Article

Response of Watermelon to Drought Stress and Its

Drought-Resistance Evaluation
Kaili Ren 1,2 , Taoxia Tang 1 , Weiping Kong 1 , Yongquan Su 1 , Yuping Wang 2 , Hong Cheng 1, * , Yonggang Yang 1 and
Xiaoqin Zhao 1

1 Institute of Vegetables, Gansu Academy of Agricultural Sciences, Lanzhou 730070, China;

[email protected] (K.R.); [email protected] (T.T.); [email protected] (W.K.); [email protected] (Y.S.);
[email protected] (Y.Y.); [email protected] (X.Z.)
2 College of Horticulture, Gansu Agricultural University, Lanzhou 730070, China; [email protected]
* Correspondence: [email protected]; Tel./Fax: +86-0931-7614722

Abstract: This study investigated the response of watermelon seedlings to drought stress by
assessing the growth, physiological, and biochemical indices using a pot-based continuous
drought method. Drought stress indices, phenotypic plasticity indices, and membership
function values were calculated, followed by a correlation analysis, principal component
analysis, and cluster analysis, to comprehensively evaluate the drought resistance of
13 watermelon genotypes. The results revealed that drought stress significantly reduced
the fresh and dry weights, root length, root area, root volume, root tips, and forks of
watermelon seedlings. Additionally, drought significantly reduced the relative water
content of leaves and increased the levels of osmotic-adjustment substances (soluble sugars,
soluble proteins, proline, and starch). Persistent drought also modulated the activities
of antioxidant enzymes (SOD, POD, and CAT), leading to oxidative stress through the
accumulation of H2 O2 . Membrane damage, indicated by a significant increase in the
MDA content and relative conductivity, was observed, adversely affecting seedling growth.
Phenotypic plasticity indices indicated that watermelon exhibits strong adaptability to
drought. Cluster analysis categorized the 13 genotypes into four groups: highly drought-
resistant (14X5), drought-resistant (LK13, JLR, HXF1, 14X4, 14X1, and 14X6), low drought-
resistant (21F05, JH1, JR3, 14X7, and 16F02), and drought-sensitive (16C07). This study
provides valuable genetic resources for breeding drought-resistant watermelon varieties.
Academic Editor: Dayong Zhang

Received: 5 March 2025 Keywords: watermelon; drought stress; drought resistance; comprehensive evaluation
Revised: 16 April 2025
Accepted: 21 April 2025
Published: 24 April 2025

Citation: Ren, K.; Tang, T.; Kong, W.; 1. Introduction

Su, Y.; Wang, Y.; Cheng, H.; Yang, Y.;
Zhao, X. Response of Watermelon to
Drought is one of the most significant meteorological disasters worldwide, with its
Drought Stress and Its Drought- severity exacerbated by climate change. Rising global temperatures have increased evapo-
Resistance Evaluation. Plants 2025, 14, ration rates, expanded arid regions, and intensified the agricultural constraints imposed
1289. https://doi.org/10.3390/ by drought [1,2]. As a major abiotic stressor, drought severely impedes plant growth and
plants14091289
development, often resulting in substantial yield losses or even plant death [3–8]. To miti-
Copyright: © 2025 by the authors. gate these effects, plants have evolved adaptive mechanisms collectively termed “drought
Licensee MDPI, Basel, Switzerland. resistance”, which encompass morphological and biochemical responses, such as enhanced
This article is an open access article
root systems, reduced transpiration, and improved water-use efficiency [4,9,10]. These
distributed under the terms and
strategies operate through three primary mechanisms: (i) escape (accelerating reproduc-
conditions of the Creative Commons
Attribution (CC BY) license
tive cycles before stress becomes lethal), (ii) avoidance (maintaining high internal water
(https://creativecommons.org/ content to prevent tissue damage), and (iii) tolerance (sustaining growth under low water
licenses/by/4.0/). availability) [11].

Plants 2025, 14, 1289 https://doi.org/10.3390/plants14091289

Plants 2025, 14, 1289 2 of 20

Drought tolerance varies significantly among plant species. Watermelon (Citrullus

lanatus (Thunb.) Matsum. et Nakai), originating from tropical Africa, has developed pro-
nounced drought adaptability through long-term evolution. Its deep-rooted system enables
efficient water extraction from soil, while moderate drought conditions can even enhance
sugar accumulation. China is the largest producer of watermelon globally, accounting for
49.14% of the world’s production area and 60.95% of the total output in 2023 [12]. Given
its economic importance and inherent drought tolerance, breeding and the promotion of
new drought-resistant varieties are crucial for sustainable agriculture. The breeding of
drought-resistant varieties is inseparable from drought-resistant germplasm resources.
Drought-resistance-evaluation methods significantly influence phenotypic assess-
ments. Prior studies have employed diverse approaches:
• PEG-simulated drought at the bud stage (e.g., rice (Oryza sativa L.), maize (Zea mays
L.), pea (Pisum sativum), foxtail millet (Setaria italica L.), soybean (Glycine max), white
poplar (Populus alba L.), Elymus nutans) [13–21],
• PEG-simulated drought at the seedling stage (e.g., wheat (Triticum aestivum L.), barley
(Hordeum vulgare), maize, sugar beet (Beta vulgaris L.), quinoa (Chenopodium quinoa
Wild.), Elymus nutans) [20,22–28],
• Pot water control at the seedling stage (e.g., cassava (Manihot esculenta Crantz),
watermelon, Gleditsia sinensis, herbaceous plants (Limonium bicolor, Agropyron mon-
golicum, Agropyron desertorum, Astragalus adsurgens, Mellilotus of ficinalis, Trifolium
repens, Medicago sativa, Glycyrrhiza uralensis, Artemisia ordosica, Suaeda glauca, Althaea
rosea, Agriophyllum squarrosum), wheat, Helleborus orientalis, chrysanthemum (Chrysan-
themum)) [29–35],
• Field natural drought screening (e.g., maize, wheat, soybean, potato (Solanum tubero-
sum L.), cotton (Gossypium), chickpea (Cicer arietinum L.)) [36–44].
Among these, pot water control at the seedling stage offers greater reproducibility and
operational feasibility for drought-resistance phenotyping.
In this study, we employed a pot water control method at the seedling stage to analyze
the physiological and biochemical responses of 13 watermelon genotypes under drought
stress. Drought stress indices and phenotypic plasticity indices were calculated, followed by
a comprehensive evaluation via correlation analysis, principal component analysis (PCA),
and cluster analysis. Our findings provide a foundation for selecting drought-resistant
watermelon genotypes to support future breeding efforts.

2. Results
2.1. Drought Injury Index of 13 Watermelon Genotypes
Under drought stress, watermelon plants exhibited wilting, leaf rolling, and browning,
with some plants dying. The drought injury index (DI) was calculated for each genotype,
with values ranging from 1.54 (14X6) to 3.65 (JR3) (Table 1). Genotypes 14X6, 14X4, 14X5,
and 14X1 showed the lowest DI values, indicating higher drought tolerance.

Table 1. Drought injury index of watermelon germplasm resources.

Germplasm DI Germplasm DI
LK13 2.83 JR3 3.65
21F05 3.18 14X1 2.21
16F02 2.86 14X4 1.68
JH1 2.95 14X5 1.95
JLR 3.19 14X6 1.54
16C07 3.20 14X7 3.45
HXF1 3.14
 spectively, over the control, whereas 34.05% and 36.90% decreases were recorded in the
fresh weights of the root and shoot, respectively, after drought treatment (Figure 1A–D).
From a detailed view of different germplasms, drought stress significantly reduced the
Plants 2025, 14, 1289 shoot and root biomass in most genotypes, except for 14X5 and 14X6, which maintained 3 of 20
relatively stable dry weights of the shoot and root (Figure 2A–D).
Root architecture was also affected. In general, except for the root diameter, drought
2.2.treatment
Effects of significantly
Drought Stress on Growth
decreased the Parameters
total root length, surface area, volume, tips, and
forks of watermelon seedlings by
As we can see in Figure 1A–D, drought39.12%, 46.04%, 52.17%,
stress 39.41%, andreduced
significantly 35.28, respectively
the growth of
(Figure 1E–J). From the perspective of different germplasms, drought stress significantly
watermelon seedlings compared with the control. Due to drought treatment, a compar-
reduced the total root length, surface area, and volume of most genotypes except for the
atively higher decrease in the seedling fresh weight was recorded than the seedling dry
total root length of HXF1 and the total root length, surface area, and volume of 14X5 (Fig-
weight. The dry
ure 3A,C,D). Theweights
number of of the
root root andforks
tips and shootof were decreased
watermelon by 19.99%
seedlings and 27.33%,
also showed a
respectively,
downward trend after drought treatment, but the degree of the decline varied with dif-in the
over the control, whereas 34.05% and 36.90% decreases were recorded
fresh weights
ferent of the
genotypes root and
(Figure 3E,F).shoot, respectively,
In addition, droughtafter
stressdrought treatment
significantly (Figure
increased 1A–D).
the av-
From a detailed
erage viewofofJH1
root diameter different
and 14X6 germplasms,
genotypes anddrought stress decreased
significantly significantly reduced the
the average
rootand
shoot diameter of LK13 and
root biomass JR3 genotypes
in most genotypes,(Figure 3B).for 14X5 and 14X6, which maintained
except
relatively stable dry weights of the shoot and root (Figure 2A–D).

A DRW B FRW C DSW D FSW

0.8 8 5 20
18
0.7
* ** *
0.6 6
4 16
**
14
0.5 3 12
0.4 4 10

0.3 2 8
6
0.2 2
1 4
0.1 2
0.0 0 0 0
CK DT CK DT CK DT CK DT
E RL F RD G RS H RV
10000 1.2 15000

** 1.1 ** **
8000 1.0 ns 12000
1500

0.9
6000 0.8 9000
1000
0.7
4000 0.6 6000
0.5 500
2000 0.4 3000
0.3
0 0.2 0 0
CK DT CK DT CK DT CK DT
I RT J RF
** **
4000 4000

3000 3000

2000 2000

1000
1000

0
CK DT CK DT

Figure 1. Box plots showing the descriptive statistics of the seedling growth traits. CK represents
normal water control treatment, and DT represents drought stress treatment. Statistical significance
was determined based on Tukey’s HSD, where ** p < 0.01, * p < 0.05, and ns means not significant. The
horizontal line and square within the box represent the median and mean, respectively. The lower
and upper limits of the box and lower and upper whiskers, represent Q1 (first quartile/25 percentile),
Q3 (third quartile/75 percentile), (Q1 − 1.5 IQR), and (Q3 + 1.5 IQR), respectively. IQR-interquartile
range. Black diamond dots on the boxes indicate the distribution of watermelon observations (DRW,
root dry weight; FRW, root fresh weight; DSW, shoot dry weight; FSW, shoot fresh weight; RL, total
root length; RD, average root diameter; RS, root surface area; RV, root volume; RT, root tips; RF, root
forks). (A) Dry weight of root; (B) Fresh weight of root; (C) Dry weight of shoot; (D) Dry weight
of shoot; (E) Total root length; (F) Average root diameter; (G) Root surface area; (H) Root volume;
(I) Root tips; (J) Root forks.

