Animal-based organic nutrition induces comparable fruit quality to that of inorganic fertigation in soilless-grown grape tomato

ABSTRACT

The aim of the present study was to determine the effect of animal-based organic nutrition and environmental parameters on tomato fruit quality, as well as to establish relations among colour and morphological values performed by the Tomato Analyzer (TA) software application. Organic tomato fruits produced by three organic nutrient solutions, which consisted of different mixtures of several OMRI certified nitrogen fertilizers and one inorganic nutrient solution (Steiner’s solution) as the control, were evaluated for their polyphenol and carotenoid content. We used Tomato Analyzer (TA) to evaluate fruit size and shape. Moreover, we implemented a digital image analysis tool, Color Test (CT), as part of the TA software application to collect and analyse fruit colour parameters. The application of organic fertilizers positively affected the total hydrolysable and condensed polyphenols of tomato fruits compared to the control. The high air temperature (>30°C) and sub-optimal light intensity negatively affected the carotene content of tomato fruits, as well as their morphological and colour attributes. Plants fed with organic solutions containing ASA + ASB + NK2SO4 showed comparable morphology and fruit colour attributes to those of the control plants that received Steiner’s nutrient solution. The results indicated that the application of organic fertilizers positively affected the total hydrolysable and condensed polyphenols of tomato fruits compared to the control. Plants fed with the Steiner’s nutrient solution exhibited the highest carotenoid content in tomato fruits. Organically produced tomatoes through animal-based fertilizer application displayed similar fruit morphology and colour attributes compared to conventionally produced tomatoes.

KEYWORDS:

• Solanum lycopersicum L
• bioactive compounds
• morphology
• colour
• image processing
• temperature

Introduction

Today’s consumers have increased their expectations for the quality of food they purchase. Tomatoes have been ranked first as a source of lycopene (71.6%), second as a source of vitamin C (12.0%), pro-vitamin A carotenoids (14.6%) and other carotenoids (17.2%), and third as a source of vitamin E (6.0%) (Garcia-Closas et al. 2004). Growing interest in polyphenols of tomato fruits is particularly connected with their antioxidant properties and possible positive health effects (Slimestad and Verheul 2009). Lycopene represents the predominant lipid-soluble compound and constitutes more than 80% of total tomato carotenoids in fully red-ripe fruits. β-carotene is of special interest for its provitamin A activity and constitutes nearly 7–10% of total tomato carotenoid synthesis (Nguyen and Schwartz 1999). Tomato fruit quality for fresh consumption (consumer acceptance) is determined by a set of attributes that describe external (size, colour, firmness) and internal (flavour, aroma, texture) properties. Colour is a key component that influences a consumer’s initial perception of quality. In recent years, nondestructive optical methods based on image analysis have been developed for determining the quality of fruits and vegetables because they require less sample preparation, do not disturb the product, and are a cost-effective and rapid technique (Shao et al. 2007). In this context, Tomato Analyzer (TA) provides objective and accurate measurements of several fruit morphological and colorimetric traits in a high-throughput and semi-automatic manner (Rodríguez et al. 2010). The TA software automatically recognises and outlines images of fruits, and the Color Test module (CT) records RGB values of each pixel of the selected object and translates them into average L*, a*, and b* values (Darrigues et al. 2008). The Color Test module implemented in TA is more precise and accurate and less expensive than other methods for analysing fruit colour. Based on the relevance of this issue, coupled with the lack of information about animal-based organic nutrition and its effect on phenotypic characteristics of fruits under greenhouse conditions, the present study aimed to apply image colour analysis for quantification of quality attributes of specialty tomatoes based on colour and shape and their relationship with bioactive compounds and environmental parameters under greenhouse conditions. Currently, no reports exist on the use of phenomic tools such as TA to study fruit shape and colour variation in organic grape tomatoes.

Materials and methods

Crop management and experimental conditions

This work was conducted in a greenhouse at the Universidad Autónoma Agraria Antonio Narro in northern México (lat. 25° 21′ N, long. 101° 02′ W, 1759 m above sea level). Grape tomato seedlings (Solanum lycopersicum L.) cv. Luciplus (Hazera Genetics Ltd.) were transplanted on 25 July 2016. This work was subjected to identical experimental management according to Guajardo-Ríos et al. (2018) to study the influence of animal-based organic nutrition and environmental parameters on morphological and fruit colour attributes, as well as the content of bioactive compounds in soilless-grown grape tomato fruits.

