Food Research 2 (6) : 564 - 571 (December 2018)

Journal homepage: http://www.myfoodresearch.com

FULL PAPER
Assessment of variability pattern of flesh color in ‘Harumanis’ mango
(Mangifera indica L.) from diverse Perlis geographical origin
1,*
Yusuf, A., 1Rahman, A.M.A., 1Zakaria, Z., 2Wahab, Z. and 3Kumar, S.V.
1
Department of Chemical Engineering Technology, Faculty of Engineering Technology, Universiti Malaysia
Perlis (UniMAP), Sungai Chuchuh, 02100 Padang Besar, Perlis.
2
Department of Mechanical Engineering Technology, Faculty of Engineering Technology, Universiti
Malaysia Perlis (UniMAP), Sungai Chuchuh, 02100 Padang Besar, Perlis.
3
Biotechnology Research Institute, Universiti Malaysia Sabah, Jalan UMS, 88400 Kota Kinabalu, Sabah,
Malaysia.
Article history:
Abstract
Received: 28 May 2018
Received in revised form: 12
‘Harumanis’ (Mangifera indica L) is one of the mango cultivars which has high market
September 2018
Accepted: 17 September 2018 value because of the excellent quality of the fruit which has attractive color, good aroma,
Available Online: 6 delicious taste and high nutritive values. In this study, fifty accessions from five different
November 2018 collection sites which belonging to North (Paya Kelubi and Chelong Balik Bukit, Padang
Besar), West (Santan, Kangar), East (Alor Ara Timur, Arau) and South (Simpang Empat,
Keywords:
Harumanis, Kangar) region of Perlis were analyzed according to flesh color traits based on their region
Flesh color, of origin. The analysis of variance using Kruskal-Wallis test resulted in significant
Geographical region, differences among the geographical region for traits of fruit flesh color such as L*, a*,
Variability
chroma and hue at (P<0.05). The correlation result shows that the intensity pattern of the
DOI: orange color of the fruit samples mesocarp was associated by an increase in the values of
https://doi.org/10.26656/fr.2017.2(6).108
a*, b* and C*, and a decrease in the values of L* and h. By performing Cluster analysis
using Ward’s method and Euclidian distance, five distinct clusters were successfully
identified. The finding shows a high distribution of ‘Harumanis’ accessions from different
locations in each distinct group. This study also reveals the relationship of variability in
fruit traits with their places of origin. However, these differences cannot be explained in
firm via morphological characterization only. Other methods such as molecular
characterization are strongly recommended.

1. Introduction carotenoid. Since mango is a climatic fruit, this factor is

postulated to play an important role in its carotenoid
Color is a vital parameter for food quality indicator accumulation. For example, ripe ‘Keitt’ mango from
since it is able to influence the consumer perception in Bahia, Brazil (hot climate) possesses more than twice the
term of flavor, sweetness and also elicit human β-carotene content than those from São Paulo (moderate
emotional feelings. In the case of mango, epidermis color climate) (Mercadante and Rodriguez-Amaya, 1998).
plays an important role in the perception of overall Thus, it can be postulated that the hot climatic condition
quality and is associated with the determination of is favourable for β-carotene accumulation and can
appropriate maturity for harvest. One of the main factors further result in fruit color variation which basically
which contribute towards the yellow-orange color of varies based on the origin of the mangoes.
mango epidermis and mesocarp is carotenoids.
Previously, the color measurement has been used for In Malaysia, the Department of Agriculture has
carotenoid estimation of ‘Manila’ and ‘Ataulfo’ mangoes registered 209 clones of mango grown in this country
(Ornelas-Paz et al., 2008). β-carotene has been which originated from India, Indonesia and other. From
emphasized as the main component that is interrelated these registered clones, some of them are possessing
with flesh color and resulting vitamin A values (Golob et high enough qualities to be recommended for general
al., 2012). plantings such as Malgoa and Apple Mango (for home
planting), MA 128, MA 162, Bahagia, MA 165, Bombay
The climatic condition has been stated as one of the Green, MA 204 and MA 205 (Ynus, 1984). Among these
major factors in the determination of accumulation of mango cultivars, ‘Harumanis’ or also known as MA 128
*Corresponding author. eISSN: 2550-2166 / © 2018 The Authors. Published by Rynnye Lyan Resources
Email: mukhlis@unimap.edu.my
 565 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571

is one the most favourable and is grown widely in Perlis. (Shahir and Visvanathan, 2014).
Therefore, the present work will focus on the evaluation Table 2. Harvest and Post-harvest process of 'Harumanis'
of ‘Harumanis’ flesh color trait variation in correlation
FULL PAPER

