Introduction

The guava fruit (Psidium guajava L.) is a member of the Myrtaceae family. Due to its abundance in vitamins C, A, and B, as well as minerals, antioxidants, and dietary fiber, it is sometimes referred to as the “Apple of the tropics” or “Poor man’s Apple” (Yousaf et al. 2020). Owing to their distinct flavor and high nutrient content, fresh guava fruits and their value-added products are in high demand in both domestic and international markets. Guava fruits contain numerous phytochemicals and bioactive substances, including ascorbic acid, catechin, apigenin, naringenin, kaempferol, lycopene, quercetin, and myrcetin. These compounds endow guava with various medicinal properties, such as antioxidant, anticancer, antidiabetic, antipyretic, antimicrobial, and hepatoprotective activities, as previously reported (Suwanwong and Boonpangrak 2021; Mathiazhagan et al. 2023). Carotenoids and anthocyanins are important antioxidant compounds with nutraceutical properties. Fruits and vegetables are the main sources of these naturally occurring coloring pigments (Mathiazhagan et al. 2021). Fruits such as tomatoes, guavas, watermelons, papayas, and mangoes have been found to contain higher levels of carotenoids and lycopene (Nwaichi et al. 2015; Karanjalker et al. 2018), whereas blueberries, strawberries, and red grapes are known to contain higher levels of anthocyanins (Fang 2015). Studies indicate that pink and red-flesh guavas have higher carotenoid and lycopene contents compared to white-flesh varieties (Singh et al. 2019), whereas purple-flesh cultivars, such as ‘Thai Maroon’ (Flores et al. 2015) and ‘Black guava’ (Senapati 2022), are good sources of anthocyanins. Consequently, colored pulp guava varieties are prized for their visual allure and nutritional richness in the market. From the perspective of the breeder, a streamlined approach for determining pigment composition to optimize time and cost efficiency is preferred, especially when handling large sample numbers. CIELAB (International Commission on Illumination) color space values (L*, a*, b*, chroma C*, and hue angle h°), measured with a color reader, for quantitatively assessing fruit color has been a common practice across various studies. These parameters have shown correlations with important pigments like carotenoids in a range of fruits including tomatoes (Pandurangaiah et al. 2020), watermelon (Ibrahim et al. 2022), papaya (Schweiggert et al. 2011), apricot (Ruiz et al. 2005), jackfruit (Saxena et al. 2012), and pumpkin (Itle and Kabelka 2009). Such correlations provide insights into the relationship between fruit color and its biochemical composition, especially regarding pigments like lycopene and carotenes.

The advantage of using color space values, such as a* (redness-greenness) or hue angle (h°), lies in their direct or indirect association with pigment content, which can help in evaluating fruit genotypes with high pigment content efficiently. By employing color space values, researchers can potentially bypass the need for labor-intensive spectrometric methods, making the assessment process quicker and more accessible. This approach highlights the significance of colorimetric analysis in fruit quality evaluation, providing a non-destructive and rapid method for assessing important biochemical constituents like carotenoids.

Earlier studies on guava utilized path coefficient and correlation analysis to evaluate the relationship and impact of different fruit characteristics on a dependent trait of interest, such as yield (Gupta and Kour 2019) and total antioxidant content (Kherwar and Usha 2016). Further, colorimeter has been employed to quantitatively estimate the color variations in guava mapping populations (Mathiazhagan et al. 2024). To our knowledge, colorimetric studies establishing association between color space values and pigment content in the guava fruits are not available. Here, an attempt has been made to understand the variation and the degree of contribution of various pigments to color space values in guava fruit pulp. For the first time correlation between colorimetric parameters (L*, a*, b*, C, and h°) and the total carotenoid, lycopene and anthocyanin contents has been established. This study examines whether color analysis can be used as an alternate screening method replacing cumbersome spectrometric methods in guava. Additionally, path coefficient analysis was performed to further explore the direct and indirect effects of various pigments on the color space a* value. The results of the correlation and path analyses are discussed here.

Materials and Methods

Plant Materials

In this study, 35 diverse guava genotypes were used, consisting of 14 germplasm and 21 hybrid progenies featuring different pulp colors (Fig. 1a). The pedigree/origin of the guava genotypes used in this study are presented in Table 1. These guava trees, aged between 3 and 6 years, were maintained in the Fruit Breeding Block of the ICAR-Indian Institute of Horticultural Research, located at Bengaluru, India, following the recommended cultural practices. To evaluate the color of the fruit pulp and quantify the levels of total carotene, lycopene, and anthocyanins, fruit samples were taken in triplicate from each genotype during the months of March and May, 2023, when the fruits were at edible ripe stage. This timing ensured that the samples were representative of the mature and edible fruit stage, allowing for accurate evaluation of color characteristics and pigment content.

Fig. 1
figure 1

a Pulp color variations of hybrid fruits of ‘Arka Poorna’ (AP) × ‘Purple Local’ (PL) used in the study. b Coordinates of CIELAB color space values

Full size image
Table 1 Genotypes and hybrids used in the study
Full size table

Color Evaluation Using Color Reader

The color variations in the freshly harvested edible ripe fruits from the 35 guava genotypes were quantitatively assessed using a color reader (NR 200 portable colour reader, 3NH Technologies Co. Ltd. China). Transverse cross-sections of the guava fruits were utilized for this purpose. A total of 35 of each genotype was evaluated in triplicate, and the average of the replicates was used to ensure accuracy and reliability in the comparisons among the samples. The color parameters were measured in the CIELAB color space, with attributes viz. L*, a*, b* and the derived C* and h° values recorded. L* represents the lightness with (+) denoting lightness and (−) denoting darkness; a* represents the red (+) or green (−) color componenet; b* represents the yellow (+) or (−) blue color component; C* represents the differences in chroma with (+) denoting brighter and (−) denoting lighter color and; h° represents the hue angle, indicating the dominant color tone (Fig. 1b). Chroma (C*) and hue (h°) are derived from a* and b* values using the equations, C* = √(a*)2 + (b*)2 and h° = arctan (b*/a*), respectively, where, C* ≥ 0; and 0 ≤ h° < 360 degrees (Azetsu and Suetake 2021).

Estimation of Total Carotene and Lycopene Contents

Total carotenoids (TC) and lycopene (TL) contents in guava fruits were estimated following the protocol by Lichtenthaler (1987). A volume of 5 g of pulp, without the peel, was ground and extracted in acetone. The extracts were separated using 15 ml of hexane and 100 ml of water in a separating funnel. The lower aqueous phase was re-extracted with hexane until it became colourless. Anhydrous Na2SO4 was added to the upper layer and the volume made up to 25 ml. The absorbance was read at 470 nm and 503 nm for carotenoids and lycopene, respectively, using an UV-Vis Spectrometer (T80+ UV/Vis Spectrometer, PG Instruments Ltd., UK). The contents of carotenoid and lycopene in the fruit pulp were estimated using the following formulae:

$$\text{Total}\,\text{carotenoids }\left(\mathrm{mg}/100\,\mathrm{g}\right)=\frac{\begin{array}{c}A_{470\mathrm{nm}}\times \mathrm{Std}.\text{value}\,\left(\upmu \mathrm{g}/A_{470\mathrm{nm}}\right)\\ \times \text{Total vol}.\text{of extract}\times 100\end{array}}{\text{Weight of the sample}\,\left(\mathrm{g}\right)\times 1000}$$
$$\text{Total}\,\text{lycopene }\left(\mathrm{mg}/100\,\mathrm{g}\right)=\frac{\begin{array}{c}A_{503\mathrm{nm}}\times \mathrm{Std}.\text{value}\,\left(\upmu \mathrm{g}/A_{503\mathrm{nm}}\right)\\ \times \text{Total vol}.\text{of extract}\times 100\end{array}}{\text{Weight of the sample}\,\left(\mathrm{g}\right)\times 1000}$$

Estimation of Total Anthocyanin Content

Total anthocyanin (TA) content in the guava fruits was estimated from 5 g of fruit pulp devoid of peel. The samples were extracted using acidified methanol containing 1% HCl as suggested by Fuleki (1969). The extracts were centrifuged at 10000 rpm for 10 min. The absorbance of the supernatant was read at 540 nm using an UV-Vis Spectrometer. The anthocyanin content of fruit pulp was estimated using the following formula:

$$\text{Total}\,\text{anthocyanin }\left(\mathrm{mg}/100\,\mathrm{g}\right)=\frac{\begin{array}{c}A_{540\mathrm{nm}}\times \mathrm{Std}.\text{value}\,\left(\upmu \mathrm{g}/A_{540\mathrm{nm}}\right)\\ \times \text{Total vol}.\text{of extract}\times 100\end{array}}{\text{Weight of the sample}\,\left(\mathrm{g}\right)\times 1000}$$

Statistical Analysis

Correlation coefficient analysis was performed at both the genotypic and phenotypic levels to assess the strength and direction of the relationship between the color space values (L*, a*, b*, C*, h°) and the total pigment contents (carotenoids, lycopenes, and anthocyanins). A linear regression analysis was performed using Microsoft Excel to model the relationship between color space values and total pigment contents. A path analysis was carried out to determine the direct and indirect effects of various pigments in guava over the color space a* value. Color a* value was regarded as a dependent or resultant variable, while the total carotenoid, lycopene, and anthocyanin contents were considered independent or causal variables. Analysis of variance, correlation and path analysis were carried out using the R program “variability” package (Popat et al. 2020).

Results

Fruit Color and Pigment Content Evaluation

Quantitative estimations based on color readings (L*, a*, b*, C*, and h°) provided valuable insights into fruit color variations among the guava genotypes studied. Due to the varied intensity in observed pulp color, color space values showed substantial variation (P < 0.01) between the genotypes. The a* values ranged from 0.43 to 37.10; however, the L* values varied from 34.47 to 80.20 (Supplementary Table 1). The pulp color b* values varied from 8.70 to 27.20. The hue angle (h°) ranged from 14.20 to 88.93, while the chroma C* values were measured between 11.48 and 47.84. Fruits with colored pulp recorded high a* and C* values, indicating vivid and intense coloration, while fruits with white pulp recorded high L* and h° values, suggesting lighter and less saturated hues. The b* values did not show significant differences between the genotypes. The guava pulp TC, TL, and TA concentrations were expressed in mg/100 g of fresh weight and varied considerably among the guava genotypes. The colored pulped genotypes used in the study recorded pigment concentration as high as 9.95 mg/100 g, 4.00 mg/100 g, and 10.80 mg/100 g of fresh weight for total carotenoids, total lycopene, and total anthocyanin, respectively. Among the genotypes examined, white pulped fruits generally exhibited lower levels of anthocyanin, lycopene, and carotenoid concentrations, while red and purple pulped genotypes showed the highest TC, TL, and TA contents. The analysis of data on total carotenoids, lycopenes, and anthocyanins showed coefficients of variation (CV%) of 113.01%, 108.67%, and 176.49%, respectively (Table 2). The CV% for color space values ranged from 21.18% (L* value) to 73.33% (a* value).

Table 2 Analysis of variance and mean estimates for different pigment contents and color space values in guava pulp
Full size table

Correlation Between Color Space Values and Pigment Contents

The correlation analysis, which included genotypic (rg) and phenotypic (rp) correlation coefficients, revealed that the pigments (TC, TL, and TA) had a strong positive correlation with that of the color space values a* and C* values (Table 3). The genotypic correlation coefficient for TC showed a positive correlation (P < 0.01) between a* (0.81) and C* (0.75). Likewise, the TL content showed a strong positive correlation with a* (0.86 at P < 0.01) and C* values (0.76 at P < 0.01), whereas TA content presented a moderately positive correlation with a* (0.52 at P < 0.01) and C* (0.43 at P < 0.05). Furthermore, there was a significant negative correlation between TA and b*, although no meaningful relationship was found between the b* value and TC or TL. All three pigment contents showed a significant negative correlation with hue angle (h°) and lightness of color (L*). With slight variations, similar trends were seen for the phenotypic correlation coefficient between the pigments and the color values. The regression analysis, indicated in terms of correlation of determination (R2), showed a positive linear relation between the color parameter a* and TC (0.70), TL (0.83) and TA (0.24). A similar pattern was observed for the chroma C*, which exhibited a positive linear relationship with the three pigment contents (Table 4). Conversely, t color L* and h° displayed a negative linear relationship with TC, TL and TA. The TC, TL and TA contents had no effects on b* values.