Environmental parameters

The environmental conditions during the experiment, including photosynthetically active radiation (PAR), air temperature (T_a) and relative humidity (RH), inside the greenhouse were measured using a HOBO® U12 data logger (Onset Computer Corp., USA). The daily averaged T_a and RH were 23.0°C and 65.0%, respectively. The photosynthetically active radiation during the daytime was 85.5 μmol m⁻² s⁻¹ and at solar noon was 109.7 μmol m⁻² s⁻¹. This study was performed under sub-optimal light intensity and continuous stressing temperatures. In this context, a high number of days with maximum temperatures ≥ 30°C were recorded in this study. The daily averaged air temperature and light intensity PAR of the cropping cycle are shown in Figure S1.

Fruit sample preparation

Four tomato fruits, selected from each fully red-ripe truss (trusses follow the order of emission in the plant, in which 1st is the first truss to appear and 8th is the last truss), from each evaluated treatment were washed and cut into halves. These organs were dried in a forced-air oven at 70°C for 72 h, and the samples were pulverised and passed through a number 20 sieve (W.S. Tyler, Inc., Mentor, OH).

Analytical RP-HPLC-ESI-MS in fruit samples

Analyses using Reverse Phase-High Performance Liquid Chromatography were performed according to Guajardo-Ríos et al. (2018). In this case, the total hydrolysable polyphenols (THP) in fruit samples were determined using Folin–Ciocalteu reagent (Ascacio-Valdés et al. 2014). The experiment was performed in triplicate, and the total hydrolysable polyphenol content was expressed in gallic acid equivalents (GAE) (as dried weight). Likewise, the total condensed polyphenols (TCP) in fruit samples were determined using a ferric reagent and HCl-butanol (Swain and Hillis 1959). The experiment was performed in triplicate, and the total condensed polyphenol content was expressed in catechin equivalents (CE) (as dried weight).

HPLC-PDA in fruit samples

Carotenoid identification was performed by HPLC (Pursuit XRs 5 C18 column, 150 × 4.6 mm) as described by Hernández-Almanza et al. (2014) with some modifications. Briefly, the analysis was determined by gradients, phase A: acetone and phase B: water (0–3 min: 75% A, 25% B; 3–6 min: 95% A, 5% B; 6–20 min: 95% A, 5% C; 20–22 min: 75% A, 25% B and 22–27 min: 75% A, 25% B), flow 1.0 mL/min and UV detector 450 nm, and the elution programme Varian WorkStation 6.9 was employed (Hernández-Almanza et al. 2017).

Fruit shape characterisation

A step-by-step protocol was used for digitalisation of grape tomato fruit and subsequent semi-automatic analysis of morphology and colour attributes using the Tomato Analyzer (TA) software package version 2.2.0.0 (Rodríguez et al. 2010). Fruit scanning, manual adjustments and morphological analyses by TA were previously reported (Brewer et al. 2006). For each truss, five fruits were harvested at the fully red-ripe stage in each evaluated treatment. Fruits were brought to the laboratory and washed immediately after harvesting. Subsequently, fruits were cut longitudinally through the centre, placed cut-side down on an HP Scanjet G3110 (Hewlett-Packard, Palo Alto, CA, USA) at a resolution of 300 dpi, which was covered with a cardboard box to minimise the effect of shadow and provide a black background, and subsequently subjected to phenotypic analyses. A total of 27 fruit shape traits organised into eight categories within the software: Basic Measurement (6), Fruit Shape Index (2), Blockiness (3), Homogeneity (3), Proximal Fruit End Shape (3), Distal Fruit End Shape (2), Asymmetry (4), and Internal Eccentricity (4), were evaluated (Table 1 and Figure S2).

Table 1. Grape tomato fruit shape descriptors studied and their description.

