Summary

Our project aims to eliminate the phytotoxicity of caffeine in spent coffee grounds (SCG) by engineering E.coli to express N-demethylases (Ndm) derived from Pseudomonas putida (P.putida) for subsequential demethylation of caffeine and for converting caffeine into xanthine neutral to plant growth, enabling SCG recycling into organic fertilizer and reducing waste. To this end, we constructed five ndm genes, and incorporated them into the T7-LacO Promoter, RBS, and T7 terminator (composite parts). And the results of part collection revealed that our composite parts cooperatively promote caffeine degradation.

1 Part list

Part Number	Type	Title	Description
BBa_25FQHKWR	Composite	NdmDA	Coding; for N1-demethylation of caffeine
BBa_25RL0R3I	Composite	NdmDB	Coding; for N3-demethylation of caffeine
BBa_258M1ORK	Composite	NdmDCE	Coding; for N7-demethylation of caffeine
BBa_257CMZCB	Composite	T7-NdmDA	Device; for N1-demethylation of caffeine
BBa_25B8BK1G	Composite	T7-NdmDB	Device; for N3-demethylation of caffeine
BBa_25FPHJPT	Composite	T7-NdmDCE	Device; for N7-demethylation of caffeine

2 Composite parts

T7-NdmDA (BBa_257CMZCB)

Functional Description:

The NdmDA construct is a bicistronic genetic circuit composed of the ndmD and ndmA genes (Figure 1), co-expressed from the pET28a plasmid. Functionally, the reductase NdmD provides essential electrons, which enables the monooxygenase NdmA to perform the N1-specific demethylation of caffeine, converting it into theobromine as the first dedicated step in the caffeine degradation pathway.

Gene Circuit Design:

Figure 1. Gene circuit of T7-NdmDA.

Characterization:

The construct of T7-NdmDA was verified by PCR (Figure 2A), and the expression of NdmD and NdmA was verified by SDS-PAGE (Figure 2B) . In our test, 20.1% of caffeine was removed from fermentation samples processed by E.coli BL21(DE3) expressing NdmDA (Figure 2C).

Figure 2. Characterization of T7-NdmDA. (A) PCR using the transformed bacterial colonies to verify the correct T7-NdmDA sequence. (B) SDS-PAGE to verify induction of NdmDA expression in E.coli BL21(DE3). (C) Caffeine degradation by BL21(DE3) expressing NdmDA. Relative concentration (measured by OD₂₇₄) of residual caffeine after 24 h of fermentation in strains harboring the indicated constructs. The control strain harbors an empty vector (EV). Data are presented as mean ± SD (standard deviation) of three biological replicates (n=3).

T7-NdmDB (BBa_25B8BK1G)

Functional Description:

The NdmDB construct is a bicistronic genetic circuit composed of the ndmD and ndmB genes (Figure 3). In this system, the reductase NdmD supplies the necessary electrons to activate the monooxygenase NdmB, which subsequently catalyzes the N3-specific demethylation of caffeine, producing paraxanthine as the key reaction.

Gene Circuits Design:

Figure 3. Gene circuit of T7-NdmDB.

Characterization:

The construct of T7-NdmDB was verified by PCR (Figure 4A), and the expression of NdmD and NdmB was verified by SDS-PAGE (Figure 4B) . In our test, 21.7% of caffeine was removed from fermentation samples processed by E.coli BL21(DE3) expressing NdmDB (Figure 4C).

Figure 4. Characterization of T7-NdmDB. (A) PCR using the transformed bacterial colonies to verify the correct T7-NdmDB sequence. (B) SDS-PAGE to verify induction of NdmDB expression in E.coli BL21(DE3). (C) Caffeine degradation by BL21(DE3) expressing NdmDB. Relative concentration (measured by OD₂₇₄) of residual caffeine after 24 h of fermentation in strains harboring the indicated constructs. The control strain harbors an empty vector (EV). Data are presented as mean ± SD (standard deviation) of three biological replicates (n=3).

