor (CSRS) was used, and a 30 mA current was applied. The calibration curve ranged from 0.3 to 25 mg/L. Samples were hundredfold diluted with ultrapure water before analysis.

2.11 Statistical analysis

All data were analyzed with Analysis of Variance (ANOVA). When ANOVA indicated significant differences, post-hoc comparisons were performed utilizing Tukey’s honestly significant difference (HSD) with an experiment wise error rate of α= 0.05. Relationships between all variables were analyzed using linear regression (α= 0.05), with correlations among variables determined using general regression with Minitab Software, version 16 (Coventry, UK). For the purposes of this work, the R² value is the percentage of variation in one variable that is explained by the variation in the correlated variable.

3 Results and discussion

3.1 Acidity

Wide ranges of acidity and significant differences were observed among cultivars for the different acidity-related postharvest quality traits explored (Table 2). Mean values of acidity variables were similar to those found in the literature. For example, the mean pH values of cultivars in the present study were similar to those reported in [3] and [26].

Table 2
Values and concentrations of juice acidity parameters, including pH, titratable acidity in percent citric acid equivalents (TA), and citrate (mM), malate (mM), and γ-Aminobutyric acid (GABA; mM) of whole fruit, aril-pressed pomegranate and store-bought commercial juice. Fruits were sourced from Winters, CA and harvested in mid-October (2014 and 2015) and mid-September (2015)

Cultivar pH TA (% in citric acid equivalents) Citrate Malate GABA

Al Sirin Nar 3.20±0.13bcde^A,B 1.10±0.24cd 63.8±7.4c 3.9±1.1bcd 0.25±0.1b

Ambrosia 3.35±0.14b 0.59±0.13de 26.8±2.2de 5.8± 1.1a 0.51±0.1ab

Blaze 3.21±0.09bcde 1.09±0.19cd 59.6±9.9c 4.1±0.50bcd 0.57±0.1ab

Desertnyi 2.84±0.07g 1.16±0.24cd 72.6±6.1c 3.0±0.54cde 0.24±0.1b

Eversweet 3.26±0.07bcd 0.28±0.08e 3.0±1.1e 4.4±0.94abc 0.22±0.1b

Golden Globe 3.77±0.09a 0.35±0.08e 15.0±2.6de 5.3±0.67ab 0.45±0.2ab

Green Globe 3.56 0.78 20.4 5.5 m.d.

Haku Botan 2.75±0.06g 2.97±0.45a 167.1±23.6a 1.6±0.61e 0.28±0.3b

Loffani 3.57 1.15 13.0 9.5 m.d.

Parfianka 2.89±0.08fg 1.39±0.33bc 71.0±7.6c 3.9±1.1bcd 0.19±0.04b

Phoenicia 2.85±0.07g 1.98±0.49b 109.2±25.6b 3.4±0.5cd 0.83±0.3a

Purple heart 3.05±0.03ef 1.41±0.30bc 77.6±8.1c 3.4±0.5cd 0.47±0.2ab

Sakerdze 3.11±0.03cde 1.09±0.27cd 62.8±11.1c 2.6±0.6de 0.18±0.1b

Wonderful 3.10±0.06de 1.37±0.39bc 75.4±20.5c 3.1±0.7cde 0.39±0.3b

Commercial juice 3.31±0.05bc m.d. 45.3±1.7cd 3.0±0.5cde^* m.d.

p-value <0.001 <0.001 <0.001 <0.001 <0.001

Cultivar	pH	TA (% in citric acid equivalents)	Citrate	Malate	GABA
Al Sirin Nar	3.20±0.13bcde^A,B	1.10±0.24cd	63.8±7.4c	3.9±1.1bcd	0.25±0.1b
Ambrosia	3.35±0.14b	0.59±0.13de	26.8±2.2de	5.8± 1.1a	0.51±0.1ab
Blaze	3.21±0.09bcde	1.09±0.19cd	59.6±9.9c	4.1±0.50bcd	0.57±0.1ab
Desertnyi	2.84±0.07g	1.16±0.24cd	72.6±6.1c	3.0±0.54cde	0.24±0.1b
Eversweet	3.26±0.07bcd	0.28±0.08e	3.0±1.1e	4.4±0.94abc	0.22±0.1b
Golden Globe	3.77±0.09a	0.35±0.08e	15.0±2.6de	5.3±0.67ab	0.45±0.2ab
Green Globe	3.56	0.78	20.4	5.5	m.d.
Haku Botan	2.75±0.06g	2.97±0.45a	167.1±23.6a	1.6±0.61e	0.28±0.3b
Loffani	3.57	1.15	13.0	9.5	m.d.
Parfianka	2.89±0.08fg	1.39±0.33bc	71.0±7.6c	3.9±1.1bcd	0.19±0.04b
Phoenicia	2.85±0.07g	1.98±0.49b	109.2±25.6b	3.4±0.5cd	0.83±0.3a
Purple heart	3.05±0.03ef	1.41±0.30bc	77.6±8.1c	3.4±0.5cd	0.47±0.2ab
Sakerdze	3.11±0.03cde	1.09±0.27cd	62.8±11.1c	2.6±0.6de	0.18±0.1b
Wonderful	3.10±0.06de	1.37±0.39bc	75.4±20.5c	3.1±0.7cde	0.39±0.3b
Commercial juice	3.31±0.05bc	m.d.	45.3±1.7cd	3.0±0.5cde^*	m.d.
p-value	<0.001	<0.001	<0.001	<0.001	<0.001

^AValues expressed as means±standard deviation (n = 5, except commercial juice, n = 3, and Green Globe, and Loffani, n = 1). ^BValues followed by different letters within a column are significantly different (P < 0.05). m.d. = missing data.