Basic measurements	Code	Units/description	Fruit shape index	Code	Units/description
Perimeter	per	Perimeter length (cm)	Fruit Shape Index I	fs I	The ratio of the maximum height to maximum width (H/W)
Area	ar	Fruit area (cm^2)	Fruit Shape Index II	fs II	The ratio of mid height to mid width (Hm/Wm)
Width at Mid-height	fl II	The width measured at ½ of the fruit’s height (cm)	Blockiness	Code	Units/description
Widest Width [Maximum Width]	fd I	The maximum horizontal distance of the fruit (cm)	Proximal Fruit Blockiness	pblk	Ratio of fruit width at the proximal end to mid width, (w1/Wm).
Height at Mid-width	fd II	The height measured at ½ of the fruit’s width (cm)	Distal Fruit Blockiness	dblk	Ratio of fruit width at the distal end to mid width, (w2/Wm)
Highest Height [Maximum Height]	fl I	The maximum vertical distance of the fruit (cm)	Fruit Shape Triangle	tri	The ratio of proximal width to distal width (w1/w2).
Homogeneity	Code	Units/description	Proximal fruit end shape	Code	Units/description
Fruit shape circular	cir	Fitting precision (R^2) of the actual shape to a circle; larger values indicate that the fruit is more circular.	Shoulder height	psh	The ratio of the average height of the shoulder points above the proximal end point to Maximum Height
Fruit shape ellipsoid	ell	Fitting precision (R^2) of the actual shape to an ellipse; larger values indicate that the fruit is more ellipsoid	Proximal end angle at 2% and 20%	Pan	The angle from the shoulder points to the site of pedicel attachment or the proximal end, where 180° is flat, greater than 180° indented, and less than 180 is tapered.
Fruit shape rectangular	rec	The ratio of maximum area inscribing the rectangle to the minimum area of the enclosing rectangle, (Sin/Sout).	Proximal fruit end indentation: area	piar	The ratio of the indentation area to the total fruit area.
Asymmetry	Code	Units/description	Internal eccentricity	Code	Units/description
Ovoid	ovo	Calculated according to the formula provided in the tomato Analyzer Manual (Rodríguez et al. 2010). The higher the value, the greater is the area of the fruit above mid height	Eccentricity	e	The ratio of the height of the internal ellipse to the Maximum Height
V. Asymmetry	ver	Fruit shape eccentric vertical asymmetry, ()/number of rows L	Proximal Eccentricity	pe	The ratio of the area of the height of the internal ellipse to the distance between the bottom of the ellipse and the top of the fruit
H. Asymmetry (obovoid)	hob	Fruit shape eccentric horizontal asymmetry, ( )/number of columns L	Distal Eccentricity	de	The ratio of the area of the height of the internal ellipse to the distance between the bottom of the ellipse and the bottom of the fruit
Width Widest Pos	wwp	The ratio of the height at which the maximum width occurs to the maximum height	Internal Fruit Shape Index	fsi	The ratio of the internal ellipse’s height to its width
Distal Fruit End Shape	Code	Units/description
Distal Angle Micro	dan I	Determines where the proximal angle is measured, ranging from 2% to 10% from the tip of the fruit
Distal Angle Macro	dan II	Determines the percentage of the perimeter from the bottom where the angle will be measured, ranging from 5% to 50%.

Note: All traits, were measured with Tomato Analyzer software version 2.2.0.0. Further details for the measurement of fruit shape traits with Tomato Analyzer can be obtained from Brewer et al. (2006), Gonzalo and van der Knaap (2008), and Rodríguez et al. (2010).

Colour test

To obtain colour standards and scanning, a standard 24-colour rendition chart (ColorChecker, X-Rite, Grand Rapids, MI) was used. Each of the 24 patches was considered an individual object and was analysed for colour. We collected RGB data and converted it to estimates of L*, a*, and b* measurements for each patch using TACT (Tomato Analyzer-Color Test) (Darrigues et al. 2008). The Color Test module calculates Hue and Chroma colour descriptors based on a* and b* (Rodríguez et al. 2010). Coordinates a* and b* indicate colour directions: + a* is the red direction, – a* is the green direction, + b* is the yellow direction, and – b* is the blue direction (Darrigues et al. 2008). The L* coordinate indicated the darkness or lightness of the colour and ranged from black (0) to white (100).

Data analysis

The data were evaluated using a one-way analysis of variance (ANOVA) in STATISTICA Version 10 (StatSoft Inc. 2013). An LSD test was applied to establish significant differences between means, with a confidence level of 95%. Additionally, a correlation matrix was performed to show the positive and negative correlations among the evaluated traits included in the study.