Root architecture was also affected. In general, except for the root diameter, drought
treatment significantly decreased the total root length, surface area, volume, tips, and
forks of watermelon seedlings by 39.12%, 46.04%, 52.17%, 39.41%, and 35.28, respectively
(Figure 1E–J). From the perspective of different germplasms, drought stress significantly
reduced the total root length, surface area, and volume of most genotypes except for the
total root length of HXF1 and the total root length, surface area, and volume of 14X5
(Figure 3A,C,D). The number of root tips and forks of watermelon seedlings also showed a
 was determined based on Tukey s HSD, where ** p < 0.01, * p < 0.05, and ns means not significant.
The horizontal line and square within the box represent the median and mean, respectively. The
lower and upper limits of the box and lower and upper whiskers, represent Q1 (first quartile/25
Plants 2025, 14, 1289 4 of 20
percentile), Q3 (third quartile/75 percentile), (Q1 − 1.5 IQR), and (Q3 + 1.5 IQR), respectively. IQR-
interquartile range. Black diamond dots on the boxes indicate the distribution of watermelon obser-
vations (DRW, root dry weight; FRW, root fresh weight; DSW, shoot dry weight; FSW, shoot fresh
downward
weight; RL, trend after
total root drought
length; treatment,
RD, average but the degree
root diameter; RS, rootof the decline
surface varied
area; RV, with different
root volume; RT,
genotypes
root tips; RF, root forks). (A) Dry weight of root; (B) Fresh weight of root; (C) Dry weight of shoot;root
(Figure 3E,F). In addition, drought stress significantly increased the average
diameter of JH1ofand
(D) Dry weight 14X6
shoot; genotypes
(E) Total and (F)
root length; significantly
Average rootdecreased theRoot
diameter; (G) average root
surface diameter
area; (H)
ofRoot
LK13 and JR3
volume; genotypes
(I) Root (Figure
tips; (J) Root forks.3B).

A B
CK DT

Fresh weight of root (g)

Dry weight of root (g)

0.6 ✱✱✱✱ 6 ✱✱✱✱

✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱
0.4 ✱✱✱✱ ✱✱✱✱ 4 ✱✱✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱
0.2 ns 2 ✱✱✱✱
ns

0.0 0
13

F1
07

7
R

13

05

02

F1
R

07

7
JH
F0

F0

JR

JH

JR

X
JX

JX
LK

LK

C
14

14

14

14

14

14

14

14

14

14
21

16

16

21

16
D
H

16

H
C 20

Fresh weight of shoot (g)

✱✱✱✱
Dry weight of shoot (g)

✱✱✱✱
3
15
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱
2 ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
10 ✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱
1 ns ns
5

0 0

13

05

02

F1
R

07

7
13

F1
07

7
R

JH

JR

X
F0

F0

JH

JR

JX
JX

LK

C
LK

X
X

14

14

14

14

14
14

14

14

14

14

21

16
21

16

16
16

H
H

Melon materials Melon materials

Figure2.2.Effects
Figure Effectsofof drought stress on
drought stress onfresh
freshand
anddry
dryweights
weightsof of watermelon
watermelon seedlings.
seedlings. (A) (A)
Dry Dry weight
weight
ofofroot.
root.(B)
(B)Fresh
Freshweight
weight of root. (C)
of root. (C)Dry
Dryweight
weightofofshoot.
shoot. (D)
(D) Fresh
Fresh weight
weight of shoot.
of shoot. CK CK represents
represents
normal
Plants 2025, 14, x FOR PEER REVIEWnormalwater
watercontrol
control treatment, andDT
treatment, and DTrepresents
representsdrought
drought stress
stress treatment.
treatment. Statistically
Statistically 5 significant
significant
of 22
impacts and interactions, determined based on 2-way ANOVA, are indicated in
impacts and interactions, determined based on 2-way ANOVA, are indicated in each panel, where each panel, where
**** p < 0.0001, ** p < 0.01, and ns means not significant. Mean values ± standard errors
**** p < 0.0001, ** p < 0.01, and ns means not significant. Mean values ± standard errors are shown. are shown.

A B
1.5
Average root diameter (cm)

8000 ✱✱✱✱ CK DT
Total root length (cm)

✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱
6000 ✱✱✱✱ ✱✱✱✱ 1.0
✱ ns ns
ns ✱ ns ns ns
ns ns ✱✱✱✱
ns
4000 ns
✱✱✱✱
ns 0.5
2000

0 0.0
13

05

02

F1
07

7
R

13

05

02

F1
07

7
R
JH

JR

X
JX

JH

JR

X
LK

JX
C

14

14

14

14

14

LK

C D
X

14

14

14

14

14
21

16

16

21

16

16

15000
Root surface area (cm2 )

✱✱✱✱ ✱✱✱✱
1500
Root volume (cm3 )

✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
10000 ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱ 1000 ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱ ✱✱✱✱
✱✱✱✱

5000
ns
✱✱✱✱ 500 ✱✱✱
ns

0 0
13

F1
13

F1

07

7
R

07

F0

F0

JH
F0

F0

JR

X
JH

JR

JX
JX

LK
LK

C
C

X
X

14

14

14

14

14
14

14

14

14

14

E F
21

16
21

16

16
16

H
H

5000 5000
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
4000 ✱✱✱✱ 4000 ✱✱✱✱
ns ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ns ✱✱✱✱
✱✱✱✱
✱✱✱✱ ns
Root forks

✱✱✱✱
Root tips

ns
3000 ns 3000 ns
✱✱ ✱✱✱
ns
ns
2000 2000
ns
ns
1000 1000

0 0
13

05

02

F1
R

07

13

F1
R

07

7
JH

JR

F0

F0

JH

JR

X
JX

JX
LK

LK

C
X

X
14

14

14

14

14

14

14

14

14

14
21

16

21

16
16

16
H

Melon materials Melon materials

Figure 3. Effect of drought stress on root architecture of watermelon seedlings. (A) Total root length.
(B)Figure
Average3. Effect
root of drought stress
diameter. on root
(C) Root architecture
surface area. (D)of Root
watermelon
volume.seedlings.
(E) Root(A) Total
tips. (F)root
Root length.
forks. CK
(B) Average root diameter. (C) Root surface area. (D) Root volume. (E) Root tips. (F)
represents normal water control treatment, and DT represents drought stress treatment. StatisticallyRoot forks. CK
represents normal water control treatment, and DT represents drought stress treatment.
significant impacts and interactions, determined based on 2-way ANOVA, are indicated in each Statistically
significant
panel, whereimpacts
**** p < and interactions,
0.0001, determined
*** p < 0.001, based
** p < 0.01, * pon< 2-way ANOVA,
0.05, and are indicated
ns means in eachMean
not significant.
panel, where **** p < 0.0001,
values ± standard error are shown. *** p < 0.001, ** p < 0.01, * p < 0.05, and ns means not significant. Mean
values ± standard error are shown.

2.3. Effect of Drought Stress on Relative Water Content and Pigment Content
Compared with the control, drought stress significantly reduced the relative water
content of leaves by 9.80% but had no significant effect on the pigment content (chloro-
 Plants 2025, 14, 1289 5 of 20

2.3. Effect of Drought Stress on Relative Water Content and Pigment Content
Compared with the control, drought stress significantly reduced the relative water
content of leaves by 9.80% but had no significant effect on the pigment content (chlorophyll
a concentration, chlorophyll b concentration, carotenoid concentration, total pigment
concentration, and content) (Figure 4A–F). Judging from the germplasms of different
genotypes, the relative water content in watermelon leaves decreased significantly under
drought stress, except for JH1, 16C07, HXF1, and 14X1 genotypes (Figure 5A). The pigment
content showed different degrees of increases and decreases under drought treatment. In
general, under drought stress, the chlorophyll a concentration, chlorophyll b concentration,
Plants 2025, 14, x FOR PEER REVIEW total pigment concentration, and content of 14X5 and 14X6 genotypes were significantly 6 of 22
higher than those of the control, while the other genotypes showed different degrees of
reductions or no significant difference (Figure 5B,C,E,F).

A RWC B Chl-a C Chl-b D Car

120 9 4.0
14
8
3.5 ns
100 ** 12 ns 7
ns
6
3.0
10
5
80
4 2.5
8
3
60 6 2.0
2

1 1.5
4
40 0
CK DT CK DT CK DT CK DT
E TPC F TP G SS H SP
9.0 8

25 2.5
*
8.5
**
ns ns 8.0 6
20 2.0
7.5
4
15 1.5 7.0

6.5
2
10 1.0 6.0

5.5
5 0.5 0
CK DT CK DT CK DT CK DT
I Pro J ST
70 4.0

60 ** 3.5 **
3.0
50
2.5
40
2.0
30
1.5
20
1.0

10 0.5

0 0.0
CK DT CK DT

Figure
Figure 4. Box
4. Box plotsshowing
plots showing the
thedescriptive
descriptivestatistics of some
statistics physiological
of some and biochemical
physiological traits of traits
and biochemical
watermelon seedlings. CK represents normal water control treatment, and DT represents drought
of watermelon seedlings. CK represents normal water control treatment, and DT represents drought
stress treatment. Statistical significance was determined based on Tukey’s HSD, where ** p < 0.01,
stress
* p treatment.
< 0.05, and Statistical
ns means not significance
significant.was
The determined
horizontal line based on Tukey
and square s HSD,
within the boxwhere ** p < 0.01, *
represent
p < 0.05, and nsand
the median means
mean,not significant.
respectively. The
The horizontal
lower and upper line andofsquare
limits within
the box the box
and lower and represent
upper the
whiskers
median andrepresent Q1 (first quartile/25
mean, respectively. percentile),
The lower Q3 (third
and upper quartile/75
limits percentile),
of the box and lower − 1.5upper
(Q1and IQR), whisk-
and (Q3 + 1.5 IQR), respectively. IQR, interquartile range. Black diamond dots on the boxes indicate
ers represent Q1 (first quartile/25 percentile), Q3 (third quartile/75 percentile), (Q1 − 1.5 IQR), and
the distribution of watermelon observations (RWC, relative water content; Chl-a, chlorophyll a
(Q3concentration;
+ 1.5 IQR), respectively. IQR, binterquartile
Chl-b, chlorophyll concentration;range. Black diamond
Car, carotenoid dots on
concentration; TPC,thetotal
boxes indicate the
pigment
distribution of watermelon
concentration; observations
TP, total pigment (RWC,
content; SS, relative
soluble water content;
sugar content; Chl-a,
SP, soluble chlorophyll
protein a concen-
content; Pro,
proline
tration; content;
Chl-b, St, starch b
chlorophyll content). (A) Relative
concentration; Car,water content;concentration;
carotenoid (B) Chl-a concentration;
TPC, total (C)pigment
Chl-b con-
concentration; (D) Carotenoid concentration; (E) Total pigment concentration; (F) Total pigment
centration; TP, total pigment content; SS, soluble sugar content; SP, soluble protein content; Pro,
content; (G) Soluble sugar content; (H) Soluble protein content; (I) Proline content; (J) Starch content.
proline content; St, starch content). (A) Relative water content; (B) Chl-a concentration; (C) Chl-b
concentration; (D) Carotenoid concentration; (E) Total pigment concentration; (F) Total pigment
content; (G) Soluble sugar content; (H) Soluble protein content; (I) Proline content; (J) Starch content.
Plants 2025, 14, x FOR PEER REVIEW 7 of 22
Plants 2025, 14, 1289 6 of 20

A100
Relative water content (%)
B 15

Chl-a concentration (mg/l)

✱✱✱✱
✱✱✱✱ ✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱ CK DT
ns
✱ ✱✱ ✱✱✱✱ ns
80 ns ✱
ns
✱✱✱✱
10 ✱✱✱✱
✱✱✱✱
60 ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ns
✱✱✱✱ ✱✱✱✱
✱✱✱✱
40 ns ✱
5
20

0 0
13

F1
R

07

13

F1
07

7
R
F0

F0

JH

JR

F0

F0

JH

JR

X
D
JX

JX
LK

LK

C
X

X
14

14

14

14

14

14

14

14

14

14
C
21

16

21

16
16

16
H

H
Carotenoid concentration (mg/l)
10 6
Chl-b concentration (mg/l)