T7-NdmDCE (BBa_25FPHJPT)

Functional Description:

The NdmDCE construct is a tricistronic operon comprising the ndmD, ndmC, and ndmE genes. (Figure 5). Within this assembly, NdmD functions as a reductase to supply electrons, while NdmC and NdmE form an enzymatic complex where NdmE, an FMN-containing reductase, transfers electrons to NdmC, a molybdenum-containing monooxygenase, enabling it to specifically catalyze the N7-demethylation of paraxanthine or theobromine, yielding 7-methylxanthine.

Gene Circuits Design:

Figure 5. Gene circuit of T7-NdmDCE.

Characterization:

The construct of T7-NdmDCE was verified by PCR (Figure 6A), and the expression of NdmD, NdmC and NdmE was verified by SDS-PAGE (Figure 6B) . In our test, 19.3% of caffeine was removed from fermentation samples processed by E.coli BL21(DE3) expressing NdmDCE (Figure 6C).

Figure 6. Characterization of T7-NdmDCE. (A) PCR using the transformed bacterial colonies to verify the correct T7-NdmDCE sequence. (B) SDS-PAGE t o verify induction of NdmDCE expression in E.coli BL21(DE3). (C) Caffeine degradation by BL21(DE3) expressing NdmDCE. Relative concentration (measured by OD₂₇₄) of residual caffeine after 24 h of fermentation in strains harboring the indicated constructs. The control strain harbors an empty vector (EV). Data are presented as mean ± SD (standard deviation) of three biological replicates (n=3).

3 Part Collection

T7-NdmA-E (UUID: bc467fd5-1956-45ff-a373-12d29481fbf4)

Functional Description:

The phytotoxicity of caffeine is mitigated through its enzymatic degradation via a sequential demethylation pathway. This process, originally characterized in P.putida, is catalyzed by a series of specific enzymes: NdmA, NdmB, NdmC, NdmD, and NdmE. In our part collection, those above-mentioned composite parts (Figure 7), T7-NdmDA (BBa_257CMZCB), T7-NdmDB (BBa_25B8BK1G), and T7-NdmDCE (BBa_25FPHJPT) were co-expressed in E.coli to establish a biological system for decaffeinating SCG. Through their coordinated, stepwise action, the entire pathway progressively removes methyl groups from caffeine, ultimately yielding the non-toxic product, xanthine.

Design:

Figure 7. Gene circuits of three T7-Ndm devices in this part collection. Three distinct polycistronic devices were assembled: (A) ndmD-ndmA for N1-demethylation, (B) ndmD-ndmB for N3-demethylation, and (C) ndmD-ndmC-ndmE for N7-demethylation. Each gene is preceded by a ribosome binding site (RBS: TTTGTTTAACTTTAAGAAGGAGA) and flanked by 6-bp spacer sequences (AATAAT).

Characterization:

The co-expression of the three plasmids NdmDA, NdmDB, and NdmDCE (Figure 8A) offers significant advantages over both individual plasmid expression and a single NdmABCDE plasmid. Compared with an individual plasmid in BL21(DE3), NdmDA (20.1% caffeine removal), NdmDB (21.7% removal), or NdmDCE (19.3% removal) alone, which only performs a single, specific demethylation step (N1, N3, or N7, respectively), the three-plasmid system (43,4% removal) reconstitutes the complete caffeine degradation pathway (Figure 8B). This enables the sequential conversion of caffeine all the way to non-toxic xanthine, thereby fully eliminating its phytotoxicity, which is the ultimate goal of our biofertilizer production.

Figure 8. Characterization of T7-NdmA-E. (A) SDS-PAGE to verify induction of Ndm expression in E.coli BL21(DE3), especially the NdmA-E (NdmDA+DB+DCE). (B) Caffeine degradation by engineered BL21(DE3) strains. Relative concentration (measured by OD₂₇₄) of residual caffeine after 24 h of fermentation in strains harboring the indicated Ndm constructs. The control strain harbors an empty vector (EV). Data are presented as mean ± SD (standard deviation) of three biological replicates (n=3).

During our evaluation of E.coli DH5α as an alternative host strain, we found that despite undetectable expression levels on SDS-PAGE (Figure 9), the expression of NdmA-E in DH5α conferred a 60.9% decrease in caffeine (Figure 10), confirming significant functional demethylation activity.