Results and discussion

Polyphenol content

The total hydrolysable and condensed polyphenols were higher in all of the organic treatments than in the control plants (p ≤ 0.05) (Table 2). All of the organic treatments exhibited more hydrolysable phenols than the control, from 5.37 to 6.65 mg, with an average of 6.2 mg of GAE/g (Table 2). Similarly, organic treatments exhibited more condensed phenols than the control treatment, from 150.85 to 218.54 mg of CE/g, with an average of 175.51 mg of CE/g (Table 2). This result was concordant with data obtained by Guajardo-Ríos et al. (2018), who found a higher level of total phenols in organically fertilised tomatoes compared to conventional tomatoes. Most studies that report measurements of the total phenolic content describe a higher phenolic concentration in organically grown fruits or vegetables. Our results are in accordance with these studies because organic tomatoes showed a higher content of polyphenols (THP and TCP) than conventional tomatoes (Table 2). On the other hand, the higher the air temperature, the higher the THP and TCP contents. In this study, the maximum biosynthesis of hydrolysable and condensed phenols occurred at 38.4°C–14.3°C (daily averaged temperature) (Table 3).

Table 2. Effect of organic and inorganic fertilisation on total hydrolysable and condensed polyphenols, lycopene and β-carotene content in grape tomato fruits under greenhouse conditions.

Treatment	Total hydrolysable polyphenols (mg GAE/g)	Total condensed polyphenols (mg CE/g)	Lycopene (mg/100 g)	β–carotene (mg/100 g)
(I) ASA + AL + NK2SO4	6.65 ± 0.82 a*	157.15 ± 12.24 b	49.66 ± 17.80 ab	34.01 ± 26.28 bc
(II) ASA + AL + NK2SO4 + MO	6.58 ± 0.66 a	218.54 ± 22.04 a	24.88 ± 3.67 b	28.00 ± 31.20 c
(III) ASA + ASB + NK2SO4	5.37 ± 2.61 b	150.85 ± 23.08 b	59.60 ± 11.71 a	81.81 ± 31.93 ab
(IV) Control	3.98 ± 1.18 c	117.19 ± 10.21 c	76.71 ± 45.90 a	100.81 ± 78.34 a
LSD (0.05)	1.15	15.89	27.47	50.28

Note: Description for each treatment (composition of nutrient solutions) according to Guajardo-Ríos et al. (2018).

*Means with the same letter letters indicate no statistically significant differences among treatments using the LSD test (p ≤ 0.05).

Table 3. Effect of organic and inorganic fertilisation on polyphenols content in relation to harvest date for each fully-red ripe truss under greenhouse conditions.

	Tomato fruit harvesting/Air temperature (max-min)
	1st truss	2nd truss	3rd truss	4th truss	5th truss	6th truss	7th truss	8th truss
Bioactive compound	Sept 16 [40.7°C–13.9°C]	Sept 24 [38.4°C–14.3°C]	Sept 30 [34.0°C–11.4°C]	Oct 08 [31.9°C–12.6°C]	Oct 16 [35.0°C–12.6°C]	Oct 23 [36.2°C–10.5°C]	Oct 30 [32.0°C–10.6°C]	Nov 07 [34.9°C–11.0°C]
THP	5.71 ± 1.23 b*	8.35 ± 2.01 a	5.37 ± 1.47 b	6.09 ± 1.15 b	4.81 ± 1.56 b	4.99 ± 1.52 b	4.66 ± 2.28 b	5.18 ± 1.46 b
TCP	161.52 ± 42.13 bc	184.94 ± 43.09 a	160.28 ± 35.41 bc	154.32 ± 41.46 bc	169.03 ± 53.63 ab	156.15 ± 59.35 bc	144.38 ± 32.31 c	156.83 ± 43.01 bc

Notes: THP = Total hydrolysable polyphenols (mg GAE/g), TCP = Total condensed polyphenols (mg CE/g). The daily averaged temperature (max-min), was calculated between each harvest period. For the first truss was only considered the daily average temperature beginning with fruit setting period.

*Means with the same letter letters within rows indicate no statistically significant differences using the LSD test (p ≤ 0.05).

Carotenoid content

Significant differences (p ≤ 0.05) were observed in lycopene and β-carotene content in tomato fruits between the conventional and organic treatments. Plants fed Steiner’s nutrient solution exhibited the highest carotenoid content in tomato fruits (Table 2). In this case, plants fed with solutions containing the organic treatments averaged 44.71 and 47.94 mg/100 g of lycopene and β-carotene content, respectively, whereas control plants recorded 76.71 and 100.81 mg/100 g for each bioactive compound at the end of the experiment, which represents an increase of 41.72% and 52.45%, respectively. All of the evaluated treatments showed higher β-carotene content in their fruits, except for treatment I (ASA + AL + NK₂SO₄), which presented higher lycopene biosynthesis (Table 2). On the other hand, no significant differences were observed in carotenoid content of grape tomatoes among the harvesting dates. However, β-carotene tended to increase with increasing lycopene content, but not to the same degree. Moreover, the higher the air temperature, the higher the β-carotene content. In this context, Brandt et al. (2006) reported that only β-carotene might be synthesised above 30°C, which has a ceiling temperature of 38°C. In this study, the daily maximum air temperature (T_a) during the cropping cycle ranged between 31.9°C and 40.7°C, and the maximum biosynthesis of carotenoid content occurred at 34°C. This prolonged, extremely high temperature may have led to the diminishing of the lycopene content among the harvesting dates. This result was concordant with data obtained by Kuti and Konuru (2005), who evaluated 40 tomato varieties under greenhouse and field conditions; a lower lycopene content was reported for cherry tomatoes grown in the greenhouse because of temperatures over 32°C in most cases.