Stage Week Orchard (small scale production)

with different locations by application of morphological 1 0-7 Pre-mature: Not yet harvested
characterization method. Mature: Fruit harvesting
Mature: Right after being harvested,
8
2. Materials and methods 2 remove the fruit wrapper, cleaned, dried
(Day 1)
and conducted artificial ripening with
2.1 Description of study area calcium carbide.
Samplings were conducted for over four months 3
8 Fruit exposure to open air to release eth-
(March, April, May and June 2017). Prior to sampling, (Day 3) ylene gas.
ecogeographical survey and plant tagging were 8
4 Fruit traits analyses
(Day 6)
performed. From the survey, five locations were selected
which covered all three main regions of Perlis as 2.3 Pulp color measurement
indicated in Table 1.
Color measurement of mangoes was assessed by
2.2 Fruit sampling strategy application of a portable colorimeter (CR-400 Minolta)
set to L*, a*, b* mode. In standard Commission
About fifty accessions were considered for Internationale de l'Eclairage (CIE), the L*a*b* color
harvesting purpose. Five dates have been chosen for space indicates different degree of color measurement, in
collection based on the usual commercial harvest time of which L* value indicates the lightness [black [L*=0] and
this variety (April till June 2017). The recommendation [L* = 100]], a* value indicates redness-greenness (red
period of ‘Harumanis fruit collection is ninety to hundred [a* = 100] and green [a* = (-100)]) and b* indicates
days after full bloom (Muhammad and Ding, 2006) and yellowness-blueness (yellow [b* = 100] and blue [b* =
followed by a post-harvest treatment as indicated in (-100)]). Both chroma and hue are derived from a*and
Table 2. Selected ‘Harumanis’ fruits need to fulfill b* using the following equations: chroma measures
several criteria for full maturity confirmation such as has intensity coloration [C = (a*) 2 + (b*) 2]1/2] and hue
outgrown shoulders, formation of a depression with angle [h = arc tan (b*/a*)] (Varakumar et al., 2011). The
ridges at the stem end and already full round grow color coordinate is able to show the variation between
Table 1. List of ‘Harumanis’ mango accessions (collections) and their collection sites.
No Accession Name Location No Accession Name Location
S1 HM-Acc-1 1 S26 HM-Acc 26 3
S2 HM-Acc-2 1 S27 HM-Acc 27 3
S3 HM-Acc-3 1 S28 HM-Acc-28 3
S4 HM-Acc-4 1 S29 HM-Acc-29 3
S5 HM-Acc-5 1 S30 HM-Acc 30 3
S6 HM-Acc-6 1 S31 HM-Acc-31 4
S7 HM-Acc-7 1 S32 HM-Acc-32 4
S8 HM-Acc-8 1 S33 HM-Acc-33 4
S9 HM-Acc-9 1 S34 HM-Acc-34 4
S10 HM-Acc-10 1 S35 HM-Acc-35 4
S11 HM-Acc-11 2 S36 HM-Acc-36 4
S12 HM-Acc-12 2 S37 HM-Acc-37 4
S13 HM-Acc-13 2 S38 HM-Acc-38 4
S14 HM-Acc-14 2 S39 HM-Acc-39 4
S15 HM-Acc-15 2 S40 HM-Acc-40 4
S16 HM-Acc-16 2 S41 HM-Acc-41 5
S17 HM-Acc-17 2 S42 HM-Acc-42 5
S18 HM-Acc-18 2 S43 HM-Acc-43 5
S19 HM-Acc-19 2 S44 HM-Acc-44 5
S20 HM-Acc-20 2 S45 HM-Acc-45 5
S21 HM-Acc-21 3 S46 HM-Acc-46 5
S22 HM-Acc-22 3 S47 HM-Acc-47 5
S23 HM-Acc-23 3 S48 HM-Acc-48 5
S24 HM-Acc-24 3 S49 HM-Acc-49 5
S25 HM-Acc-25 3 S50 HM-Acc-50 5
Location 1: Chelong Balik Bukit, Padang Besar, Location ( 6.535464⸰ [N],100.288332⸰ [E] ), Location 2: Paya Kelubi, Padang
Besar ( 6.535175⸰ [N], 100.253759⸰ [E] ), Location 3: Alor Ara Timur, Arau ( 6.456431⸰ [N], 100.263158⸰ [E] ), Location 4:
Santan, Kangar( 6.482817⸰ [N], 100.230928⸰ [E] ) and Location 5: Simpang Empat, Kangar ( 6.322332⸰ [N], 100.206032⸰ [E] ).
eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources
 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571 566

