What an ACE Protein Test Actually Tells You (and What It Doesn't)

The question every grower eventually asks

If you have ever pulled a 0–8 inch nitrate sample in April and gotten a result you did not trust, you already understand the limit of inorganic nitrogen testing. Mineral N moves, and the form determines the loss pathway: nitrate leaches with spring rain and gets denitrified in saturated zones, ammonia volatilizes after surface urea applications, and ammonium and nitrate alike get immobilized whenever a high-carbon residue lands on the soil surface. A potassium chloride (KCl) extraction tells you how much nitrate (and, depending on the lab, ammonium) is sitting in the soil at the moment you pulled the core, not how much your soil can supply across a growing season.

That gap between what is currently mineral and what is potentially mineralizable has been the central problem in agronomic nitrogen management for over a century. The autoclaved citrate extractable (ACE) protein test, included in the USDA-NRCS CEMA 216 soil health testing program as the bioavailable nitrogen indicator (USDA-NRCS, 2024), is one of the more promising attempts to fill that gap with a routine, commercially scalable measurement.

Here, we'll cover what the test is, what it actually measures, where it earns its place in a soil health package, and where the science still asks us to be careful about interpretation.

What the test is, and where it came from

The protocol itself is straightforward. A small mass of air-dried, sieved soil (typically 1.0 g) is added to 8 mL of 0.02 mol/L sodium citrate buffer at pH 7. The slurry is autoclaved at 121°C for 30 minutes, cooled, centrifuged, and the supernatant is assayed for protein content (Hurisso et al., 2018). Hurisso and colleagues used the Bradford dye-binding assay in their original validation work; the routine method adopted by the Cornell Soil Health Laboratory and most contemporary commercial labs uses the bicinchoninic acid (BCA) reaction, calibrated against bovine serum albumin (Schindelbeck et al., 2016; Moebius-Clune et al., 2016). The two assays return comparable but not identical values. The result is reported either as milligrams of protein per gram of soil or as grams of protein per kilogram of soil.

The method has an interesting backstory. Wright and Upadhyaya (1996) developed it in the mid-1990s to measure “glomalin,” a glycoprotein they believed was produced primarily by arbuscular mycorrhizal fungi. For roughly two decades, the soil health community treated the result as a fungal biomarker. Hurisso and colleagues (2018) then demonstrated through controlled spike experiments that the same extraction recovered proteins from a wide range of sources, including bacterial proteins, plant residues, and a non-trivial amount of co-extracted humic material, and that the “glomalin” framing was no longer defensible. They proposed renaming the result more honestly as autoclaved citrate extractable soil protein, or ACE protein.

So what is it, really?

The renaming to ACE protein settled the question of whether “glomalin” was a meaningful term, but it did not settle the question of what the original method actually detects. The 1996 glomalin work relied on a monoclonal antibody, MAb32B11, raised against crushed mycorrhizal spores; that antibody was the basis for the immunoassay used to identify and quantify glomalin in soil for nearly three decades. In 2025, Alptekin and colleagues at the University of Wisconsin–Madison reopened the question of what MAb32B11 binds. Using protease digestion, periodate oxidation, size exclusion chromatography, and NMR, they showed that the antibody's target is not a protein at all but a complex polysaccharide in the 511 to 600 kDa range, most likely a cell wall component of arbuscular mycorrhizal fungi. They proposed the name “glomalose” to reflect its carbohydrate nature (Alptekin et al., 2025).

The implication for routine soil testing is narrower than it might first appear. The ACE protein test as run by commercial labs is a colorimetric protein assay on a citrate-autoclave extract, and it does measure proteinaceous compounds along with well-documented humic interferences. What the Alptekin paper changes is the historical claim that the immuno-detected fraction was itself an abundant soil protein of mycorrhizal origin. The ACE protein test was already understood to extract a heterogeneous mixture of compounds, most of them non-mycorrhizal (Hurisso et al., 2018; Gillespie et al., 2011); the new work simply closes the door on the original glomalin-as-protein interpretation.