Morphological fruit descriptors

Significant differences (p ≤ 0.05) were observed in ell, pan (2%), piar, hob, e, and pe attributes between the conventional and organic treatments (Table 4 and Figure S2). On the other hand, fl II, fd I, fs I, fs II, pblk, dblk, tri, cir, rec, psh, pan (20%), dan I, dan II, ovo, ver, wwp, de, and fsi exhibited no significant effects between all of the evaluated treatments (Figure S2). Highly significant differences (p ≤ 0.001) were found for the basic fruit shape measurements (per, ar, fd II, and fl I). In this context, Treatments I (ASA + ASB + NK₂SO₄) and IV (Control) showed similar basic morphological features (per, ar, fd II, fl I) (Table 4). Furthermore, highly significant differences (p ≤ 0.001) were observed among all of the harvested fruits for per, ar, fl II, fd I, fd II, and fl I (Table 5 and Figure S2). No significant differences in fs I and fs II were found (Figure S2). Nevertheless, for these fruit shape descriptors, organic treatments and controls showed values greater than 1, which indicated elongated fruits. For pblk, dblk, and tri, no significant effects among the evaluated treatments were detected. However, the seventh truss was the least blocky (i.e. more tapered), whereas the second truss was the most blocky (dblk) (Table 5 and Figure S2). A fruit shape triangle (tri) value greater than 1 indicates that the proximal end of the fruit is wider than the distal end of the fruit, while a value less than 1 indicates that the distal end of the fruit is wider (Brewer et al. 2006). In this context, all of the harvested fruits tended to be wider at the proximal end of the fruit than at the distal end. For pan and piar descriptors, significant differences among the evaluated treatments were found, although at a lower level of significance (p ≤ 0.05). Treatment II (ASA + AL + NK₂SO₄ + MO) showed fruits more tapered (<180°) than treatment IV (control). In contrast, ASA + AL + NK₂SO₄ had fruits that were more flat (almost 180°) (p ≤ 0.05). For shoulder height (psh) and fruit shape rectangular (rec), no significant differences were detected among treatments or harvesting dates. No significant differences in dan I and dan II were found among the treatments evaluated. Nevertheless, dan I from the tip of the fruit clearly differentiated the seventh truss from the other ones; that is, it was highly pointed (Table 5). The horizontal asymmetry ovoid descriptor exhibited a value equal to 0 in all of the treatments; that is, there was more area above the horizontal axis n than below it in all of the harvest fruits; thus, we only considered horizontal asymmetry obovoid (hob) fruit descriptor in fruit phenotypic characterisation (Brewer et al. 2006). In this case, significant differences (p ≤ 0.05) were found among the four nutrient solutions evaluated for the hob descriptor (Table 4 and Figure S2). Thus, obovoid values indicated that the largest width of the fruit was well below the midpoint of the fruits (Gonzalo et al. 2009). For e and pe descriptors, significant differences among the evaluated treatments were found (p ≤ 0.05) (Table 4 and Figure S2). No significant differences for de and fsi descriptors were found among the treatments evaluated.

Table 4. Effect of organic and inorganic fertilisation on tomato fruit morphology attributes under greenhouse conditions.