the basic colors. The hue angle also able to determine the chroma values may reflect different concentrations of
intermediate colors between adjacent pairs of these basic pigments as both parameters could be correlated with the
colors. Chroma is the saturation or vividness of color. As total pigment of fruit mesocarp (Itle and Kabelka, 2009).

FULL PAPER
chromaticity increases, a color becomes more intense; as Among the five color space values, a* presents the
it decreases, a color becomes duller. Hue angle is the highest CVs especially for accessions from Simpang
basic unit of color and can be interpreted, for example, as Empat (30.27%) which is greater than 20.00%, indicated
(0 = red and 90 = yellow). Prior to use, the calorimeter a high variability (Norouzi et al., 2017).
was calibrated by using calibration with white plate (L*
= 98.15, C* = 1.92, h = 93.8, a* = 0.13 and b* = 1.92). 3.2 Analysis of variance and Bivariate Spearman
Each accession was measured as triplicates at least by correlation.
measuring both the equatorial regions of each fruit. The A significant (P<0.05) differences were observed
pulp color was measured by an internal reading at the among locations for the flesh color represented by the
central region equidistantly. The colorimeter was set to four color space value: L*, a*, chroma and hue. This
enable the light pulse to move to 3 locations of each demonstrates that the flesh color space of ‘Harumanis’
mango surface for every measurement. Thus, this will accessions was varied for different locations. This
allow the assessment of mango pulp ‘true’ color since relationship is revealed by Dunn pairwise by comparing
each measurement will represent average reading (Ayala the mean values as indicated by color space values of
-Silva et al., 2005). ‘Harumanis Paya Kelubi’ with other location (Alor Ara
2.4 Statistical analyses Timur, Santan, and Simpang Empat. A significant
difference of p-value less than 0.05 specifically for L*
All the measurements were performed in triplicate. and h was recorded between ‘Harumanis Paya Kelubi’
The values were then tabulated into a Microsoft Excel with other locations (Alor Ara Timur, Santan, and
file and SPSS IBM ver 19.0 software. Descriptive Simpang Empat) samples. This finding shows that there
analysis and test of normality were performed for is a presence of phenotypic variation in flesh color of
diversity within the variety evaluation. For descriptive ‘Harumanis’ fruits from different locations. The colors of
analysis, the values of minimum, maximum, mean and the mango flesh in this study varies from orange, orange-
coefficient of variation (CV) were computed for each yellow, yellow-orange to yellow. This could be related to
fruit traits. These descriptive statistics were important to the facts that any continuous inbreeding of the clones
assess the variation among the samples and the CV was may produce heterozygous with severe loss. One of the
vital for variability index. Due to violation towards the major mechanism which contributes towards this
normality distribution assumption, non parametric phenomenon is the interaction of gene (G), environment
Kruskal-Wallis with the multiple Dunn test was applied. (E) and G x E interaction over the population mean (h).
Non parametric Kruskal Wallis with the multiple Dunn It is also mentioned that carotenoid metabolism is a
test was purposely performed to test the significant complicated process, which is regulated by
difference of phenotype distribution in correlation with developmental stages and environmental conditions
the locations and also to test the heterogeneity effect (Zhang et al., 2015). The findings of this study are in
between the traits. Bivariate Spearman’s correlation was agreement with this as the values demonstrate that the
performed for evaluation of the correlation between the ‘Harumanis’ accessions flesh colors were varied for
phenotype and geographical region and also the different locations since different locations may possess
correlation between the traits. P-value correlation matrix different environmental condition (Itle and Kabelka,
was assessed for level of correlation indicator. 2009).