That history matters. The current value of the test does not depend on a fungal biomarker story. It depends on the broader claim that the citrate-autoclave extraction recovers a meaningful slice of the organically bound nitrogen pool, and that this slice responds to management in ways an inorganic snapshot cannot.

What the number actually represents

The vast majority of nitrogen in surface agricultural soils is held in organic forms. Stevenson (1994) places amino-acid-N and amino-sugar-N together at roughly 30–50 percent of total soil N, with the remainder partitioned among other nitrogen-containing organic compounds and a small inorganic pool that typically falls below 5 percent of the total. ACE protein is a measurement targeted at the proteinaceous fraction of that organic pool, expressed as a single number per gram of soil.

Three points deserve emphasis here.

First, the pool ACE protein measures is large in absolute terms. In the Geisseler et al. (2019) California dataset, ACE protein concentrations ranged from 1.0 to 45.2 g protein per kg of soil across 57 fields representing eight Soil Taxonomy orders. On a nitrogen basis, assuming proteins are roughly 16 percent N, ACE protein-N accounted for an average of 28 percent of total soil N across all sites in that study, but exceeded 60 percent in some high-organic-matter soils. Geisseler and coworkers attributed the implausibly high values to co-extracted humic substances interfering with the protein assay.

Second, the pool is sensitive to management. The North American Project to Evaluate Soil Health Measurements (NAPESHM), a coordinated assessment of more than 30 soil measurements at 124 long-term agricultural research sites across Canada, the United States, and Mexico (Norris et al., 2020; Bagnall et al., 2023), evaluated ACE protein among five candidate nitrogen indicators. Liptzin and colleagues (2023) reported that nitrogen indicators in the dataset increased by between 6 and 39 percent in response to decreased tillage, cover cropping, residue retention, and organic nutrient amendments. Total N and ACE protein were highlighted by Bagnall et al. (2023) as the two N-cycle indicators that responded significantly to all four of those soil health management practices. Long-term manure addition has been shown in the Knorr-Holden continuous maize experiment, the longest-running continuous corn study in North America, to significantly increase ACE protein along with several other soil health indicators (Das et al., 2023).

Third, ACE protein is conceptually a supply-side measurement. It estimates the size of a substrate reservoir that microbes can draw on for mineralization, not the rate at which mineralization is happening on a given afternoon. This is the snapshot-versus-supply distinction that opened this article.

How well does ACE protein predict nitrogen mineralization?

This is the question that decides how much weight an ACE protein number should carry in a fertilizer decision, and the honest answer is that the science is mixed.

The 2018 introduction by Hurisso and colleagues set high expectations: a useful relationship between ACE protein and organic-N substrate availability across regional datasets. Subsequent independent studies have produced a wider range of results. Geisseler and coworkers (2019), sampling 57 fields across eight soil orders in California, found that ACE protein explained only 21 percent of the variation in potential net N mineralization, performing slightly worse than total soil N alone, with much of the noise attributed to co-extracted humic substances interfering with the BCA assay.

Across a broader set of comparisons summarized in the recent literature, the R² values between ACE protein and laboratory-measured N mineralization range from approximately 0.21 to 0.76 depending on the study, the soil-climate region, and the protein quantification method (Geisseler et al., 2019; Naasko et al., 2024). The Naasko et al. (2024) Michigan dataset, drawn from a 33-year-old eight-system management intensity gradient at the Kellogg Biological Station, demonstrated that ACE protein tracked seasonal N dynamics across a wide range of cropping systems but also that the absolute values varied with sampling time and soil moisture history.

How to read your number, and what to do with it

Where ACE protein shines is as part of an interpretation framework rather than as a stand-alone diagnostic. A few practical guidelines hold up across the literature.

When ACE protein is low for your region and texture class, the soil is functioning with a small organic N reservoir. Mineralization will be modest, fertilizer N will face higher loss risk because there is less organic buffering capacity, and rotational or cover-crop investments are likely to produce the largest absolute gains in the indicator over a 3 to 5 year window. This is the classic situation where biological N supply is constrained.

When ACE protein is moderate to high for your region and texture, the soil holds a meaningful organic N pool. Pre-plant fertilizer N strategies can credibly account for some endogenous mineralization, and post-harvest residual N risk should be evaluated alongside any rate-bumping decision. A reasonable interpretation in this range is that nitrogen availability is no longer the primary biological constraint on yield.