3.1.1 pH

The mean pH for pomegranate juices ranged from 2.75±0.06 (‘Haku Botan’) to 3.77±0.09, (‘Golden Globe’, Table 2). Mean juice pH was 3.11, regardless of cultivar. There was a significant difference between the pH of fresh ‘Wonderful’ juice and commercial ‘Wonderful’ juice from concentrate, as well as differences among cultivars for pH. ‘Wonderful’ fresh juice pH was 3.10±0.06, and the commercial juice pH was 3.31±0.05.

3.1.2 Titratable acidity (TA)

The mean TA for pomegranate juices was 1.23±0.75% ranging from 0.28±0.08% (‘Eversweet’) to 2.97±0.45%, (‘Haku Botan’, Table 2). The TA of ‘Eversweet’ had only approximately 20% of the TA in the ‘Wonderful’ juice (1.37±0.39%). Cultivars with similar TA compared to ‘Wonderful’ included ‘Al Sirin Nar’, ‘Blaze’, ‘Desertnyi’, ‘Parfianka’, ‘Phoenicia’, ‘Purple Heart’, and ‘Sakerdze’. Results for TA are in agreement with other reports on pomegranate germplasm [3 –27]. The sweet-tart cultivars, ‘Phoenicia’, ‘Purple Heart’, and ‘Wonderful’ had a high acidity, with 1.98±0.49, 1.41±0.30, and 1.37±0.39% TA, respectively which was confirmed by ¹H NMR with citrate concentrations being greater than 75 mM for these cultivars. Citrate is the conjugate base of citric acid and an important flavor component in beverages and fruits contributing to the sourness and tartness of fruit [20].

3.1.3 Citrate, malate, and GABA

Citrate mean concentrations of ‘Wonderful’ fresh and commercial juices were 75.4±20.5 mM, and 45.3±1.7 Mm, respectively. There was a 56– fold difference in mean juice citrate between the sour ‘Haku Botan’ (167.1±23.6 mM) and the sweet tasting ‘Eversweet’ (3.0±1.1 mM) (Table 2). Cultivars that were significantly different from ‘Wonderful’ for malate were ‘Ambrosia’, ‘Eversweet’, ‘Golden Globe’, ‘Haku Botan’, and ‘Phoenicia’. Mean citrate and malate concentrations regardless of cultivar were 67.0±43.4 mM and 3.70±1.3 mM, respectively.

Among all cultivars, regardless of juice extraction method and harvest date, the mean malate concentration was 3.7±1.3 mM. There were significant differences among cultivars for malate. The accession with the highest level of malate was ‘Loffani’, with a concentration of 9.5 mM, whereas the lowest concentration was 1.6±0.61 mM for ‘Haku Botan’. ‘Wonderful’ juice had a mean malate concentration of 3.1±0.7 mM (P < 0.001) (Table 2). For commercial pomegranate juice, malate was in similar concentration (3.0± 0.5 mM) to that found in fresh juice of ‘Wonderful’. Cultivars differing from ‘Wonderful’ for malate were ‘Ambrosia’ and ‘Golden Globe’. Citrate concentration was not significantly different between ‘Wonderful’ and the other cultivars with the same sweet-tart juice quality i.e. ‘Al Sirin Nar’, ‘Blaze’, ‘Desertnyi’, ‘Parfianka’, ‘Phoenicia’, ‘Purple Heart’, and ‘Sakerdze’. Therefore, these other cultivars may be candidates for the beverage industries desiring ‘Wonderful’-type sweet-tart juices. ‘Haku Botan’ had the highest TA and the highest citrate concentration, which was about 60% greater than ‘Phoenicia’, the cultivar with the second highest citrate concentration. This may explain the sour flavor described in [2 , 5]. Its flavor can be described as similar to a lemon, which also has a similar range of TA, and may be favorable to the juice industry as a natural source of citric acid or a surrogate lemon. Some have likened its flavor at maturity to that of pineapple-flavored “LifeSavors” candy (personal communication).

In contrast, low acid cultivars, such as ‘Ambrosia’, ‘Eversweet’, ‘Golden Globe’, ‘Green Globe’ and ‘Loffani’ had citrate concentrations less than 40 mM and would serve as good candidates for consumers preferring less acidic, sweeter flavored fresh fruit or juices. Diverse cultivar choices may lead to increased public consumption of the fruit. If found acceptable to American consumers, low acid pomegranate cultivars could be utilized in the fresh market and in the food and beverage industries as they are in other parts of the world. Concentrations of citrate were strongly correlated with TA (P < 0.0001, R² = 0.9476) because for most cultivars, citric acid is the predominant organic acid in pomegranate juice [28]. In agreement with [3], citrate was the primary acid in pomegranate juices for all cultivars except ‘Eversweet’, which was higher in malate than citrate and had by far the lowest concentration of citrate (3.0± 1.1 mM, Fig. 1B, Table 2). This could be a factor in the fruit and juice flavor, since malic acid is less tart than citrate [20] and is the major organic acid found in apple cultivars [29]. The values for malate agreed with those reported by [3] and [30]. There was no effect of juice extraction method on the acidity parameters e.g. pH, TA, malate, and citrate concentrations.