Based on Spearman’s rank correlation tests in Table

3. Results and discussion 4, the correlation showed fair and intermediate
significant correlation between a* (rs = 0.355) and b* (rs
3.1 Descriptive statistics
= 0.789) with chroma. The chroma values could be
The data collected from the study were subjected to correlated with the total pigment of fruit mesocarp (Itle
descriptive statistical analysis as indicated in Table 3. and Kabelka, 2009) since it measures the intensity of
Overall, the +a* mean values range from 13.97 to 20.45 coloration as present by previous. These findings able to
indicated the red color direction, while +b* from 63.42 meet the previous which states that a positive correlation
to 67.66 mean values range indicated the yellow color between chroma with parameter a* and b* could be
direction (Itle and Kabelka, 2009). A range of L*, predicted as any increase in pigment would increase the
lightness (72.25) to darkness (66.80) of color, and darkness and thereby increase the saturation. L* and h
chroma mean range values indicated vividness of color gave negative correlation with C* with (rs = -0.397) and
as the mean range within 63.42 to 67.66. The L* and (rs = -0.051) (Ornelas-Paz et al., 2008). These both
eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources
 567 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571

Table 3. Descriptive statistics of hunter colour of selected ‘Harumanis’ mangoes.

Location Descriptive statistics L* a* b* Chroma (C) Hue angle (h°)
FULL PAPER

Average 70.03 16.52 65.78 67.85 75.89

Min value 65.58 13.58 61.60 64.26 73.45
1 Max value 73.17 19.62 68.25 70.08 78.35
Standard deviation 2.08 1.84 1.77 1.66 1.64
Coefficient variation (%) 2.98 11.12 2.69 2.44 2.16
Average 66.80 20.45 66.38 69.51 72.71
Min value 60.57 17.11 60.85 65.59 68.09
2 Max value 70.01 24.48 69.98 72.04 76.01
Standard deviation 2.81 2.23 2.32 1.60 2.23
Coefficient variation (%) 4.21 10.93 3.50 2.30 3.07
Average 72.75 14.10 64.15 65.73 76.06
Min value 68.37 11.73 52.22 54.08 68.37
3 Max value 75.11 19.83 70.27 71.52 79.32
Standard deviation 2.05 2.37 6.31 6.28 4.26
Coefficient variation (%) 2.82 16.83 9.83 9.56 5.60
Average 71.98 13.97 67.66 69.13 78.35
Min value 69.44 10.53 66.5 67.5 74.1
4 Max value 74.47 19.34 68.85 70.82 81.09
Standard deviation 1.77 2.66 0.66 1.02 2.10
Coefficient variation (%) 2.46 19.02 0.96 4.57 2.67
Average 72.62 14.30 63.42 64.99 77.544
Min value 60.22 2.96 52.52 54.66 73.44
5 Max value 81.44 17.84 69.26 70.64 87.13
Standard deviation 5.24 4.33 6.01 5.95 3.83
Coefficient variation (%) 7.22 30.27 9.48 9.16 4.95
Values presented are mean of triplicate analysis. Chroma(C) = [(a*) 2 + (b*) 2]1/2 and hue angle (h°) = arc tan (b*/a*)

Table 4. Hunter colour and Location Correlation Coefficient of selected ‘Harumanis’ mangoes.
Fruit Trait
Rho Spearman
L* a* b* Chroma (C) Hue angle (h°)
L* rs 1.000
a* rs -.878** 1.000
b* rs .000 .041 1.000
Chroma (C) rs -.397** .355* .789** 1.000
Hue angle (h°) rs .690** -.837** .344* -.051 1.000
**Correlation is significant at the 0.01 level (2-tailed).
*Correlation is significant at the 0.05 level (2-tailed).