✱✱✱✱

8
✱✱✱✱
4
6 ✱✱✱✱
✱✱✱✱ ✱✱✱✱
ns ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱
✱✱✱✱ ns ✱✱✱✱ ns ✱✱✱✱ ✱✱✱✱
4 ✱✱✱✱ ✱✱✱ ns ns
ns ✱✱✱✱
ns ns ns ns 2
2

0 0
13

F1
R

07

13

F1
07

7
R
F0

F0

JH

JR

F0

F0

JH

JR

X
JX

JX
LK

LK

C
X

X
14

14

14

14

14

14

14

14

14

14
21

16

21

16
16

16
H

H
E30 F
Total pigment concentration (mg/l)

Total pigment content (mg/g·FW)

3
✱✱✱✱ ✱✱✱✱

✱✱✱✱
20 ✱✱✱✱ 2 ✱✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ns ✱✱✱✱ ✱✱✱✱ ns
✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ns
ns ns ns
10 1

0 0
13

F1
R

07

13

F1
R

07

7
F0

F0

JH

JR

F0

F0

JH

JR

X
JX

JX
LK

LK

C
X

X
14

14

14

14

14

14

14

14

14

14
21

16

21

16
16

16
H

H
Melon materials Melon materials

Effect of
Figure5.5. Effect
Figure of drought
droughtstress onon
stress relative water
relative content
water and pigment
content contentcontent
and pigment in watermelon leaves.
in watermelon
(A) Relative water content. (B) Chl-a concentration. (C) Chl-b concentration. (D) Carotenoid
leaves. (A) Relative water content. (B) Chl-a concentration. (C) Chl-b concentration. (D) Carotenoid con-
centration. (E) Total pigment concentration. (F) Total pigment content. CK represents normal water
concentration. (E) Total pigment concentration. (F) Total pigment content. CK represents normal
control treatment, and DT represents drought stress treatment. Statistically significant impacts and
water control treatment, and DT represents drought stress treatment. Statistically significant im-
interactions, determined based on 2-way ANOVA, are indicated in each panel, where **** p < 0.0001,
pacts
*** pand interactions,
< 0.001, determined
** p < 0.01, * p < 0.05,based
and nson means
2-way ANOVA, are indicated
not significant. in each
Mean values ±panel, where
standard ****
error
p are
< 0.0001,
shown. *** p < 0.001, ** p < 0.01, * p < 0.05, and ns means not significant. Mean values ± standard
error are shown.
2.4. Effect of Drought Stress on the Contents of Osmotic Adjustment Substances
When
2.4. Effect droughtStress
of Drought stresson
occurs in plants,
the Contents the osmotic-adjustment
of Osmotic substances tend to
Adjustment Substances
increase to reduce osmotic stress and improve the plant tolerance to
When drought stress occurs in plants, the osmotic-adjustment substancesdrought stress.
tend The
to in-
results showed that compared with the control, drought stress significantly increased the
crease to reduce osmotic stress and improve the plant tolerance to drought stress. The
content of soluble sugar, soluble protein, proline, and starch by 5.16%, 24.12%, 56.780%,
results showed that compared with the control, drought stress significantly increased the
and 34.46%, respectively (Figure 4G–J). From a detailed view of different germplasms,
content of soluble sugar, soluble protein, proline, and starch by 5.16%, 24.12%, 56.780%,
drought treatment significantly increased the soluble sugar content in JXR, 14X5, and 14X6
and 34.46%, respectively (Figure 4G–J). From a detailed view of different germplasms,
genotypes, significantly increased the soluble protein content in LK13, 16F02, JH1, HXF1,
drought treatment significantly increased the soluble sugar content in JXR, 14X5, and 14X6
14X1, 14X6, and 14X7 genotypes, and significantly increased the starch content in 21F05,
genotypes, significantly increased the soluble protein content in LK13, 16F02, JH1, HXF1,
JXR, 16C07, HXF1, JR3, 14X1, 14X4, 14X5, 14X6, and 14X7 genotypes (Figure 6A,B,D). It is
14X1, 14X6, and 14X7 genotypes, and significantly increased the starch content in 21F05,
worth noting that except for the JXR genotype, drought treatment significantly increased
JXR, 16C07, HXF1,
the proline contentJR3, 14X1,
of the other14X4, 14X5, 14X6,
watermelon and 14X7
genotypes genotypes
(Figure 6C). (Figure 6A,B,D). It is
worth noting that except for the JXR genotype, drought treatment significantly increased
2.5.proline
the Antioxidant Enzyme
content of theActivities and Oxidative
other watermelon Stress
genotypes (Figure 6C).
The activities of superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT),
and the contents of hydrogen peroxide (H2 O2 ) and superoxide anion (O2− ) in watermelon
leaves were measured on the 7th day of continuous drought. The results showed that
compared with the control, drought stress significantly reduced SOD activity (decreased
by 20.52%) and increased POD activity and the H2 O2 content (increased by 29.55% and
Plants 2025, 14, 1289 7 of 20

Plants 2025, 14, x FOR PEER REVIEW

21.29%, respectively) but had no significant effect on CAT activity and the O2−
8 of 22
content
(Figure 7A–E).

A B

Soluble protein content (mg/g)

10
Soluble sugar content (mg/g)

8
ns ✱✱✱✱ CK DT
✱ ✱✱✱✱ ✱✱✱✱
8 ns ns ns ns ns ns ns ns ✱✱✱✱ ns ns
✱✱✱✱
6
6 ✱✱✱✱
ns ns ✱✱✱✱
4 ✱✱✱
4 ns ✱✱ ns
ns

2 2

0 0
13

F1
R

07

7
F0

F0

JH

JR

13

F1
R

07

7
JX
LK

JH
F0

F0

JR

X
14

14

14

14

14

JX
LK
21

16

C
16

X
H

14

14

14

14

14
C60

21

16
D4

16

H
✱✱✱✱
Proline content (μg/g)

Starch content (mg/g)

✱✱✱✱
✱✱✱✱ ✱✱✱✱
3 ✱✱✱ ✱✱ ✱✱✱
✱✱✱✱
ns ns ns ✱✱✱✱
40 ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ns
✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱
✱✱✱✱ 2
✱✱✱✱ ✱✱✱✱
✱✱✱✱
20
1

0 0
13

F1

13

F1
R

07

R
1

07

7
F0

F0

F0

F0
JH

JR

JH

JR

X
JX

JX
LK

LK
C

C
X

X
14

14

14

14

14

14

14

14

14

14
21

16

21

16
16

16
H

H
Melon materials Melon materials

Figure
Figure6. Effects
6. Effects of drought
of drought stressstress on osmotic-adjustment
on osmotic-adjustment substances
substances in watermelon
in watermelon leaves.
leaves. (A) Sol-
(A) Soluble sugar content. (B) Soluble protein content. (C) Proline content. (D)
uble sugar content. (B) Soluble protein content. (C) Proline content. (D) Starch content. CK repre- Starch content.
CK represents
sents normal
normal water watertreatment,
control control treatment, and DT represents
and DT represents drought
drought stress stressStatistically
treatment. treatment. sig-
Statisti-
cally
Plants 2025, 14, x FOR PEER REVIEW significant impacts and interactions, determined based on 2-way ANOVA, are
nificant impacts and interactions, determined based on 2-way ANOVA, are indicated in each panel, indicated
9 ofin22each
panel,
wherewhere
**** p <**** p < 0.0001,
0.0001, *** p <**0.001,
*** p < 0.001, ** *p p<<0.01,
p < 0.01, p <ns
0.05,*and 0.05, andnot
means ns significant.
means not Mean
significant.
valuesMean
±
values ± standard error
standard error are shown. are shown.

A SOD B C POD CAT D H2O2

2.5. Antioxidant Enzyme Activities
30 and Oxidative Stress

350 ** 600 * ns 30 **
25
The
500
activities of superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT),
300
20
and the 400
contents of hydrogen peroxide (H2O2) and superoxide 20
anion (O2−) in watermelon
250
leaves300 were measured on the 157th day of continuous drought. The results showed that
200 10
compared200 with the control, drought stress significantly 10reduced SOD activity (decreased
150 5
by 20.52%)
100 and increased POD activity and the H2O2 content (increased by 29.55% and
100 0 0
CK DT21.29%, respectively)
CK but
DT had no significantCK effect
DT on CAT activityCK and DT the O2− content
E O2- F MDA G EC
1400 (Figure 30 7A–E). 1.2
ns **
1200
From the perspective of different 1.0
**
germplasm resources, drought stress changed the
1000
activities
20 of SOD, POD, and CAT, 0.8
and the enzyme activities of some genotypes increased,
800

600
while the enzyme activities of some genotypes decreased. Under drought stress, POD ac-
0.6

400
tivity 10was significantly increased in the LK13, 21F05, 16F02, JH1, 16C07, and 14X1 geno-
0.4
200 types but significantly decreased in the JXR genotype (Figure 8B). At the same time, CAT
0
CK DT
activity 0
underCK
drought stress
DT
was
0.2
significantly
CK
increased
DT
in the 16C07, JR3, 14X1, 14X4,
14X6, and 14X7 genotypes and significantly decreased in the 16F02, JH1, and 14X5 geno-
Figure Box plots
types7.(Figure showing
8C). However, the descriptive
drought stress statistics of some physiological
significantly reduced the and SODbiochemical
activity of traits
the of
Figure 7. Box
watermelon plots showing
seedlings. CK the descriptive
represents normal statistics
water of some treatment,
control physiological andandDTbiochemical
represents traits
drought
21F05, 16F02, JXR, 16C07, HXF1, JR3, 14X4, 14X5, and 14X6 genotypes (Figure 8A). What
of watermelon
stress treatment.seedlings. CK
Statistical represents normal
significance water control
was determined treatment,
based and DT
on Tukey’s represents
HSD, wheredrought
** p < 0.01,
is more, drought stress significantly increased the H2O 2 content in watermelon leaves ex-
stress treatment. Statistical significance was determined based on
* p < 0.05, and ns means not significant. The horizontal line and square within the 2− Tukey s HSD, where ** p < 0.01, *
box represent
cept for the 14X1, 14X4, and 14X6 genotypes, and significantly increased the O content
the median and mean, respectively. The lower and upper limits of the box and lower andthe
p < 0.05, and ns means not significant. The horizontal line and square within the box represent upper
in watermelon leaves, except for the 21F05, 14X1, 14X5, and 14X6 genotypes (Figure 8D,E).
median represent
whiskers and mean,Q1 respectively. The lower
(first quartile/25 and upper
percentile), Q3limits of quartile/75
(third the box and lower and upper
percentile), (Q1 − whisk-
1.5 IQR),
The increase in the active oxygen content indicated that drought stress caused oxidative
and ers(Q3
represent Q1 (first
+ 1.5 IQR), quartile/25IQR,
respectively. percentile), Q3 (third
interquartile range.quartile/75 percentile),
Black diamond dots(Q1
on −the
1.5boxes
IQR), indicate
and
stress.
the(Q3 + 1.5 IQR), of
distribution respectively.
watermelon IQR, interquartile(SOD,
observations range.superoxide
Black diamond dots onactivity;
dismutase the boxes indicate
POD, the
peroxidase
distribution of watermelon observations (SOD, superoxide dismutase 2 −
activity;
activity; CAT, catalase activity; H2 O2 , hydrogen peroxide content; O , superoxide anion content; POD, peroxidase ac-
MDA,tivity;malonaldehyde
CAT, catalase activity;
content;HREC,
2O2, hydrogen peroxide content;
relative conductivity). 2−
(A)OSOD, superoxide
activity; anion
(B) POD content;
activity;
(C)MDA,
CAT malonaldehyde
activity; (D) H2content; REC,(E)
O2 content; O2− content;
relative conductivity).
(F) MDA (A) SOD activity;
content; (B) POD conductivity.
(G) Relative activity; (C)
CAT activity; (D) H2O2 content; (E) O2− content; (F) MDA content; (G) Relative conductivity.
From the perspective of different germplasm resources, drought stress changed the
A400 B600and the enzyme activities of some genotypes increased,
activities of SOD, POD, and CAT,
D activity （ U /g・ F W ）

