India, being enriched with diverse agro-ecological environments, is the second largest producer of wheat (Triticum aestivum L.) worldwide (MoA & FW, 2019), which is cultivated during the Rabi season (Ramadas et al., 2020). It requires an optimal mean temperature ranging from 20 to 25°C (Tripathi and Mishra, 2017). However, the average global temperature is reported to be increasing at a rate of 0.18°C every decade (Hansen et al., 2012). Worldwide, a significant part of wheat areas are experiencing heat stress, and it is estimated that the production of wheat will decrease by 4–6 million tons for every 1°C rise in temperature. There will be a 9%–25% profit loss for rainfed wheat when the temperature rises by 2–3.5°C (Aggarwal et al., 2009; Mukherjee et al., 2019). Several studies have recorded the remarkable influence of high temperatures on wheat (Satorre and Slafer, 1999) and also studied the impact of growing season temperatures in major wheat-producing regions (Asseng et al., 2015; Liu et al., 2016; Hatfield and Dold, 2018; Demirhan, 2020). In 2021, every month experienced warmer temperatures than the average, as stated in the 2021 Global Climate Report (NOAA, National Centers for Environmental Information, 2021) and there was a 15%–20% yield loss observed by CIMMYT’s on-farm experiments in 2022 (Bentley et al., 2022).

Wheat yield, corresponding grain size, and weight will be affected due to intense temperature changes during the sensitive growth stages of flowering, anthesis, and milking (Talukder et al., 2014). In order to accommodate the adverse physiological and biochemical changes that occur due to stress factors, wheat improvement approaches have been proposed to bridge the yield gaps and improve productivity by developing climate-resilient wheat genotypes (Kumar et al., 2019). Several management strategies including changes in agronomic practices such as sowing time, irrigation pattern, nutrient management, timely weeding, and resistant cultivars have been suggested (Singh et al., 2017). Nowadays, genetic engineering and manipulation by molecular breeding are being explored to provide tolerance to several biotic and abiotic stresses, as they offer a more reliable solution than conventional breeding (Varshney et al., 2011).

Currently, the plant microbiome has gained attention just like the human microbiome in maintaining homeostasis (Hirt, 2020; Wagner, 2022) under normal and stress conditions. Characterizing the core microbial communities of the plant and identifying their role are critical inputs in alleviating various stress factors. The modern breakthrough in sequencing technologies has facilitated extensive investigations on microbial communities in the plant vicinity such as the rhizosphere, phyllosphere, and endosphere of important crops and model plant species (Redford and Fierer, 2009; Bulgarelli et al., 2012; Lebeis et al., 2012; Bodenhausen et al., 2013; Shakya et al., 2013). The significance of plant microbiomes and their role in stress tolerance has also been explored by several multi-omics techniques such as whole-genome and metagenomic analyses (Bai et al., 2015; Bulgarelli et al., 2015).

Several research studies lay emphasis on contemplating the use of plant growth-promoting (PGP) microorganisms to enhance the ability of host plants to sustain their productivity under stress conditions (Verma et al., 2015; Torbaghan et al., 2017; Albdaiwi et al., 2019; Ripa et al., 2019; Atieno et al., 2020; Rana et al., 2020). Several endophytic bacteria such as Pseudomonas, Bacillus, Enterobacter, Burkholderia, and Azospirillum colonize various plant species, and their promising benefits have been assessed over the years (Pieterse et al., 2012; Zipfel and Oldroyd, 2017; Afzal et al., 2019; Song et al., 2021; Suman et al., 2021; Zhang et al., 2021). Like other plant organs, the seed-harbored microbial endophytes have paramount potential, by establishing their existence in the host plant, generation after generation, through vertical transmission (Cope-Selby et al., 2017; Shade et al., 2017). The seed endophytes apparently foster seed germination and promote plant growth and development by utilizing various direct and indirect mechanisms, under several biotic and abiotic stress conditions (Santoyo et al., 2016; Shahzad et al., 2017; Rodríguez et al., 2018; Shearin et al., 2018).