There was evidence for significant differences among cultivars for GABA, with ‘Phoenicia’ having the highest concentration (Table 2). ‘Wonderful’ had a GABA concentration of 0.39± 0.3 mM and this value was over 50% lower than GABA concentration of ‘Phoenicia’ (0.83± 0.3 mM). No other cultivars were significantly different from ‘Wonderful’ for GABA concentration. Concentrations of GABA were significantly different among cultivars, with ‘Phoenicia’ having higher GABA concentrations than ‘Wonderful’ and several other cultivars (Table 2). This is the first report of differences among cultivars in the USDA pomegranate germplasm for GABA. It has been reported that GABA can affect flavor sensory attributes of fruit [21], so it is important that these flavor-influencing juice constituents are characterized in germplasm and taken into consideration when analyzing sensory panel results, planning parentage for crosses in breeding, and evaluating fruit quality in breeding populations. ‘Wonderful’, the industry standard, has been described as having bitterness by trained panelists [3], but the identity of the chemical compounds of the fruit that cause this perceived bitterness have not been confirmed. The more we know about the metabolic profiles of these cultivars, the more likely breeders can use this metabolomic data coupled with sensory panel results to develop improved germplasm and breed for flavors that are perceived as less bitter and thus more acceptable to consumers.

3.2 Sugars and maturity index

3.2.1 TSS

The mean TSS for pomegranate juices regardless of cultivar was 16.5± 0.83% and ranged from 15.5± 0.29% (‘Green Globe’) to 17.4± 0.88%, (‘Blaze’, Table 3). ‘Wonderful’ had a mean of 17.1± 0.45% TSS. ‘Al Sirin Nar’, ‘Blaze’, and ‘Wonderful’ had higher TSS than ‘Golden Globe’. Growers in California, USA typically harvest fruit after reaching at least 15% TSS. ‘Al Sirin Nar’ and ‘Blaze’ had the highest level, with 17.3± 0.4% and 17.4± 0.9% TSS respectively (Table 3). These values were not significantly different than ‘Wonderful’, which had 17.1± 0.5% TSS. The mean TSS value of ‘Golden Globe’, was the lowest (15.5± 0.3% TSS) and was significantly different than ‘Wonderful’, ‘Al Sirin Nar’, and ‘Blaze’ (P = 0.002) (Table 3). TSS results agree with [3], who reported a range of 15.9–17.7% TSS for the accessions that they studied from the NCGR pomegranate collection.

Table 3
Values and concentrations of juice sugar and harvest parameters, including TSS (% total soluble solids), Maturity Index (sugar to acid ratio; TSS:TA) and glucose (mM) and fructose (mM) of whole fruit, aril-pressed pomegranate and store-bought commercial juice. Fruits were sourced from Winters, CA and harvested in mid-October (2014 and 2015) and mid-September (2015)

Cultivar TSS Maturity Index Glucose Fructose

Al Sirin Nar 17.3±0.38a^A,B 16.3±3.2d 404.5±22.3a 438.5±16.4a

Ambrosia 16.7±0.81ab 29.3±5.8c 409.4±27.7a 441.3±34.6a

Blaze 17.4±0.88a 16.4±3.6d 422.6±45.5a 434.2±51.9a

Desertnyi 16.6±0.91ab 15.1±4.2d 385.8±28.8a 409.2±26.6ab

Eversweet 16.0±0.55ab 60.6±11.4a 394.0±20.0a 440.3±19.50a

Golden Globe 15.5±0.29b 45.3±8.9b 375.5±31.3ab 437.6±28.5a

Green Globe 16.1 20.6 435.3 484.0

Haku Botan 15.9±1.0ab 5.5±0.89d 301.6±53.7b 335.4±55.7b

Loffani 17.0 14.8 489.3 553.8

Parfianka 16.0±0.72ab 12.1±2.9d 368.4±47.0ab 403.6±29.2ab

Phoenicia 16.4±0.82ab 8.8±2.4d 373.9±42.5ab 389.6±36.0ab

Purple heart 16.8±0.50ab 12.5±2.8d 379.8±22.6a 404.3±24.4ab

Sakerdze 16.5±0.46ab 16.0±4.2d 389.5±37.7a 413.5±32.4a

Wonderful 17.1±0.45a 13.5±4.4d 417.1±23.5a 444.3±42.0a

Commercial juice m.d. m.d. 369.8±1.4ab 374.8±3.8ab

p-value 0.002 <0.001 <0.001 <0.001

Cultivar	TSS	Maturity Index	Glucose	Fructose
Al Sirin Nar	17.3±0.38a^A,B	16.3±3.2d	404.5±22.3a	438.5±16.4a
Ambrosia	16.7±0.81ab	29.3±5.8c	409.4±27.7a	441.3±34.6a
Blaze	17.4±0.88a	16.4±3.6d	422.6±45.5a	434.2±51.9a
Desertnyi	16.6±0.91ab	15.1±4.2d	385.8±28.8a	409.2±26.6ab
Eversweet	16.0±0.55ab	60.6±11.4a	394.0±20.0a	440.3±19.50a
Golden Globe	15.5±0.29b	45.3±8.9b	375.5±31.3ab	437.6±28.5a
Green Globe	16.1	20.6	435.3	484.0
Haku Botan	15.9±1.0ab	5.5±0.89d	301.6±53.7b	335.4±55.7b
Loffani	17.0	14.8	489.3	553.8
Parfianka	16.0±0.72ab	12.1±2.9d	368.4±47.0ab	403.6±29.2ab
Phoenicia	16.4±0.82ab	8.8±2.4d	373.9±42.5ab	389.6±36.0ab
Purple heart	16.8±0.50ab	12.5±2.8d	379.8±22.6a	404.3±24.4ab
Sakerdze	16.5±0.46ab	16.0±4.2d	389.5±37.7a	413.5±32.4a
Wonderful	17.1±0.45a	13.5±4.4d	417.1±23.5a	444.3±42.0a
Commercial juice	m.d.	m.d.	369.8±1.4ab	374.8±3.8ab
p-value	0.002	<0.001	<0.001	<0.001

^AValues expressed as means± standard deviation (n = 5, except commercial juice, n = 3, and Green Globe, and Loffani, n = 1).^BValues followed by different letters within a column are significantly different (P < 0.05). m.d. = missing data.