When ACE protein is very high, especially in soils with substantial organic matter or recent manure history, the value should be interpreted with caution. Co-extracted humic substances inflate the colorimetric reading, and the apparent ACE protein value may overstate the truly mineralizable pool (Geisseler et al., 2019). In manured systems, pairing ACE protein with potentially mineralizable N (PMN), respiration, or POXC tightens the picture considerably.

Where ACE protein fits in the broader soil health conversation

The 2023 Soil Health Institute synthesis from Bagnall and colleagues drew on the full NAPESHM dataset of 124 long-term agricultural research sites to recommend a three-indicator minimum suite for routine soil health assessment: soil organic carbon, aggregate stability via slaking image recognition, and 24-hour carbon mineralization potential (Bagnall et al., 2023). ACE protein did not make that minimum cut.

That is worth understanding rather than glossing over. The omission was not because ACE protein is uninformative. Both total N and ACE protein were flagged in that synthesis as highly responsive N-cycle indicators. The minimum-suite framework grouped indicators by ecosystem service and selected a single indicator per group. Because total N and SOC were strongly correlated (r² ≈ 0.92 in the NAPESHM dataset), and ACE protein was in turn strongly correlated with total N, the authors concluded that organic carbon could serve as a reasonable surrogate for the nutrient-cycling ecosystem service in a continental-scale minimum suite. CEMA 216, by contrast, was designed not as a continental minimum suite but as a fuller diagnostic panel, and it explicitly retains ACE protein as the dedicated bioavailable nitrogen indicator alongside SOC, aggregate stability, respiration, and POXC (USDA-NRCS, 2024).

Two methodological developments are worth watching. The first is the Soil Health Assessment Protocol and Evaluation (SHAPE) scoring system being deployed by NRCS, which compares raw ACE protein values to distributions from soils of similar texture and climate, returning a 0 to 100 score rather than an absolute number (USDA-NRCS, 2022). This is the same logic Saurav Das has championed in the Cropland Reference Ecological Unit framework for soil health targets, and it is the right direction for a measurement whose absolute values are deeply context-dependent. The second is the development of pedotransfer-function approaches that predict ACE protein from cheaper and more reproducible measurements such as total N, texture, and SOC (Naasko et al., 2024). These methods may eventually make a “predicted ACE protein” available alongside the laboratory-measured value for routine packages, the same way the Cornell Soil Health Laboratory already reports a predicted protein index in its standard CASH package (Cornell Soil Health Laboratory, 2024).

The ACE protein test is not a perfect oracle for nitrogen management, and treating it as one will produce frustration. What it is, and what it does well, is provide a routinely measurable estimate of the size of the soil's organically bound nitrogen reservoir. That reservoir governs the long-term capacity of a field to supply nitrogen biologically, to buffer against fertilizer mistakes, and to recover from disturbance. Read in regional context, paired with companion indicators, and tracked across years of management change, ACE protein answers a question that nitrate and ammonium tests structurally cannot answer: how much nitrogen does your soil have to give?