Table 5. Minimum and maximum value of five quantitative and one qualitative key descriptors used for morphological
classification of 50 Harumanis mango samples separately for the five identified clusters based on z score standardization.
Morphological Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 Cluster 6
characteristics (n = 5) (n = 11) (n = 13) (n = 9) (n = 11) (n = 1)
3, 4, 5, 8, 9, 25, 28, 31, 33,
2, 6, 13, 14, 1, 7, 26, 29,
11, 12, 16, 24 10, 15, 21, 27, 34, 36, 39, 40,
Accesion (HM-Acc-) 17, 18, 19, 20, 38, 41, 43, 47 50
and 30 32, 35, 45 and 49, 49, 22 and
42, 44 and 37 and 48
46 23
Location 2 and 3 1, 2, 4 and 5 1, 2, 3, 4 and 5 1, 3, 4 and 5 3, 4 and 5 5
Hue angle, h (⸰) 68.09-71.65 73.03-74.1 74.59-76.83 77.17-78.35 78.75-81.09 87.13
Orange yellow
Orange and Yellow orange
Flesh color Orange yellow and yellow Yellow orange Yellow
orange yellow and yellow
orange

eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources

 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571 568

correlation values indicates that the intensity of the classification provided by Royal Horticultural Society’s
orange color of the mesocarp was associated with an Colour Chart score in combination with previous mango
increase in the values of a*, b* and C*, and a reduction flesh color classification as attained by African bush

FULL PAPER
in the values of L* and h. An identical performance was mango trees (Irvingia species) in West Africa (Romaric
reported by several mango cultivars such as ‘Dashehari’, et al., 2013), these sub-clusters were classified into
Nam Dok Mai’, Kaew’, Mahahanaka’, ‘Manila’ and several color groups as shown in Table 5. This color
‘Ataulfo (Ornelas-Paz et al., 2008). The correlation value analysis is conducted to examine any presence of
between h which and a* presents a high negative phenotypic variation of flesh color among the
correlation (rs= -0.837) with a significant level of p- ‘Harumanis’ accessions from different locations. The
value less than 0.05. The negative correlation observed color analysis is determined by comparing the value of
between h and the a* estimated in this investigation hue angle with the color wheel which able to show an
proposes that as hue angles decrease and a* increase, abstract illustrative organization of color hues around a
carotenoid concentrations would increase. These inverse circle (Stancil and Jordan, 1985).
correlation between h and a* indicates the expression of
a deep orange-yellow color and related to high β- Cluster 1 grouped ‘Harumanis’ five mango samples
carotene content (V´azquez-Caicedo et al., 2004). This from the East (Alor Ara Timur, Arau) and North (Paya
result is strongly in agreement with the previous Kelubi, Padang Besar) region of Perlis which have the
correlation result of ‘Langra’ mango (Gill, 2017), since it densest flesh color (orange and orange-yellow) as the
also shows a high negative correlation (r =-0.955) mangoes scored lowest h range values from 68.09⸰ to
between h and a*. Each of the four reports viewed 71.65⸰. The greatest mean of h value was presented by
colorimetric investigation as a fitting estimator of cluster 5 which consists of populations from three
carotenoid concentration (Itle and Kabelka, 2009). These locations (3, 4 and 5) and showing yellow-orange and
discoveries are as per those of Godoy and Rodriguez- yellow fruit flesh according to the color wheel. This is in
Amaya (1989), who showed that the most vital agreement with previous findings which indicated that
carotenoid of mango is all-trans-β-carotene, with 48 to low values of h in peel or pulp indicates a deep orange-
84% of aggregate carotenoid content, depending on yellow color (V´azquez-Caicedo et al., 2004).
cultivar and fruit maturity stage (Mercadante and Color analyses of ‘Harumanis’ were performed on
Rodriguez-Amaya, 1998). the fruit flesh as this cultivar usually retains its green
3.3 Clustering analysis color even if it is physiologically mature and ripe.
Mango has a high content of carotenoids in mesocarp
By performing hierarchical clustering and overall tissues which responsible for the intense yellow color
profile analyses, two distinct groups were identified. The and this attribute is an important role in determining the
clustering analysis was constructed by utilizing the color fruit quality (Muhammad and Ding, 2006). In mango, as
space value of h as shown in Figure 1 and Table 5. Hue ripening progressed, the color of pulp turned orange,
angle was selected as a key descriptor in this analysis which is attractive and engaging for consumption.
since it is identified as one of the methods of reporting Beforehand, for commercialization purpose, a high
color which able to exhibit the estimation of visual bright red shading of fruit like tomato cultivars have
attributes (Mclellan et al., 1994). The first main cluster been associated with a strong positive impact among the
included 29 accessions and was divided into three sub- buyers, whereas a light (low red shading have been
clusters. The first sub-cluster contained five accessions related with a negative reaction by the shopper (Ayala-
while 11 accessions included in the second sub-cluster. Silva et al., 2005).
Third sub-cluster contained most of the population study
with 13 accessions. Meanwhile, about 21 accessions Content and compositions of carotenoid play a
were subjected into second main clusters by dividing into critical part in the formation of fruit color. Color changes
other three sub-clusters. The fourth sub-cluster consist of during fruit ripening include the conversion of
nine accessions, whilst cluster 5 included about eleven chloroplast to chromoplast. As a result of the loss of
accessions. The ‘Accession 50’ which subjected into photosynthetic capacity of the chloroplast, thylakoid
cluster 6 was excluded from this study since it was structures become sites for the accumulation of
identified as an extreme outlier. This clustering analysis carotenoids in the fruit cells. For instance, the
able to visualize the high distributional behavior which chlorophyll content is moderately high in the green
indicated by the five populations of ‘Harumanis’ in natural products, while the red and orange organic
Perlis. products are portrayed by high lycopene and carotene
substance, separately (Li et al., 2018). As mentioned
By referring to the standard of mango flesh color before, the most important carotenoid in mango is all-

eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources

 569 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571
FULL PAPER

Figure 1. Final dendogram showing five clusters as a result of cluster analysis (Ward method, Euclidian distance, z-score
standardization of variables using two color space value key descriptor (h) on 50 ‘Harumanis’ mangoes samples collected from
North (Paya Kelubi and Chelong Balik Bukit, Padang Besar), West (Santan, Kangar), East (Alor Ara Timur, Arau) and South
(Simpang Empat, Kangar) region of Perlis. The cutting line for the cluster formation is marked as a dotted line. Cluster six was
excluded as it identified as extreme outlier.

trans- β-carotene representing 48 to 84% of total strongly in agreement with a previous report on ‘Langra’
carotenoid content, depending on the cultivar and fruit mango (Gill, 2017) which showed a high negative
maturity stage (Mercadante and Rodriguez-Amaya, correlation result between a* and h. The inverse
1998). correlation between h and a* expresses the deep orange-
yellow color and a high β-carotene content (V´azquez-
Among the color space parameters, a high a* values Caicedo et al., 2004). This relationship well explained
reflect a high β-carotene content (V´azquez-Caicedo et the color pattern attained by the flesh of ‘Harumanis'
al., 2004.). This study found that the a* values are samples in Cluster 1 which has the lowest h range among
inversely correlated with the value of h. This result was the sample clusters and the densest orange color. This
eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources
 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571 570

result is consistent with previous studies which exposed environmental factor in the ‘Harumanis’ cultivar flesh
that any increase in the intensity of yellowness of the color trait variation which usually associated with the
mesocarp was accompanied by an increase in a*, b* and carotenoid accumulation. Earlier, it is also reported that,