3.2.2 Maturity index

The MI, which is the TSS:TA ratio, mean ranged from 5.5± 0.89 (‘Haku Botan’) to 60.6± 11.4 (‘Eversweet’ Table 3). ‘Wonderful’ had a mean MI of 13.5± 4.4. Cultivars that were significantly different from ‘Wonderful’ for MI included sweet cultivars: ‘Ambrosia’, ‘Eversweet’, and ‘Green Globe’. The mean MI regardless of cultivar was 16.5± 0.8. The MI, which is the TSS:TA ratio, was highly variable among the cultivars in this study, which agrees with ranging values reported in [3]. This wide range of MI is a result of classifying pomegranate fruit from cultivars that have a diverse range of acidity. The pomegranate cultivars in the higher acidity range have a MI below 20 and the lower acid cultivars have a MI higher than 20. Therefore, it is critical that MIs are developed separately for different pomegranate types so that growers know the optimal time to harvest. Pomegranate cultivars that fit the criteria for ‘Wonderful’ in terms of MI are ‘Al Sirin Nar’, ‘Blaze’, ‘Desertnyi’, ‘Haku Botan’, ‘Parfianka’, ‘Phoenicia’, ‘Purple Heart’, and ‘Sakerdze’ (Table 3). These varieties could be good candidates for the juice industry, with the exception of ‘Haku Botan’ since it has a high acidity and lack of pigment and desirable color in the fruit (Table 1).

3.2.3 Glucose and fructose

Among the ¹H-NMR signals, only two sugars were identified: glucose and fructose (Fig. 1). Sucrose, as reported by [9], was not detected in these pomegranate juices. Concentrations of sugars were significantly different among the pomegranate accessions (Table 3), and in agreement with [30] for both glucose and fructose. There were significant differences for glucose and fructose, with the majority of cultivars having significantly higher concentrations than ‘Haku Botan’ (Table 3), including ‘Wonderful’. There were also no differences among cultivars for glucose:fructose ratio, which had a mean ratio of 0.926 (data not shown). The juice extraction method had no significant effect on glucose, which agrees with [30].

The mean concentration of glucose regardless of cultivar was 385.2 mM and ranged by cultivar from 301.6± 53.7 mM (‘Haku Botan’) to 422.6± 45.5 mM (‘Blaze’). The commercial ‘Wonderful’ juice concentration of glucose was 369.8± 1.4 mM and this value was similar to fresh squeezed ‘Wonderful’, which had 417.1± 23.5 mM glucose. The only cultivar with significantly lower glucose than ‘Wonderful’ was ‘Haku Botan’, the ornamental cultivar with no red pigment.

For fructose, the mean concentration was 416± 43.8 mM regardless of cultivar and juice type. ‘Haku Botan’ had the lowest fructose concentration (335.4± 55.7 mM) and ‘Loffani’ juice had the highest 553.8 mM (note there was only one sample of ‘Loffani’ tested). ‘Wonderful’ cultivar, which had one of the highest fructose concentrations, and the commercial juice had 444.3± 42.0 mM, and 374.8± 3.8 mM fructose, respectively. ‘Haku Botan’ was the only cultivar with a significantly lower fructose concentration than ‘Wonderful’ fresh juice, but it was not different from commercial ‘Wonderful’ juice.

3.3 Other juice quality traits

3.3.1 Antioxidant activity (AA)

When data were pooled, there were no significant differences in AA among pomegranate cultivars in this study. Mean AA regardless of cultivar was 39.20± 0.73%. All AA values were similar to the industry standard, ‘Wonderful’, which produced a mean value of 39.13± 0.62% for inhibition of DPPH. The juice extraction method had a significant effect on antioxidant activity, with aril-pressed juice having a lower DPPH inhibition percentage than peel-pressed juice (Fig. 2).

Fig.2

Comparison of the extraction methods (aril-pressed versus peel-pressed) for the 14 cultivars. Parameters measured using refractometry were total phenolics (Panel A) and antioxidant activity (Panel B). Parameters measured using ¹H NMR included glutamine (Panel C) and glutamate (Panel D). Results are displayed as mean± standard error. Asterisk (*) indicates significant difference between juice extraction methods (P < 0.05). Double asterisk (**) indicates significant difference between juice extraction methods (P < 0.01).

3.3.2 Total phenolics (TP)

The cultivars in this study had large quantities of polyphenolic compounds in the juice, regardless of juicing method. There were significant differences between juicing method, with aril-pressed juice having significantly lower quantities TP than peel-pressed juices (Fig. 2). Aril-pressed juice had a mean TP of 3606± 356 mg/L GAE and peel-pressed juice had mean TP of 4464± 72 mg/L GAE, with the peel-pressed juicing method increasing TP by almost 25%. There were no significant differences among cultivars for TP (Table 4). Due to the relatively low concentrations of these individual compounds, they were neither quantified nor identified by ¹H NMR in this investigation. Mean TP regardless of juicing methods and cultivar was 3955± 676 mg/L GAE. ‘Wonderful’ had mean TP 4475± 824 mg/L GAE.

