From glomalin to glomalose: unraveling the molecular identity of the MAb32B11 antigen

Introduction

Arbuscular mycorrhizal (AM) fungi belong to Glomeromycotina and form mutualistic relationships with over 70% of terrestrial plants, enhancing nutrient and water uptake and promoting plant growth (Leigh et al., 2009; Wang et al., 2017; Kakouridis et al., 2022). They enhance plant tolerance to biotic and abiotic stresses, including drought and heavy metals (Begum et al., 2019; Shi et al., 2023). Arbuscular mycorrhizal fungi also serve as a valuable carbon sink in the soil, storing over 10% of carbon from fossil fuel emissions, mainly as fungal necromass (Schweigert et al., 2015; Hawkins et al., 2023). This fungal necromass forms a scaffold stabilizing soil aggregates and promoting soil aggregation within mineral-associated organic matter (Irving et al., 2021; Hawkins et al., 2023). Many of these benefits AM fungi provide are attributed to a substance they produce and release into the soil called ‘glomalin’. The abundance of glomalin in the soil has been correlated with many plant and soil health benefits, such as improved plant tolerance to abiotic stresses, increased soil aggregation, increased soil water retention, and carbon sequestration (Wright & Upadhyaya, 1996; Rillig, 2004; Zou et al., 2014; Zhang et al., 2017). This substance was discovered in 1996 by Dr. Wright after raising a monoclonal antibody named MAb32B11 against crushed Rhizophagus irregularis spores. The antibody was reported to be an IgM and monoclonal by Wright et al. (1996). The reactivity of the antibody was tested against AM fungi hyphae residing on plant roots using an immunofluorescence assay. Further efforts were also made to discover the nature of the antigen that reacts to MAb32B11 antibody, and an unusual glycoprotein was suggested to be the potential antigen for this antibody. Although glomalin was predicted to be a glycoprotein, the substance was unaffected by many harsh conditions that would degrade most proteins, such as autoclaving, sodium dodecyl sulfate (SDS), or even protease treatments (Wright et al., 1996; Wright & Upadhyaya, 1996).

Although always considered a protein, the definition of glomalin changed over time to reflect variations in the purification and quantification procedures used by different groups and their lack of specificity. Glomalin-related soil protein (GRSP) was coined to reflect that many proteins besides the MAb32B11 antigen were purified in the extraction procedure. GRSP was further separated into easily extractable GRSP to define the soil proteins collected after a single autoclave cycle at 121°C with 20 mM sodium citrate pH 7.0 and total GRSP to define soil proteins collected after multiple autoclave cycles at 121°C with 50 mM sodium citrate pH 8.0 until the extract becomes ‘straw-colored’ (Wright et al., 1996; Wright & Upadhyaya, 1998; Rillig, 2004; Irving et al., 2021). Although the addition in terminology eased confusion, it incorrectly implies that the extraction process is specific for glomalin; previous work has observed co-extracted compounds and proteins of nonmycorrhizal origin soil glomalin extract (Nichols & Wright, 2005; Rosier et al., 2006; Schindler et al., 2007; Gillespie et al., 2011). The contradiction in using a nonspecific extraction process to define one unknown molecule emphasizes the knowledge gap in glomalin's molecular nature. A similar expansion in terminology was introduced to reflect nuances in glomalin quantification techniques. Two main methods have been used to quantify glomalin in the high-temperature citrate extracts (HTCEs): enzyme-linked immunosorbent assay (ELISA) using MAb32B11 and total protein quantification using the Bradford or Lowry methods. Initially, an indirect ELISA was designed with MAb32B11 to quantify glomalin, and this GRSP fraction was later termed immunoreactive soil proteins (IRSP; Wright et al., 1996; Wright & Upadhyaya, 1998; Rillig, 2004). In response to limited MAb32B11 availability, the lack of pure standard preventing absolute quantification, and the assumption of glomalin existing in the environment as an abundant soil protein, the Bradford assay followed by HTCE became a standard method for glomalin quantification in publications, and this GRSP fraction was termed Bradford-reactive soil proteins (BRSP). However, simplicity brings additional difficulties as the Bradford assay is both nonspecific and influenced by contaminants in soil, such as lipids, humic acids, and sugars (Pande & Murthy, 1994; Nichols & Wright, 2005; Banik et al., 2009; Irving et al., 2021). The differences between the methods raise questions about their ability to equally and reliably detect glomalin.

Over the years, various protein candidates have been proposed for glomalin. Early descriptions of glomalin described it as an iron-associated glycoprotein with N-linked oligosaccharides similar to hydrophobins (Wright et al., 1996, 1998, 2000). In 2006, heat shock protein 60 (HSP60) from the AM fungus, R. irregularis, was proposed as the glomalin gene product based on predicted compatibility between the MAb32B11 binding sequence and the HSP60 sequence as well as observed antibody affinity for nonglycosylated RiHSP60 expressed in vitro (Gadkar & Rillig, 2006). However, while glomalin extract from the soil is known to be a heterogeneous mixture of compounds and proteins, RiHSP60 was not detected when the glomalin extracts were analyzed with mass spectrometry (Gillespie et al., 2011). Similarly, in a proteomics study of R. irregularis mycelia, HSP60 was not detected (Murphy et al., 2020). In addition, the widespread conservation of the MAb32B11 epitope across fungi raised questions about the molecular nature of glomalin (Irving et al., 2021).

In this study, we aimed to revisit the nature of the MAb32B11 antigen and, more generally, glomalin. We showed significant discrepancies between the Bradford and ELISA techniques for quantifying glomalin. Although some fungal samples had detectable BRSPs, no IRSPs were detected. Conversely, we did not detect any BRSPs in hyphal and spore samples from AM fungi despite a strong and dose-dependent IRSP signal. These results made us question the molecular nature of glomalin as a protein. We expressed in vivo and purified the RiHSP60 protein, showing that MAb32B11 does not cross-react with this protein. Furthermore, we observed a strong correlation between the amount of carbohydrates in AM extracts and the ELISA signal for glomalin. We analyzed the glomalin extract from R. irregularis hyphae and spores using size exclusion chromatography (SEC); we identified that the molecule is most likely a complex carbohydrate ranging in 511–600 kDa.