Fruit shape descriptors	Treatment
Basic measurements	(I) ASA + AL + NK2SO4	(II) ASA + AL + NK2SO4 + MO	(III) ASA + ASB + NK2SO4	(IV) Control
per	9.49 ± 0.94 ab*	9.39 ± 0.87 b	9.69 ± 0.61 ab	9.86 ± 0.87 a
ar	6.24 ± 1.05 ab	6.18 ± 1.01 b	6.42 ± 0.63 ab	6.74 ± 1.12 a
fd II	3.07 ± 0.29 ab	3.05 ± 0.27 b	3.17 ± 0.23 ab	3.24 ± 0.28 a
fl I	3.12 ± 0.29 b	3.09 ± 0.27 b	3.22 ± 0.23 ab	3.29 ± 0.28 a
Homogeneity
ell	0.79 ± 0.004 ab	0.79 ± 0.01 a	0.79 ± 0.005 ab	0.78 ± 0.004 b
Proximal fruit end shape
pan (2%)	179.63 ± 6.58 a	172.94 ± 6.98 b	177.33 ± 5.11 ab	179.85 ± 5.38 a
piar	0.004 ± 0.004 ab	0.001 ± 0.002 b	0.003 ± 0.002 ab	0.006 ± 0.01 a
Asymmetry
hob	10.02 ± 3.52 ab	8.12 ± 1.52 b	9.76 ± 1.94 ab	10.59 ± 1.67 a
Internal eccentricity
E	0.79 ± 0.004 ab	0.79 ± 0.01 a	0.79 ± 0.005 ab	0.78 ± 0.004 b
Pe	0.89 ± 0.001 a	0.89 ± 0.001 ab	0.89 ± 0.001 ab	0.89 ± 0.001 b

Notes: Description for each fruit shape atributes as given in Table 1. Description for each treatment (composition of nutrient solutions) according to Guajardo-Ríos et al. (2018).

*Means with the same letter letters within rows indicate no statistically significant differences using the LSD test (p ≤ 0.05).

Table 5. Effect of organic and inorganic fertilisation on grape tomato fruit descriptors in relation to harvest date for each fully-red ripe truss under greenhouse conditions.

	Tomato fruit harvesting/Air temperature (max-min)
	1st truss	2nd truss	3rd truss	4th truss	5th truss	6th truss	7th truss	8th truss
Fruit shape descriptors	Sept 16 [40.7°C–13.9°C]	Sept 24 [38.4°C–14.3°C]	Sept 30 [34.0°C–11.4°C]	Oct 08 [31.9°C–12.6°C]	Oct 16 [35.0°C–12.6°C]	Oct 23 [36.2°C–10.5°C]	Oct 30 [32.0°C–10.6°C]	Nov 07 [34.9°C–11.0°C]
per	10.51 ± 0.49 a*	10.48 ± 0.47 a	9.54 ± 0.41 c	10.15 ± 0.32 ab	9.22 ± 0.55 c	9.71 ± 0.29 bc	8.62 ± 0.14 e	8.63 ± 0.62 de
ar	7.49 ± 0.67 a	7.50 ± 0.74 a	6.07 ± 0.46 c	6.91 ± 0.43 ab	5.88 ± 0.64 bc	6.40 ± 0.36 ab	5.08 ± 0.24 d	5.83 ± 0.74 cd
fl II	2.68 ± 0.12 a	2.70 ± 0.14 a	2.46 ± 0.05 b	2.64 ± 0.06 a	2.39 ± 0.13 b	2.45 ± 0.07 b	2.19 ± 0.11 c	2.16 ± 0.11 c
fd I	2.71 ± 0.12 a	2.73 ± 0.14 a	2.50 ± 0.06 b	2.68 ± 0.07 a	2.42 ± 0.13 b	2.49 ± 0.08 b	2.22 ± 0.11 c	2.19 ± 0.10 c
fd II	3.42 ± 0.21 a	3.40 ± 0.17 a	3.06 ± 0.23 bc	3.24 ± 0.16 ab	3.03 ± 0.17 bcd	3.21 ± 0.16 ab	2.89 ± 0.04 cd	2.80 ± 0.22 d
fl I	3.49 ± 0.20 a	3.44 ± 0.18 a	3.11 ± 0.24 bc	3.28 ± 0.17 ab	3.06 ± 0.19 bcd	3.26 ± 0.14 ab	2.94 ± 0.04 cd	2.86 ± 0.22 d
dblk	0.55 ± 0.02 ab	0.58 ± 0.02 a	0.54 ± 0.04 ab	0.54 ± 0.03 ab	0.52 ± 0.02 bc	0.52 ± 0.04 bc	0.48 ± 0.04 c	0.53 ± 0.03 ab
dan I	148.30 ± 8.82 ab	152.65 ± 4.95 a	133.89 ± 15.90 ab	145.20 ± 10.15 ab	144.17 ± 5.24 ab	150.81 ± 31.32 a	129.17 ± 7.61 b	138.91 ± 6.19 ab

Notes: Description for each fruit shape atributes as given in Table 1. The daily averaged temperature (max-min), was calculated between each harvest period. For the first truss was only considered the daily average temperature beginning with fruit setting period.