Table 4
Antioxidant activity (% inhibition of DPPH), total phenolics (in gallic acid equivalents (GAE) mg/L), potassium (mg/L), glutamate (mM), glutamine (mM), and ethanol (mM) of fresh whole fruit, aril-pressed pomegranate juice and store-bought commercial juice. Fruits were sourced from the USDA-ARS National Clonal Germplasm Repository, in Davis, CA, and harvested in mid-October (2014 and 2015) and mid-September (2015)

Cultivar Antioxidant Activity Total Phenolics Potassium Glutamate Glutamine Ethanol

Al Sirin Nar 39.14±0.76^A 4186±912 1999±197ab^B 0.40±0.08abcd 1.34±0.39ab 0.55±0.13bc

Ambrosia 39.42±0.34 3801±425 1491±200cdef 0.61±0.15abcd 2.68±1.64ab 0.71±0.42abc

Blaze 39.22±0.31 4007±440 1903±135abc 0.57±0.11abcd 1.54±0.52ab 1.40±0.79a

Desertnyi 39.30±0.80 3533±224 1419±119def 0.33±0.15bcd 1.01±0.62ab 0.23±0.13c

Eversweet 39.24±0.42 3394±364 1282±114f 0.59±0.21abcd 2.50±0.47ab 0.40±0.18bc

Golden Globe 38.84±1.15 4142±775 1346±91f 0.73±0.32ab 1.99±0.71ab 0.60±0.35abc

Green Globe 37.9 3583 1385 0.8 6.6 n.d.

Haku Botan 39.23±1.08 4387±792 2301±244a 0.28±0.06cd 0.68±0.19b 0.14±0.20c

Loffani 38.6 3301 1851 1.5 7.8 0.21

Parfianka 38.48±1.56 3757±491 1359±226ef 0.56±0.07abcd 2.17±0.42ab 0.44±0.20bc

Phoenicia 39.61±0.35 3956±603 1793±153bcde 0.78±0.30a 3.00±1.59a 0.71±0.20abc

Purple heart 38.97±0.58 4439±697 1942±223ab 0.59±0.13abcd 1.67±0.68ab 0.82±0.59abc

Sakerdze 39.42±0.58 3314±306 1930±378ab 0.26±0.14d 0.96±0.83ab 0.25±0.13c

Wonderful 39.13±0.62 4475±824 1845±153bcd 0.53±0.35abcd 2.01±1.82ab 1.11±0.47ab

Commercial juice m.d. m.d. m.d. 0.77±0.02abc 1.63±0.09ab n.d.

p-value NS NS <0.001 0.001 0.008 <0.001

Cultivar	Antioxidant Activity	Total Phenolics	Potassium	Glutamate	Glutamine	Ethanol
Al Sirin Nar	39.14±0.76^A	4186±912	1999±197ab^B	0.40±0.08abcd	1.34±0.39ab	0.55±0.13bc
Ambrosia	39.42±0.34	3801±425	1491±200cdef	0.61±0.15abcd	2.68±1.64ab	0.71±0.42abc
Blaze	39.22±0.31	4007±440	1903±135abc	0.57±0.11abcd	1.54±0.52ab	1.40±0.79a
Desertnyi	39.30±0.80	3533±224	1419±119def	0.33±0.15bcd	1.01±0.62ab	0.23±0.13c
Eversweet	39.24±0.42	3394±364	1282±114f	0.59±0.21abcd	2.50±0.47ab	0.40±0.18bc
Golden Globe	38.84±1.15	4142±775	1346±91f	0.73±0.32ab	1.99±0.71ab	0.60±0.35abc
Green Globe	37.9	3583	1385	0.8	6.6	n.d.
Haku Botan	39.23±1.08	4387±792	2301±244a	0.28±0.06cd	0.68±0.19b	0.14±0.20c
Loffani	38.6	3301	1851	1.5	7.8	0.21
Parfianka	38.48±1.56	3757±491	1359±226ef	0.56±0.07abcd	2.17±0.42ab	0.44±0.20bc
Phoenicia	39.61±0.35	3956±603	1793±153bcde	0.78±0.30a	3.00±1.59a	0.71±0.20abc
Purple heart	38.97±0.58	4439±697	1942±223ab	0.59±0.13abcd	1.67±0.68ab	0.82±0.59abc
Sakerdze	39.42±0.58	3314±306	1930±378ab	0.26±0.14d	0.96±0.83ab	0.25±0.13c
Wonderful	39.13±0.62	4475±824	1845±153bcd	0.53±0.35abcd	2.01±1.82ab	1.11±0.47ab
Commercial juice	m.d.	m.d.	m.d.	0.77±0.02abc	1.63±0.09ab	n.d.
p-value	NS	NS	<0.001	0.001	0.008	<0.001

^AValues expressed as means±standard deviation; m.d. = missing data; n.d. = not detected. ^BValues followed by different letters within a column are significantly different (P < 0.05). NS = not significant (P > 0.05).

3.3.3 Potassium concentration

Juice potassium (K) levels were on average determined to be 1.71± 0.44 g/L, ‘Haku Botan’ was found to have the highest K concentration in the juice, 2301± 244 mg/L, and ‘Parfianka’ the lowest (1359± 226 mg/L) (Table 4). ‘Wonderful’ had a mean K concentration of 1845± 153 mg/L. Cultivars with significantly lower K than ‘Wonderful’ included ‘Eversweet’, ‘Golden Globe’, ‘Haku Botan’, and ‘Parfianka’. Pomegranate juice is known to have high concentrations of potassium [9]. Additionally, K is known to affect juice flavor and consumer perception of acidity due to its interactions with organic acids, which influences buffering capacity [20]. Because K is a nutrition labeling factor, it is important for industry to know if a cultivar used for human consumption has significantly less K than the industry standard. However, most of the sweet-tart cultivars that could possibly be candidates for commercial juice had similar mean K quantities compared to ‘Wonderful’, which was similar to values reported for ‘Ruby’ pomegranate in [26]. The genotypic variation in K among pomegranate cultivars was also reported in [31], who reported similar, though higher, potassium values for pomegranate juice. The only exception to this was ‘Parfianka’, which had significantly lower quantities of K compared to ‘Wonderful’. For fresh market fruit, it is important to note that sweet or (low-acid) cultivars have less mean K than sweet-tart and tart cultivars. The values of K reported herein demonstrate that pomegranate is a K-rich food and beverage, and can provide a large proportion of the 4.7 g of K suggested as the daily nutritional requirement for adults.

