Caffeine (1,3,7-trimethylxanthine), theophylline (1,3-dimethylxathine), and theobromine (3,7-dimethylxanthine) are the most common and most well-known methylxanthines, and these compounds are primarily found in coffee, tea, and chocolate, respectively (Andreeva et al., 2012, Fredholm, 2010). Methylxanthines are of growing interest due to their diverse pharmacological applications, including their ubiquitous ability to cross the blood-brain barrier and act as adenosine receptor antagonists (Hayallah et al., 2011, Monteiro et al., 2016; Nivedita Singh, 2018). Caffeine is most commonly known for its stimulating and awakening effects (Snel et al., 2004). Theobromine and theophylline are both known for acting as smooth muscle relaxants, and they have been used in the treatment of asthma (Simons et al., 1985; Smit, 2011; Zhang et al., 2010) and as a bronchodilator to treat asthma and chronic obstructive pulmonary disease (COPD) (Barnes, 2006, Barnes, 2013, Hayallah et al., 2011, Zhang et al., 2010), respectively. Less common and less well-known natural methylxanthines include paraxanthine (1,7-dimethylxanthine), 1-methylxanthine, 3-methylxanthine, and 7-methylxanthine. In contrast to caffeine, theophylline, and theobromine, these compounds are far less abundant in nature but are equally as relevant to the medical field. Paraxanthine and 1-methylxanthine, for example, can both be found in the human body during the degradation of caffeine by the liver, as around 80% of ingested caffeine is first metabolized to approximately 80-84% paraxanthine, which can then be further broken down to 1-methylxanthine (Kot and Daniel, 2008, Victorino et al., 2021). Paraxanthine has been investigated for potential neuroprotective properties and has shown a strong potential to treat and prevent Parkinson’s Disease (Guerreiro et al., 2008, Janitschke et al., 2021, Xu et al., 2010).

1-Methylxanthine has a variety of potential uses and medical applications, starting with its strong antioxidant properties. Like caffeine, 1-methylxanthine is able to scavenge hydroxyl radicals through the Fenton reaction, thereby providing protection to cells from the damage caused by most reactive oxygen species (Lee, 2000, Santos et al., 2010). Additionally, 1-methylxanthine has been found to enhance the radio-sensitization of human colorectal cancer cells and lung cancer cells at concentrations almost non-toxic to the target tumor cells by inhibiting the repair of double stranded breaks in the DNA induced by radiation (Jeong et al., 2009, Youn et al., 2009). Derivatives of 1-methylxanthine have also been explored for potential medical applications. Specifically, 3,7-dihydro-1H-purine-2,6-dione has been designed and synthesized to act as a corticotropin-releasing-factor (CRF) receptor antagonist. Various endocrine and psychiatric disorders (Gilligan et al., 2000), such as depression (Nemeroff et al., 1984), anxiety (Owens & Nemeroff, 1991), and post-traumatic stress disorder (PTSD), have been linked to the hypersecretion of CRFs. Selective antagonism of these CRF receptors by derivatives of 1-methylxanthine are promising therapeutics to help treat these diseases (Hartz et al., 2004). In addition, 1-methylxanthine has been found to be less toxic than caffeine with a median lethal dose (LD50) of 510 mg 1-methylxanthine/kg body weight in mice (Alstot et al., 1973) compared to an LD50 of only 100 – 360 mg caffeine/kg body weight (Boyd, 1959).

Unfortunately, traditional chemical routes of methylxanthine synthesis can be quite complicated, non-selective, and potentially hazardous (Gulevskaya and Pozharskii, 1991, He et al., 2006, Müller and Sandoval-Ramírez, 1995). The direct synthesis of 1-methylxanthine, for example, requires six steps, involves hazardous chemicals such as methanolic hydrogen chloride gas, and results in a yield of only 20% (Black & Gatto, 1989). Due to the variation in accessibility of methylxanthines both naturally and synthetically, the cost of purchasing these compounds is extremely broad. Caffeine can be purchased in bulk for as little as $0.05 per gram and theophylline for $0.26 per gram, whereas a compound such as 1-methylxanthine can cost as much as $918 per gram, and is not available in bulk sizes (prices according to Sigma Aldrich, June 2023, Table S1).

We have previously developed methods for the biosynthetic production of both paraxanthine and 7-methylxanthine from caffeine using metabolically engineered Escherichia coli expressing combinations of the N-demethylase genes ndmABCDE, which were identified and characterized from Pseudomonas putida CBB5 (Mock et al., 2022, Mock et al., 2022a, Mock and Summers, 2023). Expression of ndmABCDE facilitates the metabolism of caffeine and theophylline and enables the bacterium to survive on these compounds as a sole source of carbon and nitrogen (Summers et al., 2011, Yu et al., 2009). NdmA and NdmB are responsible for the sequential N1- and N3-demethylation of both caffeine and theophylline (Summers et al., 2012, Summers et al., 2011). NdmCDE forms a soluble complex wherein NdmC is responsible for the N7-demethylation of 7-methylxanthine to xanthine, NdmD is a reductase that facilitates the electron transfer from NADH for all of the demethylation reactions, and NdmE provides structural support (Summers et al., 2013).

Crystal structures have been successfully obtained for NdmA and NdmB (Kim et al., 2019), which has facilitated the targeted mutagenesis of NdmA (Mills et al., 2021). The resulting mutants, NdmA3 and NdmA4, were successfully modified to alter the enzyme specificity by mimicking the active site of NdmB. The mutations resulting in NdmA3 conferred increased activity towards the N3-methyl group of caffeine, and the mutations resulting in NdmA4 enabled the mutant to convert caffeine to paraxanthine more efficiently than was previously observed by NdmB (Mills et al., 2021). Optimization of paraxanthine production led to the generation of E. coli strain MBM019, a strain that is also capable of generating 7-methylxanthine via paraxanthine through the demethylation of caffeine (Mock et al., 2022, Mock et al., 2022a). Furthermore, we tested these whole cell biocatalytic processes, including the collection and purification of these compounds, at milligram scales to demonstrate the potential of these processes as safe, efficient, and cost-effective alternatives to chemical synthesis for industrial production. The goal of this study was to expand upon the versatility of these N-demethylase genes and the efficiency of strain MBM019 by establishing a method for the sustainable production of 1-methylxanthine from the low-cost substrate theophylline. In this project, we replaced the ndmA4 genes found in MBM019 with ndmA3 to generate the E. coli strain MBM020, capable of enhanced and efficient degradation of theophylline to 1-methylxanthine (Fig. 1). We further describe the gram-scale production and purification of 1-methylxanthine, demonstrating the highest reported yields of biocatalytically-produced methylxanthines to date.