*Means with the same letter letters within rows indicate no statistically significant differences using the LSD test (p ≤ 0.05).

Fruit colour attributes

No significant differences for L*, a*, b*, Hue and chroma values were found among the treatments evaluated; however, significant differences (p ≤ 0.05) were observed among all of the harvested dates (Table 6). The increase in the a* value is known to be directly associated with lycopene synthesis, contrary to L* values, which decrease at the full-ripe stage (loss of greenness). In this case, our results exhibited averaged L*, a*, and b* values from 47.18 to 47.24, 18.63 to 19.23, and 28.74 to 29.46 (organic treatments versus inorganic treatments, respectively). For instance, the a* value was significantly higher in the sixth and seventh truss (p ≤ 0.05), which indicates more red colouration intensity (redness) (Table 6). On the other hand, the hue parameter, the most usable colour index of the CIELab colour system, is also closely correlated to the lycopene content of tomato fruits (Hertog et al. 2007). In this context, the hue value of tomato fruits was significantly lower (more red colour) in the fifth truss than in the second truss. A hue of 180° represents pure green and a hue of 0°, pure red (Shewfelt et al. 1988). We also found significant differences (p ≤ 0.05) among the harvesting dates relative to the lightness value (L*) (Table 6).

Table 6. Effect of organic and inorganic fertilisation on fruit colour atributes in relation to harvest date for each fully-red ripe truss under greenhouse conditions.

	Tomato fruit harvesting/Air temperature (max-min)
	1st truss	2nd truss	3rd truss	4th truss	5th truss	6th truss	7th truss	8th truss
Fruit colour atributes	Sept 16 [40.7°C–13.9°C]	Sept 24 [38.4°C–14.3°C]	Sept 30 [34.0°C–11.4°C]	Oct 08 [31.9°C–12.6°C]	Oct 16 [35.0°C–12.6°C]	Oct 23 [36.2°C–10.5°C]	Oct 30 [32.0°C–10.6°C]	Nov 07 [34.9°C–11.0°C]
L*	47.20 ± 1.40 b	49.20 ± 0.63 a*	45.86 ± 0.38 bc	47.51 ± 0.52 ab	44.54 ± 0.60 c	46.91 ± 1.90 b	49.16 ± 1.87 a	47.19 ± 1.28 b
a*	18.77 ± 1.01 ab	18.68 ± 2.70 b	16.63 ± 0.58 c	18.44 ± 0.73 bc	18.78 ± 0.67 ab	20.68 ± 0.36 a	19.41 ± 1.59 ab	18.87 ± 0.76 ab
b*	28.91 ± 1.61 bc	29.36 ± 0.77 bc	26.60 ± 0.37 e	28.40 ± 1.32 cd	27.05 ± 0.39 de	30.41 ± 0.92 ab	31.68 ± 1.37 a	28.98 ± 1.53 bc
Hue	58.37 ± 2.09 ab	60.49 ± 2.11 a	60.64 ± 0.83 a	58.23 ± 0.44 ab	54.71 ± 1.38 c	57.28 ± 0.87 bc	58.84 ± 3.20 ab	58.26 ± 1.96 ab
Chroma	35.25 ± 1.64 cd	35.85 ± 2.03 bc	32.38 ± 0.64 e	34.61 ± 0.62 cd	33.65 ± 0.50 de	37.38 ± 0.99 ab	37.65 ± 0.30 a	35.34 ± 1.43 cd

Note: The daily averaged temperature (max-min), was calculated between each harvest period. For the first truss was only considered the daily average temperature beginning with first fruit setting period.

*Means with the same letter letters within rows indicate no statistically significant differences using the LSD test (p ≤ 0.05).