Furthermore, PGP microbes are being considered an inexpensive alternative to resource-consuming agrochemicals in sustainable agriculture. Several kinds of commercial microbial formulations are available that specifically target nutrient mobilization, hormone production, and/or biocontrol of different pathogens (Berg et al., 2020). Such formulations are based on either a single or a consortium of microbial isolates and are not specific to crops. Hence, oftentimes, formulations showing excellent results under laboratory conditions fail to do so under field conditions. Wheat-associated microbial partnerships have been explored in several studies. Still, there are no evident reports on culturable and unculturable seed endophytic microbiomes of heat sensitivity variable wheat varieties. The epiphytic, endophytic, and rhizospheric bacterial diversity of wheat growing in six agroclimatic zones in India has been elucidated, and more than 200 diverse isolates were identified as PGP isolates (Verma et al., 2015; Suman et al., 2016; Verma et al., 2016; Verma and Suman, 2018; Sharma et al., 2022). Metagenomics of wheat rhizosphere has been studied for abiotic stress management and clearly indicated the future implications of beneficial bacteria in the rhizosphere to combat stress conditions (Ahlawat et al., 2018). The role of epiphytic pink-pigmented methylotrophic bacteria in wheat (T. aestivum) has been determined to enhance seed germination and seedling growth through the production of indole acetic acid (IAA) (Meena et al., 2012). Chakraborty et al. (2013) have shown the potential of osmotic stress-tolerant bacteria in water stress amelioration and plant growth promotion in wheat plants.

Seed microbiomes/endophytes are of particular interest as they are transmitted from generation to generation. By being seed-borne, these endophytes assure their presence in new plants (Truyens et al., 2015). Seed microbiomes have diverse interactions and are predicted to be an important biological resource for sustainable agriculture (Barret et al., 2015; Lugtenberg et al., 2016). In this study, the seed microbiomes of three wheat varieties having contrasting heat sensitivity indices (HSIs) were explored for culturable and unculturable microbial community composition and their probable beneficial functions, with the ultimate aim to identify beneficial microbiomes tolerant to high temperatures. Selected heat-tolerant PGP seed endophytes together with the unculturable microbiome would be formulated as small microbial communities to mitigate stress conditions in plants.

Endophytes are the key class of plant symbionts subsisting inside the cells, such as roots (rhizosphere), leaves (phylloplane), stems (laimosphere and caulosphere), fruits (carposphere), seeds (spermosphere), and flowers (anthosphere) (Clay and Holah, 1999; Lindow and Brandl, 2003; Saikkonen et al., 2004; Shahzad et al., 2017), without causing any diseases (Brader et al., 2017), and are bonded together throughout their life cycle. The diversity of endophytic bacteria in the seeds of wheat varieties having contrastive heat sensitivity was assessed by culturable and unculturable approaches in this study. The culturable diversity showed that the heat^S variety held a larger population of bacteria compared to the heat^T variety, which might be due to the selection pressure offered by the different genotypes over the bacterial communities. Moreover, the susceptible genotypes probably require assistance to sustain their life, which diverse groups of microorganisms could provide either by enhancing the plant nutrient acquisition and biocontrol or by improving the genetic makeup of the plants through certain stress-responsive genes and molecular chaperones. Yet, there is no proof confirming this cause until now. However, it was stated that seed endophytes have a wide extent of colonization and could be found either in lesser or greater numbers (Mundt and Hinkle, 1976; Rijavec et al., 2007). Also, epiphytic, endophytic, and rhizospheric bacterial diversity of wheat growing in six agroclimatic zones in India has been determined by various studies (Verma et al., 2013; Verma et al., 2014; Verma et al., 2015; Suman et al., 2016; Verma et al., 2016; Verma and Suman, 2018) and more than 200 diverse isolates were identified as PGP isolates. In this study, the endophytic bacteria from three contrasting wheat varieties were isolated using four different growth media, whereas the NA media held the highest population ranging from 12×10² to 57×10² cfu g⁻¹ of seeds. This was in contrast with the observations by Robinson et al. (2016), where the serially diluted samples of surface-sterilized seeds rendered no bacterial colonies. Furthermore, in rice seeds, the population of bacteria was detected up to 3.5 × 10⁵ cfu g⁻¹ of fresh tissue (Hardoim et al., 2012), and the aerial tissues had been expeditiously inhabited by the seed-borne endophytes. Likewise, Sai Prasad et al. (2021) and Manias et al. (2020) reported that it would be sensible to use different growth media for the isolation of endophytes, especially for analyzing abundance and diversity.

