Plant Science 319 (2022) 111251

Contents lists available at ScienceDirect

Plant Science
journal homepage: www.elsevier.com/locate/plantsci

Overexpression of cowpea NAC transcription factors promoted growth and

stress tolerance by boosting photosynthetic activity in Arabidopsis
Richa Srivastava a, Yuriko Kobayashi b, Hiroyuki Koyama b, Lingaraj Sahoo a, *, 1
a
Department of Biosciences and Bioengineering, Indian Institute of Technology Guwahati, Assam, 781039, India
b
Faculty of Applied Biological Sciences, Gifu University, 1-1, Yanagido, Gifu 501-1193, Japan

A R T I C L E I N F O A B S T R A C T

Keywords: ATAF-like NAC transcription factors are bonafide regulators of stress-signaling. However, their overexpression
Cowpea NAC transcription factor often exerts growth-retardation by activating ABA-hypersensitivity, chloroplast-degradation, or carbon-
Growth/tolerance balance starvation. To improve tolerance to multiple stress complying with growth sustainability, we examined two
Improved growth
ATAF orthologs, VuNAC1 and VuNAC2, isolated from a drought-hardy cowpea genotype, for a harmonized
Multiple stress tolerance
Photosynthetic activity
regulation of stress and growth signaling. The genes were induced by dehydration, NaCl, polyethylene glycol,
Stomatal density heat, cold, ABA, and light. Analysis of the promoter-elements and regulatory network corroborated the inte­
gration of circadian, hormonal, stress, developmental, and nutrition signals, being VuNAC1/2 the central
transcriptional-switch interfacing growth and stress responses. The constitutive gene overexpression in Arabi­
dopsis resulted in an improved embryonic, rosette, and inflorescence growth, under optimum as well as limiting
nutrition, in association with increased photosynthetic activity and stomatal-density. The transgenic seedlings
manifested tolerance to dehydration, salinity, aluminum, cadmium, and H2O2 toxicity, in addition to ABA-
mediated seed dormancy and hypersensitivity. The soil-grown plants survived severe drought and hypersalini­
ty by maintaining the water-status and membrane integrity through the accumulation of stress protectants, such
as proline, glutathione, and ascorbate. Unlike their orthologs from other species, VuNAC1/2 conferred tolerance
to multiple abiotic stresses in line with improved growth attributes via regulation of photosynthetic controls and
nutritional balance, suggesting growth being a crucial component of stress-tolerance and recovery. Such unique
stress-responsive transcription factors, which also confer photosynthetic gain, could be sustainable biotechno­
logical tools for developing stress-tolerant crops and translating the improved growth into yield without unin­
tended trade-offs.

1. Introduction
Plants encounter multiple stresses under field conditions making it
challenging to extrapolate the tolerance mechanisms for the combined
effects [1]. Diverse breeding strategies have been employed but ach­
ieved only partial success due to the polygenic trait of the tolerance
mechanism [2]. One approach to confer tolerance to multiple stress
tolerances would be genetic engineering of transcriptional programs
that plays a crucial role in stress tolerance in plants. Recent studies
uncovered that several transcription factors (TFs) regulate multiple
stress tolerance together [3]. For example, STOP1, a critical regulator of
aluminum tolerance, can also control biotic and abiotic stress responses
by exerting pleiotropic effects through target genes such as PGIP1,
HsfA2, GDHs, ALMT1, etc [4,5]. Many other TF families, such as MYB,
bZIP, DREB, NAC, etc., are also emerging as central regulators of stress
responses, hence ideal biotechnological tools for developing multiple
stress tolerance in crops [6–8]. For instance, overexpression of
OsHBP1b, a bZIP TF, controls diverse abiotic stresses in rice through
enhanced antioxidant and photosynthesis [9]. The ectopic expression of
OsMYBR1 TF, a member of the rice MYB family, conferred tolerance to
drought and chromium stress in Arabidopsis [10]. Overexpression by
genetic engineering of such TFs would be one useful approach to confer
multiple stress tolerance of plants.
The ATAF (Arabidopsis Transcription Activation Factor) members of
the NAC family have been identified and ascribed for their role in
various abiotic and biotic stress responses in the Arabidopsis model and

* Corresponding author.
E-mail address: ls@iitg.ac.in (L. Sahoo).
1
https://orcid.org/0000–0001-8103–2714

https://doi.org/10.1016/j.plantsci.2022.111251
Received 3 February 2022; Received in revised form 7 March 2022; Accepted 10 March 2022
Available online 14 March 2022
0168-9452/© 2022 Elsevier B.V. All rights reserved.
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 1. Identification of stress-responsive NAC genes from cowpea and their time-course induction under abiotic stresses, stress hormones, and light. (A) Degenerate
sequences of ATAF-like NAC members from Arabidopsis, rice, and soybean (encircled in red, left) were used as a template to amplify stress-inducible orthologs from
15-day-old cowpea seedlings of a drought-hardy genotype (Kannanado White) treated with 20% of PEG and 200 mM of NaCl (middle). The clones obtained were
aligned with the ESTs available for cowpea NAC genes and ATAF orthologs (ATAF1/2, ANAC032, and ANAC032) to study their phylogenetic relationships (right). (B)
Time-dependent expression analysis by semi-quantitative RT-PCR of 15-day-old cowpea seedlings showed strong induction of VuNAC1 and VuNAC2 by dehydration,
PEG, NaCl, and aluminum. (C) The fold-change in the VuNAC1/2 transcripts in response to stress was determined by qRT-PCR. The housekeeping VuUbq1
(XM_028074892) expression was used to normalize the gene expression values. The experiment was repeated more than three times to obtain similar outcomes. The
letters a, b, c, etc., indicate significant differences in the stress responses at different time intervals at p < 0.05 (Tukey’s test). (D) VuNAC1 and VuNAC2 showed
strong induction by heat, cold, abscisic acid (ABA), methyl jasmonate (MeJA), and (E) Rapid and transient induction at 6:00 AM time-point (start of diurnal
photoperiod) indicated circadian induction of VuNAC1/2 by light.

edible crops such as rice and soybean [11–13]. A single ATAF member
can serve multiple roles, e.g., ATAF1 controls drought tolerance, chlo­
roplast maintenance, senescence cascades, photosynthesis, ABA
biosynthesis, carbon starvation, and developmental processes in Arabi­
dopsis, mediated by proteins such as RD22/RD29A (RESPONSIVE TO
DESICCATION 22/29A), COR47 (cold regulated 47), NCED3 (NINE-­
CIS-EPOXYCAROTENOID DIOXYGENASE 3), GLK1 (GOLDEN2-LIKE 1),
SnRK1 (SNF1-Related Protein Kinase 1), etc [14,15]. This suggests that
genetic engineering of ATAF would be one promising approach in mo­
lecular breeding of stress tolerance. The nature of the NAC TF is majorly
dictated by the conserved NAC subdomains, determining the set of
target genes and the interacting TF co-partners. However, perturbation
in ABA-signalling can lead to detrimental crosstalk between stress re­
sponses and growth. For instance, authentic ATAF1 overexpression in
Arabidopsis enhanced tolerance towards drought response but led to
dwarf and short primary root phenotypes, increasing sensitivity to ABA,
salt, and oxidative stress in Arabidopsis [11]. Moreover, the ABA signal
may activate the carbon starvation signal due to the ATAF1-SnRK
interaction [14]. Similarly, overexpression of ATAF homolog in rice
(OsNAC6) improved tolerance to high salinity but retarded the growth of
transgenic rice [12]. These results suggest that genetic engineering of
ATAF-like proteins would require procedures that escape over­
expression’s adverse effects, including cross-talk of ABA signaling.
There are several strategies to minimize the negative effects caused
by overexpression of stress-responsive genes. For example, substituting
the constitutive 35 S CaMV promoter with an inducible RD29A promoter
improved drought tolerance by alleviating the growth anomalies asso­
ciated with DREB1A expression [16]. This strategy might be suitable for
improving tolerance to short-duration intermittent stress treatment, but
the growth may never recover after persistent and aggravated stress
conditions in fields, resulting in massive yield penalties [17]. Ap­
proaches like gene-stacking involving co-expression of stress-responsive
and growth-regulating genes to compensate for the trade-offs are com­
plex [18]. Another suitable approach could be the overexpression of
versatile TFs, playing dual roles by amalgamating transcriptional net­
works of stress tolerance with growth improvement. Our previous
studies found two ATAF-like NAC TFs encoded by VuNAC1 and VuNAC2
cloned from the drought-resilience cowpea (Vigna unguiculata L. Walp)
genotype, accelerated growth and improved tolerance to multiple
growth-inhibiting stresses by reprogramming energy-producing
biosynthetic pathways when expressed heterologously in a yeast sys­
tem [19]. Indeed the transformed strains survived starvation and man­
ifested enhanced stress tolerance under limited nutrition. Such ATAF
homologs isolated from robust genetic sources hold potential for sus­
tainable improvement of stress tolerance by boosting growth traits,
hence offering dual benefits [20].
This study examined the induction of the VuNAC1 and VuNAC2 on
the subjection of multiple abiotic stresses and validated their functional
roles in the signaling of plant stress and growth processes. When over­
expressed constitutively in Arabidopsis, the VuNAC1/2 conferred
improved growth, yield, and tolerance to various abiotic stresses such as
drought, salinity, and metal toxicity under optimum as well as limiting
nutrition. The balanced growth and stress phenotype could be deter­
mined by several possible factors such as improved photosynthetic ac­
tivity or suppression of carbon-starvation signals, hormonal
perturbation, etc., triggered by VuNAC1/2, unlike their Arabidopsis
homolog [14,15]. To investigate that, we characterized the photosyn­
thetic parameters, stomatal abundance, and ABA-sensitivity in relation
to nutrition availability. Also, we proposed a foundational regulatory
model based on orthologous integrative gene regulatory networks.
2. Materials and methods
2.1. Preparation of expression vectors and data sources
The VuNAC1/2 genes were cloned from a drought-hardy cowpea
genotype (Kannando White) [21], as described in our earlier work [19].
The complete ORF sequences were submitted to Genbank with accession
numbers OM141592 and OM141593 (https://www.ncbi.nlm.nih.gov/
genbank/). The 35 S:VuNAC1/2 constructs used to generate the over­
expressor Arabidopsis lines were prepared by cloning the 888 bp ORFs at
the SfiI site of pBEAB, a T-DNA vector (provided from Gifu University,
Japan). The empty vector was used as the control. The constructs were
mobilized in the EHA105 strain of Agrobacterium tumefaciens,