3.3.4 Glutamate and glutamine

The mean concentrations of glutamate ranged from 0.28± 0.06 mM for ‘Haku Botan’ to 1.50 mM for ‘Loffani’ (Table 4). ‘Wonderful’ had mean glutamate concentration of 0.53± 0.35 mM, and the commercial juice had 0.77± 0.02 mM. No cultivars had significantly different glutamate compared to ‘Wonderful’, but there were differences among other cultivars. The mean concentrations of glutamine ranged from 0.68± 0.19 mM for ‘Haku Botan’ to 3.00± 1.59 mM for ‘Phoenicia’. ‘Wonderful’ had mean glutamine concentration of 2.01± 1.82 mM, and the commercial juice had 1.63± 0.09 mM. The single sample of ‘Loffani’ had a glutamine concentration of 7.8 mM. ‘Phoenicia’ had significantly higher glutamine than ‘Haku Botan’. No cultivars were significantly different from ‘Wonderful’ for glutamine.

This study presents the first known data on glutamate and glutamine concentrations in the juices of different pomegranate cultivars as measured by ¹H NMR. Amino acids are important components of the human diet. Glutamic acid can be an important compound in food because it is responsible for the umami taste in foods and can only be detected when it is in its free form in the food product [32]. Additionally, it has been reported that glutamate can also contribute sourness to fruit flavors [20]. The values for glutamate reported for pomegranate herein are greater than values reported for grape juice [33], indicating increased importance for consumers regarding marketing fraud in the pomegranate juice industry. There were significant differences among the cultivars in terms of glutamate and glutamine (Table 4). ‘Loffani’ had the highest concentration of glutamate and glutamine, but there was only one sample tested of this cultivar. ‘Phoenicia’ had significantly higher glutamate concentrations than many cultivars, especially ‘Haku Botan’ and ‘Sakerdze’ (P = 0.001) but was not different than ‘Wonderful’ (Table 4). Juice extraction method had a significant effect on the glutamate concentration (P = 0.0150), with aril-pressed juice (0.58 mM) having more glutamate than peel-pressed juice (0.44 mM) (Fig. 2).

There were significant differences among pomegranate cultivars for glutamine, with ‘Phoenicia’ having significantly higher concentrations than ‘Haku Botan’. ‘Phoenicia’ did not have significantly higher glutamate when compared to ‘Wonderful.’ The juice extraction method also had a significant effect on glutamine (P = 0.015), with aril-pressed juice having greater concentration of glutamine than peel-pressed juice, with mean values of 2.1 and 1.4 mM, respectively, for all cultivars combined. Glutamine is the amino acid of the highest concentration in the human blood [34] and is a nonessential amino acid. The typical blood-red color of pomegranate juice coupled with the nutritional profiles described herein may play into the folklore regarding pomegranate juice as a “blood tonic” [35] and as a medicinal plant [8].

3.3.5 Ethanol

Almost every cultivar had ethanol in fresh juice. Mean ethanol concentration was 0.61± 0.49 mM regardless of cultivar. For the concentration of ethanol, ‘Blaze’ had the highest mean concentration measured (1.40± 0.79 mM), and this was significantly different from ‘Al Sirin Nar’, ‘Desertnyi’, ‘Eversweet’, ‘Haku Botan’, ‘Parfianka’, and ‘Sakerdze’. Ethanol was not detected in ‘Green Globe’ juice and it was also absent in the 100% ‘Wonderful’ commercial juice from concentrate. The mean value of ethanol for fresh squeezed ‘Wonderful’ was 1.1± 0.47 mM.

Ethanol is of importance in the food and beverage industries because it is an indicator of anaerobic respiration and metabolism in postharvest fruit products. Along with other alcohols, ethanol can contribute to off flavors and odors or even enhance the flavor of fruit if concentrations are low [36]. The freshly expressed juice had ethanol values that were generally far less than the levels prohibited by countries for religious and food safety reasons, although some cultures permit no measurable quantity of ethanol in beverages. The mean value for ’Blaze’ was significantly greater than ‘Al Sirin Nar’, ‘Desertnyi’, ‘Evertsweet’, ‘Haku Botan’, ‘Parfianka’, ‘Sakerdze’ (P < 0.001). ‘Wonderful’ ethanol concentration (1.11± 0.47 mM) was approximately an order of magnitude higher than ethanol reported in [3], however, values for all other cultivars were in agreement with previous studies, with ethanol values less than 1.0 mM (Table 4).

3.4 Regression analysis

Regression analysis indicated AA was weakly correlated with TP (P < 0.004, R² = 0.1305) and glutamate was found to be strongly correlated with glutamine (P < 0.001, R² = 0.7475) because they share biochemical pathways, such as the PCA cycle. Malate was correlated with citrate (P < 0.001, R² = 0.478), as they are both in the citric acid cycle. Both glucose and fructose were individually correlated with TSS (P < 0.001, R² = 0.3438, P < 0.001, R² = 0.2180, respectively), which is logical considering the primary disaccharide in pomegranate is sucrose, which would be reduced to its constituents, glucose and fructose when the juice is expressed from the fruit. Glucose was correlated with fructose (P < 0.001, R² = 0.8290).

3.5 Commercial juice

Commercial juice samples analyzed in this study had values similar to freshly pressed ‘Wonderful’ fruit juice for many parameters quantified by ¹H NMR (Tables 2, 3, and 4). However, there were some parameters that were significantly different between commercially-packaged and -sold ‘Wonderful’ fruit juice and freshly pressed pomegranate juice. Both pH and ethanol were significantly different between fresh ‘Wonderful’ juice and commercial juice. This is likely due to processing of the juice, which can include freezing during storage and the addition of pomegranate-based products into the juice. All other measured juice quality traits and metabolites were statistically the same for freshly pressed and commercial ‘Wonderful’ juice.