Thereafter, the bacterial colonies showing different morphological characteristics were purified and subjected to cultural characterization to evaluate their capability to grow at a wider range of environmental conditions in this study, while several abiotic stress factors ranging from high temperature, drought, and salinity to oxidative stress and heavy metal toxicity are inhibitory to the plants’ growth. The ability of osmotic stress-tolerant bacteria to promote plant growth and ameliorate the water stress in wheat has been deciphered by Chakraborty et al. (2013). However, it was demonstrated that most of the WSEB isolates from three different varieties were able to grow under diverse circumstances. The WSEB isolates from heat^T varieties were well established under an increasing temperature range when compared to the heat^S variety. The ability of these endophytes might increase the bounds of temperature tolerance of heat^T varieties (Dastogeer et al., 2022). This was in line with the studies that growth of spot inoculated purified microbial isolates on medium plates under selection pressure of some chemicals or incubation conditions is considered as tolerant to respective stress condition. (Zhang et al., 2006; Fan et al., 2018).

PGP endophytic microorganisms help regulate the growth of plants through the production of various phytohormones such as auxins, gibberellins, and cytokinins (Santoyo et al., 2016). The production of IAA by the epiphytic pink-pigmented methylotrophic bacteria in wheat has been proven to enhance seed germination and seedling growth (Meena et al., 2012). Studies have reviewed that auxin, as an effector molecule, regulates the interaction between bacteria and plants, as well as between bacteria (Lambrecht et al., 2000; Spaepen and Vanderleyden, 2011). The endophytes also aid in solubilizing the unavailable form of nutrients such as phosphorus, potassium, iron, and zinc and make them available to plants by facilitating absorption. Moreover, the siderophore produced by the endophytic microbes benefits the plants through both direct and indirect mechanisms, i.e., by promoting iron acquisition as well as by increasing the competition for available iron and defending against pathogenic organisms (Bouizgarne, 2013). However, most of the heat^T WSEB endophytes isolated from heat^T and heat^S varieties had the tendency to produce more IAA and solubilize phosphorus, while some were able to grow in an N-free medium and solubilize zinc phosphate. Relatively few WSEB isolates could solubilize potassium and zinc oxide and produce siderophores.

Subsequently, the 16S rRNA gene of all heat^T WSEB isolates was amplified and identified in a total of 10 different genera, namely, Alcaligenes, Bacillus, Brachybacterium, Enterobacter, Pantoea, Priestia, Pseudomonas, Staphylococcus, Stenotrophomonas, and Streptomyces, with more diversity in heat^S V1 ( Figure 9 ). Three genera, viz., Bacillus, Priestia, and Stenotrophomonas, were common in all three varieties. Some genera were specific to particular varieties, such as Alcaligenes, Pantoea, and Streptomyces specific to V1, Enterobacter specific to V2, and Brachybacterium specific to V3. Then, the phylogenetic tree was constructed, having a 1,000 bootstrap value by the neighbor-joining method. However, similar studies have been conducted in which phylogenetic trees and distance matrices were made by maximum likelihood algorithms using the MEGA software (Thomas, 2011; Fan et al., 2018; Shan et al., 2018; Sánchez-Cruz et al., 2019). The bacterial communities extensively found as seed endophytes belonged to the phylum Proteobacteria. Other phyla like Actinobacteria, Firmicutes, and Bacteroidetes were encountered to a lesser degree (Hardoim et al., 2015). The most commonly reported genera from the plant seeds include Bacillus and Pseudomonas, while Paenibacillus, Micrococcus, Staphylococcus, Pantoea, and Acinetobacter have also been reported (Truyens et al., 2015; Herrera et al., 2016). Large amounts of endospore-forming Gram-positive bacteria are capable of surviving inside the seeds for a longer period (Luna et al., 2010). Likewise, in this study, Bacillus and Priestia (derived from Bacillus) together contributed to more than 50% of the identified WSEB isolates, which might be attributed to the fastidious nature of their growth and the spore-forming ability to survive.