2
R. Srivastava et al.
Fig. 1. (continued).
independently using tri-parental mating through the pRK2013 helper
plasmid. The detailed primer information was listed in Supplementary
Table S12. For their phylogenetic clustering, the ESTs available for the
cowpea NAC family and Arabidopsis NAC genes were downloaded from
the Plant TFDB 4.0 database (http://planttfdb.gao-lab.org/index.php)
[22]. The promoter sequences were retrieved from the NCBI database
(https://www.ncbi.nlm.nih.gov), and the regulatory elements were
identified using the PLACE tool (https://www.dna.affrc.go.jp/PLACE/?
action=newplace) [23].
3
Plant Science 319 (2022) 111251
2.2. Expression analysis and stress treatment
Healthy cowpea seeds (Kannando White) were germinated by soaking
in water for two days in the dark followed by a long-day photoperiod
condition (16 hr of light, 8 hr of dark) at 28 ◦ C, with white light illu­
mination (110 µmol photons m− 2s− 1). Four-day-old germinated seed­
lings were transferred to a gauzed hydroponics container supplied with
modified Hoagland hydroponics media for approximately ten days until
the first trifoliate leaves expanded [24]. For various stress treatments,
the hydroponics media was supplemented with 20% PEG 6000, 200 mM
NaCl, 50 µM of AlCl3 (pH 5.0), 50 µM ( ± ) ABA, and 50 µM methyl
R. Srivastava et al.
jasmonate (MeJA), individually. For heat and cold treatments, the plants
were transferred to the respective growth chambers maintained at 45 ◦ C
and 4 ◦ C. For dehydration stress, the leaves were dried at 26 ◦ C until
10% fresh weight loss. The trifoliate leaves from the stressed and the
control plants were sampled at the indicated time intervals for RNA
extraction using the RNeasy mini kit (Qiagen), followed by cDNA syn­
thesis (Applied Biosystems, USA). The semi-quantitative and quantita­
tive real-time PCR was performed in triplicates for each sample on a
Thermocycler Dice, Real-time system II (Takara, Japan). VuUbq1
(Genbank accession: XM_028074892) was used as a reference to deter­
mine the relative expression. The experiments were repeated three times
with similar results. The statistically significant stress response (p <
0.05) at different time-interval/treatment was calculated using post hoc
Tukey HSD (honest significant difference) (https://astatsa.com/
OneWay_Anova_with_TukeyHSD/).
2.3. Growth conditions for stress analysis
For the study in the seedling stage, healthy seeds of wild type Ara­
bidopsis (Col-0) and T2 generation transgenic seeds were sterilized in 1%
(v/v) sodium hypochlorite for 10 min, followed by washing with de-
ionized water and kept for stratification at 4 ◦ C for 48 hrs. The seeds
were transferred to plates containing 0.5X MS media supplemented with
1% (w/v) sucrose (added as indicated), 0.8% agar, (pH 5.8), and grown
in long-day photoperiod condition (16 hr of light, 8 hr of dark) at 25 ◦ C,
4
Plant Science 319 (2022) 111251
Fig. 2. Ectopic overexpression of VuNAC1 and
VuNAC2 in Arabidopsis. (A) Antibiotic resis­
tance of T2 generation seeds of three indepen­
dent transgenic lines (L1, L2, and L3) confirmed
the T-DNA transfer. (B) Quantification of the
transgene transcripts by qRT-PCR indicated
overexpression of VuNAC1/2 regulated by 35 S
promoter in the transgenic lines w.r.t. to vector
transformed control (pBEAB) and wildtype
(WT) seedlings. The housekeeping actin2
(NM_001338358.1) was used to normalize the
expression values. The experiment was
repeated more than three times to obtain
similar outcomes. Different letters indicate sig­
nificant differences (p < 0.05, post hoc Tukey
HSD) between the genotypes. (C) Amplification
of the transgene ORF (1.5 kb) and selection
marker gene (NptII) by genomic PCR also vali­
dated T-DNA integration in the VuNAC1/2
overexpressing transgenic lines and the vector-
transformed control.
with white light illumination (110 μmol photons m− 2 s− 1). The growth
media was supplemented with the chemicals as per the indicated con­
centration for various stress treatments. The experiments were per­
formed with three different lines (50 seedlings each) and repeated more
than thrice to get similar outcomes. For stress analysis of soil-grown
plants, 1-week old germinated seedlings were transferred to soilrite
mix and grown for four weeks while supplementing MS media every
week. For imposing nutrient stress, only water was supplied throughout
the vegetative growth of 4 weeks. For the drought assay, water was
withdrawn, and plants were maintained at 40% humidity for a week
until recovery. To inflict high salinity, 400 mM of NaCl solution was
supplied for a week. The experiments were performed with three
different lines (25 plants each) and repeated more than thrice to get
similar results. The significant difference between various parameters of
different genotypes (p < 0.05) was calculated using post hoc Tukey HSD.
2.4. Visualization of stomata by FESEM
The leaf sections were collected from a similar position in different
samples. The tissue was pre-fixed with 2.5% glutaraldehyde (w/v) so­
lution overnight. The samples were then dehydrated with an ethanol
gradient (10%, 20%, 30%, 50%, and 70%, once for 10 min at each step),
and then immersed in 100% ethanol (twice, 30 min each step). One-cm-
sections were mounted on carbon tape with their adaxial surface facing
upwards. The specimen was coated with gold and analyzed at 2000X
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 3. Characterization of growth phenotype of transgenic seedlings under normal and limiting growth conditions. (A) Embryonic (one week after germination) and
juvenile (3–4 weeks after germination) phenotypes were studied in optimum media (0.5X MS + 1% w/v sucrose) and limiting growth media lacking any carbon
supplement (0.5X MS). The transgenic seedlings (35 S:VuNAC1/2-Lx) showed accelerated germination and improved post-germination growth than the vector-
transformed control under both normal and limiting nutrition. The contrasting and unmasked growth phenotype was more apparent during the early germina­
tion stage under limiting growth conditions. (B) The growth parameters such as maximal leaf blade size (cm) and primary root length (cm) were increased in the
transgenic lines (L1, L2, and L3). All data are mean ± SE (n = 6). Different letters indicate significant differences between genotypes (p < 0.05, post hoc Tukey HSD).

5
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 4. Analysis of vegetative and reproductive growth traits off soil-grown plants under normal conditions. (A) The transgenic lines (L1, L2, and L3) displayed
improved rosette growth after 4-weeks. (B) Richer inflorescence growth was observed in 8-week old transgenic lines (top), producing more siliques resulting in
greater seed-yield (bottom). (C) The evaluation of rosette leaf size (cm), inflorescence height (cm), and siliques count indicated improved growth and yield attributes
in the transgenic plants. All data are mean ± SE (n = 6). Different letters indicate significant differences between genotypes (p < 0.05, post hoc Tukey HSD).

magnification using Field Emission Scanning Electron Microscopy
(FESEM) (Gemini 300, Zeiss, Germany), operating under an accelerating
voltage of 3 kV.
2.5. Gene regulatory models
The orthologous Arabidopsis NAC TFs were used to predict the
VuNAC1/2 coexpression network generated by ATTED II v. 10.1
(https://atted.jp/) [25]. The genes in the constructed network were
subjected to ontology analysis to annotate the ATAF-like VuNAC TFs
using the panther tool (http://go.pantherdb.org/genelistanalysis.do)
[26]. The upstream regulators were predicted using the RnR (Regulatory
Network Research) database of Arabidopsis T87 culture cells (http:
//webs2.kazusa.or.jp/kagiana/rnr0912/indexff.html).
The rest of the methodologies used in the study were described in the
Supplementary material.
3. Results
3.1. ATAF-like VuNAC1/2 transcription factors isolated from wild
cowpea genotype were induced in response to light, ABA, and multiple
abiotic stresses
In our previous study, we isolated two NAC genes (VuNAC1 and
VuNAC2) from a non-commercialized drought-hardy cowpea genotype
(Kannanado White) treated with NaCl and PEG-induced stress [19], using
a degenerate PCR approach seeded by the conserved regions of drought
and salt-responsive genes from various model plants such as Arabi­
dopsis, rice, and soybean [11,13,27] (Fig. 1A). The phylogenetic study
indicated close similarity of the cowpea NAC genes with the
stress-responsive ATAF subgroup in Arabidopsis (ATAF1, ATAF2,
ANAC03, and ANAC102). The complete ORFs (888 bp each) were
retrieved by the RACE PCR, encoding a protein containing 295 amino
acid residues. The signature NAC domain of 160 amino acid residues
was present in the N-terminal part, displaying a sequence identity of
93.5% (VuNAC1) and 85.8% (VuNAC2), with ATAF1 domain [11].
However, the overall protein sequence identity of 70.9% and 60.6% due
to high divergence in the C-terminal regions carrying crucial regulatory
motifs, suggested a possibly indifferent behavior from their orthologs in
other plants. The localization study of GFP-fused protein in tobacco
epidermal cells showed that the VuNAC1/2 proteins reside in the nu­
cleus (Supplementary Fig. S1). Our earlier studies have demonstrated
the DNA binding and transcriptional abilities of the proteins in their
native dimeric state [19].
The analysis of the gene promoters revealed over-representation of
the key cis-regulatory elements known for hormone signaling (ABA and
auxin), TF binding (MYB, WRKY, and NAC), abiotic stress responses
(dehydration, osmotic stress, hypersalinity, and cold), nutrition signals
(sugar, iron, and copper), light regulation, and expression in cytochrome
and mesophyll tissues in Arabidopsis and other model species

6
R. Srivastava et al. Plant Science 319 (2022) 111251

Table 1
Evaluation of growth, photosynthetic and yield parameters of the transgenic lines under optimum growth conditions and nutrition stress.
Growth Parameter Photosynthetic Parameter Yield
parameter
4-week-old seedling in plate 8-week-old plant in soil 8-week-old plant in soil 12-week-old
plant
Germination Seedling size Rosette size Plant CO2 assimilation rate Chlorophyll (aþb) Siliques per
rate (%) (in cm) (in cm) height (in (in µmol m⁻⁻2 s⁻⁻1) (in ug/ml) plant
cm)

Normal condition pBEAB 98.3 ± 2.6 3.1 ± 0.05 2.3 ± 0.15 16.9 ± 0.15 11.5 ± 0.39 259.2 ± 20.3 20.0 ± 2.0
VuNAC1-L1 99.4 ± 1.2 6.2 ± 0.11 4.9 ± 0.10 30.0 ± 2.00 22.7 ± 0.41 336.3 ± 03.8 52.0 ± 2.5
VuNAC1-L2 98.3 ± 1.6 6.1 ± 0.18 4.5 ± 0.12 25.6 ± 2.00 22.6 ± 0.25 341.7 ± 15.7 53.3 ± 1.5
VuNAC1-L3 99.3 ± 0.8 6.1 ± 0.10 4.4 ± 0.80 24.7 ± 0.60 15.0 ± 0.55 325.7 ± 10.8 51.3 ± 1.2
VuNAC2-L1 99.4 ± 1.1 5.6 ± 0.05 3.8 ± 0.21 26.7 ± 1.50 19.3 ± 0.29 357.1 ± 07.1 46.6 ± 3.1
VuNAC2-L2 99.5 ± 1.1 5.6 ± 0.01 3.3 ± 0.10 25.3 ± 1.15 16.9 ± 0.52 347.0 ± 25.5 43.3 ± 1.2
VuNAC2-L3 99.0 ± 0.5 5.6 ± 0.15 3.3 ± 0.25 24.5 ± 3.20 13.5 ± 0.10 335.8 ± 08.6 44.0 ± 2.0
Nutrition stress pBEAB 83.6 ± 2.4 3.2 ± 0.06 1.4 ± 0.10 12.6 ± 0.57 04.4 ± 0.25 199.6 ± 06.8 11.3 ± 1.2
VuNAC1-L1 97.3 ± 1.1 4.2 ± 0.12 3.8 ± 0.21 25.3 ± 0.80 16.5 ± 0.41 281.5 ± 20.5 42.5 ± 2.5
VuNAC1-L2 98.4 ± 1.5 4.0 ± 0.13 3.3 ± 0.10 24.0 ± 0.58 17.8 ± 0.25 310.5 ± 08.1 43.3 ± 2.8
VuNAC1-L3 97.5 ± 0.6 3.9 ± 0.10 3.2 ± 0.30 23.0 ± 0.50 13.3 ± 0.60 300.5 ± 10.2 42.7 ± 2.5
VuNAC2-L1 97.8 ± 1.1 3.7 ± 0.14 3.6 ± 0.06 22.7 ± 1.66 15.4 ± 0.29 296.1 ± 02.1 36.6 ± 2.9
VuNAC2-L2 98.6 ± 1.3 3.6 ± 0.12 3.1 ± 0.12 23.3 ± 0.76 14.0 ± 0.52 293.2 ± 10.8 38.7 ± 3.2
VuNAC2-L3 97.0 ± 0.9 3.6 ± 0.95 3.1 ± 0.15 24.0 ± 0.80 11.6 ± 0.60 310.5 ± 15.8 40.3 ± 0.6

(Supplementary Table S2). The time-course expression study indicated
that strong induction of VuNAC1/2 in response to dehydration (10% loss
in fresh weight), salt (200 mM NaCl), and osmotic stress (20% PEG). The
genes showed early (~3 hrs) and persistent stress response with a
marked accumulation of mRNA transcripts with the maximum relative
fold change recorded after 24 hrs (Figs. 1B and C). Under aluminum
stress (50 µM, acidic pH 5.0), we found the strongest induction at 3 hrs
in shoot tissue, which decreased gradually, indicating an early but
transient response. Whereas, the genes showed late expression after 24
hrs in roots, suggesting their differential regulation in shoots and roots
during the aluminum response (Supplementary Fig. S3). In addition, the
genes exhibited prominent induction in response to ABA (50 µM),
methyl jasmonate (50 µM), cold (4 ◦ C), and heat stress (45 ◦ C), as shown
in Fig. 1D. Interestingly, the genes also showed rapid and transient
response on exposure with diurnal light, which diminished within an
hour (Fig. 1E). This explained the peculiar expression profile of control
plants at 12 hrs where VuNAC1/2 showed transient induction irre­
spective of any stress treatment (Fig. 1B). Our study suggested a crucial
role of VuNAC1/2 TFs in both growth and stress response via light,
nutrition, and multiple stress signaling. As indicated by the promoter
regulatory elements (Supplementary Table S2) and rapid induction by
light exposure (Fig. 1E), VuNAC1/2 TFs were stipulated to play indis­
pensable roles in photosynthesis and nutrition assimilation by signaling
sugar metabolism, phosphate starvation response, and ion uptake.
3.2. The ectopic overexpression of VuNAC1/2 improved embryonic and
rosette growth under optimum and limited nutrition availability
To characterize the growth-associated role of VuNAC1/2 in detail, we
generated at least three independent lines (L1, L2, and L3), ectopically
expressing the VuNAC1/2 TFs in Arabidopsis, governed by a constitutive
35 S CaMV promoter (Fig. 2 and Supplementary Fig. S4). The growth of the
transgenic lines was analyzed at different developmental stages such as
embryonic, rosette, and flowering under both optimum (sucrose supple­
mented) and limited growth conditions (without any carbon source), to
prevent any masking of phenotype due to rich exogenous nutrition (Fig. 3
and Fig. 4). Under optimum growth conditions, both the control (pBEAB)
and transgenic seedlings (35 S:VuNAC1/2) germinated to produce green
cotyledonary leaves after one week (Fig. 3A). But, the transgenic seedlings
showed early sprouting, increased maximal leaf-blade size and root length,
more expanded cotyledonary leaves with emerging juvenile leaves.
Further, when grown in limiting media, the one-week-old seedlings pro­
duced lighter green and less expanded leaves. Despite a slower germinate
rate, the transgenic seedlings produced expanded leaves and roots to reach
the post-germination stage. For the control seedlings. The primary leaves
and roots were still not fully emerged. Even so, after four weeks, the
transgenic seedlings displayed increased leaf size and primary root length,
in addition to denser lateral roots and newly emerging leaves, as compared
to the control (Fig. 3B and Supplementary Fig. S5). The phenotypes of the
three lines (L1, L2, and L3) were comparable in line with the transgene
expression level.
When the gain-of-function phenotype was studied for vegetative-
stage plants (4-week-old) under optimum conditions, the transgenic
plants displayed richer rosette growth with increased plant height, leaf
count, leaf area, and rosette diameter (Fig. 4A and Table 1). Later,
during the reproductive stage (8-week-old), the transgenic plants
showed a thicker flowering axis, more branched inflorescence, and
increased production of siliques, hence improved seed yield (Fig. 4B).
Also, the degree of the improved phenotype in different lines was
correlated to the expression value (L1 > L2 > L3 for VuNAC1 and
L1 ~ L2 ~ L3 for VuNAC2) (Fig. 4C).
Moreover, when the plants were inflicted with nutrition stress for a
month, the transgenic plants grew expanded rosette leaves and healthy
reproductive growth. In contrast, the wild type plants showed poor growth
and impeded development, having smaller rosette base and delayed
emergence of the floral axis (Fig. 5A). The phenotype study indicated
impeded seed-dormancy, accelerated post-embryonic growth, and refined
vegetative and reproductive growth in Arabidopsis due to overexpression
of VuNAC1/2 TFs, suggesting richer seed reserve and improved nutrition
sequestration in the transgenic seedlings under optimum as well as limiting
nutrition availability.
3.3. The transgenic lines exhibited improved photosynthetic activity and
stomatal abundance
The induction of transcripts by light and the associated regulatory
elements in the promoter suggested circadian or light-mediated regu­
lation of VuNAC1/2 TFs. Moreover, the improved growth of the trans­
genic plants and their sustenance under nutrition limitation indicated
improved photosynthetic activity, which not only dictates the biomass
growth but also the yield potential [28]. To assess the photosynthetic
activity, the gas exchange and chlorophyll fluorescence parameters were
recorded under optimum growth and after two weeks of nutritional
stress (Fig. 5B). Higher CO2 assimilation rate (A), stomatal conductance
(Gsw), quantum yield of photosystem II (φ PSII), and electron transport
rate (ETR), as compared to the vector transformed control, indicated an

7
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 5. Evaluation of photosynthetic efficiency under limited nutrition and stomatal density under drought. (A) The healthier rosette growth of VuNAC1/2 over­
expressing lines (L1 and L2) than the vector-transformed control (pBEAB), after a 4-weeklong nutritional deficiency (left), indicated improved tolerance to starvation.
The right panel (control) shows the weekly growth of L1 transgenic lines under optimum conditions. (B) The enhanced gas exchange and chlorophyll fluorescence
parameters in the transgenic lines such as carbon assimilation rate (A), stomatal conductance (Gsw), photosystem II quantum yield (φ PSII), and electron transport
rate (ETR), indicated improved photosynthetic activity under nutritional deficiency as well as normal conditions. (C) Under control conditions, the stomatal density
was higher in the transgenic plants (top), supporting higher gas exchange during optimum growth conditions and stress recovery. However, the stomatal pores were
smaller in the transgenic lines suggesting tight regulation of water or gas exchange. Under drought stress (bottom), the stomatal apertures were closed, indicating
controlled evapotranspiration in all genotypes. All data are mean ± SE (n = 4). Different letters indicate significant differences between genotypes (p < 0.05, post
hoc Tukey HSD).

8
R. Srivastava et al.
Fig. 5. (continued).
enhanced photosynthetic activity in transgenic plants.
One of the possible factors directly translated into increased photo­
synthetic activity and biomass growth is stomatal abundance [29]. As
per the reports, NAC TFs such as ZmNAC49 and VvNAC17 regulate
stomatal density in maize and grapevine [30,31]. The microscopic
visualization of the adaxial (upper) surface of leaf epidermis revealed a
noteworthy higher stomatal density (per square mm of lea tissue) in the
transgenic plants (Fig. 5C). This suggested that VuNAC1/2 promotes
stomatal development. Although the smaller stomatal openings indicate
tight regulation of the stomatal activity in the transgenic plants, yet the
higher stomatal density could dictate an overall increase in CO2 assim­
ilation and stomatal conductance to improve the photosynthetic activ­
ity. Moreover, the size of the stomatal aperture is crucial to minimize
water loss and optimize CO2 uptake to maintain leaf water potential
during water-deficit conditions [32]. Under drought stress, the stomatal
pores were closed in both transgenic and control leaves, indicating
controlled transpiration. Even so, the higher stomatal abundance in
transgenic lines may support their post-stress recovery through
improved CO2 assimilation. Like ZmNAC49 and VvNAC17, VuNAC1/2
may be involved in the direct expression of the genes associated with
stomatal density and development, such as FAMA and MUTE. Thus,
VuNAC1/2 strikes a balance between carbon gain and water-use effi­
ciency by optimizing stomatal activity.
9
Plant Science 319 (2022) 111251
3.4. The transgenic seedlings displayed ABA and auxin insensitivity at
germination and post-germination stage
The transgenic seedlings manifested improved germination rates
under normal and stressed conditions. Also, VuNAC1/2 showed promi­
nent induction by ABA, which plays a critical role in maintaining and
releasing seed dormancy. Dynamic changes in ABA level or sensitivity
can elicit signaling to regulate developmental checkpoints that transi­
tion dormancy to germination through radicle emergence and post
germinative seedling establishment [33]. To examine the role of
ABA-inducible VuNAC1/2 in dormancy breaking, we studied germina­
tion rate under high exogenous ABA levels, 1.0 µM, 2.0 µM, and 3.0 µM
(Fig. 6A). The transgenic seeds displayed a higher germination rate
reflecting insensitivity to ABA, while the germination of control seeds
was inhibited with increasing ABA levels. Besides dormancy, the
post-germinative seedling establishment is finely tuned by ABA levels. It
can suppress growth in germinated seedlings by inhibiting the expres­
sion of enzymes required for cell-wall loosening, expansion, cell-wall
biosynthesis, and other structural proteins. The high levels of ABA
imposed growth inhibition and induced senescence in the control
seedlings, more apparently in the limited growth conditions (Fig. 6B). In
contrast, the transgenic seedlings showed tolerance to ABA sensitivity.
Similarly, a somewhat lower range of auxin concentrations acceler­
ates root growth, but the application of high concentrations can retard
both shoot and root growth due to pathological toxicity [34]. In the
sucrose supplemented optimum growth media, 2.5 µM and 5.0 µM of
IBA triggered lateral root formation but no substantial change in the
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 6. A. Study of ABA response during seed germination. The transgenic seedlings showing an insignificant effect on germination rates due to increased
exogenous ABA levels suggested ABA insensitivity, whereas the vector-transformed seedlings with reduced germination rates indicated ABA-induced dormancy. All
data are mean ± SE (n = 6). Different letters indicate significant differences between genotypes (p < 0.05, post hoc Tukey HSD). Fig. 6 B. Study of ABA and auxin
response on post-germinative growth. The transgenic seedlings showed lesser toxicity to high ABA levels (left), whereas the control seedlings displayed growth
inhibition and death (more apparent under limiting growth media). The growth was reduced at high IBA levels (7.5 µM) in the control seedlings, while the transgenic
lines showed denser lateral root formation. All data are mean ± SE (n = 6). Different letters indicate significant differences between genotypes (p < 0.05, post hoc
Tukey HSD).

primary root length. However, at higher levels (7.5 µM), the shoot and
root growth of control seedlings was inhibited, whereas the transgenic
seedlings continued growth, suggesting auxin insensitivity. Without
sucrose, the IBA-triggered root formation with increasing levels (up to
5.0 µM) was more apparent, but at higher concentrations, IBA caused
growth inhibition in control seedlings (data not shown). Even so, the
transgenic seedlings produced lengthier roots regardless of any auxin
supplement.
3.5. The overexpression of VuNAC1/2 conferred tolerance to multiple
abiotic stress such as drought, salinity, aluminum, cadmium, and H2O2
stress in Arabidopsis
To characterize VuNAC1/2 role in stress response, we examined the
growth phenotype of the transgenic seedlings under varying doses of
stress inducers supplemented in optimum and limiting growth media.
The transgenic seedlings exhibited improved germination rates and
seedling establishment under PEG and NaCl-induced stress (Supple­
mentary Fig. S6). Further, when stress was imposed during the post-
germinative stage, the transgenic seedlings exhibited improved sur­
vival and growth maintenance indicating tolerance to osmotic stress,
salinity, aluminum and cadmium toxicity, and oxidative stress (Fig. 7A
and B). The one-week-old transgenic seedlings could survive up to 5.0%
PEG, 150 mM NaCl, 15.0 µM AlCl3, 10.0 µM CdCl2, 7.5 µM H2O2,
whereas the control seedlings showed severe growth inhibition of shoot
and root, and failed to survive after 3–4 days of stress treatment
(Fig. 7A). The stress tolerance and sensitive phenotypes of the transgenic
and control seedlings, respectively, were more contrastingly appeared
under the limited growth conditions (Fig. 7B). The results indicated that
VuNAC1/2 overexpression conferred tolerance to multiple abiotic
stresses in transgenic lines, and their ability to maintain growth under
starvation favors stress mitigation and recovery.
3.6. The transgenic plants showed improved water status, membrane
integrity, and ascorbate-glutathione content under drought and salinity
To study the response in severe and prolonged drought and high
salinity, we examined the growth performance and stress recovery of
soil-grown mature plants (Fig. 8). All the plants showed chlorosis and
wilting due to severe desiccation and ion toxicity perpetuated by
400 mM NaCl aqueous solutions (Fig. 8A). Compared to the unstressed
control plants, continuing their healthy vegetative growth, the rosette
growth was inhibited due to stress, resulting in reduced leaf size.
However, the transgenic plants exhibited recommendable recovery of
vegetative growth, gradually acquiring the flowering state and ensuing
seed production, indicating tolerance to drought and salt stress (Fig. 8A
and Supplementary Fig. S7). In comparison, the control plants inflicted
by drought failed to resume their growth. The plants could not withstand
the toxicity generated by sequestered Na+ ions and showed outright
chlorosis resulting in death.
Since several biochemical and physiological adaptations are essen­
tial for counteracting the stress effect, we further examined the water
status, membrane integrity, and antioxidant content in the transgenic
lines (Fig. 8B). The transgenic leaves showed more relative water con­
tent (RWC) than the vector-transformed control under drought and salt
stress, indicating better water status and osmotic potential. The lower
electrolyte leakage rate accompanied by lower MDA content in the
transgenic tissue was the hallmark of more intact membrane biology
than the control tissue, which indicated severe membrane disruption.
Proline plays versatile roles as a compatible solute relieving cellular
osmotic potential and heavy metal detoxifier. The transgenic plants
manifested higher content during drought and salt stress. Furthermore,
ascorbate and glutathione make the heart of the redox hub, also regu­
lating light energy dissipation, biosynthesis of zeaxanthin and vitamin,
and signal transduction. Under both the normal and stressed conditions,
the transgenic plants accumulated higher contents of glutathione and

10
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 6. (continued).

11
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 7. Examining the role of VuNAC1/2 in multiple abiotic stresses. The one-week-old Arabidopsis seedlings were characterized in response to multiple abiotic
stresses exerted in the presence of (A) optimum and (B) limiting growth media. The healthier seedling growth of transgenic lines indicated better tolerance to
dehydration (PEG-induced), salinity (NaCl), metal toxicity (AlCl3 and CdCl2), and oxidative stress (H2O2), whereas the control seedlings showed stress-induced
growth inhibition. The contrasting phenotype was more apparent under limiting nutrition. All data are mean ± SE (n = 4). Different letters indicate significant
differences between genotypes (p < 0.05, post hoc Tukey HSD).

ascorbate, indicating the increased antioxidant level to maintain the
redox balance. Our results showed that the VuNAC1/2 TFs control
osmoregulation by maintaining tissue water relations, membrane
integrity, and osmolyte accumulation. Moreover, they ease the oxidative
burden by promoting the production of ascorbate and glutathione.
Indeed, the accumulation under both the stressed and normal conditions
suggested a direct role of VuNAC1/2 in glutathione and ascorbate
metabolism. However, drought-induced accumulation indicated the
involvement of other factors in proline synthesis.
3.7. Regulatory model for VuNAC1/2 TFs
Integrative gene regulatory networks (iGRN) covering orthologous
genes of model plants can provide a foundational framework for
understanding and inferring the gene regulation of plants with unknown
mechanisms [35]. The upstream regulators predicted based on the
Arabidopsis orthologs suggested that the VuNAC1/2 TFs perceive the
signal of several other proteins such as HSFA2 (combined
stress-response and ascorbate signaling) [36], RRTF1 (oxidative stress
response) [37], COBRA (cell-expansion), MYB21 (photo-morphogen­
esis), ANAC025 (pollen and seed development) [38], flower develop­
ment (PISTILLATA) [39], and PHL1 and HRS1 (nitrogen, phosphate and
iron starvation and breaking of ABA-mediated seed dormancy) [40–42]
(Table 2). Further, we generated a gene interactome by convoluting
experimental data of multiple co-expression studies of the Arabidopsis
orthologs under different developmental and environmental conditions.
The orthologous interactome speculated the involvement of VuNAC1/2
in metabolic pathways such as glutathione, starch and sucrose, hormone

12
R. Srivastava et al.
Fig. 7. (continued).
signaling, and circadian rhythms (Fig. 9). The functional roles of genes
in the coexpressing network corroborated with the phenotypes of the
transgenic plants, suggesting the interaction or coordination of
VuNAC1/2 TFs with many stress and hormone signaling proteins,
transporters, and metabolic enzymes (Supplementary Table S8).
Combining the upstream regulatory model and coordinating gene
network suggested that the VuNAC1/2 might be induced by stress,
hormone, nutrition, light, and developmental signals in a feedback loop
(Table 2 and Fig. 9).
4. Discussion
Several NAC TFs, such as ATAF in Arabidopsis and their orthologs in
rice, soybean, etc., are induced by diverse stress-related signals and
improve crop stress tolerance [11–13]. However, the constitutive
overexpression of critical stress-regulating NAC proteins such as ATAF1,
OsNAC6, TsNAC1, and EsNAC1 can compromise the plant growth due to
perturbations in intertwined hormonal signaling or metabolic rerouting
to affect beneficial growth processes [11,12,43,44]. The
13
Plant Science 319 (2022) 111251
non-commercialized cowpea genotypes, producing rich forage biomass
under harsh conditions, could be an ideal source to hunt NAC genes
supporting growth and yield sustenance while stress tolerance, which
may be lost in the domesticated genotypes [21,45]. Our previous studies
reported two NAC TFs (VuNAC1 and VuNAC2) cloned from
drought-resilience cowpea (Vigna unguiculata L. Walp) genotype that
improved both growth and stress tolerance in rich and minimal growth
conditions in yeast by reprogramming energy and antioxidant producing
pathways such as biosynthesis of purine, vitamin B complex, methio­
nine, glutathione [19]. Such ATAF homologs isolated from robust ge­
netic sources hold potential for sustainable improvement of stress
tolerance in plants without growth penalty [20]. Our study of untapped
cowpea ATAF orthologs (VuNAC1 and VuNAC2), cloned from a
drought-resilience genotype (Kannanado White), showed that their
ectopic expression in Arabidopsis (governed by 35 S CaMV promoter),
conferred drought and salt tolerance, with simultaneous improvement
of rosette growth and seed-yield, under both normal and
nutrition-limiting conditions (Table 1). Hence, constitutive over­
expression of functional ATAF orthologs from cowpea conferred
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 8. Recovery from severe drought and salinity and the associated physiological and biochemical changes. (A) One-week desiccation (water-withdrawal at 40%
relative humidity) and high-salinity (400 mM) were imposed on five-week-old soil-grown plants. The transgenic plants manifested recovery and resumed the
vegetative and inflorescence growth within a week after stress was lifted, whereas the control plants failed to survive. (B) Higher relative-water content (RWC %),
lesser electrolyte leakage rate (%) and malondialdehyde (MDA) content, and higher accumulation of proline, glutathione, and ascorbate under drought and salt stress
indicated better stress-adaptation of the transgenic lines. All data are mean ± SE (n = 6). Different letters indicate significant differences between genotypes
(p < 0.05, post hoc Tukey HSD).

14
R. Srivastava et al. Plant Science 319 (2022) 111251

Table 2
Upstream regulators predicted by orthologous gene network in Arabidopsis.
Gene ID PR Function
Score

RRTF1 At4g34410 100 regulates redox homeostasis [37]

DIV2 At5g04760 96.7 negatively regulates salt response [51]
MYB48 At3g46130 97.8 regulates drought tolerance and flavonol biosynthesis [52]
HsFA2 At2g26150 98.2 regulates heat and combined stress tolerance, ascorbate signaling [36]
PHL1 At5g29000 98.2 involved in phosphate and iron starvation response [42]
HRS1 At1g13300 97.3 integrates nitrate and phosphate starvation responses and adaptation of root architecture depending on nutrient availabilities and
negatively regulates dormancy [40,41]
MYB28 At5g61420 98.7 primary regulator of glucosinolate biosynthesis to protect against insects
SAP6 At3g52800 99.4 regulates stress response
COBRA At5g60920 99.4 necessary for oriented cell expansion
MYB103 At1g63910 99.2 involved in tapetum development, callose dissolution, and exine formation anthers
PISTILLATA At5g20240 98.4 required for petals and stamen development [39]
ANAC025 At1g61110 98.3 regulates anther, pollen and seed development [38]
ATHB-8 At4g32880 99.2 involved in the regulation of vascular development [53]
MYB21 At3g27810 99.1 involved in photomorphogenesis in the light [54]
RGA1 At2g01570 97.8 represses GA-induced vegetative growth and floral initiation [55,56]
CYP79B2 At4g39950 99.5 converts tryptophan to indole-3-acetaldoxime, a precursor for tryptophan-derived glucosinolates and indole-3-acetic acid (auxin) [57]

Source: http://webs2.kazusa.or.jp/kagiana/rnr0912/indexff.html.

multiple stress tolerance overcoming the growth aberrations (Fig. 3,
Fig. 4, Fig. 7, and Fig. 8), unlike their orthologs in Arabidopsis and rice.
These unique and multifaceted NAC genes could be promising biotech­
nological tools for the sustainable engineering of stress tolerance in
crops by overcoming any growth/tolerance trade-offs (Fig. 10).
Both VuNAC1 and VuNAC2 were inducible by various abiotic stimuli,
including dehydration, salinity, osmotic stress, ABA, showing persistent
expression characteristics similar to ATAF1. Moreover, these cowpea
orthologs were also induced by aluminum and light (Fig. 1B, C, and D).
Their ectopic overexpression in Arabidopsis conferred tolerance to
drought, salinity, aluminum, cadmium, and H2O2 (Fig. 7 and Fig. 8).
Indeed, overexpression of VuNAC1/2 also improved embryonic, rosette,
and inflorescence growth, resulting in enhanced seed yield in Arabi­
dopsis (Fig. 3, Fig. 4, and Table 1). The overexpression characteristics
suggested that these key stress regulators are also involved in the tran­
scriptional tuning of the growth-controlling entities influencing cell
proliferation, developmental fates, and floral transition, besides the
expression of downstream stress-responsive genes. The concomitant
increase in photosynthetic activity, stomatal abundance, and ABA
insensitivity, seemed to play a pivotal part in the stress tolerance
without growth inhibition (Fig. 5B, Fig. 5C, and Fig. 6A). The ortholo­
gous gene network (Fig. 9), and prediction of upstream regulators
(Table 2), proposed the regulon for the VuNAC1/2 TFs. This establishes
that different stress and growth signals mediated through overlapping
transmission components such as light, hormone, nutrition, ROS, etc.,
converge at VuNAC1/2 being the central transcriptional regulators.
Previous studies reported that ATAF1 coupled stress-signaling with
photosynthesis-related transcriptional cascades, executing repression of
the GLK genes required for chloroplast development resulting in
photosynthetic suppression [14]. Unlike ATAF1, the overexpression of
the VuNAC1/2 TFs increased rosette size and boosted photosynthetic
activity (Fig. 4A and Fig. 5B). It suggested that the VuNAC1/2 do not
suppress photosynthesis-associated genes (PAGs), like ATAF1. The
concomitant improved photosynthetic activity might support the energy
requirements due to efficient sequestration of photosynthates such as
sucrose even when no exogenous carbon was supplemented (Fig. 5A and
Fig. 5B). In addition, VuNAC1/2 overexpression yielded an enhanced
stomatal density in Arabidopsis (Fig. 5C), one of the possible factors
resulting in increased photosynthetic activity and biomass growth [29].
The VuNAC1/2 TFs might be rather involved in the activation of genes
that synthesis photosynthesis apparatus control the stomatal develop­
ment, unlike the ZmNAC49, which repressed ZmMUTE, ZmSDD1,
ZmFAMA, etc., to reduce stomatal density in maize [30]. Also, some
unknown mechanisms, like the ABA-dependent reduction of stomatal
density in VvNAC17 overexpressing in grapevine, can be involved [31].
Further studies are needed to investigate the role of VuNAC1/2 in
controlling photosynthesis and stomatal development.
Moreover, the VuNAC1/2 overexpression did not exert carbon star­
vation and ABA hypersensitivity phenotypes, like ATAF1 [15]. Instead,
the transgenic plants exhibited tolerance to nutrition starvation imposed
at seedling and mature stages (Fig. 3A, and Fig. 5A). In contrast, the
transgenic seedlings showed insensitivity to ABA-induced seed
dormancy and post-embryonic growth inhibition due to ABA, NaCl, and
PEG-induced stress (Fig. 6A, Fig. 6B, and Supplementary Fig. S6). Our
earlier studies found that the transgenic yeast strains manifested
increased cellular biomass production, longer life-span, accumulated
energy currencies (ATP, NADH), and survived under adenine-limitation
and minimal nutrition [19]. Similarly, this study in Arabidopsis indi­
cated that the nutrition sufficiency of the transgenic plants could
possibly be due to improved uptake of phosphate and nitrogen [46],
regulated by PHL1 and HRS1, as implied by the predicted regulators and
promoter elements (Table 2 and Supplementary Table S2). Moreover,
the photosynthetic gain and its sink regulation could also play a crucial
role in maintaining the carbon/nitrogen balance to result in an optimal
seedling establishment as well as increased rosette biomass and seed
yield [47,48] (Fig. 3 and Fig. 4). Further, improved growth could also
play a key role in stress survival and recovery. Besides, the transgenic
plants also accumulated stress metabolites, which was also evident in
our preliminary study in yeast (ascorbate, flavonoids, and glutathione)
[19]. Taken together, our study indicated that VuNAC1/2 TFs play a
crucial role in boosting growth and nutrition assimilation, besides
improving stress tolerance and recovery, unlike other ATAF1 orthologs.
One of the major factors accounting for this dissimilar behavior
could be the different regulatory domains present in the VuNAC1/2
protein. The conserved N-terminal domains of VuNAC1/2 TF hold sig­
nificant similarity with Arabidopsis ATAF1 (≥ 85%), implying similar
target gene sets. But the mode of transcriptional tuning (activation/
repression) and functional versatility also depend on the transcriptional
regulatory regions (TRR) residing in the divergent C-terminal regions.
The uniqueness of the C-terminal regions of VuNAC1/2 TFs (~30%
protein length) may be responsible for the indifferent behavior from
their orthologous counterparts through dissimilar protein interactions.
Indeed, the regulon associated with VuNAC1/2 could be similar to the
ATAF-like gene-regulatory network in Arabidopsis (Fig. 9A and Fig. 9B).
However, the distinct signal perception to launch a positive or negative
stress-response could determine the distinct functions. Besides, the over-
representation of both ABA-responsive elements (ABRE) as well as
ACGTATERD1, DRECRTCOREAT, and CBFHV in the gene-promoters

15
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 9. Proposed regulatory network. (A) The co-expression network (directly connecting at least two Arabidopsis orthologous members, ATAF1/2, ANAC032, and
ANAC102) and the metabolic pathways governed by the regulon. (B) The gene-interactome around Arabidopsis orthologs indirectly connecting at least two co-
expressing orthologs (indicated by grey lines) or protein-protein interactions (indicated by red lines). (C) The gene ontology (GO) annotation of the network pre­
dicted the cellular, molecular and biological function associated with the VuNAC1/2 TFs.

16
R. Srivastava et al. Plant Science 319 (2022) 111251

Fig. 10. A model to balance growth trade-off. Constitutive overexpression VuNAC1/2 conferred multiple stress tolerance accompanied by a simultaneous
improvement of growth mediated by increased photosynthetic activity, nutrition homeostasis, and ROS tolerance, unlike their orthologs in Arabidopsis (ATAF1) and
rice (OsNAC6). The upward and downward arrows indicate an increase and decrease in phenomena. A balance between ABA-mediated stress and growth signaling
seemed to play a crucial role in overcoming the trade-off. VuNAC1/2 could be a single-gene biotechnological tool to restore yield while achieving tolerance to
multiple and persistent abiotic stresses.

indicated both ABA-dependent and/or ABA-independent regulation of Acknowledgments

VuNAC1/2. The mixed execution of ABA-dependent stress tolerance
with pleiotropic ABA-independent growth responses could counteract
the growth penalty, despite constitution expression throughout different
developmental stages [49].
In conclusion, the cowpea ATAF orthologs (VuNAC1/2) converge
multiple stress responses and basal growth signaling. The TFs regulate
stress and growth responses via overlapping hormonal and nutritional
signaling such as ABA-mediated pathways, photosynthesis, remodeling
of antioxidants and carbon metabolism (Fig. 10). However, the detailed
regulatory model is still unknown. Genes like VuNAC1/2 that confer
tolerance to multiple abiotic stresses and ameliorate the shrinking
vegetative growth and reproductive yield by improving the energy sta­
tus should be focused on for sustainable drought and salinity tolerance
in cash crops harmonized with the yield improvement. They might also
play a crucial role in protection against pH-dependent phytotoxicity of
aluminum and other toxic metals, which are major constraints for root
development in acidic soil [50]. The gain-of-phenotype study in Arabi­
dopsis and the inference thereof laid the foundation to implement the
strategy in crops to accomplish dual objectives, i.e., multi-stress resil­
ience and yield improvement. As the unfavorable climatic conditions are
becoming persistent, a ubiquitous and constitutive expression of unique
VuNAC1/2 TFs secure the yield potential of crops without any growth
trade-off. Further understanding of the underlying molecular mecha­
nisms can bring new insights to harmonized cross-talk between stress
adaptation and growth. Also, the functional role of these TFs in native
cowpea species and related legumes for sustainable stress improvement
remains to be explored in the future.
Declaration of Competing Interest
The author(s) declare no competing interests.
This work was supported by a research grant from the Program
Support Grant Phase-II from the Department of Biotechnology, Gov­
ernment of India to L.S. (BT/PR13560/COE/34/44/2015).
Author statement
L.S., Y.K., and H.K. designed and oversaw the research plan. R.S. has
performed the experiments and wrote the manuscript. The final version
of the article was corrected and approved by all the authors.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the
online version at doi:10.1016/j.plantsci.2022.111251.
References
[1] R. Mahalingam, Consideration of combined stress: a crucial paradigm for
improving multiple stress tolerance in plants, in: Combined Stresses in Plants,
Springer, 2015, pp. 1–25.
[2] S. Ashkani, M.Y. Rafii, M. Shabanimofrad, G. Miah, M. Sahebi, P. Azizi, F.
A. Tanweer, M.S. Akhtar, A. Nasehi, Molecular breeding strategy and challenges
towards improvement of blast disease resistance in rice crop, Front. Plant Sci. 6
(2015) 886.
[3] R. Srivastava, S. Kumar, Y. Kobayashi, K. Kusunoki, P. Tripathi, Y. Kobayashi,
H. Koyama, L. Sahoo, Comparative genome-wide analysis of WRKY transcription
factors in two Asian legume crops: Adzuki bean and Mung bean, Sci. Rep. 8 (2018)
1–19.
[4] T. Enomoto, M. Tokizawa, H. Ito, S. Iuchi, M. Kobayashi, Y.Y. Yamamoto,
Y. Kobayashi, H. Koyama, STOP1 regulates the expression of HsfA2 and GDH s that
are critical for low-oxygen tolerance in Arabidopsis, J. Exp. Bot. 70 (2019)
3297–3311.
[5] A. Sadhukhan, Y. Kobayashi, S. Iuchi, H. Koyama, 2021. Synergistic and
antagonistic pleiotropy of STOP1 in stress tolerance, Trends in Plant Science,
(2021).

17
R. Srivastava et al.
[6] R. Shahzad, S. Jamil, S. Ahmad, A. Nisar, Z. Amina, S. Saleem, M.Z. Iqbal, R.M.
Atif, R.D. Thompson, X. Wang, 2020. Harnessing the Potential of Plant
Transcription Factors in Developing Climate-Smart Crops: Future Prospects,
Challenges, and Opportunities, Saudi Journal of Biological Sciences, (2020).
[7] E.H. Baillo, R.N. Kimotho, Z. Zhang, P. Xu, Transcription factors associated with
abiotic and biotic stress tolerance and their potential for crops improvement, Genes
10 (2019) 771.
[8] S. Singh, H. Koyama, K.K. Bhati, A. Alok, The biotechnological importance of the
plant-specific NAC transcription factor family in crop improvement, J. Plant Res.
(2021) 1–21.
[9] P. Das, N. Lakra, K.K. Nutan, S.L. Singla-Pareek, A. Pareek, A unique bZIP
transcription factor imparting multiple stress tolerance in Rice, Rice 12 (2019)
1–16.
[10] P. Tiwari, Y. Indoliya, A.S. Chauhan, V. Pande, D. Chakrabarty, Over-expression of
rice R1-type MYB transcription factor confers different abiotic stress tolerance in
transgenic Arabidopsis, Ecotoxicol. Environ. Saf. 206 (2020), 111361.
[11] Y. Wu, Z. Deng, J. Lai, Y. Zhang, C. Yang, B. Yin, Q. Zhao, L. Zhang, Y. Li, C. Yang,
Dual function of Arabidopsis ATAF1 in abiotic and biotic stress responses, Cell Res.
19 (2009) 1279–1290.
[12] H. Takasaki, K. Maruyama, S. Kidokoro, Y. Ito, Y. Fujita, K. Shinozaki,
K. Yamaguchi-Shinozaki, K. Nakashima, The abiotic stress-responsive NAC-type
transcription factor OsNAC5 regulates stress-inducible genes and stress tolerance in
rice, Molecular Genetics, Genomics 284 (2010) 173–183.
[13] H. Jin, F. Huang, H. Cheng, H. Song, D. Yu, Overexpression of the GmNAC2 gene,
an NAC transcription factor, reduces abiotic stress tolerance in tobacco, Plant Mol.
Biol. Report. 31 (2013) 435–442.
[14] P. Garapati, G.-P. Xue, S. Munné-Bosch, S. Balazadeh, Transcription factor ATAF1
in Arabidopsis promotes senescence by direct regulation of key chloroplast
maintenance and senescence transcriptional cascades, Plant Physiol. 168 (2015)
1122–1139.
[15] P. Garapati, R. Feil, J.E. Lunn, P. Van Dijck, S. Balazadeh, B. Mueller-Roeber,
Transcription factor Arabidopsis activating factor1 integrates carbon starvation
responses with trehalose metabolism, Plant Physiol. 169 (2015) 379–390.
[16] Y. Yang, H.H.J. Al-Baidhani, J. Harris, M. Riboni, Y. Li, I. Mazonka, N. Bazanova,
L. Chirkova, S. Sarfraz Hussain, M. Hrmova, DREB/CBF expression in wheat and
barley using the stress-inducible promoters of HD-Zip I genes: impact on plant
development, stress tolerance and yield, Plant Biotechnol. J. 18 (2020) 829–844.
[17] R. Dolferus, To grow or not to grow: a stressful decision for plants, Plant Sci. 229
(2014) 247–261.
[18] M. Kudo, S. Kidokoro, T. Yoshida, J. Mizoi, M. Kojima, Y. Takebayashi,
H. Sakakibara, A.R. Fernie, K. Shinozaki, K. Yamaguchi-Shinozaki, A gene-stacking
approach to overcome the trade-off between drought stress tolerance and growth
in Arabidopsis, Plant J. 97 (2019) 240–256.
[19] R. Srivastava, L. Sahoo, Cowpea NAC Transcription Factors Positively Regulate
Cellular Stress Response and Balance Energy Metabolism in Yeast via
Reprogramming of Biosynthetic Pathways, ACS Synthetic Biology, (2021).
[20] K.P. Kulkarni, R. Tayade, S. Asekova, J.T. Song, J.G. Shannon, J.-D. Lee,
Harnessing the potential of forage legumes, alfalfa, soybean, and cowpea for
sustainable agriculture and global food security, Front. Plant Sci. 9 (2018) 1314.
[21] M. Ishiyaku, H. Aliyu, Field evaluation of cowpea genotypes for drought tolerance
and Striga resistance in the dry savanna of the North-West Nigeria, Int. J. Plant
Breed. Genet. 7 (2013) 47–56.
[22] J. Jin, F. Tian, D.-C. Yang, Y.-Q. Meng, L. Kong, J. Luo, G. Gao, PlantTFDB 4.0:
toward a central hub for transcription factors and regulatory interactions in plants,
Nucleic Acids Res. (2016) gkw982.
[23] K. Higo, Y. Ugawa, M. Iwamoto, T. Korenaga, Plant cis-acting regulatory DNA
elements (PLACE) database: 1999, Nucleic Acids Res. 27 (1999) 297–300.
[24] D.R. Hoagland, D.I. Arnon, The water-culture method for growing plants without
soil, Circ. Calif. Agric. Exp. Station 347 (1950).
[25] T. Obayashi, Y. Aoki, S. Tadaka, Y. Kagaya, K. Kinoshita, ATTED-II in 2018: a plant
coexpression database based on investigation of the statistical property of the
mutual rank index, Plant Cell Physiol. 59 (2018) e3-e3.
[26] H. Mi, D. Ebert, A. Muruganujan, C. Mills, L.-P. Albou, T. Mushayamaha, P.
D. Thomas, PANTHER version 16: a revised family classification, tree-based
classification tool, enhancer regions and extensive API, Nucleic Acids Res. 49
(2021) D394–D403.
[27] J.S. Jeong, Y.S. Kim, M.C. Redillas, G. Jang, H. Jung, S.W. Bang, Y.D. Choi, S.H. Ha,
C. Reuzeau, J.K. Kim, OsNAC5 overexpression enlarges root diameter in rice plants
leading to enhanced drought tolerance and increased grain yield in the field, Plant
Biotechnol. J. 11 (2013) 101–114.
[28] A.J. Simkin, P.E. López-Calcagno, C.A.J.Jo.E.B. Raines, Feeding the world:
improving photosynthetic efficiency for sustainable crop production, J. Exp. Bot.
70 (2019) 1119–1140.
[29] K. Sakoda, W. Yamori, T. Shimada, S.S. Sugano, I. Hara-Nishimura, Y. Tanaka,
Higher stomatal density improves photosynthetic induction and biomass
production in Arabidopsis under fluctuating light, Front. Plant Sci. 11 (2020) 1609.
[30] Y. Xiang, X. Sun, X. Bian, T. Wei, T. Han, J. Yan, A. Zhang, The transcription factor
ZmNAC49 reduces stomatal density and improves drought tolerance in maize,
J. Exp. Bot. 72 (2021) 1399–1410.
[31] Y.-l Ju, X.-f Yue, Z. Min, X.-h Wang, Y.-l Fang, J.-x Zhang, VvNAC17, a novel stress-
responsive grapevine (Vitis vinifera L.) NAC transcription factor, increases
18
Plant Science 319 (2022) 111251
sensitivity to abscisic acid and enhances salinity, freezing, and drought tolerance in
transgenic Arabidopsis, Plant Physiol. Biochem. 146 (2020) 98–111.
[32] L.T. Bertolino, R.S. Caine, J.E. Gray, Impact of stomatal density and morphology on
water-use efficiency in a changing world, Front. Plant Sci. 10 (2019) 225.
[33] A.R. Kermode, Role of abscisic acid in seed dormancy, J. Plant Growth Regul. 24
(2005) 319–344.
[34] M. Alarcón, J. Salguero, P.G. Lloret, Auxin modulated initiation of lateral roots is
linked to pericycle cell length in maize, Front. Plant Sci. 10 (2019) 11.
[35] I. De Clercq, J. Van de Velde, X. Luo, L. Liu, V. Storme, M. Van Bel, R. Pottie,
D. Vaneechoutte, F. Van Breusegem, K. Vandepoele, Integrative inference of
transcriptional networks in Arabidopsis yields novel ROS signalling regulators,
Nat. Plants 7 (2021) 500–513.
[36] A. Nishizawa, Y. Yabuta, E. Yoshida, T. Maruta, K. Yoshimura, S. Shigeoka,
Arabidopsis heat shock transcription factor A2 as a key regulator in response to
several types of environmental stress, Plant J. 48 (2006) 535–547.
[37] M. Matsuo, J.M. Johnson, A. Hieno, M. Tokizawa, M. Nomoto, Y. Tada, R. Godfrey,
J. Obokata, I. Sherameti, Y.Y. Yamamoto, High REDOX RESPONSIVE
TRANSCRIPTION FACTOR1 levels result in accumulation of reactive oxygen
species in Arabidopsis thaliana shoots and roots, Mol. Plant 8 (2015) 1253–1273.
[38] T. Kunieda, N. Mitsuda, M. Ohme-Takagi, S. Takeda, M. Aida, M. Tasaka,
M. Kondo, M. Nishimura, I.J.T.P.C. Hara-Nishimura, NAC family proteins NARS1/
NAC2 and NARS2/NAM in the outer integument regulate embryogenesis in
Arabidopsis, Plant Cell 20 (2008) 2631–2642.
[39] C.D. Mara, V.F. Irish, Two GATA transcription factors are downstream effectors of
floral homeotic gene action in Arabidopsis, Plant Physiol. 147 (2008) 707–718.
[40] C. Wu, J. Feng, R. Wang, H. Liu, H. Yang, P.L. Rodriguez, H. Qin, X. Liu, D. Wang,
HRS1 acts as a negative regulator of abscisic acid signaling to promote timely
germination of Arabidopsis seeds, PLoS One 7 (2012), e35764.
[41] A. Medici, A. Marshall-Colon, E. Ronzier, W. Szponarski, R. Wang, A. Gojon, N.
M. Crawford, S. Ruffel, G.M. Coruzzi, G. Krouk, AtNIGT1/HRS1 integrates nitrate
and phosphate signals at the Arabidopsis root tip, Nat. Commun. 6 (2015) 1–11.
[42] M. Bournier, N. Tissot, S. Mari, J. Boucherez, E. Lacombe, J.-F. Briat, F. Gaymard,
Arabidopsis ferritin 1 (AtFer1) gene regulation by the phosphate starvation
response 1 (AtPHR1) transcription factor reveals a direct molecular link between
iron and phosphate homeostasis, J. Biol. Chem. 288 (2013) 22670–22680.
[43] C. Liu, Q. Sun, L. Zhao, Z. Li, Z. Peng, J. Zhang, Heterologous expression of the
transcription factor EsNAC1 in Arabidopsis enhances abiotic stress resistance and
retards growth by regulating the expression of different target genes, Front. Plant
Sci. 9 (2018) 1495.
[44] C. Liu, B. Wang, Z. Li, Z. Peng, J. Zhang, TsNAC1 is a key transcription factor in
abiotic stress resistance and growth, Plant Physiol. 176 (2018) 742–756.
[45] R. Srivastava, L. Sahoo, Balancing yield trade-off in legumes during multiple stress
tolerance via strategic crosstalk by native NAC transcription factors, J. Plant
Biochem. Biotechnol. (2021) 1–22.
[46] L. Nilsson, R. Müller, T.H. Nielsen, Dissecting the plant transcriptome and the
regulatory responses to phosphate deprivation, Physiol. Plant. 139 (2010)
129–143.
[47] Z.-L. Zheng, Carbon and nitrogen nutrient balance signaling in plants, Plant Signal.
Behav. 4 (2009) 584–591.
[48] A. Podgórska, R. Mazur, M. Ostaszewska-Bugajska, K. Kryzheuskaya, K. Dziewit,
K. Borysiuk, A. Wdowiak, M. Burian, A.G. Rasmusson, B. Szal, Efficient
photosynthetic functioning of Arabidopsis thaliana through electron dissipation in
chloroplasts and electron export to mitochondria under ammonium nutrition,
Front. Plant Sci. 11 (2020) 103.
[49] D.L. Des Marais, T.E. Juenger, Pleiotropy, plasticity, and the evolution of plant
abiotic stress tolerance, Ann. N. Y. Acad. Sci. 1206 (2010) 56–79.
[50] M. Moreno-Alvarado, S. García-Morales, L.I. Trejo-Téllez, J.V. Hidalgo-Contreras,
F.C. Gómez-Merino, Aluminum enhances growth and sugar concentration, alters
macronutrient status and regulates the expression of NAC transcription factors in
rice, Front. Plant Sci. 8 (2017) 73.
[51] Q. Fang, Q. Wang, H. Mao, J. Xu, Y. Wang, H. Hu, S. He, J. Tu, C. Cheng, G. Tian,
AtDIV2, an RR-type MYB transcription factor of Arabidopsis, negatively regulates
salt stress by modulating ABA signaling, Plant Cell Rep. 37 (2018) 1499–1511.
[52] Y. Wang, Q. Wang, M. Liu, C. Bo, X. Wang, Q. Ma, B. Cheng, R. Cai, Overexpression
of a maize MYB48 gene confers drought tolerance in transgenic arabidopsis plants,
J. Plant Biol. 60 (2017) 612–621.
[53] S. Baima, M. Possenti, A. Matteucci, E. Wisman, M.M. Altamura, I. Ruberti,
G. Morelli, The Arabidopsis ATHB-8 HD-zip protein acts as a differentiation-
promoting transcription factor of the vascular meristems, Plant Physiol. 126 (2001)
643–655.
[54] B. Shin, G. Choi, H. Yi, S. Yang, I. Cho, J. Kim, S. Lee, N.C. Paek, J.H. Kim, P.
S. Song, AtMYB21, a gene encoding a flower-specific transcription factor, is
regulated by COP1, Plant J. 30 (2002) 23–32.
[55] X. Fu, N.P. Harberd, Auxin promotes Arabidopsis root growth by modulating
gibberellin response, Nature 421 (2003) 740–743.
[56] H. Yu, T. Ito, Y. Zhao, J. Peng, P. Kumar, E.M. Meyerowitz, Floral homeotic genes
are targets of gibberellin signaling in flower development, Proc. Natl. Acad. Sci.
101 (2004) 7827–7832.
[57] Y. Zhao, A.K. Hull, N.R. Gupta, K.A. Goss, J. Alonso, J.R. Ecker, J. Normanly,
J. Chory, J.L. Celenza, Trp-dependent auxin biosynthesis in Arabidopsis:
involvement of cytochrome P450s CYP79B2 and CYP79B3, Genes Dev. 16 (2002)
3100–3112.