Math-Corrected Enzyme Ratios: Descriptors of Allocation and Tools for Comparability—Not Proof of Limitation

Abstract

I thank Mori (2025) for the thoughtful and constructive comment on my recent article (Puissant 2025). The article showed that ratios of log-transformed enzyme activities are unit-dependent and tend to converge toward 1, falsely suggesting a global C:N:P 1:1:1 pattern—a signal previously reported and widely interpreted as biological (Sinsabaugh et al. 2008, 2009), but arising here from mathematics rather than biology. The remedy is to compute direct ratios from raw activities (or vector metrics from raw shares), or to apply logs only after forming the ratio. This removes the mathematical artifact. By itself, however, it neither validates any threshold nor ensures that enzyme stoichiometry diagnoses nutrient limitation. On this broader point, I largely agree with Mori (2025).

The broader debate over the usefulness of enzyme stoichiometry ratios is valuable. Several studies have raised critiques of inferring nutrient limitation from these ratios (e.g., Rosinger et al. 2019; Mori et al. 2023), while others report support under specific designs—particularly when ratios are interpreted temporally (weeks-scale integration) and corroborated with independent evidence (Moorhead et al. 2023; Kunito et al. 2024). As noted by Mori (2025), ecoenzymatic theory still lacks comprehensive validation, and multiple arguments (e.g., terminal-step control, multiple enzymes per nutrient, lack of uniform responses to nutrient additions) urge caution.

Why enzyme ratios still matter—and why fixing the math is important?

Measured activities are highly sensitive to assay conditions—substrate identity and concentration, pH, ionic strength, temperature, dilution, incubation time—yielding substantial between-lab and protocol variance for the same nominal enzyme (Nannipieri et al. 2018; Greenfield et al. 2021). When multiple enzymes are assayed under the same conditions within a study, many multiplicative scale factors act jointly; expressing activity as ratios (proportions)—i.e., ecoenzymatic ratios—helps cancel shared effects and improves within-study comparability, analogous to housekeeping-gene normalization in omics or intensity scaling in spectroscopy. Very extreme, distribution-incoherent ratios can also flag potential assay inconsistencies.

Ecoenzymatic ratios index extracellular enzyme production as an investment in acquiring specific substrates, which can decouple from instantaneous nutrient limitation under stress or constraint. Fundamental knowledge of soil enzyme regulation remains limited: alkaline phosphatases are classically induced under phosphate deprivation, whereas β-glucosidase and leucine aminopeptidase often show substrate-induced expression superimposed on variable constitutive baselines, with carbon- and nitrogen-catabolite effects that differ among taxa and contexts. Clarifying how microbial community composition, climate (temperature, moisture), and substrate/nutrient availability interact to govern enzyme production is therefore a priority. Ecoenzymatic ratios can aid understanding of enzyme production when calculated correctly and linked to gene expression and product turnover (e.g., CAZyme and phosphatase transcripts, product fluxes).

Enzyme stoichiometry ratios should not be used as stand-alone tests of microbial nutrient limitation, and correcting the log-ratio artefact should not be taken as conceptual validation for such use. Nevertheless, fixing the calculation is necessary: when computed appropriately, ratios remain useful for improving comparability across assays, probing the drivers of microbial enzyme allocation, advancing fundamental enzymology, and ultimately informing microbial trait-based and biogeochemical models.

The author declares no conflicts of interest.

This article is a Response to a Letter to the Editor by Taiki Mori https://doi.org/10.1111/gcb.70519 regarding Jérémy Puissant https://doi.org/10.1111/gcb.70228.