4 Conclusions

This work demonstrates the high level of phenotypic diversity of pomegranate juices that exists in approximately 5% of the USDA available pomegranate germplasm collection and compares these accessions to the most popular commercial pomegranate juice in the United States. This study is the first of its kind comparing juice quality of the industry standard, ‘Wonderful’, with 13 other NCGR pomegranate cultivars using ¹H NMR coupled with other analytical techniques to assess differences among pomegranate cultivars for metabolic, physicochemical, and nutritional traits. The results indicated a great complexity of juice quality and nutritional differences among the cultivars analyzed, with many fitting the profile of ‘Wonderful’, but others differing greatly from this industry standard. There were striking differences between the commercial juice and fresh-squeezed juices as well as significant differences between juice extraction methods. Potassium, malate, citrate, and GABA concentrations varied widely among cultivars, which can affect the flavor, bioactivity, and nutritional composition of these fruits and their juices. This work also presents further evidence that pomegranate is a dietary source of potassium and for certain nonessential amino acids. As a replacement, alternate or substitute candidate for ‘Wonderful’ in juice markets, ‘Al Sirin Nar’, ‘Blaze’, ‘Desertnyi’, ‘Parfianka’, ‘Phoenicia’, ‘Purple Heart’, and ‘Sakerdze’, all meet a host of juice quality parameters and mostly fit the nutritional profile of ‘Wonderful’. These varieties should be considered for breeding and further investigation via cultivar trials to determine phenotypic traits (phenomes) important to growers (e.g. yield, establishment, pest resistance, drought tolerance, etc.) and sensory panels to determine preferences of consumers.

Funding

The authors report no funding.

Conflict of interest statement

The authors declare that there exists no financial, personal interest or belief that could affect the objectivity of this manuscript. The authors declare that there exists no conflict of interest regarding the publication of this research.

Footnotes

Acknowledgments

The authors would like to acknowledge and thank the USDA-ARS National Clonal Germplasm Repository for Tree Fruit and Nut Crops and Grapes, especially Jeff Moersfelder, for supplying the fruit used in this research. This work was partially supported by the University of California, Riverside Hellman Fellowship to Zhenyu Jia.

The supplementary material is available in the electronic version of this article: .

References

Stover

, Mercure

. The pomegranate: A new look at the fruit of paradise. HortScience. 2007;42(5):1088–92.

Chater

, Merhaut

, Jia

, Arpaia

, Mauk

, Preece

. Effects of site and cultivar on consumer acceptance of pomegranate. J Food Sci. 2018;83(5):1389–95.

Beaulieu

, Lloyd

, Preece

, Moersfelder

, Stein-Chisholm

, Obando-Ulloa

. Physicochemical properties and aroma volatile profiles in a diverse collection of California-grown pomegranate (Punica granatum L.) germplasm. Food Chem. 2015;181:354–64.

Chater

. Register of New Fruit and Nut Cultivars List 47: Pomegranate. HortScience. 2014;49(4):415.

Preece

, Chater

, Moersfelder

. Register of new fruit and nut cultivars list 48: Pomegranate. HortScience. 2016;51(6):645.

Chater

, Merhaut

, Jia

, Mauk

, Preece

. Fruit quality traits of ten California-grown pomegranate cultivars harvested over three months. Sci Hort. 2018;237:11–19.

Day , KR . The California Pomegranate Industry: Trends and Production Insights; 2013. Available from: http://www.crec.ifas.ufl.edu.

Lansky

, Newman

. Punica granatum (pomegranate) and its potential for prevention and treatment of inflammation and cancer. J Ethnopharmacol. 2007;109(2):177–206.

Zhang

, Krueger

, Durst

, Lee

, Wang

, Seeram

, Heber

. International multidimensional authenticity specification (IMAS) algorithm for detection of commercial pomegranate juice adulteration. J Agric Food Chem. 2009;57(6):2550–7.

10.

Silva

. Total ascorbic acid determination in fresh squeezed orange juice by gas chromatography. Food Control. 2005;16(1):55–8.

11.

Justesen

, Knuthsen

, Leth

. Quantitative analysis of flavonols, flavones, and flavanones in fruits, vegetables and beverages by high-performance liquid chromatography with photo-diode array and mass spectrometric detection. J Chromatogr A. 1998;799(1):101–10.

12.

Gil

, Tomás-Barberán

, Hess-Pierce

, Holcroft

, Kader

. Antioxidant activity of pomegranate juice and its relationship with phenolic composition and processing. J Agric Food Chem. 2000;48(10):4581–9.

13.

Kraszni

, Marosi

, Larive

. NMR assignments and the acid–base characterization of the pomegranate ellagitannin punicalagin in the acidic pH-range. Anal Bioanal Chem. 2013;405(17):5807–16.

14.

Cuny

, Vigneau

, Le Gall

, Colquhoun

, Lees

, Rutledge

. Fruit juice authentication by ¹H NMR spectroscopy in combination with different chemometrics tools. Anal Bioanal Chem. 2008;390(1):419–27.

15.

Ogrinc , NK , Košir

, Spangenberg

, Kidrič

. The application of NMR and MS methods for detection of adulteration of wine, fruit juices, and olive oil. A review. Anal Bioanal Chem. 2003;376(4):424–30.

16.

Spraul

, Schütz

, Humpfer

, Mörtter

, Schäfer

, Koswig

, Rinke

. Mixture analysis by NMR as applied to fruit juice quality control. Magn Reson Chem. 2009;47(S1):S130–S137.