Figure 9. SDS-PAGE to verify induction of Ndm expression in E.coli DH5α and BL21(DE3).

Figure 10. Comparative analysis of caffeine degradation between BL21(DE3) and DH5α. Relative concentration of residual caffeine after fermentation with BL21(DE3) or DH5α strains harboring the indicated Ndm constructs.

Application:

To apply our method in SCG detoxification, we probed for the best fermentation conditions. They are as follows (Figure 11),

Temperature: 28°C;

Duration: 48 h;

Best Carbon source: Starch 1 g/L;

Best Nitrogen source: NaNO3 2 g/L

Best ZnSO4 concentration: 0.1 g/L;

Best FeSO4 concentration: 0.1 g/L;

Best MgSO4 concentration: 0.5 g/L;

Figure 11. optimization of fermentation conditions. Relative concentration of residual caffeine in cultures of DH5α harboring the NdmA-E construct at the indicated temperatures (A) after indicated time of fermentation (B). the optimization of concentration of carbon source corn starch (C), nitrogen source NaNO3 (D), and inorganic salts, ZnSO4 (E) FeSO4 (F) and MgSO4 (G).

Next, we cultured the engineered DH5α in the optimal medium with sterile SCG under those optimized conditions. Surprisingly, DH5α expressing NdmA-E achieved approximately 90% caffeine degradation in sterile SCG (Figure 12), demonstrating its practical applicability.

Figure 12. Validation of caffeine degradation in SCG. Relative concentration of residual caffeine in sterile SCG after treatment with engineered DH5α (NdmA-E) under optimized fermentation conditions for 48 h. The control was treated with sterile medium only.

Discussion:

In contrast to a single plasmid expressing all five genes (NdmA-E) in one operon, the three-plasmid strategy mitigates the high metabolic burden and potential genetic instability associated with large, complex constructs. By distributing the genes across three separate vectors, the system reduces the risk of transcriptional attenuation, improper protein folding, and inclusion body formation in the BL21(DE3) strain. This modular approach often leads to more balanced expression and higher overall enzymatic activity, as evidenced by the superior caffeine degradation efficiency achieved in the E. coli DH5α strain with the three-plasmid system.

Our modular, three-plasmid expression system represents a strategic and efficient approach to reconstituting the multi-step caffeine degradation pathway in a heterologous host. By distributing the five essential ndm genes across three devices, T7-NdmDA, T7-NdmDB, and T7-NdmDCE, we successfully minimized the metabolic burden and genetic instability often associated with large, polycistronic operons. This design not only alleviates issues such as transcriptional attenuation and translational inefficiency but also reduces the risk of protein misfolding and inclusion body formation, which were observed in the BL21(DE3) strain under high-level induction.

Despite undetectable protein bands on SDS-PAGE, the functional expression of these enzymes in E.coli DH5α highlights that moderate “leaky” expression can be sufficient, and in some cases preferable, for achieving high enzymatic activity. This may be attributed to better protein folding, proper cofactor assembly, and reduced cellular stress under lower expression conditions. The resulting 60.9% caffeine degradation in DH5α and up to 90% in sterilized SCG under optimized fermentation conditions underscore the practical potential of this system for real-world application.

From an environmental and agricultural perspective, our system offers a sustainable and scalable alternative to conventional decaffeination methods, which often involve toxic solvents or energy-intensive processes. By enabling the safe reuse of spent coffee grounds as organic fertilizer, this approach supports circular economy principles, reduces organic waste, and promotes green agriculture.

Furthermore, the genetic parts we have developed and characterized, including the three composite devices, are fully documented and submitted to the iGEM Registry. These parts provide a modular toolkit for future iGEM teams or researchers working on metabolic pathways involving multi-enzyme complexes, particularly those requiring balanced expression and manageable genetic load.

In conclusion, our work not only demonstrates a functional biological solution to coffee waste valorization but also offers a genetically tractable and functionally validated system that aligns with both synthetic biology standards and sustainable development goals.