In general, barely a limited number of potential bacteria are allowed to enter and colonize inside parts of the plant, which confer the specificity of endophytic rather than rhizospheric microbiome. The endophytic microbiome advance their connection mainly during seed germination and early development (Johnston-Monje et al., 2014), making an integrated network in the plant interior, which remains persistent through all the phases of the plant growth cycle (Podolich et al., 2015). Unraveling the probable beneficial functions of the plant microbiome by high-throughput sequencing approaches together with the culture-dependent approaches would pave the way towards understanding their interactions that could be labored to improve plant growth and health (Mendes et al., 2013; Suman et al., 2021). The composition of seed microbiome from drought-tolerant and drought-susceptible wheat lines was evaluated under rainfed and drought conditions by using culture-dependent and metagenomic methods (Hone et al., 2021). The assembly of bacterial and fungal microbiota in domesticated and wild wheat species was regulated by the selection factors such as plant habitat and host genetics (Bulgarelli et al., 2012; Edwards et al., 2015). These differences showed greater degrees of variance in the community structure (Özkurt et al., 2020). Likewise, our study has clearly depicted that the fungal OTUs were significantly higher than the bacterial OTUs, and also the bacterial communities differed according to the hosts’ ability to tolerate heat stress, and this is not very surprising as the stress tolerance pressure imparts a higher selective force on the community assembly (Peiffer et al., 2013; Mendes et al., 2014; Ofek-Lalzar et al., 2014; Coleman-Derr et al., 2016).

The evolution of bacterial communities in rhizospheric samples from wheat fields was studied by Donn et al. (2015). Ahlawat et al. (2018) deduced the eventual significance of rhizospheric bacteria to combat stress conditions by studying the metagenomics of wheat rhizosphere. Germida and Siciliano (2001) have shown that the taxonomic diversity of bacteria congregated to the roots of modern, recent, and ancient cultivars of wheat. A total of 30 different OTUs that belonged to Alphaproteobacteria, Betaproteobacteria, Deltaproteobacteria, Gammaproteobateria, Actinobacteria, Bacilli, Clostridia, and uncultivable bacteria were identified from the wheat rhizosphere (Velázquez-Sepúlveda et al., 2012). Similarly, this investigation revealed the distribution of bacterial OTUs in three dominant phyla, viz., Proteobacteria, Actinobacteria, and Firmicutes, along with minimum contribution from several other phyla such as Acidobacteria, Aquificae, Bacteroidetes, Chlorobi, Chloroflexi, Cyanobacteria, Chrysiogenetes, Deferribacteres, Deinococcus-Thermus, Elusimicrobia, Fibrobacteres, Fusobacteria, Gemmatimonadetes, Lentisphaerae, Nitrospirae, Planctomycetes, Spirochaetes, Synergistetes, Tenericutes, Thermotogae, and Verrucomicrobia.

Considering the class-level distribution, the most abundant class was Gammaproteobacteria, followed by Alphaproteobacteria (Hardoim et al., 2015; Rahman et al., 2018) and Bacilli (Comby et al., 2016). It was also noted from studies on barley, rice, bean, and maize endosperms that Proteobacteria, Actinobacteria, and Firmicutes were usually found to be dominant (Kaga et al., 2009; Ruiza et al., 2011). This was in agreement with the current study in which the Gammaproteobacteria had the highest number of OTUs, while contrasting evidence showed the distinct dominance of Actinobacteria and Clostridia as the second and third highest, respectively, followed by Bacilli, Betaproteobacteria, and Alphaproteobacteria. The unclassified classes derived from phyla Proteobacteria and Cyanobacteria also offered approximately 250 OTUs per sample, whereas Deltaproteobacteria came next to the unclassified group.