FULL PAPER
C* values and a subsequent decrease in L* and h values other than environmental conditions, transcriptional
(Ornelas-Paz et al., 2008). regulation of the structural carotenoid pathway gene
during developmental stages, specifically ripening part,
From the analyses of variance, it is concluded that could also influence the carotenoid metabolism. (Ma et
there is a presence of environmental influences towards al., 2018). This is in accordance with past research
the flesh color space values with significant (P<0.05) which reported that carotenoid metabolism is a
differences among locations. This finding is reciprocal complicated process which is regulated by both the
with the cluster analysis results as there is a high developmental stages and environmental conditions
distribution of ‘Harumanis’ accessions from different (Zhang et al., 2015). Evidence in support of this
locations in each distinct group. In an earlier report, explanation is carotenoid accumulation patterns in
carotenoid content in flesh are varied among cultivars, different cultivars and during fruit development and
while carotenoids compositions were generally stable ripening is correlated with the expression of key
(Ornelas-Paz et al., 2008). It is well archived that the biosynthetic genes such as LCYB, LCYE, PSY, PDS,
content and composition of carotenoid development ZDS, BCH and ZEP (Ma et al., 2018).
regulation is influenced by environmental stimuli
(Cazzonelli and Pogson, 2010). Among the
environmental factor, light has been identified as one of 4. Conclusion
the important elements which able to regulate carotenoid In summary, there are significant differences in
metabolism in plants (Zhang et al., 2015). Meanwhile, a ‘Harumanis’ accessions flesh color different locations. In
former study revealed that hot climatic condition plays addition, ‘Harumanis’ accessions which belong to
an imperative part in the accumulation of carotenoids different color groups were successfully identified. This
such as ‘Keitt’ mango from Bahia (hot climate) identification may be useful for selection of ‘Harumanis’
(Mercadante and Rodriguez-Amaya, 1998). This is individuals with desirable characteristics as valuable
complemented with the earlier discoveries which genetic resources specifically for the breeding program.
mention that practically, light intensity was correlated Consequently, this evaluation and assessment can be a
with temperature, particularly in the open air (He et al., good basis for the establishment of ‘Harumanis’ seed
2018) and it was also demonstrated that the temperature program since the selection of good offspring is
was at significant linear to the light intensity (Gou et al., important as trait like the color is able to influence the
2015). fruit markets.

For this analysis, the surface meteorological

temperature observation data (January till June 2017) Acknowledgement
obtained from the Malaysian Meteorological Department The authors would like to acknowledge the financial
was used involving two regions which are Chuping support provided by the Ministry of Education Malaysia
representing orchards in Chelong Balik Bukit, Paya under the Fundamental Research Grant Scheme (FRGS)
Kelubi, Santan and Alor Ara Timur and Kangar (FRGS/1/2016/WAB01/UNIMAP/03/2).
representing orchards in Simpang Empat. The trend
shows a consistent and not much difference in the
temperature between the two regions with a average References
temperature of (28.4⸰C) for Kangar and (27.7⸰C) for Ayala-Silva, T., Schnell, R.J., Meerow, A.W.,
Chuping. Meanwhile, the high distribution of the Winterstein, M., Cervantes, C. and Brown, J.S.
‘Harumanis’ accessions from different locations (2005). Determination of Color and Fruit Traits of
expresses the presence of phenotype variation for the Halfsib Families of Mango ( Mangifera Indica L.).
fruit flesh color trait and thus indicated high variability Proceedings of the Florida State Horticultural
between the individuals. This can be clearly seen in Society, 118, 253-257.
Table 5, ‘Harumanis Simpang Empat’ samples (HM41, Cazzonelli, C.I. and Pogson, B.J. (2010). Source to sink:
HM42, HM43, HM44, HM45, HM46, HM47, HM48, regulation of carotenoid biosynthesis in plants.
HM49, and HM50) were highly distributed into distinct Trends in Plant Science, 15(5), 266-274. https://
groups which further do not represent a significant doi.org/10.1016/j.tplants.2010.02.003
relationship of differences in fruit flesh color with the
Gill, P.P.S. (2017). Transitions in mesocarp colour of
environmental condition. Thus, it can be postulated that
mango fruits kept under variable temperatures.
the temperature of Perlis regions is not a crucial
Journal of Food Science and Technology, 54(13),
eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources
 571 Yusuf et al. / Food Research 2 (6) (2018) 564 - 571

4251–4256. https://doi.org/10.1007/s13197-017- postharvest ripening of “Manila” and “Ataulfo”

2894-z mango fruit and relationship with carotenoid content
determined by liquid chromatography-APcI+-time-
FULL PAPER