17.

Pomeranz

. A comprehensive strategy to overhaul FDA authority for misleading food labels. Am J L & Med. 2013;39(4):617–47.

18.

Belton

, Colquhoun

, Kemsley

, Delgadillo

, Roma

, Dennis

, Sharman

, Holmes

, Nicholson

, Spraul

. Application of chemometrics to the 1 H NMR spectra of apple juices: Discrimination between apple varieties. Food Chem. 1998;61(1):207–13.

19.

Hasnaoui

, Mars

, Ghaffari

, Trifi

, Melgarejo

, Hernandez

. Seed and juice characterization of pomegranate fruits grown in Tunisia: Comparison between sour and sweet cultivars revealed interesting properties for prospective industrial applications. Ind Crops Prod. 2011;33(2):374–81.

20.

Kader

. Flavor quality of fruits and vegetables. J Sci Food Agric. 2008;88(11):1863–8.

21.

Clausen

, Pedersen

, Bertram

, Kidmose

. Quality of sour cherry juice of different clones and cultivars (Prunus cerasus L.) determined by a combined sensory and NMR spectroscopic approach. J Sci Food Agric. 2011;59(22):12124–30.

22.

Tezcan

, Gültekin-Özgüven

, Diken

, Özçelik

, Erim

. Antioxidant activity and total phenolic, organic acid and sugar content in commercial pomegranate juices. Food Chem. 2009;115(3):873–7.

23.

Ogg

, Kingsley

, Taylor

. WET, a T 1-and B 1-insensitive water-suppression method for in vivo localized 1 H NMR spectroscopy. J Magn Reson. 1994;Series B104:1–10.

24.

Claridge

High-resolution NMR techniques in organic chemistry (Vol. 27). Oxford, United Kingdom: Elsevier; 2016. 11–433 P.

25.

Pidgeon

Tutorials for the biomedical sciences: Animations, simulations, and calculations using Mathematica

New York, NY, USA: John Wiley & Sons, Inc; 1996. 227–65.

26.

Fawole

, Opara

. Changes in physical properties, chemical and elemental composition and antioxidant capacity of pomegranate (cv. Ruby) fruit at five maturity stages. Sci Hort. 2013;150:37–46.

27.

Tehranifar

, Zarei

, Nemati

, Esfandiyari

, Vazifeshenas

. Investigation of physico-chemical properties and antioxidant activity of twenty Iranian pomegranate (Punica granatum L.) cultivars. Sci Hort. 2010;126(2):180–5.

28.

Melgarejo

, Salazar

, Artes

. Organic acids and sugars composition of harvested pomegranate fruits. Eur Food Res Techno. 2000;211(3):185–90.

29.

Eisele

, Drake

. The partial compositional characteristics of apple juice from 175 apple varieties. J Food Compos Anal. 2005;18(2):213–21.

30.

Mphahlele

, Stander

, Fawole

, Opara

. Effect of fruit maturity and growing location on the postharvest contents of flavonoids, phenolic acids, vitamin C and antioxidant activity of pomegranate juice (cv. Wonderful). Sci Hort. 2014;179:36–45.

31.

Hepaksoy

, Aksoy

, Can

, Ui

. Determination of relationship between fruit cracking and some physiological responses, leaf characteristics and nutritional status of some pomegranate varieties. Série A. Séminaires Méditerranéens. 2000;42:81–6.

32.

Kurihara

. Glutamate: From discovery as a food flavor to role as a basic taste (umami). Am J Clin Nutr. 2009;90(3):719–22.

33.

Garde-Cerdán

, Arias-Gil

, Marsellés-Fontanet

, Ancín-Azpilicueta

, Martín-Belloso

. Effects of thermal and non-thermal processing treatments on fatty acids and free amino acids of grape juice. Food Control. 200;18(5):473–9.

34.

Brosnan

. Interorgan amino acid transport and its regulation. J Nutr. 2003;133(6):2068–2072S.

35.

Jurenka

. Therapeutic applications of pomegranate (Punica granatum L.): A review. Altern Med Rev. 2008;13(2):128.

36.

Shi

, Goldschmidt

, Goren

, Porat

. Molecular, biochemical and anatomical factors governing ethanol fermentation metabolism and accumulation of off-flavors in mandarins and grapefruit. Postharvest Biol Technol. 2007;46(3):242–51.

Juice quality traits,potassium content,and 1 H NMR derived metabolites of 14 pomegranate cultivars

Abstract

BACKGROUND:

OBJECTIVE:

METHODS

RESULTS

CONCLUSIONS:

Keywords

1 Introduction

2 Materials and methods

2.1 Plant materials

2.3 pH measurement

2.4 Titratable acidity (TA)

2.5 TSS

2.6 Maturity index (MI)

2.7 Antioxidant activity (AA)

Equation 1. Antioxidant activity

2.8 Total phenolics

2.9 1H NMR experimental parameters and quantification steps

2.2.9.2 Equation 3. Calculation of the compound concentration, based on the DSS internal standard concentration, and the T1 corrected area per proton for DSS and the compound

2.10 Ion chromatography

2.11 Statistical analysis

3 Results and discussion

3.1 Acidity

3.1.2 Titratable acidity (TA)

3.1.3 Citrate, malate, and GABA

3.2 Sugars and maturity index

3.2.1 TSS

3.2.3 Glucose and fructose

3.3 Other juice quality traits

3.3.1 Antioxidant activity (AA)

3.3.4 Glutamate and glutamine

3.3.5 Ethanol

3.4 Regression analysis

3.5 Commercial juice

4 Conclusions

Funding

Conflict of interest statement

Footnotes

Acknowledgments

References

2.9 ¹H NMR experimental parameters and quantification steps