When comparing the presence of Zn-mobilizing species such as Adhaeribacter, Janthinobacterium, Massilia, and Pseudomonas in rhizospheric and bulk soil of wheat, it was observed that the relative abundance was higher in the rhizosphere that could articulate the local community structure to improve the plant growth by mobilizing nutrients (Wang et al., 2021). Acinetobacter was highly prominent, while Pantoea, Pseudomonas, and Paracoccus were also detected from the endosperm of T. aestivum cv. “Hondia”. Acinetobacter, Micrococcus, and Staphylococcus were also viewed by the endosperm as endophytes (Truyens et al., 2015; Soldan et al., 2019). Distinctively, in our study, Clostridium was obtained as the predominant genus from heat^S V1, while Stenotrophomonas was predominant in heat^T V2 and V3, followed by Enterobacter, Bacillus, and Streptomyces. Moreover, the variation in relative abundance and specificity of bacterial communities was also found between the tested heat^T and heat^S varieties, where a total of 36 different genera were specific only to the tolerant variety as well as 25 different genera specific to the susceptible variety. An interesting feature with this specificity was that the unique genera (Thermomicrobium, Acetivibrio, Methylophaga, Fervidobacterium, and Thermotoga) found in heat^T varieties were mainly of thermophilic nature, while those particular to the heat^S variety were mesophilic. The number of OTUs for Beggiatoa and Pantoea varies rapidly between the heat^T and heat^S varieties.

Hone et al. (2021) showed that the microbial diversity and abundance of wheat seed microbiome would vary between the drought-tolerant and -susceptible lines under drought and rainfed conditions, in which the lines subjected to drought had a greater Shannon index, implying more diversity. Jochum et al. (2019) also recognized the improved alpha and beta diversities of wheat microbiome after successive drought stress. In contrast, there are other studies that suggest a decline in the diversity of sorghum and wheat microbiomes (Le Cocq et al., 2017; Safin et al., 2018), and the Shannon diversity of modern wheat cultivars was higher than that of primitive landraces (Gholizadeh et al., 2022). However, in this study, the H–index of heat^T V2 was higher than V1 and V3 in the metagenomic composition, while in the case of culturables, the H–index was found to be very low in V2, moderate in V3, and the highest in heat^S V1. The diversity of shoot endophytic communities (Sobs and Chao 1) was relatively less than that of the root communities (Liu et al., 2017). In addition, it is believed that the reduction in Shannon index and evenness denotes a lesser diversity, while more species richness and greater diversity are believed to be crucial for managing stress conditions due to higher metabolic rates (Nautiyal and Dion, 2008). Likewise, this study demonstrated high evenness in the heat^S variety compared to the other heat^T varieties. Hence, the diversity indices as a whole show that there is a greater abundance and diversity in V2 followed by V1 and V3. This study also depicts that the culturable diversity is very high in heat^S V1 than in heat^T varieties, which could be related to the selection pressure on microbial communities under stress conditions and also due to selective isolation practices.

To the best of our knowledge, this study is the first to demonstrate the microbial community composition of wheat varieties with variable heat sensitivity by employing both culturable and unculturable approaches. The findings have significantly shown that there is a remarkable variation in the culturable as well as unculturable microbial abundance and diversity. Even though the population and diversity of culturable bacteria are higher in the heat^S variety, the preferential demand for survival under heat-stressed conditions contributes to the lesser diversity of heat^T varieties. Moreover, a notable number of unculturable bacteria that are thermophilic by nature, such as Thermomicrobium, Acetivibrio, Methylophaga, Fervidobacterium, and Thermotoga, are associated with the tolerant varieties that could impart resilience to the plants. Furthermore, the heat sensitivity variable varieties from different agro-climatic zones can be explored to a greater extent to better understand the interaction of microbial communities with the plant, their composition, and the mechanisms by which they stipulate tolerance to heat stress. In the future, manipulating the crops’ efficiency through microbiome approaches would result in phenomenal changes to sustainable agriculture under changing climatic conditions.

The original contributions presented in the study are publicly available at NCBI for 16S rRNA gene sequences of Wheat Seed Endophytic Bacteria under accession numbers OP782593 - OP782627 and Metagenome data under BioProject PRJNA944920.

AK: Methodology, Investigation, Data curation, and Writing—Original draft preparation. AS: Conceptualization, Funding acquisition, Methodology, and Writing—Reviewing and Editing. PS: Visualization and Validation. PS: Wheat seed resources. SG and DP: Operation support. All authors contributed to the article and approved the submitted version.