A randomized, covariate-controlled trial in 54 lactating Holstein dairy cows, followed by whole-animal respiration-chamber measurement — evaluating enteric methane emissions, milk production, energy utilization, and the rumen microbiome.
Suppressing methanogenesis is not the same as putting the spared hydrogen to work. This trial is designed to test both halves at once.
Enteric fermentation in ruminants is a major anthropogenic methane source, and it represents a direct energetic loss — roughly 2–12% of dietary gross energy leaves the animal as methane rather than becoming milk or muscle. Reducing it is therefore both an environmental and a productivity objective.
Several dietary mitigation strategies — 3-nitrooxypropanol (3-NOP), nitrate, lipids, tannins, essential oils, and Asparagopsis spp. — have shown varying degrees of success. But their efficacy depends heavily on diet, dose, delivery method, and animal-specific factors. More importantly, suppressing methanogenesis alone does not guarantee that the reducing equivalents spared from methane formation are redirected into productive fermentation. The hydrogen has to go somewhere. If it simply accumulates, fermentation itself is inhibited.
Propionate is the most attractive destination. It is the major glucogenic volatile fatty acid absorbed by ruminants, and the succinate and acrylate pathways that produce it consume reducing equivalents — whereas acetate formation releases them, making them available to methanogens again. Increasing propionate formation therefore does two things at once: it withdraws metabolic hydrogen from the methanogens and it increases the supply of glucogenic substrate to the animal.
Iron cycles readily between ferric [Fe(III)] and ferrous [Fe(II)] states, which allows Fe(III) to act as an alternative electron acceptor in anaerobic environments. Iron-reducing microorganisms compete with methanogens for hydrogen, and work in sediments, soils, wastewater, and anaerobic digestion has shown that ferric iron can suppress methanogenesis by altering microbial electron flow and redox conditions.
Iron-sulfur clusters extend that principle into biology. They are integral cofactors of ferredoxins, hydrogenases, respiratory enzymes, and numerous microbial oxidoreductases, where they function as highly efficient electron-transfer catalysts. That redox versatility is the foundation TM0415 is built on.
TM0415 is a stabilized amorphous iron-sulfur cluster (AISC) developed to modulate electron flow within the rumen. The proposed mode of action is that it behaves as a redox-active electron-transfer catalyst cycling between Fe(III) and Fe(II) — intercepting reducing equivalents that would otherwise feed hydrogenotrophic methanogenesis, and handing them to alternative hydrogen-utilizing pathways, principally propionate production.
Supplementation with TM0415 will decrease enteric methane production by redirecting reducing equivalents from methanogenesis toward propionate synthesis — without compromising rumen fermentation.
That final clause is the part that makes this a real test rather than a demonstration. A compound can lower methane and still be a failure if it depresses intake, disrupts fermentation, or costs the animal energy. The protocol is built to detect that.
The design below reflects the current protocol. Methodology is still being finalized with the research team, and specifics may be refined before enrollment opens.
The study runs in two sequential phases. Phase 1 is a production-scale trial at the Cornell Dairy Research and Education Center (CDREC). Phase 2 moves a subset of the same animals into whole-animal respiration chambers at the Large Animal Research and Teaching Unit (LARTU) for precision energy and nitrogen balance.
Fifty-four multiparous, mid-lactation Holstein cows will be enrolled in a randomized complete block design. Eligible cows must have no active clinical disease or postpartum health disorder within the previous 30 days and four functional teats.
Before treatment begins, animals pass through two preparatory periods: a 3-week acclimation to the Calan Broadbent individual-intake system and the GreenFeed gas-emission system, then a 3-week covariate period on a common control diet to establish per-animal baselines. Cows are then blocked by energy-corrected milk yield and parity before random assignment — so that treatment groups are matched on productivity before the intervention starts, and each cow serves as her own reference.
Basal total mixed ration with no AISC supplementation.
n = 18Control diet supplemented with TM0415 at a low fixed inclusion rate, without prior fermentation.
n = 18Control diet supplemented with TM0415 at the same rate, following prior fermentation.
n = 18The third arm is deliberate. It isolates whether pre-fermentation of the feed changes how TM0415 performs — a question that matters directly for how the product would be delivered on a working farm, and one that a simple treated-versus-untreated design could not answer.