✱✱✱✱ CK DT
D activity （ U /g・ F W ）

✱✱✱✱
✱✱✱✱ ✱✱✱✱
300 ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱ 400
ns ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ns
ns ns ✱✱✱✱ ns ns ns
✱✱✱✱ ✱✱✱✱ ✱ ✱✱
200 ns
ns
200
ns
100
 1200
1.0
**
1000
20
0.8
800

600
0.6
Plants 2025, 14, 1289 10 8 of 20
400
0.4
200

0 0 0.2
CK DT CK DT CK DT
while the enzyme activities of some genotypes decreased. Under drought stress, POD
activity was significantly increased in the LK13, 21F05, 16F02, JH1, 16C07, and 14X1
Figure 7. Box plots showing the descriptive statistics of some physiological and biochemical traits
genotypes but significantly decreased in the JXR genotype (Figure 8B). At the same time,
of watermelon seedlings. CK represents normal water control treatment, and DT represents drought
CAT activity
stress treatment.under drought
Statistical stresswas
significance was significantly
determined based increased in thewhere
on Tukey s HSD, 16C07,
** p <JR3,
0.01,14X1,
*
14X4, 14X6, and 14X7 genotypes and significantly decreased in the 16F02, JH1,
p < 0.05, and ns means not significant. The horizontal line and square within the box represent the and 14X5
genotypes
median and(Figure 8C). However,
mean, respectively. The lowerdrought stress
and upper significantly
limits of the box andreduced
lower andthe SOD
upper activity
whisk-
ofers
the 21F05, Q1
represent 16F02,
(first JXR, 16C07,
quartile/25 HXF1, JR3,
percentile), 14X4,quartile/75
Q3 (third 14X5, and 14X6 genotypes
percentile), (Figure
(Q1 − 1.5 IQR), and8A).
What is more, drought stress significantly increased the
(Q3 + 1.5 IQR), respectively. IQR, interquartile range. Black diamond dots H O content in watermelon
2 2on the boxes indicate the
leaves except
distribution for the 14X1,
of watermelon 14X4, and
observations 14X6
(SOD, genotypes,
superoxide andactivity;
dismutase significantly increased
POD, peroxidase ac- the
2− content
Otivity; in watermelon
CAT, catalase activity; Hleaves, except for
2O2, hydrogen the 21F05,
peroxide content;14X1, 14X5, and anion
O2−, superoxide 14X6 content;
genotypes
MDA, malonaldehyde
(Figure 8D,E). The increasecontent;
inREC, relativeoxygen
the active conductivity).
content (A) SOD activity;
indicated that(B) POD activity;
drought (C)
stress caused
CAT activity;
oxidative (D) H2O2 content; (E) O content; (F) MDA content; (G) Relative conductivity.
stress. 2−

A400 B600

POD activity （ U /g・ F W ）

✱✱✱✱ CK DT
SOD activity （ U /g・ F W ）

✱✱✱✱
✱✱✱✱ ✱✱✱✱
300 ✱✱✱✱
✱✱✱✱
✱✱✱✱ ✱✱✱✱ 400
ns ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ns
ns ns ✱✱✱✱ ns ns ns
✱✱✱✱ ✱✱✱✱ ✱ ✱✱
200 ns
ns
200
ns
100

0 0 13

05

02

F1
R

07

7
13

F1
07

7
R

JH

JR

X
F0

F0

JH

JR

JX
JX

F
LK

C
LK

X
X

14

14

14

14

14
14

14

14

14

14

21

16
21

16

16
16

H
H

C D 30
CAT activity (μmol/min/g·FW)

H2 O2 content (μmol/g·FW)

✱✱✱✱ ns
250 ✱✱✱✱

✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱

✱✱✱✱ ns ✱✱✱✱
200 ✱✱✱✱
✱✱✱✱ ns
20 ✱✱✱✱ ✱
150
✱✱✱✱
✱✱✱✱
20 ns ✱✱✱✱ ✱✱✱ ✱✱
✱✱✱✱ ns ✱
15 ns 10
ns
10
5
0 0
13

F1
R

07

7
13

F1
07

7
R

F0

F0

JH

JR

X
F0

F0

JH

JR

JX
JX

LK

C
LK

X
X

14

14

14

14

14
14

14

14

14

14

21

16
21

16

16
16

H
H

E2500 Melon materials

O2- content (nmol/g·FW)

✱✱✱✱
✱✱✱✱
2000

1500
✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱
1000 ns
✱✱✱✱

500 ns
ns ns

0
13

F1
07

7
R
F0

F0

JH

JR

X
JX
LK

14

14

14

14

14
21

16

16

Melon materials

Figure 8. Effects of drought stress on antioxidant system in watermelon leaves. (A) SOD activity.
(B) POD activity. (C) CAT activity. (D) H2 O2 content. (E) O2− content. CK represents normal water
control treatment, and DT represents drought stress treatment. Statistically significant impacts and
interactions, determined based on 2-way ANOVA, are indicated in each panel, where **** p < 0.0001,
*** p < 0.001, ** p < 0.01, * p < 0.05, and ns means not significant. Mean values ± standard error
are shown.

2.6. Membrane Damage

Malonaldehyde (MDA) and relative conductivity are important indexes to measure
membrane damage. The results showed that drought stress significantly increased the
MDA content and relative conductivity by 39.08%and 30.41%, respectively, compared
with the control (Figure 7F,G). From the perspective of different germplasms, drought
treatment increased the MDA content and relative conductivity of all watermelon genotypes
(Figure 9A,B). Among them, the MDA content reached a significant level, except for the
14X4 and 14X6 germplasms, and the relative conductivity was significant except for the
14X4 germplasm.
 with the control (Figure 7F,G). From the perspective of different germplasms, drought
treatment increased the MDA content and relative conductivity of all watermelon geno-
types (Figure 9A,B). Among them, the MDA content reached a significant level, except for
Plants 2025, 14, 1289 the 14X4 and 14X6 germplasms, and the relative conductivity was significant except for9 of 20
the 14X4 germplasm.

A 30 B1.0
MDA content （nmol/g·FW）
CK DT

Relative conductivity (%)

✱✱✱✱
✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱
0.8 ✱✱✱✱ ✱✱✱✱
✱✱✱✱ ns ✱✱✱✱ ✱✱✱✱
20 ✱✱✱✱
✱✱✱✱ ✱✱ ns ✱✱✱✱
✱✱✱✱ ✱✱✱✱
✱✱✱✱ ns 0.6

0.4
10

0.2

0 0.0
13

F1
R

07

13

F1
R

07

7
F0

F0

JH

JR

F0

F0

JH

JR

X
JX

JX
LK

LK

C
14

14

14

14

14

14

14

14

14

14
21

16

16

21

16
H

16

H
Melon materials Melon materials

Figure9.9.Effects
Figure Effects of
of drought stress on
drought stress onmembrane
membranedamagedamage ofof watermelon
watermelon leaves.
leaves. (A) (A)
MDA MDA content.
content.
(B) Relative conductivity. CK represents normal water control treatment, and DT
(B) Relative conductivity. CK represents normal water control treatment, and DT represents represents drought
stress treatment.
drought Statistically
stress treatment. significant
Statistically impacts
significant and and
impacts interactions, determined
interactions, determinedbased
basedon on 2-way
2-
ANOVA, are indicated in each panel, where **** p < 0.0001, ** p < 0.01, and ns means
way ANOVA, are indicated in each panel, where **** p < 0.0001, ** p < 0.01, and ns means not sig- not significant.
Mean values
nificant. Mean± standard error areerror
values ± standard shown.
are shown.

2.7. Drought Stress Index and Phenotypic Plasticity Index

2.7. Drought Stress Index and Phenotypic Plasticity Index
The drought stress index of 27 watermelon indexes under drought stress was calcu-
The drought stress index of 27 watermelon indexes under drought stress was calcu-
lated. The results are shown in Table 2. The drought stress indexes of 13 indexes (including
lated. The results are shown in Table 2. The drought stress indexes of 13 indexes (includ-
the
ingdry
theand
dryfresh weights
and fresh of theofroot,
weights dry and
the root, dry fresh weights
and fresh of the
weights of shoot, totaltotal
the shoot, rootroot
length,
root surface
length, root area, root
surface area,volume, root tips,
root volume, root root
root tips, forks, relative
forks, water
relative content,
water content,chlorophyll
chloro-
a phyll
concentration, total pigment concentration, and SOD activity)
a concentration, total pigment concentration, and SOD activity) were were less
lessthan
than1,1,and
14and
indexes (average
14 indexes root diameter,
(average chlorophyll
root diameter, b concentration,
chlorophyll carotenoid
b concentration, carotenoidconcentration,
concen-
soluble sugar content, soluble protein content, proline content, starch content,
tration, soluble sugar content, soluble protein content, proline content, starch content, MDA con-
tent, relative conductivity, POD activity, 2 −
MDA content, relative conductivity, POD CAT activity,
activity, H2 O2 content,
CAT activity, O content,
H2O2 content, O2− con-and
DI) were
tent, andgreater
DI) were than 1. The
greater variation
than in the drought
1. The variation stress index
in the drought was abundant,
stress index was abundant,and the
and the coefficient
coefficient of variation
of variation ranged ranged from 6.979%
from 6.979% to 121.634%.
to 121.634%. AmongAmong
them,them,
the the coeffi- of
coefficient
cient of variation
variation of the
of the starch starchwas
content content was theand
the largest, largest, and the coefficient
the coefficient of variation
of variation of
of the soluble
the soluble sugar content
sugar content was the smallest. was the smallest.
In order to clarify the phenotypic plasticity of 27 traits of watermelon under drought
stress,
Table the phenotypic
2. Drought plasticity
stress index index wasplasticity
and phenotypic calculated, and
index the value was between 0.206
(PPI).
and 0.909 (Table 2). The phenotypic plasticity index of 15 traits (total root length, root
surface Drought Stress Index
Traitsarea, root volume, root tips, root forks, chlorophyll a concentration, chlorophyll PPIb
Min Max AVE SE CV
concentration, total pigment concentration, soluble protein content, proline content,
starch
DRW content, POD
0.712activity, CAT
0.897activity, O 2− content, and DI) were larger, ranging from
0.807 0.069 8.503 0.206
FRW 0.521 0.868 0.688 0.135 19.603 0.400
DSW 0.589 0.867 0.736 0.113 15.333 0.320
FSW 0.523 0.786 0.643 0.090 14.030 0.334
RL 0.315 0.943 0.632 0.195 30.880 0.666
RD 0.729 1.323 1.009 0.146 14.471 0.449
RS 0.288 0.819 0.559 0.161 28.783 0.648
RV 0.224 0.726 0.497 0.145 29.240 0.691
RT 0.221 1.045 0.630 0.243 38.586 0.788
RF 0.250 1.752 0.739 0.374 50.613 0.857
RWC 0.670 1.015 0.907 0.097 10.693 0.340
Chl-a 0.691 1.712 0.944 0.263 27.854 0.597
Chl-b 0.598 2.786 1.041 0.552 53.053 0.786
Car 0.793 1.566 1.016 0.198 19.481 0.493
TPC 0.689 1.921 0.979 0.311 31.729 0.641
SS 0.934 1.196 1.056 0.074 6.979 0.219
SP 1.035 2.224 1.297 0.313 24.105 0.534
Pro 1.043 8.649 2.219 2.017 90.911 0.879
St 0.916 10.033 2.001 2.434 121.634 0.909
MDA 1.011 1.943 1.440 0.315 21.853 0.479
Plants 2025, 14, 1289 10 of 20

Table 2. Cont.

Drought Stress Index

Traits PPI
Min Max AVE SE CV
REC 1.055 1.553 1.304 0.135 10.329 0.321
SOD 0.538 1.057 0.821 0.174 21.177 0.491
POD 0.710 3.204 1.379 0.752 54.520 0.778
CAT 0.448 2.876 1.176 0.692 58.828 0.844
H2O2 0.954 1.708 1.256 0.238 18.976 0.441
O2− 0.929 1.956 1.378 0.324 23.548 0.525
DI 1.542 3.650 2.757 0.685 24.838 0.578
Note: Min, Max, AVE, SE, and CV represent the minimum, maximum, mean, standard error, and coefficient of
variation of the drought stress index (DRW, root dry weight; FRW, root fresh weight; DSW, shoot dry weight;
FSW, shoot fresh weight; RL, total root length; RD, average root diameter; RS, root surface area; RV, root
volume; RT, root tips; RF, root forks; RWC, relative water content; Chl-a, chlorophyll a concentration; Chl-b,
chlorophyll b concentration; Car, carotenoid concentration; TPC, total pigment concentration; SS, soluble sugar
content; SP, soluble protein content; Pro, proline content; St, starch content; MDA, malonaldehyde content; REC,
relative conductivity; SOD, superoxide dismutase activity; POD, peroxidase activity; CAT, catalase activity; H2 O2 ,
hydrogen peroxide content; O2− , superoxide anion content; DI, drought injury index).

In order to clarify the phenotypic plasticity of 27 traits of watermelon under drought

stress, the phenotypic plasticity index was calculated, and the value was between 0.206
and 0.909 (Table 2). The phenotypic plasticity index of 15 traits (total root length, root
surface area, root volume, root tips, root forks, chlorophyll a concentration, chlorophyll
b concentration, total pigment concentration, soluble protein content, proline content,
starch content, POD activity, CAT activity, O2− content, and DI) were larger, ranging from
0.525 to 0.909, among which the phenotypic plasticity index of the starch content was the
largest. From the phenotypic plasticity index of each index, watermelon has strong drought
adaptability and can better adapt to environmental water changes.

2.8. Correlation Analysis

The correlation analysis of the drought stress index of 27 traits of watermelon showed
that there were different degrees of correlation between the drought stress indexes of
each trait (Figure 10). The shoot fresh weight was significantly positively correlated
with the root dry weight, root fresh weight, shoot dry weight, and proline content but
significantly negatively correlated with the MDA content, H2 O2 content, and DI. And
the proline content was significantly positively correlated with the shoot fresh weight,
average root diameter, chlorophyll a concentration, chlorophyll b concentration, carotenoid
concentration, and total pigment concentration but significantly negatively correlated
with MDA content and DI. In addition, the DI was significantly negatively correlated
with dry and fresh weights of the root, dry and fresh weights of the shoot, chlorophyll
a concentration, chlorophyll b concentration, total pigment concentration, and proline
content (the correlation coefficients were −0.71, −0.75, −0.75, −0.61, −0.58, −0.60, −0.58,
and −0.73, respectively) but significantly positively correlated with the MDA content,
relative conductivity, H2 O2 , and O2− content (the correlation coefficients were 0.77, 0.78,
0.60, and 0.59, respectively).
 fresh weights of the root, dry and fresh weights of the shoot, chlorophyll a concentration,
chlorophyll b concentration, total pigment concentration, and proline content (the corre-
lation coefficients were −0.71, −0.75, −0.75, −0.61, −0.58, −0.60, −0.58, and −0.73, respectively)
Plants 2025, 14, 1289 but significantly positively correlated with the MDA content, relative conductivity, H2O , 20
112of
and O content (the correlation coefficients were 0.77, 0.78, 0.60, and 0.59, respectively).
2−

Figure
Figure 10.10.Correlation
Correlationanalyses
analyses of
of drought resistanceindex
drought resistance indexofofwatermelon
watermelon under
under drought
drought stress.
stress. In In
thethe
lower
lowerpanel,
panel,the
theredredand
and blue
blue circles indicatepositive
circles indicate positiveandandnegative
negative correlations,
correlations, respectively,
respectively,
with
withincreasing
increasingsize
sizereflecting
reflectingaa higher coefficient.The
higher coefficient. Theupper
upperpanel
panel shows
shows thethe correlation
correlation coefficient
coefficient
of of
the
therelated
related traits. * indicate
traits. * indicate significant
significant p ≤(DRW,
at p ≤at0.05 0.05 (DRW,
root dryroot dryFRW,
weight; weight;
root FRW, root fresh
fresh weight;
weight;
DSW, shoot dry weight; FSW, shoot fresh weight; RL, total root length; RD, average root diameter;root
DSW, shoot dry weight; FSW, shoot fresh weight; RL, total root length; RD, average
diameter;
RS, root RS, rootarea;
surface surface
RV, area; RV, rootRT,
root volume; volume; RT, RF,
root tips; rootroot
tips; RF, root
forks; RWC, forks; RWC,
relative waterrelative
content;water
content; Chl-a, chlorophyll
Chl-a, chlorophyll a concentration;
a concentration; Chl-b, chlorophyll
Chl-b, chlorophyll b concentration;
b concentration; Car, carotenoid Car, carotenoid
concentra-
concentration;
tion; TPC, total pigment concentration; SS, soluble sugar content; SP, soluble protein content; content;
TPC, total pigment concentration; SS, soluble sugar content; SP, soluble protein Pro,
Pro, proline
proline content;
content; St, starch
St, starch content;
content; MDA,MDA, malonaldehyde
malonaldehyde content;content; REC, conductivity;
REC, relative relative conductivity;
SOD,
SOD, superoxide
superoxide dismutase
dismutase activity;
activity; POD,POD, peroxidase
peroxidase activity;
activity; CAT,CAT, catalase
catalase activity;
activity; H2OH22, O 2 , hydrogen
hydrogen
peroxide content; O 2− , superoxide anion content; DI, drought injury index).
peroxide content; O2−, superoxide anion content; DI, drought injury index).

2.9. Principal Component Analysis

2.9. Principal Component Analysis
In order to reduce the redundancy of data, the drought stress indexes of 27 traits were
In order to reduce the redundancy of data, the drought stress indexes of 27 traits were
analyzed via principal component analysis. According to the principle of an eigenvalue
analyzed via principal component analysis. According to the principle of an eigenvalue
greater than 1, five principal components were extracted, and the original 27 single indexes
were transformed into five new independent comprehensive indexes (PC1, PC2, PC3,
PC4, and PC5). The eigenvalues of each principal component factor, the load matrix of
the original index, and the contribution rate to the phenotype are shown in Table 3. The
contribution rates of PC1, PC2, PC3, PC4, and PC5 factors were 33.900%, 25.108%, 13.067%,
8.681%, and 6.959%, respectively, and the cumulative contribution rate reached 87.715%.
Among them, PC1 was mainly closely related to the plant biomass (dry and fresh weights of
the root, dry and fresh weights of the shoot), relative water content, pigment concentration
(chlorophyll a, chlorophyll b, carotenoid, and total pigment concentration), soluble sugar
content, proline content, MDA content, relative conductivity, SOD activity, H2 O2 content,
and DI. PC2 was mainly related to the root architecture (total root length, average root
diameter, root surface area, root volume, root tips, root forks) and pigment concentration
(chlorophyll a, chlorophyll b, carotene, and total pigment concentration). PC3 was closely
Plants 2025, 14, 1289 12 of 20

related to the relative water content, soluble sugar content, soluble protein content, starch
content, relative conductivity, CAT activity, and O2− content. PC4 was closely related to
the POD activity and H2 O2 content, and PC5 was closely related to the starch content, SOD
activity, and O2− content.

Table 3. Component matrix and the cumulative contribution rate of principal components.

PC1 PC2 PC3 PC4 PC5

Evaluation Traits Load Load Load Load Load
Weight Weight Weight Weight Weight
Capacity Capacity Capacity Capacity Capacity
DRW 0.686 0.075 0.092 0.014 −0.217 −0.062 0.446 0.19 −0.394 −0.21
FRW 0.72 0.079 0.489 0.072 −0.147 −0.042 −0.276 −0.118 0.064 0.034
DSW 0.767 0.084 0.271 0.04 −0.24 −0.068 0.214 0.091 −0.219 −0.117
FSW 0.708 0.077 0.128 0.019 −0.226 −0.064 0.169 0.072 −0.401 −0.214
RL 0.115 0.013 0.952 0.14 0.235 0.067 0.015 0.007 −0.005 −0.003
RD 0.368 0.04 −0.626 −0.092 −0.055 −0.016 −0.416 −0.178 0.086 0.046
RS 0.151 0.016 0.961 0.142 0.134 0.038 −0.052 −0.022 0.074 0.039
RV 0.219 0.024 0.922 0.136 0.014 0.004 −0.138 −0.059 0.111 0.059
RT −0.073 −0.008 0.934 0.138 0.157 0.044 0.135 0.058 −0.153 −0.081
RF 0.409 0.045 0.768 0.113 0.4 0.113 0.066 0.028 0.251 0.134
RWC −0.549 −0.06 0.088 0.013 −0.529 −0.15 −0.12 −0.051 −0.427 −0.227
Chla 0.788 0.086 −0.541 −0.08 0.168 0.048 0.096 0.041 0.124 0.066
Chlb 0.811 0.089 −0.501 −0.074 0.13 0.037 0.013 0.005 0.049 0.026
Car 0.71 0.078 −0.647 −0.095 0.239 0.068 0.065 0.028 0.063 0.033
TPC 0.797 0.087 −0.545 −0.08 0.164 0.047 0.063 0.027 0.089 0.047
SS 0.662 0.072 0.111 0.016 0.553 0.157 0.027 0.011 0.001 0
SP 0.286 0.031 −0.018 −0.003 −0.705 −0.2 −0.336 −0.143 0.391 0.208
Pro 0.862 0.094 −0.376 −0.055 −0.095 −0.027 −0.074 −0.031 −0.068 −0.036
St 0.411 0.045 0.315 0.046 0.559 0.158 0.158 0.067 0.509 0.271
MDA −0.763 −0.083 −0.169 −0.025 0.139 0.039 −0.119 −0.051 0.443 0.236
REC −0.5 −0.055 −0.441 −0.065 0.632 0.179 −0.162 −0.069 −0.16 −0.085
SOD −0.524 −0.057 −0.025 −0.004 −0.409 −0.116 0.344 0.147 0.505 0.269
POD −0.309 −0.034 −0.035 −0.005 −0.304 −0.086 0.847 0.361 −0.048 −0.026
CAT 0.048 0.005 0.201 0.03 −0.673 −0.191 −0.472 −0.201 0.096 0.051
H2 O2 −0.631 −0.069 −0.298 −0.044 0.201 0.057 0.526 0.224 0.055 0.029
O2− −0.349 −0.038 0.036 0.005 0.549 0.156 −0.468 −0.2 −0.514 −0.274
DI −0.898 −0.098 −0.173 −0.026 0.289 0.082 −0.08 −0.034 −0.116 −0.062
Eigen values 9.153 6.779 3.528 2.344 1.879
Variance contribution (%) 33.900 25.108 13.067 8.681 6.959
Cumulative variance
87.715
contribution (%)
Note: DRW, root dry weight; FRW, root fresh weight; DSW, shoot dry weight; FSW, shoot fresh weight; RL, total
root length; RD, average root diameter; RS, root surface area; RV, root volume; RT, root tips; RF, root forks; RWC,
relative water content; Chl-a, chlorophyll a concentration; Chl-b, chlorophyll b concentration; Car, carotenoid
concentration; TPC, total pigment concentration; SS, soluble sugar content; SP, soluble protein content; Pro, proline
content; St, starch content; MDA, malonaldehyde content; REC, relative conductivity; SOD, superoxide dismutase
activity; POD, peroxidase activity; CAT, catalase activity; H2 O2 , hydrogen peroxide content; O2− , superoxide
anion content; DI, drought injury index.

2.10. Comprehensive Evaluation of Drought Resistance

Based on the results of the principal component analysis, the membership function
values of five comprehensive indexes were calculated (Table 4). The weights of the five
principal component factors were calculated according to the contribution rate, and the
weights of PC1 to PC5 were 0.3865, 0.2862, 0.1490, 0.0990, and 0.0793, respectively. Accord-
ing to the formula, the comprehensive evaluation value D of the drought resistance of each
watermelon genotype was calculated. The drought resistance of 13 watermelon resources
was ranked according to the D value, and the drought resistance from strong to weak was
as follows: 14X5, 14X1, 14X6, 14X4, HXF1, LK13, JLR, 21F05, JH1, 14X7, JR3, 16F02, and
16C07 (Table 4).
According to Ward’s method of squared deviations, the D value of watermelon was
analyzed via cluster analysis. As shown in Figure 11, 13 watermelon genotypes were
clustered into 4 categories. The first category was the highly-drought-resistant germplasm,
including 14X5; the second category was the drought-resistant germplasms, including
LK13, JLR, HXF1, 14X4, 14X1, and 14X6; the third category was the low-drought-resistant
germplasms, including 21F05, JH1, JR3, 14X7, and 16F02; the fourth category was the
drought-sensitive germplasm, including 16C07.
 2.10. Comprehensive Evaluation of Drought Resistance
Plants 2025, 14, 1289 Based on the results of the principal component analysis, the membership 13 of 20function
values of five comprehensive indexes were calculated (Table 4). The weights of the five
principal component factors were calculated according to the contribution rate, and the
Table 4. Membership function values of five main factors and D-value ranks of drought tolerance.
weights of PC1 to PC5 were 0.3865, 0.2862, 0.1490, 0.0990, and 0.0793, respectively. Ac
Germplasm U(X1 ) cordingU(Xto2 ) the formula,
U(X3 ) the comprehensive
U(X4 ) evaluation
U(X5 ) value D of the Tolerance
D-Value drought resistance
Rank
of each watermelon genotype was calculated. The drought resistance of 13 watermelon
LK13 0.26 0.74 0.40 1.00 −0.10 0.46 6
21F05 0.06
resources
0.35
was ranked according to
0.54
the D value,1.22
0.61
and the drought 0.36
resistance8 from strong to
16F02 0.31 weak was0.35 as follows: 0.4514X5, 14X1, 14X6, 14X4,
0.48 HXF1, LK13,
−0.16 JLR, 21F05, 12
0.32 JH1, 14X7, JR3
JH1 0.10 0.33 0.56 0.55 0.97 0.35 9
16F02, and 16C07 (Table 4).
JLR 0.26 0.75 0.75 0.12 −0.38 0.41 7
16C07 0.22 According
0.46 to Ward0.66 s method of squared −deviations,
0.26 2.29 the D value of watermelon
0.16 13 wa
HXF1 0.19 1.00 0.66 0.00 0.29 0.48
analyzed via cluster analysis. As shown in Figure 11, 13 watermelon genotypes were clus 5
JR3 0.00 0.43 0.78 0.15 0.92 0.33 11
14X1 0.32 tered into
0.74 4 categories. 0.00 The first 0.08
category was2.43 the highly-drought-resistant
0.54 2 germplasm
14X4 0.53 0.88 14X5; the0.31
including second category0.29 was the−0.44 drought-resistant 0.50 4
germplasms, including
14X5 0.67 0.94 1.00 0.57 2.72 0.95 1
14X6 1.00 LK13, JLR,
0.00 HXF1, 14X4, 0.56 14X1, and0.2314X6; the third 0.14 category was 0.50 the low-drought-resistan
3
14X7 0.11 germplasms,
0.05 including0.61 21F05, JH1,0.16 JR3, 14X7,2.11 and 16F02; 0.33 the fourth category
10 was the
drought-sensitive germplasm,
Note: U(X1 ) to U(X5 ) represent including
the membership 16C07.
function values of the first principal component to the fifth
principal component, respectively. The D-value is the comprehensive evaluation value.

Figure 11. Systematic clustering of 13 watermelon genotypes based on D-values.

Figure 11. Systematic clustering of 13 watermelon genotypes based on D-values.
3. Discussion
Table 4. Membership
Global climate function values
change has of five
brought mainextreme
about factors and D-value
drought, andranks of droughtof
the constraints tolerance.
drought on agriculture have also intensified. Breeding and the promotion of new drought- Tolerance
Germplasm
resistant varietiesU(X ) mostU(X
are 1the 2)
economical U(Xeffective
and 3) U(X4)
measures. U(X
The 5) D-Value
breeding of drought-
Rank
resistant varieties is inseparable from drought-resistant germplasms. Therefore, the identifi-
LK13 0.26 0.74 0.40 1.00 −0.10 0.46 6
cation methods of plant drought resistance (simple and easy-to-operate), the plant response
to drought stress (the physiological and biochemical), and the comprehensive evaluation
of plant drought resistance are very important for the screening and identification of
drought-tolerant germplasms.

3.1. Identification Method of Plant Drought Resistance

The drought resistance of plants is affected by the identification methods. At present,
the identification methods of plant drought resistance include PEG-simulated drought
at the bud stage [13–21], PEG-simulated drought at the seedling stage [20,22–28], the pot
water control method at the seedling stage [29–35], and the field natural identification
method [36–44]. In this paper, the drought resistance of 13 different genotypes of water-
Plants 2025, 14, 1289 14 of 20

melon was identified using the pot water control method at the seedling stage. However,
different identification methods have their own advantages and disadvantages. The field
identification method is closer to the field environment but is affected by uncontrollable
rainfall and humidity. The study of Cai et al. (2020) [22] showed that there was no signifi-
cant difference in the adverse effects of PEG-simulated drought stress and water control
drought stress on barley growth. However, the decrease in the soil water content caused
by drought is a slow process, while PEG treatment rapidly causes osmotic stress, so PEG
cannot fully simulate drought stress conditions. The pot water control method at seedling
are suitable for the laboratory and greenhouse, which can effectively and accurately control
water. Therefore, it is more referential and operable to identify the drought tolerance of
watermelon using the pot water control method at the seedling stage.

3.2. Response of Plants to Drought Stress

Drought stress inhibits plant growth, and plants respond to drought stress through
a series of morphological, physiological, and biochemical adaptive evolution. The root
is the sensor of plants, which senses osmotic stress under drought stress and plays an
important role in the mechanism of plant drought resistance. Mahmood et al. (2022) [45]
showed that the root length, root volume, and root number are key indicators of drought
resistance in cotton. The study of Guo et al. (2024) [46] showed that the root architecture
of drought-resistant cotton varieties showed a significant increase in the average length
of all lateral roots and a significant decrease in the average lateral root emergence angle,
while the drought-sensitive cotton varieties showed the opposite trend. The results of this
study showed that drought significantly inhibited the growth of watermelon roots and
significantly reduced the total root length, root volume, root area, root tips, and forks. Due
to the pot water control method used in this paper, drought stress has seriously inhibited
the root growth of all tested watermelon genotypes when scanning the roots on the 7th day
of continuous drought.
The changes in the osmotic-adjustment substance contents and antioxidant enzyme
activity are the key indicators of plant responses to drought stress [31,45]. Previous studies
have shown that the contents of osmotic-adjustment substances (soluble protein and
proline) in Gleditsia sinensis and oak (Quercus) were significantly increased under drought
stress [31,47]. The results of this study showed that drought stress significantly increased
the contents of soluble sugar, soluble protein, proline, and starch, which was consistent
with previous studies.
The activity of antioxidant enzymes determines the level of ROS, and high levels of
ROS lead to plant membrane damage. Liu et al. (2023) [31] and Xiong et al. (2022) [47]
reported that drought stress significantly increased the activities of SOD, POD, and CAT in
Gleditsia sinensis and oak, while Islam et al. (2020) [48] reported that the activities of SOD,
POD, and CAT in sugar beet decreased significantly under drought stress. However, the
results of this study showed that the antioxidant enzyme activities of different watermelon
genotypes showed different levels of increase or decrease. The reason for this result is that
on the one hand, different genotypes of watermelon have different resistances to drought
stress, but on the other hand, this paper only sampled and measured on the 7th day of
continuous drought, so the result is reasonable. Drought stress causes oxidative stress
and membrane damage in plants. Wang et al. (2024) [49] showed that drought stress
significantly increased the accumulation of ROS and MDA contents in tobacco (Nicotiana
tabacum L.) seedlings. What is more, Liu et al. (2023) [31] and Xiong et al. (2022) [47] also
reported that the MDA content in Gleditsia sinensis and oak increased with the aggravation
of drought. This is consistent with the results of this study that drought stress significantly
increased the H2 O2 and MDA contents of all watermelon genotypes.
Plants 2025, 14, 1289 15 of 20

3.3. Comprehensive Evaluation of Plant Drought Resistance

Plant drought resistance is a comprehensive biological trait controlled by multiple
genes. A single index cannot directly reflect the drought resistance of plants. Therefore, a
comprehensive evaluation of plant drought resistance using multiple indicators is one of
the effective methods to identify plant drought resistance. Badr et al. (2020) [13] identified
the drought resistance of maize and used the frequency of the 5% optimal traits and 5%
worst traits to screen high-resistance and high-sensitivity maize germplasms. Liu et al.
(2023) [31] used the drought resistance index (drought stress index in this paper) as the
main index, combined with growth, leaf morphology, and photosynthetic physiological
indexes, to evaluate and identify the drought resistance of Gleditsia sinensis. Zhang et al.
(2021) [50] calculated the membership function values of six indexes and evaluated the
drought resistance of Iris germanica. Li et al. (2023) [51] transformed 13 physiological
and biochemical indexes into four independent comprehensive indexes via PCA and
evaluated the drought resistance of lettuce using three evaluation methods (D value, CDC,
and WDC), and the results showed that there was no significant difference among the
three evaluation methods. In this paper, the drought stress index was calculated, and
the drought stress index was analyzed via correlation analysis and PCA. The 27 traits
were transformed into five principal component factors, and the membership function
value and comprehensive evaluation D value were calculated. And the D value was
used to evaluate the drought resistance. Finally, the 27 traits of 13 watermelon resources
were comprehensively evaluated for drought resistance, and one highly-drought-resistant
germplasm and six drought-resistant germplasms were screened.
For the identification of watermelon drought resistance, He et al. (2023) [30] calculated
the membership function value of the relative change rate of 13 traits of watermelon
germplasms and comprehensively evaluated the drought resistance of watermelon. The
physiological and biochemical traits of watermelon investigated in this paper are as many
as 27, which are more comprehensive and representative.
In addition, this paper calculated the drought stress index, phenotypic plasticity index,
and membership function value, carried out correlation analysis, PCA, and cluster analysis,
and finally obtained the comprehensive evaluation D value [45]. The evaluation method
can more truly and effectively reflect the drought resistance of plants.

4. Materials and Methods

4.1. Plant Materials
In this study, 13 watermelon genotypes with diverse genetic backgrounds preserved
in the germplasm resource bank of the Vegetable Research Institute of Gansu Academy
of Agricultural Sciences were used as materials. Detailed material information is shown
in Table 5.

Table 5. Thirteen watermelon germplasms with different genotypes.