Golob, P., Farrell, G. and Orchard, J.E. (2012). Crop

Post-Harvest : Science and Technology, Vol. 1: of-flight mass spectrometry. Postharvest Biology
Principles and Practice. USA: Wiley-Blackwell. and Technology, 50(2–3), 145–152. https://
doi.org/10.1016/j.postharvbio.2008.05.001
Gou, Z., Lau, S.S.-Y. and Qian, F. (2015). Comparison
of mood and task performance in naturally-lit and Romaric, V., Ronald, G.V.D. and Marc, S.M.S. (2013).
artificially-lit environments. Indoor and Built Morphological characterization of African bush
Environment, 24(1), 27–36. https:// mango trees (Irvingia species) in West Africa.
doi.org/10.1177/1420326X13507792 Genetic Resources and Crop Evaluation, 60(4), 1597
-1614. https://doi.org/10.1007/s10722-013-9969-0
He, Q., Yang, H. and Hu, C. (2018). Effects of
temperature and its combination with high light Shahir, S. and Visvanathan, A.R. (2014). Maturity
intensity on lipid production of Monoraphidium measurement of mango and banana as related to
dybowskii Y2 from semi-arid desert areas. ripening. Journal of Trends in Bioscience, 7(9), 741–
Bioresource Technology, 265, 407–414. https:// 744.
doi.org/10.1016/j.biortech.2018.06.044 Stancil, W.C. and Jordan, D. (1985). Precise color
Itle, R.A. and Kabelka, E.A. (2009). Correlation communication. Book of papers. National Technical
Between Lab Color Space Values and Carotenoid Conference, p. 33-35. USA: AATCC
Content in Pumpkins and Squash (Cucurbita spp.). V´azquez-Caicedo, A.L., Neidhart, S. and Carle, R.
HortScience, 44(3), 633–637. (2004). Postharvest ripening behavior of nine Thai
Li, F., Song, X., Wu, L., Chen, H., Liang, Y. and Zhang, mango cultivars and their suitability for industrial
Y. (2018). Scientia Horticulturae Heredities on fruit applications. ISHS Acta Horticulturae, (645), 617–
color and pigment content between green and purple 625. https://doi.org/10.17660/
fruits in tomato. Scientia Horticulturae, 235, 391– ActaHortic.2004.645.81
396. https://doi.org/10.1016/j.scienta.2018.03.030 Varakumar, S., Kumar, Y.S. and Reddy, O.V.S. (2011).
Ma, X., Zheng, B., Ma, Y., Xu, W., Wu, H. and Wang, Carotenoid composition of mango (mangifera indica
S. (2018). Carotenoid accumulation and expression l.) wine and its antioxidant activity. Journal of Food
of carotenoid biosynthesis genes in mango flesh Biochemistry, 35(5), 1538–1547. https://
during fruit development and ripening. Scientia doi.org/10.1111/j.1745-4514.2010.00476.x
Horticulturae, 237, 201–206. https:// Ynus, N. (1984). Mango Cultivation in Malaysia. The
doi.org/10.1016/j.scienta.2018.04.009 Archives of The Rare Fruit Council of Australia.
Mclellan, M.R., Lind, L.R. and Kime, R.W. (1994). Hue Retrieved from website: http://rfcarchives.org.au/
Angle Determination and Statistical for Next/Fruits/Mango/MangoCultivation9-84.htm
Multiquadrant Hunter L, a, b Data. Journal of Food Zhang, L., Ma, G., Yamawaki, K., Ikoma, Y.,
Quality, 18(3), 235–240. https://doi.org/10.1111/ Matsumoto, H., Yoshioka, T. and Kato, M. (2015).
j.1745-4557.1995.tb00377.x Effect of blue LED light intensity on carotenoid
Mercadante, A.Z. and Rodriguez-Amaya, D.B. (1998). accumulation in citrus juice sacs. Journal of Plant
Effects of ripening, cultivar differences, and Physiology, 188, 58–63. https://doi.org/10.1016/
processing on the carotenoid composition of mango. j.jplph.2015.09.006
Journal of Agricultural and Food Chemistry, 46(1),
128-130. https://doi.org/10.1021/jf9702860
Muhammad, H. and Ding, P. (2006). Cellular Changes
During Ripening Mango of Harumanis. Malaysian
Journal of Microscopy, 3, 19-24.
Norouzi, E., Erfani-Moghadam, J., Fazeli, A. and
Khadivi, A. (2017). Morphological variability within
and among three species of Ziziphus genus using
multivariate analysis. Scientia Horticulturae, 222,
180–186. https://doi.org/10.1016/
j.scienta.2017.05.016
Ornelas-Paz, J.D.J., Yahia, E.M. and Gardea, A.A.
(2008). Changes in external and internal color during

eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources