Plants, as metaorganisms, are associated with diverse microbes from various niches (such as the rhizosphere, spermosphere, phyllosphere, and endosphere). The interaction between plant and microbes may be either casual or intimate, but they all furnish an evolving seed microbiome that may carry over to other developmental stages found often in plants growing in natural situations (Hardoim et al., 2015). Bacterial endophytes are all bacteria that may or may not be effectively cultured and can colonize a plant internally without causing apparent harm or creating any obvious exterior structures (Azevedo and Araujo, 2007). Endophytes with plant growth–promoting (PGP) traits are generally the most effective in promoting growth because they live inside plant tissues and can communicate with the host plant and exert their beneficial effect much more effectively (Giauque et al., 2019). Protected from the outside environment, endophytes are much less susceptible to the soil’s frequent chemical–physical biotic and abiotic variations. The beneficial action of these bacteria is primarily expressed by directly promoting nutrient absorption via modulation of plant hormone levels (Marchut-Mikolajczyk et al., 2018). Seed-borne endophytes lay the foundation for the development of plant microbiomes as the pioneer colonizers, and seeds serve as carriers for the advantageous microorganisms for multigenerational preservation of these beneficial relationships (Truyens et al., 2015). Association of seed and endophytes have been documented in a variety of crops including rice (Cottyn et al., 2001), maize (Rijavec et al., 2007), bean (López-López et al., 2010), wheat (Kuźniar et al., 2020), and tomato (Xu et al., 2014). The collection of factors that go into the construction and organization of seed microbiomes is quite complex, with microbes drawn from not only the soil microbiome (Hardoim et al., 2012) but also dispersal agents, pollinator, and floral microbiomes (Gandolfi et al., 2013). Only specialized endophytic strains are able to colonize and flourish in reproductive plant organs (Truyens et al., 2015), and particularly, bacteria with competitive and adaptive colonization traits could dwell in seeds (Okunishi et al., 2005). Through a variety of processes, such as the solubilization of phosphate (Berg, 2009), synthesis of phytohormones such as indole-3-acetic acid (Defez et al., 2017), production of 1-aminocyclopropane deaminase (Glick, 2005), nitrogen fixation (Shabanamol et al., 2018), siderophore generation, and inhibition of phytopathogenic microbes (Hong and Park, 2016), seed bacterial endophytes have been reported to enhance plant development and health. Across a wide range of plant taxa including wheat, Brassica, rice, and maize, seed-associated bacteria have been identified using next-generation sequencing (NGS) techniques and are found largely within the bacterial phyla Acidobacteria, Proteobacteria, Actinobacteria, Firmicutes, Bacteroidetes, and Deinococcus-Thermus (Johnston-Monje and Raizada, 2011; Liu et al., 2012) with Gammaproteobacteria (Hardoim et al., 2015; Herrera et al., 2016) found to be the most prevalent type of bacteria at the class level, followed by Alphaproteobacteria (Hardoim et al., 2015) and Bacilli (Comby et al., 2009). Saving seeds with their indigenous microbes offers an enormous potential to safeguard, explore, and exploit microbial diversity and their untapped metabolic potential for plant, human, and environmental health.

The citrus fruit is a well-known source of biocompounds crucial for human nutrition and among one of the most commercially significant evergreen subtropical fruit crops in the world (Penyalver et al., 2022). North-eastern India is recognized as a significant hotspot of citrus biodiversity renowned as the “Treasure House of Citrus Genetic Wealth” attributable to the region’s special blend of different soil, physiographic, and climatic conditions (Malik et al., 2013). Domestication of wild plant species has been a historical process that has lasted since the dawn of agriculture. In order to meet changing human demands, wild species are domesticated and diversification of the seed microbiome occurs as a consequence of domestication (Abdullaeva et al., 2021). The northeastern region of the country harbors several wild, semiwild, and domesticated species including Citrus indica, C. jambhiri, C. grandis, C. macroptera, C. latipes, and C. megaloxycarpa aside from the commercially cultivated species like Citrus reticulata (Khasi mandarin) and C. sinenis. The rapid destruction of forests in the northeastern region has endangered these naturally occurring citrus species and the microbiota associated with them. These species may have great potential in the improvement of Indian Citrus industry by being a source of genes for combating biotic and abiotic stresses like pests and disease resistance, especially resistance to greening disease in its natural habitat (Yang and Ancona, 2022; Jing et al., 2023), tolerance to psorosis and exocortis virus (Sharma et al., 2004), cold tolerance or resistance (Hazarika, 2012), and drought resistance (Pandey et al., 2008). In citrus, the seed-to-seedling phase is very crucial as the seedling is commonly used as rootstock onto which scion varieties are grafted. Numerous ethnomedic benefits of citrus species, which have been used by the local population for centuries, have been discovered. Traditional medicines frequently employ the peel of the C. grandis fruit to treat epilepsy, edema, and cough (Anmol et al., 2021); fresh fruit, juice, and dried powder from C. indica are used for curing viral diseases, jaundice, and deadly communicable diseases like small pox (Malik et al., 2006). C. macroptera fruits are used medicinally to alleviate digestive disorders and stomach pain (Malik et al., 2013), whereas C. latipes fruits are used to treat viral infections, digestive problems, and jaundice (Malik et al., 2013). In a recent study, microbial profiling was carried out for five different biocompartments, viz. bulk soil, root endosphere, flush, rhizosphere, and flower) of citrus from a single California orchard (Xi et al., 2023). However, except for the work by Newcombe et al. (2018) reporting the presence of one cultivable endophyte per seed in the Rutaceae plant family and the study highlighting Cutibacterium and Acinetobacter as the most prevalent bacterial genera in seeds and shoots of C. limon (Faddetta et al., 2021), seed microbiota of citrus species remain largely unexplored. The present study aimed to explore the culturable members of the seed microbiome of C. reticulata Blanco, C. grandis (L.) Osbeck, C. latipes Tanaka, C. megaloxycarpa Lushaigton, C. jambhiri Lush, C. sinensis (L.) Osbeck, C. macroptera Montr, and C. indica Tanaka collected from the Purvanchal Himalaya region, carried out 16S rRNA–based taxonomic identification, and screened them for multifaceted PGP traits.

The Food and Agriculture Organization of the United Nations estimated that during the past century, agricultural crops have lost approximately 75% of their genetic diversity because of widespread desertion of genetically diverse traditional crops in favor of genetically uniform modern crop types. According to the concept of a “plant holobiont,” a plant is a coevolved species conglomerate made up of an interkingdom community of bacterial, archaeal, fungal, and other eukaryotic species (Vandenkoornhuyse et al., 2015). The domestication and cultivation of agricultural plants may have a detrimental effect on the beneficial microbial community of plants. According to Pérez-Jaramillo et al. (2016), the “back to the roots” outlook calls for investigating the microbiomes of wild relatives of agricultural plant species and their natural environments in order to identify beneficial bacteria that may have been lost during domestication and reintroduce them in seeds of current crop species. The northeastern region is renowned as a treasure house of citrus germplasm and is home to several landraces of the citrus species, many of which remain unexploited due to the lack of commercial cultivation (Singh and Singh, 2006). In the current work, healthy fruits of various citrus species were collected from different parts of northeastern India. Microbial strains were isolated from the surface-sterilized seeds of the fruits, which were further screened for features that promote plant growth. 16S rRNA–based molecular characterization was carried out for taxonomic identification of the seed microbiome.

A total of 36 microbial isolates were recovered from seeds of several citrus species, including 2 from Sikkim orange seeds, 2 from pummelo seeds, 5 from khasi papeda seeds, 9 from sour pummelo seeds, 2 from rough lemon seeds, 3 from sweet orange seeds, 5 from khasi mandarin seeds, 4 from melamesian papeda seeds, and 4 from Indian wild orange seeds. The colony isolates displayed a variety of shapes, of which the majority were circular, while a small number were irregular, most had glistening appearances and dull appearances were rare, flat to elevated surfaces, and white colony colors for the majority of strains, with cream, yellow, and orange pigments for a few. Faddetta et al. (2021) reported 60 endophytes from C. limon seeds that were streaked on a nutrient agar plate on the basis of distinctive colony morphology and pigmentation. Numerous studies have reported the association of microorganisms with the seeds of various crops like maize, rice, wheat, rice cut grass, beans, chickpea, cucurbits, and tomato (Johnston-Monje and Raizada, 2011; Verma et al., 2017;. Herrera et al., 2016; Khalaf and Raizada, 2016; Walitang et al., 2017; Morella et al., 2019). Different strains of epiphytic bacteria such as Bacillus amyloliquefaciens and Pseudomonas parafulva have been found to be associated with the fruits of several citrus species, including C. unshiu, C. reticulata, and C. sinensis and tend to have positive effect on plant health (Jing et al., 2023). PGPR isolated from the rhizospheres of different citrus species was prepared in an aqueous suspension and then used as a foliar spray and soil-drench, which led to increased crop growth (Chakraborty et al., 2016).

The auxin family member that has been investigated the most is indole-3-acetic acid, and tryptophan is the primary IAA precursor in five of the major pathways used by bacteria to produce auxin (Gamalero and Glick, 2011). The primary effects of bacterial IAA on plants include increased lateral and adventitious rooting, which improves nutrient uptake and root exudation, which, in turn, promotes bacterial growth on the roots (Thakuria et al., 2004). The synthesis of IAA-like compounds by the 36 isolates in the current investigation ranged from 4.8 ± 0.3 to 187.3 ± 18.1 µg ml^-1 h^-1 in the presence of L-tryptophan as a precursor among which CME6-1, CI11-4, CME6-9, and CG2-1 are the highest producers. The results were consistent with those of Verma et al. (2017), who reported that two isolates of rex rice (Oryza sativa L.) seed endophytic bacteria, E. asburiae and P. dispersa, produced 70.81 ± 0.98 and 7.98 ± 0.58 µg ml^-1 IAA, respectively. Faddetta et al. (2021) reported IAA production ability by three microbial endophytes isolated from C. limon seeds. Several other researchers have also noted the capacity of seed microbial endophytes to produce IAA in a variety of crops, including cucurbits (Khalaf and Raizada, 2016), rice (Walitang et al., 2017), marama beans (Chimwamurombe et al., 2016), maize (Sandhya et al., 2017), and wheat (Herrera et al., 2016).

Phosphate solubilization is one of the most crucial bacterial physiological features in soil biogeochemical cycles and in plant growth promotion by PGPB. In the current study, the capacity of dissolution of insoluble phosphates was determined on Pikovskaya’s media amended individually with various inorganic phosphate complexes, which include Ca₃(PO₄)_2, AlPO_4, FePO₄ and Zn₃(PO₄)_2. In the Ca₃(PO₄)₂-amended Pikovskaya’s agar plate, 14 isolates produced solubilization zones with solubilization ranging from 1.31 ± 0.2 to 14.84 ± 0.3 µg P_i ml^-1 h^-1. Nine bacterial isolates created a solubilization zone with a degree of solubilization between 1.54 ± 0.02 and 4.61 ± 0.2 µg P_i ml^-1 h^-1 in FePO₄-amended Pikovskaya’s media, while nine bacterial isolates produced a solubilization zone with a degree of solubilization ranging from 1.74 ± 0.3 to 3.61 ± 0.4 µg P_i ml^-1 h^-1 in AlPO₄-amended Pikovskaya’s media. The Zn₃(PO₄)₂ dissolution ranged between 2.44 ± 0.03 and 13.16 ± 0.13 µg P_i ml^-1 h^-1 among 12 isolates that showed a solubilization zone in Zn₃(PO₄)₂-amended Pikovskaya’s agar plate. The findings were supported by Thokchom et al. (2014), who reported that on Ca₃(PO₄)₂-supplemented medium, 54.8% of mandarin orange rhizobacterial strains could solubilize insoluble P with phosphate solubilization ranging from between 6.2 ± 1.1 and 267.9 ± 24.1 µg ml^-1. The ability of seed endophytes to solubilize phosphate has been documented in numerous papers Endophytes isolated from Rex Rice seed was found with phosphate solubilization properties (Verma et al., 2017). Johnston-Monje and Raizada (2011) reported 63 isolates obtained from Zea mays that solubilized phosphate. Seed-associated endophytes of cucurbit family reported 21% of strains, primarily Bacillus, lactic acid bacteria and Enterobacteriaceae, showed phosphate solubilization on Ca₃(PO₄)₂-amended plates (Khalaf and Raizada, 2016). The outcomes also agreed with those of Faddetta et al. (2021), who discovered that bacteria cultivated from C. limon seeds could flourish on solid media containing various sources of inorganic phosphate (Ca-, Fe-, and Al-). Gluconacetobacter diazotrophicus was reported to solubilize Zn₃(PO₄)₂ (Saravanan et al., 2007). The primary strategy used by phosphate- solubilizing bacteria to dissolve inorganic P is the production of low-molecular-weight organic acids, which bind phosphate with their hydroxyl and carboxyl groups, chelating cations, and contributing soil acidification, both of which result in the release of soluble phosphate (Rodriguez et al., 2004). Phosphorus is stored in plant seeds and pollen in form of phytate as a primary inositol source. Many seeds, notably those from grains and legumes, store phosphorus mostly in the form of phytate, which promotes the growth of seed endophytes during germination. Bean seed phytate solubilizers include Acinetobacter, Bacillus, Streptomyces, and Rhizobium endophyticum (López-López et al., 2010). Three bacteria isolated from surface-sterilized seeds of rice have been reported to act as phytate solubilizers (Verma et al., 2017), and isolated seed microbial endophytes of C. limon were able to grow on organic phosphate, i.e., Na-phytate-amended agar plates (Faddetta et al., 2021). The results of the current study revealed nine isolates with solubilizing capacities for sodium phytate ranging from 12.17 ± 0.5 to 18.00 ± 0.8 µg P_i ml^-1 h^-1.

Citrus decline is thought to be a condition with numerous contributing variables. However, a clear causal involvement of zinc nutrition as a contributor to citrus decline/blight have been hypothesized by many. Zinc-solubilizing bacteria are potential substitutes for zinc supplements because they transform applied inorganic zinc into usable forms. In this study, the zinc solubilization efficiency of bacterial isolates was investigated on modified Bunt and Rovira medium amended separately with two insoluble zinc sources, namely, ZnO and ZnSO_4. Seven isolates were able to solubilize ZnO (solubilization efficiencies ranged from 64 to 111.68), with isolate CJ7-1 having the largest solubilization diameter of 86 mm. Eight isolates solubilize ZnSO4 (solubilization efficiencies ranged from 60.4 to 290.45, with the isolate CME6-5 (B. altutidinis) having the largest solubilization diameter of 164.4 mm. The findings were consistent with those of Saravanan et al. (2004) who reported 17 bacterial isolates with strain ZSB-O-1 (Bacillus sp.) producing a clearing zone of 2.80 cm on ZnSO₄ and 1.80 cm of clearing zone with ZnO-containing medium, and G. diazotrophicus strain produced a solubilization zone diameter of 36 mm (Saravanan et al., 2007).

Since nitrogen makes up approximately 1.5%–2.0% of a plant’s dry matter and 16% of its total protein, plants need more nitrogen than other organisms. Different parts of plant act as microniches for the colonization of endophytic diazotrophs, where without competition with other microorganisms and absence of soil limitations present in the rhizosphere, the endophytes fix nitrogen (Verma et al., 2017). In the present study, 36% of the total microbe isolates were able to fix N₂ in N-free bromothymol blue media distinguished by a shift in the medium’s color from green to blue. Other studies have also reported the N₂ fixation ability by seed endophytes of Tylosema esculentum (Chimwamurombe et al., 2016), cucurbit family (Khalaf and Raizada, 2016) and Zea species (Johnston-Monje and Raizada, 2011).

Siderophore production is a vital characteristic of endophytic bacteria. These bacteria enhance plant growth and health as they synthesize siderophores that chelate iron in an iron-limited condition making it accessible for the plants and unavailable for phytopathogens (Mercado-Blanco and Bakker, 2007). In the present study, nine isolates of the total isolated strains were able to synthesize siderophores on CAS blue agar medium evident by formation of a halo zone around the colony. The findings were in line with the results of Chimwamurombe et al. (2016) who reported 13 isolates of T. esculentum seed positive for siderophore production and Herrera et al. (2016) who found six bacterial strains isolated from wheat seed positive for siderophore production.

HCN toxicity is influenced by its capacity to inhibit cytochrome c oxidase as well as other significant metalloenzymes (Nandi et al., 2017). While the ability of bacteria to make HCN is often considered as a negative factor, it has been suggested that HCN-producing endophytic bacteria may have important applications for the biocontrol of weeds, soil phytopathogenic fungi, and nematodes. In this study, only three isolates produced HCN in amounts that were positive. Phragmites australis seed endophytes (Verma et al., 2018), maize seed and root endophytes (Sandhya et al., 2017), and endophytes from nodules and roots of chickpea and pea (Maheshwari et al., 2019) have all been documented to be capable of producing HCN.

A fascinating feature in the current study is that the genus Bacillus appears to make up the culturable core seed microbiota across all the examined citrus species. A distinctive characteristic of seed endophytes seen in prior studies is the capacity to produce endospores, which are assumed to protect the inhabitants of the seed from internal changes (Truyens et al., 2015). Munir et al. (2020) reported that B. velezensis, Curtobacterium luteum, Microbacterium testaceum, and B. subtilis as the most prevalent endophytes in the leaves of 24 citrus types, with B. subtilis occurring most frequently in healthy plants. In the leaf tissues of numerous citrus rootstocks, the predominated bacterial species was found to be Pantoea agglomerans and B. pumilus (Araújo et al., 2001). Proteobacteria, Actinobacteria, Bacteroidetes, and Firmicutes were the most prevalent bacterial phyla in the microbial endophytic community of C. limon seeds. (Faddetta et al., 2021). Further, the genus Bacillus has been reported as common member of seed microbiota in different crops such as rice (Cottyn et al., 2001), gurana (e Silva et al., 2016), beans (López-López et al., 2010), and cucurbit family (Khalaf and Raizada, 2016). Members of the genus Bacillus is functionally diverse and one of the most widely utilized bacteria in the field of agrobiotechnology. It is well known that members of the genus Bacillus possess a multitude of beneficial attributes that aid plants either directly or indirectly by acquisition of nutrients, phytohormone production, and shielding plants from biotic and abiotic stress. In this study, various isolates obtained from the citrus seed belonging mainly to the Bacillus genus showed multifaceted plant growth promotion traits like nitrogen fixation, phosphate and zinc solubilization, and IAA production. Nitrogen fixation ability has been reported by B. pumilus and B. altitudinis that were isolated from the rhizosphere of maize and rice, respectively (Habibi et al., 2014). There were 16 of the 19 distinct Bacillus strains screened for the nifH gene that were found to be positive (Yousuf et al., 2017). According to Sharma et al. (2013), many species of Bacillus including B. megaterium M510, which was isolated from the maize rhizosphere in the eastern Himalayan region, were found to be capable of solubilizing iron and aluminum phosphate in addition to moderately solubilizing tricalcium phosphate. B. subtilis, B. thuringiensis, B. aryabhattai, and B. tequilensis are well-known zinc solubilizers (Singh et al., 2017). The isolates isolated from the chickpea rhizosphere that were highly similar to B. altitudinis displayed PGP traits including IAA, ammonia, siderophores, and phosphate solubilization (Kushwaha et al., 2021). K. rhizophila, isolated from citrus seed in this study was found to be a high IAA producer, while K. rhizophila Y1 isolated from maize rhizosphere soil exhibited growth-promoting characteristics such as phosphate solubilization and IAA synthesis (Li et al., 2020).

Microbial strains with PGP properties have the potential to be used in the development of bioinoculants, biofertilizers, and biopreparations that improve crop plant growth and yield. They are committed to being used in organic/natural farming as they increase crop yields enhancing the uptake of nutrition and improving plant vigor, combat pathogens, and increase plant abiotic resistance while conferring habitat-adapted tolerance to the host plant (Pylak et al., 2019). Endophytic isolates were evaluated as bioinoculants for potato tubers and recommended as biofertilizers to reduce dependency on chemical fertilizers (Kuźniar et al., 2019). Works have been focused on the evaluation of compatibility of the different PGP strains and prospecting the optimal combinations of the isolates to develop into bioinoculants for commercialization. Bacillus strains have been most frequently utilized as the common commercially available bioinoculants till date (Kuźniar et al., 2019). For instance, Bacillus species have successfully been used as priming agents in crops such as potato (Gururani et al., 2013), rice, mung bean, and chickpea (Chakraborty et al., 2011). As such, additional research is needed to translate the in vitro PGP activities of the isolates into pot experiments in greenhouse and field trials for improved crop yield.

Our study clearly demonstrated that citrus seeds are an important source of beneficial seed microbiomes and that members belonging to genus Bacilli seem to be the core member as common dominant bacteria across eight citrus species tested. The culturable members of the citrus seed microbiome possessed multifaceted PGP traits like dissolution of insoluble phosphate and zinc complexes, mineralization of the organic phosphate complex, production of IAA-like substances, siderophore-like substances and HCN, and nitrogen fixation. Isolates CG2-1, CME6-1, CME6-4, CME6-5, CME6-8, CME6-9, CJ7-1, CMA10-1, CI11-3, and CI11-4 are promising candidate bioinoculants for development of microbial consortium having multifaceted PGP traits for nutritional benefits of nitrogen, phosphorus, and zinc and IAA hormonal benefits to citrus crops for better fitness in acid soils. Bacterial seed endophytes may be highly conserved in plant species to specific regions, necessitating the exploration of species from different regions. Additionally, research can be conducted using different growth media other than LB to increase the number of cultivable seed microbiomes from different citrus species. The isolated beneficial strains and their consortium will be tested further in greenhouse and field trials under natural environmental conditions.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

DT conceived and designed the study, validated results, and supervised the final draft of manuscript. SS, CC, and PU executed the experiments and wrote the first draft of manuscript. SH, PD, AS, and BL helped in citrus sample collection and checking the draft manuscript. All authors contributed to the article and approved the submitted version.