Peel Covering
Name Fruit Shape Flesh Color Seed Morphology Type Abbreviation
Type
Long Ke 13 Ellipse Green peel Red Brown small seed Big fruit size LK13
2021F05 Roundness Green peel Pink dark brown middle seed Small fruit type 21F05
2016F02 Roundness Green peel Red dark brown middle seed Small fruit type 16F02
Jin Hua 1 Ellipse Green peel Red Brown small seed Big fruit size JH1
Jiao Li Ren Roundness Green peel Yellow Black small seed Small fruit type JLR
2016C07 Roundness Green peel Red Black small seed Small fruit type 16C07
Hua Xin F1 Roundness Green peel Pink Black small seed Small fruit type HXF1
JR3 Roundness Black peel Yellow Black middle seed Seed watermelon JR3
2014X1 Roundness Green peel Faint yellow Red middle seed Wild watermelon 14X1
Plants 2025, 14, 1289 16 of 20

Table 5. Cont.

Peel Covering
Name Fruit Shape Flesh Color Seed Morphology Type Abbreviation
Type
2014X4 Roundness Green peel Faint yellow Yellow medium seed Wild watermelon 14X4
2014X5 Roundness Green peel White White medium seed Wild watermelon 14X5
2014X6 Roundness Green peel White Yellow medium seed Wild watermelon 14X6
2014X7 Roundness Walnut peel White Yellow big seed Wild watermelon 14X7

4.2. Material Planting and Drought Stress Treatment

The experiment was carried out in the greenhouse of Langou Village, Jiuhe Town,
Lanzhou City. Watermelon seeds were soaked for germination on 23 March 2024 and sowed
in pots (12.5 cm in diameter/11.2 cm in height) on 25 March 2024. The drought-stress
treatment was started on 25 April 2024 (when the seedlings grew to the four-leaf stage). The
experiment used natural drought treatment, that is, continuous drought treatment using
a pot water control. During the seedling period, normal water management was carried
out. When the seedlings grew to the four-leaf stage, the seedlings with uniform growth
were selected for treatment. Sixty seedlings with the same growth were selected from each
material and divided into control and treatment groups, with 30 seedlings in each group.
The treatment group was not watered for 7 days, while control plants were maintained
at 75% field capacity. The growth and physiological and biochemical parameters were
measured at the end of the stress period.

4.3. Measuring Indicators and Methods

4.3.1. Drought Injury Grade and Drought Injury Index
The classification of the drought injury grade was based on He et al. (2023) [30], and
the five levels of S0–S4 in the literature were changed to S1–S5 in turn.

Drought injury index (DI) = (1 × S1 + 2 × S2 + 3 × S3 + 4 × S4 + 5 × S5)/total number of plants, (1)

where S1–S5 are the number of plants corresponding to drought levels 1 to 5. The greater
the average drought injury index, the worse the drought resistance.

4.3.2. Seedling Growth Index

Three seedlings were randomly selected from each treatment. The seedlings were
divided into aboveground and underground parts from the cotyledon internodes, and the
weights of the shoot and root were weighed using a one-thousandth electronic scale. The
root fresh weight was weighed after washing the root and wiping the surface moisture.
The dry weight of the material was deactivated at 105 ◦ C for 30 min and baked at 80 ◦ C for
24 h to a constant weight.

4.3.3. Root Architecture

Three seedlings were randomly selected from each treatment, and the roots were
washed with clean water, and then, the root architecture was analyzed using a plant root
analyzer HM-GXONE.

4.3.4. Physiologic and Biochemical Index

Relative water content: Three leaves of the same node were selected for each
treatment, and the fresh weight (FW), saturated weight (TW; saturated leaf after full
immersion in water), and dry weight (DW) were weighted. Relative water content
(RWC) = (FW − DW)/(TW − DW) × 100%.
Plants 2025, 14, 1289 17 of 20

Relative conductivity: Three leaves of the same node were selected for each treatment,
and 10 leaf discs were taken with a 0.8 cm puncher and placed in a test tube, and then,
5 mL of deionized water was added. Put it into the vacuum pump to extract air for
10 min, and measure the initial conductance S1 after shaking well. After boiling the water
bath for 30 min, the conductivity S2 was measured after cooling to room temperature.
The conductivity of deionized water was used as the blank S0. Relative conductivity
(REC) = (S1 − S0)/(S2 − S0) × 100%.
Three seedlings were randomly selected from each treatment, and one fully expanded
leaf below the growth point was picked up and frozen in liquid nitrogen for the detection
of physiological and biochemical indexes. The chlorophyll content was determined using
ethanol extraction method.
SOD, POD, CAT, MDA, H2 O2 , O2− , soluble protein, soluble sugar, proline, and
starch were measured using the kits of Suzhou Grace Biotechnology Co., Ltd. (Suzhou,
China), and the product numbers were G0102F, G0108F, G0106F, G0110F, G0112F, G0116F,
G0417F, G0501F, G0111F96, and G0507F, respectively. All were operated according to the
kit instructions.

4.4. Comprehensive Evaluation of Drought Resistance

The drought stress index, phenotypic plasticity index, membership function value,
and comprehensive evaluation D value were calculated.
The drought stress index (DS) was the relative value of the treatment and control, and
the formula was as follows:
X
DS = DS (2)
XCK
where XDS represents the value of the drought stress treatment, and XCK represents the
value of the normal water control.
The phenotypic plasticity index (PPI) is the difference between the maximum drought
stress index and the minimum drought stress index divided by the maximum drought
stress index, and the formula was as follows:

MaxDS − MinDS
PPI = (3)
MaxDS

A PCA was performed on the drought stress index of all traits, and then, its member-
ship function value U(Xj ) was calculated:

X j − Xmin
U Xj = , j = 1, 2, . . . . . . , n (4)
Xmax − Xmin

where Xj is the jth principal component value of each variety, and Xmax and Xmin are
maximum and minimum values of jth principal component value, respectively.

Pj
Wj = n , j = 1, 2, . . . . . . , n (5)
∑ j =1 Pj

where Wj represents the importance of the jth principal component, that is, the weight; Pj
represents the contribution rate of the jth principal component.
The D value is the comprehensive evaluation value. The higher the D is, the material
is indicated as having greater comprehensive drought resistance. The calculation formula
for the D value is as follows:
n
∑


D= U X j × Wj , j = 1, 2, . . . . . . , n (6)
j =1
Plants 2025, 14, 1289 18 of 20

4.5. Statistics
Microsoft Excel was used for data collation, and SPSS 26 was used for significance
analysis, PCA, etc. Prism 9 software was used to make the histogram, and Origin 2024
software was used to analyze the related heat map and cluster map.

5. Conclusions
(1) This paper clarified the growth and physiological and biochemical responses of water-
melon seedlings under drought stress. Compared to the control, drought stress signif-
icantly reduced the fresh and dry weights, root length, root area, root volume, root
tips, and forks of watermelon seedlings. What is more, drought significantly reduced
the relative water content of leaves and increased the levels of osmotic-adjustment
substances (soluble sugars, soluble proteins, proline, and starch). Persistent drought
also modulated the activities of antioxidant enzymes (SOD, POD, and CAT), leading to
oxidative stress through the accumulation of H2 O2 , resulting in oxidative stress. Mem-
brane damage, indicated by a significant increase in the MDA content and relative
conductivity, was observed, adversely affecting seedling growth.
(2) The 13 watermelon genotypes were clustered into 4 categories. The first category
was a highly-drought-resistant germplasm, including 14X5; the second category
was drought-resistant germplasms, including LK13, JLR, HXF1, 14X4, 14X1, and
14X6; the third category was low-drought-resistant germplasms, including 21F05,
JH1, JR3, 14X7, and 16F02; the fourth category was a drought-sensitive germplasm,
including 16C07. This study provides the material basis for watermelon drought
resistance breeding.

Author Contributions: W.K. and Y.S. designed the research; K.R. and T.T. conducted the experiments;
Y.W., Y.Y. and X.Z. performed data analysis; K.R. wrote the manuscript; H.C. reviewed and edited the
manuscript. All authors have read and agreed to the published version of the manuscript.

Funding: This work was supported by the Major Science and Technology Projects of Gansu Provincial
(Grant Number: 24ZDNA005), the National Natural Science Foundation of China (Grant Number:
32460762), the Key Research and Development Program of Gansu (Grant Number: 25YFNA026), and
the Ministry of Agriculture and Rural Affairs of China (Grant Number: CARS-25).

Data Availability Statement: All relevant data are within the paper.

Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Zhu, Y.L.; Liu, Y.; Kong, X.H.; Wang, X.X.; Zhang, M.Q.; Hong, X.W.; Chen, H.P.; Sun, J.Q. Research progress and prospect on the
drought, heatwave, and compound drought and heatwave events in China. Trans. Atmos. Sci. 2025, 48, 26–36. [CrossRef]
2. Chen, Y.N.; Li, Z.; Fang, G.H.; Li, Y.P. Global drought variation and its adaptation. Sci. Technol. Rev. 2025, 1–6. Available online:
http://kns.cnki.net/kcms/detail/11.1421.N.20241212.1623.006.html (accessed on 20 April 2025).
3. Cao, Y.; Yang, W.; Ma, J.; Cheng, Z.; Zhang, X.; Liu, X.; Wu, X.; Zhang, J. An Integrated framework for drought stress in plants. Int.
J. Mol. Sci. 2024, 25, 9347. [CrossRef] [PubMed]
4. Hura, T.; Hura, K.; Ostrowska, A. Drought-stress induced physiological and molecular changes in plants. Int. J. Mol. Sci. 2022,
23, 4698. [CrossRef] [PubMed]
5. Li, Y.Y.; Zhao, X.; Yue, L.; Qiu, Y. Research progress and prospect on the interaction between plant root exudates and rhizosphere
microorganisms under drought stress. J. Arid Meteorol. 2024, 42, 325–337. [CrossRef]
6. Mukherjee, A.; Dwivedi, S.; Bhagavatula, L.; Datta, S. Integration of light and ABA signaling pathways to combat drought stress
in plants. Plant Cell Rep. 2023, 42, 829–841. [CrossRef]
7. Zhang, C.Q.; Li, L.L. Citespace-based analysis of the effects of high temperature and drought on plant growth and metabolism. J.
Mt. Agric. Biol. 2024, 43, 53–60. [CrossRef]
8. Zhang, H.; Zhu, J.; Gong, Z.; Zhu, J.K. Abiotic stress responses in plants. Nat. Rev. Genet. 2022, 23, 104–119. [CrossRef]
Plants 2025, 14, 1289 19 of 20