Diets will be isonitrogenous, built primarily from corn silage, grass haylage, ground corn, soybean meal, and vitamin-mineral supplements, with forage at approximately 50% of dietary dry matter. Formulation uses the Cornell Net Carbohydrate and Protein System within AMTS Nutrition software to meet or exceed requirements for lactating cows — including metabolizable methionine, metabolizable lysine, a Lys:Met ratio near 2.6:1, and a ruminal unsaturated fatty acid load below 3.5%.
Critically, the diets will not contain monensin, yeast, or yeast-derived additives — removing the confounders that could otherwise be credited with a methane or performance effect. Cows are fed once daily as a TMR and milked three times daily, with water ad libitum and refusals held between 5 and 10% of feed offered.
The measurement plan spans four levels — the gas leaving the animal, the milk it produces, the fermentation inside the rumen, and the microbial community driving it.
| Domain | Measurements | Method / frequency |
|---|---|---|
| Enteric gas | CH₄, CO₂, H₂ emissions | Three GreenFeed units (C-Lock Inc.), continuous through the experimental period |
| Intake & body | Dry matter intake, body weight, body condition score, rumination activity | Daily; automated weighing/imaging system and SCR collars |
| Milk | Yield, fat, protein, lactose, somatic cell count; fatty acid profile if fat effects appear | Six consecutive milkings, twice weekly, at covariate and weeks 2, 4, 6, 8 |
| Digestibility | Apparent total-tract digestibility of DM, OM, starch, NDF, fat; urinary N excretion | uNDF240 internal marker; 3-day urine and fecal collection in a 36-cow subset |
| Rumen | VFA concentrations, propionate, acetate-to-propionate ratio, pH, protozoal dynamics | Rumen fluid collected ~5 h post-feeding |
| Blood | Plasma propionate, glucose, non-esterified fatty acids (NEFA) | Caudal vein, covariate period and week 8 |
| Microbiome | Community structure, functional potential, active gene expression, metabolic pathways | Integrated metagenomic + metatranscriptomic sequencing; pre-treatment, weeks 4 and 8 |
The microbiome component is what makes the mechanism testable rather than inferred. By pairing metagenomics with metatranscriptomics, the study can distinguish which organisms are present from which are active — and specifically track the abundance and activity of methanogenic archaea, iron-reducing microorganisms, and propionate-producing bacteria against each other.
GreenFeed measures emissions well, but it samples. To close the energy balance properly you need a sealed chamber, and that is what Phase 2 provides.
A subset of 18 cows — six per treatment — will be transported to the Large Animal Research and Teaching Unit in staggered enrollment. After a 3-day facility acclimation in tie stalls and a further 2-day acclimation inside the chambers, each animal undergoes 2 consecutive days of intensive sampling. Three cows, one from each treatment, are evaluated simultaneously. Gas exchange is recorded every 10 minutes, giving continuous CH₄, CO₂, O₂, and H₂ measurement across the full 48-hour window.
During the chamber period, total feces and urine are quantitatively collected. Representative samples of feces, urine, feed, and milk are analyzed for gross energy by bomb calorimetry and for carbon and nitrogen by C/N analyzer. From those inputs the study derives:
This is the measurement that answers the question the whole platform rests on: if methane goes down, where does the energy actually end up? Blood and rumen sampling mirror the Phase 1 panel, so the two phases can be read together.
Data will be analyzed using the MIXED procedure of SAS, with fixed effects of treatment, time, and their interaction, and cow as a random effect. Measurements from the pre-experimental covariate period enter the model as covariates. Significance is declared at P ≤ 0.05, with tendencies at 0.05 < P ≤ 0.10. Calculated variables include energy-corrected milk, component yields, feed efficiency, methane yield (g/kg DMI), and methane intensity (g/kg milk or ECM).
Reporting methane as both yield and intensity matters: a product that lowers absolute methane by suppressing intake would show up as a failure on intensity, which is exactly the discipline a productivity claim requires.
Animal health is monitored continuously and any clinical illness or veterinary intervention is documented. Cows requiring treatment for more than three consecutive days are removed from the experiment and excluded from the final statistical analyses. All procedures are conducted under Cornell University Institutional Animal Care and Use Committee (IACUC) guidelines.
TerraMetis is the study sponsor. Sponsors have the opportunity to review experimental diets and animal selection prior to study initiation. Conduct of the trial, measurement, and statistical analysis rest with the Cornell research team.
We will publish the outcome of this study whether or not it supports our hypothesis. A validation program is only worth running if the result is allowed to be negative.
Collaborator: Eric Boyd Denbow
If you're a B2B partner, climate investor, or technical reviewer, we can walk you through the protocol in detail and notify you when results are released.
Get in touch →