References

1. Alptekin, B., Hirsch, H., Kleven, B., King, L., McLimans, C., Daniels, D., Irving, T., Floss, D., & Ané, J.-M. (2025). From glomalin to glomalose: unraveling the molecular identity of the MAb32B11 antigen. New Phytologist, 247(5), 2328–2341. https://doi.org/10.1111/nph.70253
2. Bagnall, D. K., Rieke, E. L., Morgan, C. L. S., Liptzin, D. L., Cappellazzi, S. B., & Honeycutt, C. W. (2023). A minimum suite of soil health indicators for North American agriculture. Soil Security, 10, 100084. https://doi.org/10.1016/j.soisec.2023.100084
3. Das, S., Liptzin, D., & Maharjan, B. (2023). Long-term manure application improves soil health and stabilizes carbon in a continuous maize production system. Geoderma, 430, 116338. https://doi.org/10.1016/j.geoderma.2023.116338
4. Geisseler, D., Miller, K., Leinfelder-Miles, M., & Wilson, R. (2019). Use of soil protein pools as indicators of soil nitrogen mineralization potential. Soil Science Society of America Journal, 83(4), 1236–1243. https://doi.org/10.2136/sssaj2019.01.0012
5. Gillespie, A. W., Farrell, R. E., Walley, F. L., Ross, A. R. S., Leinweber, P., Eckhardt, K.-U., Regier, T. Z., & Blyth, R. I. R. (2011). Glomalin-related soil protein contains non-mycorrhizal-related heat-stable proteins, lipids and humic materials. Soil Biology & Biochemistry, 43(4), 766–777. https://doi.org/10.1016/j.soilbio.2010.12.010
6. Hurisso, T. T., Moebius-Clune, D. J., Culman, S. W., Moebius-Clune, B. N., Thies, J. E., & van Es, H. M. (2018). Soil protein as a rapid soil health indicator of potentially available organic nitrogen. Agricultural & Environmental Letters, 3(1), 180006. https://doi.org/10.2134/ael2018.02.0006
7. Hurisso, T. T., & Culman, S. W. (2021). Is autoclaved citrate-extractable (ACE) protein a viable indicator of soil nitrogen availability? In D. L. Karlen, D. E. Stott, & M. M. Mikha (Eds.), Soil Health Series: Volume 2 — Laboratory Methods for Soil Health Analysis. ASA, CSSA, SSSA. https://doi.org/10.1002/9780891189831.ch10
8. Liptzin, D., Norris, C. E., Cappellazzi, S. B., Bean, G. M., Cope, M., Greub, K. L. H., Rieke, E. L., Tracy, P. W., et al. (2023). An evaluation of nitrogen indicators for soil health in long-term agricultural experiments. Soil Science Society of America Journal, 87(4), 868–884. https://doi.org/10.1002/saj2.20558
9. Moebius-Clune, B. N., Moebius-Clune, D. J., Gugino, B. K., Idowu, O. J., Schindelbeck, R. R., Ristow, A. J., van Es, H. M., Thies, J. E., Shayler, H. A., McBride, M. B., Kurtz, K. S. M., Wolfe, D. W., & Abawi, G. S. (2016). Comprehensive Assessment of Soil Health: The Cornell Framework, Edition 3.2. Cornell University, Geneva, NY.
10. Naasko, K., Martin, T., Mammana, C., Murray, J., Mann, M., & Sprunger, C. (2024). Soil protein: A key indicator of soil health and nitrogen management. Soil Science Society of America Journal, 88(1), 89–108. https://doi.org/10.1002/saj2.20600
11. Norris, C. E., Bean, G. M., Cappellazzi, S. B., Cope, M., Greub, K. L. H., Liptzin, D., Rieke, E. L., Tracy, P. W., Morgan, C. L. S., & Honeycutt, C. W. (2020). Introducing the North American Project to Evaluate Soil Health Measurements. Agronomy Journal, 112(4), 3195–3215. https://doi.org/10.1002/agj2.20234
12. Schindelbeck, R. R., Moebius-Clune, B. N., Moebius-Clune, D. J., Kurtz, K. S., & van Es, H. M. (2016). Cornell Soil Health Laboratory: Comprehensive Assessment of Soil Health Standard Operating Procedures. Cornell University, Ithaca, NY.
13. Stevenson, F. J. (1994). Humus Chemistry: Genesis, Composition, Reactions (2nd ed.). John Wiley & Sons, New York.
14. USDA-NRCS. (2022). Conservation Evaluation and Monitoring Activity 216: Soil Health Testing (FY22 technical document). United States Department of Agriculture, Natural Resources Conservation Service. https://www.nrcs.usda.gov
15. USDA-NRCS. (2024). FY25 CEMA 216 Soil Health Testing technical document. United States Department of Agriculture, Natural Resources Conservation Service.
16. Wright, S. F., & Upadhyaya, A. (1996). Extraction of an abundant and unusual protein from soil and comparison with hyphal protein of arbuscular mycorrhizal fungi. Soil Science, 161(9), 575–586. https://doi.org/10.1097/00010694-199609000-00003