Table 3 Phenotypic (rp) and genotypic correlation coefficients (rg) of color values, total carotenoid, lycopene, and anthocyanin contents in guava genotypes
Full size table
Table 4 Linear regression (R2) between color space values and pigment contents
Full size table

Path Analysis

The path coefficient analysis was conducted to understand the degree of contribution of different pigments on the color a* values of guava pulp. In this analysis, the path analysis, with color a* value as dependant variable, revealed that each of the three pigments viz. carotenoids, lycopenes and anthocyanins contributed to the color a* value of pulp in different proportions. The phenotypic path coefficients showed a positive direct effect for TC (0.30), TL (0.511) and TA (0.129) on a* value of pulp (Fig. 2a). Among these, the fruit TL content had the greatest direct impact on the color a* values. The phenotypic residual effects stood at 0.249. According to the genotypic path coefficient analysis, TC (−0.144) and TA (−0.069) had negative direct effects, while TL (1.105) showed a positive direct effect on color a* (Fig. 2b). On the other hand, the TC (1.018) and TA (0.632) indirectly influenced color a* through the TL content. The estimate of the genotypic residual effect was 0.127. The path analysis demonstrated the influence of TC, TL and TA, either directly or indirectly through TL, on the color a* value of guava pulp.

Fig. 2
figure 2

Path coefficient analysis. a Phenotypic path coefficient analysis and b Genotypic path coefficient analysis with color a* value as dependant variable. TC total carotene, TL total lycopene, TA total anthocyanin, R residual effect

Full size image

Discussion

Guava fruits are rich in many bioactive compounds, with ascorbic acid and lycopene being the most common ones (Nwaichi et al. 2015). Carotenoids, in particular lycopene, have numerous health benefits due to their antioxidant potential. Hence, coloured pulp guava fruits are a good dietary source of lycopene (Bin-Jumah et al. 2022). Colored pulp guava fruits are nutritionally superior to the white pulped fruits, thus breeding for colored fruits has gained more significance. The traditional methods of estimation of pigment contents in guava fruits are laborious and time consuming. Hence, an alternate method with ease of operation can facilitate the screening of hundreds of hybrids and mapping populations for carotenoids, lycopene, and anthocyanin contents.

Pigment Content Versus Color Space Values

The color reader-based values are important indicators of color characteristics such as intensity, vividness, and color angle within a sample. In CIELAB’s L*, a*, b* based color space systems, the chromaticity increases as the color deepens, but decreases when the color becomes lighter. This relationship is crucial in understanding the color characteristics of guava fruits, especially considering that red- or pink-fleshed guava fruits contain higher concentrations of pigments such as carotenoids, lycopene, and anthocyanin compared to white-fleshed guava varieties (Singh et al. 2019). A similar pattern was observed in our study, which showed that genotypes with white pulp exhibited very low levels of TC, TL, and TA, while fruits with red and purple pulp had higher concentrations of these pigments in different ratios (Supplementary Table 1). This aligns with the findings by Gowda (2023) and Ankad et al. (2023), who reported significant higher levels of lycopene and anthocyanin in red, pink, or purple guava pulp, whereas low levels in white guava pulp. Importantly, in our study, we observed that the changes in pigment contents were reflected by the corresponding changes in color space values. Specifically, we observed the lightness of pulp (L*) and hue angle (h°) in the order, white > red > purple fruits. The increase in pigment content in the red and purple pulps was reflected by a corresponding decrease in L* and h° values, but higher a* and C* values. Similar observations were made by Li et al. (2017), who observed a reduction in L* and h° values of white fleshed guava and a vice versa in red fleshed fruits. Likewise, orange-colored apricots were found to have low L* and h° values, but higher a* and C* values (Ruiz et al. 2005). Overall, the differences in color space values observed in guava fruits are due to the various ratios of the three pigments carotenoids, lycopenes, and anthocyanins.