9. Gupta, A.; Rico-Medina, A.; Caño-Delgado, A.I. The physiology of plant responses to drought. Science 2020, 368, 266–269.
[CrossRef]
10. Hura, T.; Dziurka, M.; Hura, K.; Ostrowska, A.; Dziurka, K. Different allocation of carbohydrates and phenolics in dehydrated
leaves of triticale. J. Plant Physiol. 2016, 202, 1–9. [CrossRef]
11. Basu, S.; Ramegowda, V.; Kumar, A.; Pereira, A. Plant adaptation to drought stress. F1000Research 2016, 5, 1554. [CrossRef]
[PubMed]
12. FAOSTAT. Data/QCL. Available online: https://www.fao.org/faostat/en/ (accessed on 25 January 2025).
13. Badr, A.; El-Shazly, H.H.; Tarawneh, R.A.; Börner, A. Screening for drought tolerance in maize (Zea mays L.) germplasm using
germination and seedling traits under simulated drought conditions. Plants 2020, 9, 565. [CrossRef]
14. Hou, S.; Qin, H.B.; Li, M.; Wang, H.G.; Mu, Z.X. Identification of agronomic traits and drought tolerance at germination stage of
44 local foxtail millet varieties from Shanxi Province. Agric. Res. Arid Areas 2024, 42, 26–32. [CrossRef]
15. Huang, Y.X.; Liu, X.; Yan, S.J.; Li, N.; Zhang, X.W.; Liu, Z.; Wang, W.P. Evaluation and screening of rice germplasm resources for
drought resistance in germination stage. Hybrid Rice 2024, 39, 27–35. [CrossRef]
16. Li, L.; Shen, B.Y.; Zhang, T.J.; Yang, T.; Liu, R.; Zong, X.X. Evalation and screening of pea (Pisum Sativum) germplasm resources for
drought resistance during germination stage. J. Plant Genet. Resour. 2017, 18, 778–785. [CrossRef]
17. Xie, J.G.; Wang, M.L.; Zhang, Y.F.; Meng, F.F.; Zheng, Y.H.; Li, G.; Sun, X.M.; Fan, X.H.; Yang, Z.Y.; Wang, S.M.; et al. QTL mapping
and related gene mining of drought tolerance at germination stage in soybean. Chin. J. Oil Crop Sci. 2025, 1–10. [CrossRef]
18. Zhao, M.; Yang, S.; Jiang, H.W.; Xie, J.G.; Zhou, R.; Zheng, L.P.; Meng, F.F.; Wang, S.M. Identification and comprehensive
evaluation of drought-tolerance soybean germplasm resources at bud stage. Soybean Sci. 2023, 42, 701–709.
19. Zhao, X.; Liu, Z.; Li, H.; Zhang, Y.; Yu, L.; Qi, X.; Gao, H.; Li, Y.; Qiu, L. Identification of Drought-tolerance genes in the
germination stage of soybean. Biology 2022, 11, 1812. [CrossRef]
20. Long, J.; Dong, M.; Wang, C.; Miao, Y. Effects of drought and salt stress on seed germination and seedling growth of Elymus
nutans. PeerJ 2023, 11, e15968. [CrossRef]
21. Vuksanović, V.; Kovačević, B.; Kebert, M.; Pavlović, L.; Kesić, L.; Čukanović, J.; Orlović, S. In vitro selection of drought-tolerant
white poplar clones based on antioxidant activities and osmoprotectant content. Front. Plant Sci. 2023, 14, 1280794. [CrossRef]
22. Cai, K.; Chen, X.; Han, Z.; Wu, X.; Zhang, S.; Li, Q.; Nazir, M.M.; Zhang, G.; Zeng, F. Screening of worldwide barley collection for
drought tolerance: The assessment of various physiological measures as the selection criteria. Front. Plant Sci. 2020, 11, 1159.
[CrossRef] [PubMed]
23. Hou, L.Y.; Chen, Y.H.; Wang, Y.C.; Hong, D.X.; Wang, Y.X.; Chen, Z.; Wu, S.J.; Dong, Y.H. Influence of drought stress duration on
quinoa physiological characteristics and drought tolerance evaluation at seedling stage. J. Nucl. Agric. Sci. 2024, 38, 2237–2246.
24. Wang, L.; Deng, J.; Zhang, Z.; Zhao, M.W.; Che, X.Y.; Wang, G.Y.; Guo, X.; Zhang, H.Y.; He, L.; Weng, J.F.; et al. Identification and
evaluation of drought tolerant germplasm resources at seedling stage of maize under PEG stress. Crops 2024, 40, 43–53. Available
online: http://zwzz.chinacrops.org/CN/10.16035/j.issn.1001-7283.2024.04.006 (accessed on 20 April 2025).
25. Yang, D.D.; Han, X.; Kong, X.X.; Zhao, P.F.; Jin, J.M.; Zhao, G.X.; Su, Y.Z.; Zhao, G.J. Identification and screening of drought
tolerance in 76 winter wheat varieties (lines) during seedling stage. China Seed Ind. 2024, 77–81. [CrossRef]
26. Mohi-Ud-Din, M.; Hossain, M.A.; Rohman, M.M.; Uddin, M.N.; Haque, M.S.; Ahmed, J.U.; Hossain, A.; Hassan, M.M.;
Mostofa, M.G. Multivariate analysis of morpho-physiological traits reveals differential drought tolerance potential of bread wheat
genotypes at the seedling stage. Plants 2021, 10, 879. [CrossRef]
27. Rida, S.; Maafi, O.; López-Malvar, A.; Revilla, P.; Riache, M.; Djemel, A. Genetics of germination and seedling traits under drought
stress in a MAGIC population of maize. Plants 2021, 10, 1786. [CrossRef]
28. Tan, W.; Li, W.; Li, J.; Liu, D.; Xing, W. Drought resistance evaluation of sugar beet germplasms by response of phenotypic
indicators. Plant Signal Behav. 2023, 18, 2192570. [CrossRef]
29. Chen, M.; Chen, L.; Li, S.J.; Xie, H.H.; Song, Y. Effects of drought stress on physiological characteristics and stomatal morphology
of different cassava seedlings. J. Sichuan Agric. Univ. 2024, 42, 1049–1056+1084. [CrossRef]
30. He, Y.P.; Yin, L.J.; Ding, X.L.; Wang, C.X.; Hou, Y.J.; Ma, Y.H.; Yu, R.; Yue, Z.; Yang, J.Q.; Zhang, X.; et al. Identification and
screening of drought resistance indexes of 25 watermelon germplasms at seedling stage. J. Northwest A F Univ. Nat. Sci. Ed. 2023,
51, 49–59. [CrossRef]
31. Liu, F.; Zhao, Y.; Wang, X.; Wang, B.; Xiao, F.; He, K. Physiological response and drought resistance evaluation of Gleditsia sinensis
seedlings under drought-rehydration state. Sci. Rep. 2023, 13, 19963. [CrossRef]
32. Xu, A.Y.; Cao, B.; Xie, Y. Physiological-ecological responses of twelve herbaceous plant species under drought stress and
evaluation of their drought resistance when planted in coal producting basis in arid windy and sandy areas. Acta Prataculturae
Sinica 2020, 29, 22–34. [CrossRef]
33. Wang, J.; Zhang, X.; Han, Z.; Feng, H.; Wang, Y.; Kang, J.; Han, X.; Wang, L.; Wang, C.; Li, H.; et al. Analysis of physiological
indicators associated with drought tolerance in wheat under drought and re-watering conditions. Antioxidants 2022, 11, 2266.
[CrossRef] [PubMed]
Plants 2025, 14, 1289 20 of 20

34. Zhao, L.; Jin, H.D.; Cao, X.Y.; Deng, W.H.; Du, L.J. Physiological response to drought stress and drought resistance of six Helleborus
orientlis cultivars. Chin. J. Appl. Ecol. 2023, 34, 2644–2654. [CrossRef]
35. Zhou, Y.X.; Li, P.T.; Su, J.S.; Wang, H.B.; Fang, W.M.; Chen, F.D.; Zhang, F. Evaluation of drought resistance in 37 accessions of
chrysanthemum related species. J. Nanjing Agric. Univ. 2023, 46, 1060–1068. [CrossRef]
36. Ahmed, S.F.; Ahmed, J.U.; Hasan, M.; Mohi-Ud-Din, M. Assessment of genetic variation among wheat genotypes for drought
tolerance utilizing microsatellite markers and morpho-physiological characteristics. Heliyon 2023, 9, e21629. [CrossRef] [PubMed]
37. Bao, X.; Hou, X.; Duan, W.; Yin, B.; Ren, J.; Wang, Y.; Liu, X.; Gu, L.; Zhen, W. Screening and evaluation of drought resistance
traits of winter wheat in the North China Plain. Front. Plant Sci. 2023, 14, 1194759. [CrossRef]
38. Martínez, I.; Muñoz, M.; Acuña, I.; Uribe, M. Evaluating the drought tolerance of seven potato varieties on volcanic ash soils in a
medium-term trial. Front. Plant Sci. 2021, 12, 693060. [CrossRef]
39. Shojaei, S.H.; Mostafavi, K.; Omrani, A.; Illés, Á.; Bojtor, C.; Omrani, S.; Mousavi, S.M.N.; Nagy, J. Comparison of maize genotypes
using drought-tolerance indices and graphical analysis under normal and humidity stress conditions. Plants 2022, 11, 942.
[CrossRef]
40. Sun, F.; Chen, Q.; Chen, Q.; Jiang, M.; Gao, W.; Qu, Y. Screening of key drought tolerance indices for cotton at the flowering and
boll setting stage using the dimension reduction method. Front. Plant Sci. 2021, 12, 619926. [CrossRef]
41. Tiwari, P.N.; Tiwari, S.; Sapre, S.; Tripathi, N.; Payasi, D.K.; Singh, M.; Thakur, S.; Sharma, M.; Tiwari, S.; Tripathi, M.K.
Prioritization of physio-biochemical selection indices and yield-attributing traits toward the acquisition of drought tolerance in
chickpea (Cicer arietinum L.). Plants 2023, 12, 3175. [CrossRef]
42. Yang, J.; Han, D.X.; Wang, Y.J.; Li, M.D.; Xi, H.J.; Abulati, A.; Liang, X.L. Drought Tolerance Identification and Evaluation of 26
Maize Inbred Lines under Natural Drought Condition in Xinjiang. Xinjiang Agric. Sci. 2019, 56, 1388–1396.
43. Zhou, Y.Q.; Yang, Y.Z.; Zhou, W.Q.; Lian, X.R.; Zhang, Y.J.; Wang, X.R.; Kou, S.R.; He, H.J.; Liu, Z.X.; Wang, X.J. Evaluation and
selection of drought resistance inbred lines of maize under drought stress. Agric. Res. Arid Areas 2020, 38, 211–217. [CrossRef]
44. Zhao, X.Z.; Xu, J.Y.; Yu, L.L.; Gu, Y.Z.; Liu, W.X.; Qiu, L.J. Field drought tolerance evaluation and excellent germplasm
identification of soybean germplasm. Soybean Sci. 2020, 39, 825–832. [CrossRef]
45. Mahmood, T.; Iqbal, M.S.; Li, H.; Nazir, M.F.; Khalid, S.; Sarfraz, Z.; Hu, D.; Baojun, C.; Geng, X.; Tajo, S.M.; et al. Differential
seedling growth and tolerance indices reflect drought tolerance in cotton. BMC Plant Biol. 2022, 22, 331. [CrossRef] [PubMed]
46. Guo, C.; Zhu, L.; Sun, H.; Han, Q.; Wang, S.; Zhu, J.; Zhang, Y.; Zhang, K.; Bai, Z.; Li, A.; et al. Evaluation of drought-tolerant
varieties based on root system architecture in cotton (Gossypium hirsutum L.). BMC Plant Biol. 2024, 24, 127. [CrossRef]
47. Xiong, S.; Wang, Y.; Chen, Y.; Gao, M.; Zhao, Y.; Wu, L. Effects of drought stress and rehydration on physiological and biochemical
properties of four oak species in China. Plants 2022, 11, 679. [CrossRef]
48. Islam, M.J.; Kim, J.W.; Begum, M.K.; Sohel, M.A.T.; Lim, Y.S. Physiological and biochemical changes in sugar beet seedlings to
confer stress adaptability under drought condition. Plants 2020, 9, 1511. [CrossRef]
49. Wang, X.D. Evaluation of growth, physiological response, and drought resistance of different flue-cured tobacco varieties under
drought stress. Front. Plant Sci. 2024, 15, 1442618. [CrossRef]
50. Zhang, J.; Huang, D.; Zhao, X.; Zhang, M. Evaluation of drought resistance and transcriptome analysis for the identification of
drought-responsive genes in Iris germanica. Sci. Rep. 2021, 11, 16308. [CrossRef]
51. Li, J.; Abbas, K.; Wang, L.; Gong, B.; Hou, S.; Wang, W.; Dai, B.; Xia, H.; Wu, X.; Lü, G.; et al. Drought resistance index screening
and evaluation of lettuce under water deficit conditions on the basis of morphological and physiological differences. Front. Plant
Sci. 2023, 14, 1228084. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.