Correlations between variables

A Pearson correlation matrix for the 36 variables showed both high positive and negative correlations among the traits included in the analysis. In this case, a negative significant correlation was observed between THP and the basic fruit shape traits (per, ar, fd II, and fl I). Moreover, the air temperature (T_a) exhibited a significant positive correlation with these basic fruit shape measurements throughout the entire harvesting period. On the other hand, TCP content exhibited a high correlation with lycopene content, with a negative and significant value (p ≤ 0.05). In contrast, we found a strong positive correlation between TCP and fruit eccentricity at p ≤ 0.05. Furthermore, a significant negative correlation (r = −0.996) was found between lycopene content and e (p ≤ 0.05), indicating that higher eccentricity values were related to lower lycopene content. There was a linear relationship between lycopene and β-carotene content with a* and b*, respectively. In this context, Sacks and Francis (2001) reported that these bioactive compounds are characterised by a* and b* in the CIELab colour system. Lycopene and β-carotene exhibited high positive correlations with per (p ≤ 0.05). Moreover, β-carotene exhibited a strong negative and positive correlation (r = −0.97 and 0.98, respectively) with two internal eccentricity descriptors (de and fsi) (Figure S2). On the other hand, width at mid-height (fl II) and maximum width (Fd I) exhibited a strong positive correlation with chroma (r = 0.98), which indicates higher saturation or vividness at the midpoint of the fruit.

In conclusion, we can state that organically produced tomatoes exhibited higher levels of polyphenols compared to conventionally produced tomatoes, which may offer potential health benefits. All of the fruit shape descriptors were highly influenced by air temperature throughout the entire harvesting period. Plants fed with organic solutions containing ASA + ASB + NK₂SO₄ showed comparable morphology and fruit colour attributes, as well as a comparable carotenoid content, as that of the control plants that received Steiner’s nutrient solution; therefore, we conclude that organic fertilisation, mainly using ASA, ASB and NK₂SO₄, may be a potential substitute for inorganic fertilisation in terms of fruit quality.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Dr. Carlos Javier Lozano-Cavazos is a full-time professor and researcher in the Department of Plant Breeding (Protected Horticulture Area) at the Universidad Autónoma Agraria Antonio Narro, Saltillo, Coah. His research is mainly focused on hydroponic cultivation of greenhouse vegetable species, organic plant nutrition and crop biofortification.

Dr. Luis A. Valdez-Aguilar is a senior full-time researcher and professor at Universidad Autónoma Agraria Antonio Narro. His research is focused on the hydroponic cultivation of greenhouse ornamental and vegetable species, nutrient interactions, design of fertilization programs for ornamentals and hydroponic systems and water quality for irrigation of greenhouse crops. He received his PhD in horticulture at Texas A&M University in 2004 and has published 76 papers.

Dr. Luis Ibarra-Jiménez is a professor-researcher in the Department of Agricultural Plastics at the Centro de Investigación en Química Aplicada, Saltillo, Coah, México, since 1980 until now. His research is oriented to vegetable production and physiology, especially with Solanaceous and Cucurbit species, including plastic mulches, row covers/tunnel, high tunnels, and shade house conditions. He has published 60 refereed articles.

Dr. Juan A. Ascacio-Valdés is a full professor in the Food Research and Technology Department, School of Chemistry, Universidad Autónoma de Coahuila, Saltillo, Coahuila, México. His experience area is in production, recovery, purification, identification and characterization of bioactive compounds using bioprocesses.

Dr. Adalberto Benavides-Mendoza is a researcher in the Department of Horticulture at the Universidad Autónoma Agraria Antonio Narro. His research interest is led towards plant nutrition, with a focus on the nutritional quality of vegetables and fruits and crop biofortification.

Dr. Cristobal N. Aguilar-González is a researcher in the Food Research Department (School of Chemistry) at the Universidad Autónoma de Coahuila, Saltillo Campus, Coahuila, México. His research is oriented towards bioprocesses and bioproducts; this includes microbial, enzymatic and chemical technologies for production/recovery of bioactive compounds for agro-industrial, pharmaceutical, cosmetic and food industries.

Dr. Oscar Guajardo-Ríos is a Doctoral graduate of the Universidad Autónoma Agraria Antonio Narro in protected agriculture, and worked on the effect of organic fertilization on yield and quality of tomato under greenhouse conditions.

ORCID

Carlos Javier Lozano-Cavazos http://orcid.org/0000-0003-2838-8476

Luis Alonso Valdez-Aguilar http://orcid.org/0000-0002-2510-1962

Luis Ibarra-Jiménez http://orcid.org/0000-0002-2070-756X

Juan Alberto Ascacio-Valdés http://orcid.org/0000-0001-6595-863X

Adalberto Benavides-Mendoza http://orcid.org/0000-0002-2729-4315

Cristóbal Noé Aguilar-González http://orcid.org/0000-0001-5867-8672

Additional information

Funding

This work was supported by PRODEP-SEP (Programa para el Desarrollo Profesional Docente, para el Tipo Superior-Secretaría de Educación Pública) under [grant number DSA/103.5/14/11033].