Correlation Between Pigment Content and Color Values

The color space values (L*, a*, C and h°) showed a strong positive correlation with the pigments, TC and TL whereas a moderate positive correlation with TA. Values L* and h° were negatively correlated to pigment concentration. This is consistent with findings from a study by Goisser et al. (2020), wherein the lycopene content of tomatoes was found to have a linear relationship with the values of L*, a*, C*, and derived (a*/b*), with the exception of the b* values. Similarly, Guo et al. (2021) observed a high correlation of a* values with the total carotenoid content and expression of carotenoid biosynthesis pathway genes in pepper. Comparably in peach, lower pulp h° value indicated a higher amount of carotene in the fruit pulp (Matias et al. 2014). Comparable results were reported in apricots (Ruiz et al. 2005), pumpkins and squash (Itle and Kabelka 2009), and tomatoes (Pandurangaiah et al. 2020).

Notably, the correlation and regression estimates between color a* and TA (rg = 0.52; R2 = 0.24) was moderate since color a* values are primarily influenced by red color rather than the purple color. Studies in plum, and red wine grapes, demonstrated that a* values did not have any relationship with TA since the fruits were purple or blue in color (Rampácková et al. 2021; Rolle and Guidoni 2007). Unlike tomatoes, where lycopene is the major pigment, guava pulp color is affected by a variety of pigments. Therefore, based on correlation analysis color, a* values may serve as more useful indirect indicators of TC and TL pigment content in guava pulp than total anthocyanins.

Path Analysis: Effects of Color Space Values on Pigment Content

Despite being significant, the results of a correlation analysis do not always show the contribution of traits to coloration in guava pulp. Through examination of direct and indirect effects, path analysis enables us to comprehend the strength of a relationship between traits. Remarkably, TL had the largest positive direct effect observed on the a* value in both the genotypic (1.105) and phenotypic path analysis (0.511). This suggests that lycopene content strongly influences the redness or greenness of guava pulp, indicating its crucial role in determining fruit coloration. Interestingly, TC and TA had a negative direct effect in the genotypic level and a positive direct effect in the phenotypic level. The total carotene and anthocyanin showed an indirect influence on the color a* value of guava pulp through total lycopene (Fig. 2). The findings imply that each of the three pigments plays a significant role in determining the guava pulp’s color a* value, with lycopene being the major contributor.

Estimation of Pigment Content in Pulp Based on Color a* Values

There is a need to develop and adopt a technique for simple and reproducible estimation of guava pulp pigment concentration to facilitate faster evaluation of hybrids generated in breeding programs. This is the first report in guava, to analyse the association between and the effects of various pigment contents on the color space values of pulp. The results strongly suggest the application of color a* value as indirect selection method for total carotenoid and lycopene contents in coloured guava fruits due to its strong correlation and direct effects on pigment contents as evidenced in path analysis. Higher color a* value correspond to high carotene and lycopene content in guava pulp. Spectrometric methods require sample extraction and estimation of the light-sensitive pigments like carotenes and lycopenes. The objective evaluation based on color a* values can serve as a simpler alternative to spectrometric methods and can produce consistent results. For broader application of these findings, further research is required to examine the relationship between the color space values of guava processed products and total lycopene and carotenoid contents. This extension of the study would provide valuable insights into the suitability of color a* values for evaluating processed guava products and further enhance the utility of this indirect selection method in guava breeding and quality assessment efforts.

Conclusion

It is well known that the color of guava pulp is due to the presence of lycopenes, carotenes and to a minor extent by anthocyanins. This study establishes a relationship between the color a* value and the total carotenoids, lycopene, and anthocyanin contents in guava fruit pulp color. Path analysis demonstrated that total lycopene contributed directly to color a* value, while total carotene and anthocyanin contributed indirectly through lycopene. Overall results showed that lycopene contributed significantly to the colored pulp in guavas (direct effects), highlighting its importance as a primary determinant of fruit coloration. The findings facilitate the adoption of colorimetric a* values as an indirect selection criterion for total carotene, and lycopene pigment content. This approach offers a faster and less expensive method for estimation of pigments and will facilitate the breeding of antioxidant-rich guava varieties with desirable color characteristics. Overall, the research contributes to a better understanding of the factors influencing guava pulp coloration and offers practical implications for guava breeding programs aiming to develop varieties with enhanced nutritional quality and visual appeal.