Whose Mitochondria Are Being Measured? MeScreen, Nanolive, and the Strongest Case for Blood-Based Testing

Edited by Samael A. Tejada, Founder & Editor-in-Chief

This is the second article in our mitochondrial-testing series. The first examined the Bioenergetic Health Index panels, the AONM/MMD menu, and the ProgeneDx GENIE transcriptomic test, and found none of them clinically validated. This article does three things: it makes the strongest honest case for why serious doctors use blood-based mitochondrial testing anyway; it takes MeScreen apart bolt by bolt; and it geeks out on the imaging technology that may eventually change how mitochondrial function is studied.

First, the steelman: why good doctors reach for blood-based testing

It is easy to grade the commercial tests harshly — the first article in this series did, on the evidence — and stop there. It is more useful to ask why thoughtful clinicians, including academic researchers with real reputations, keep using and developing these methods. The answer is that the underlying science is not fringe. It is unfinished, which is a different thing.

The core idea is sound. Mitochondria are not just power plants; they are sensors. Immune cells rewire their metabolism when they activate — effector T cells and inflammatory macrophages shift toward glycolysis, memory T cells and regulatory macrophages toward oxidative phosphorylation — a field called immunometabolism with a solid literature behind it (O'Neill and colleagues, Nature Reviews Immunology, 2016). If that is true, then the respiration of blood cells is a genuine readout of something real: the metabolic state of your immune system. White blood cells also circulate everywhere, surveying tissues in a way a muscle biopsy cannot.

The best surrogacy study is genuinely encouraging. In a 2019 JCI Insight study from the University of Pittsburgh (Braganza and colleagues), platelet respiration in older adults (32 participants, average age 88) correlated with the same people's muscle-biopsy respiration (r = 0.595) and with phosphorus magnetic resonance spectroscopy of muscle ATP production (r = 0.643) — the closest thing to a gold standard the field has. Proton leak in platelets also tracked walking speed. In nonhuman primates, blood-cell respiration has been associated with skeletal-muscle, cardiac-muscle, and even frontal-cortex mitochondrial respiration. Blood is not a perfect window into tissue. In the best studies, it is a real window.

The disease signals are consistent in direction, at group level. Platelets from people with diabetic kidney disease show reduced complex-I-supported respiration even before albumin appears in urine, worsening as kidney disease progresses (2025, Molecular and Cellular Biochemistry). In ME/CFS, a Newcastle group found every oxygen-consumption parameter lower in patients' blood cells than in controls, with maximal respiration the most consistent difference (Tomas and colleagues, PLoS ONE, 2017). In fibromyalgia, the Bioenergetic Health Index was 22.1% lower in 50 patients than in 20 controls, and tracked fatigue and pain scores (Macchi and colleagues, Scientific Reports, 2024). Naviaux's 2016 PNAS metabolomics study of ME/CFS (84 participants) found a hypometabolic signature that separated patients from controls with 94–96% accuracy — inside that sample.

And the practical logic is real. A muscle biopsy is painful, expensive, and unscalable; you cannot biopsy a patient every quarter to see whether treatment is working. A blood draw you can repeat. Clinicians caring for people with disabling fatigue — patients for whom standard medicine often has no objective measure at all — are not foolish for wanting a number they can trend in the same patient over time, provided everyone understands what the number is and is not.

So why did the first article still grade the commercial tests D? Because of the gap between group-level association and individual diagnosis, and three hard problems that sit inside the steelman itself:

1. The signals do not always point the same way. In sepsis, some studies find higher blood-cell respiration in non-survivors, others higher in survivors, others nothing. A measurement that can mean opposite things in the same condition is a research tool, not a clinical compass.
2. The measurement is fragile. Delay processing a blood sample, or freeze it, and respiration values shift by tens of percent — in one study, cryopreservation tripled non-mitochondrial respiration. A 2026 review calls the lack of standardised protocols — with identical samples varying two- to three-fold between laboratories — "the single most important barrier to clinical translation."
3. The decisive study has not been run. The Bioenergetic Health Index was proposed in 2014 explicitly as a future predictive assay. Twelve years on, the prospective study showing a blood score predicts who gets sick, or who responds to treatment, still has not been done. The suramin autism trial that cell-danger-response enthusiasts cite enrolled 10 boys, showed effects that faded as the drug cleared, and its industry follow-up missed its primary outcome.

That is the honest shape of the preliminary science: a plausible mechanism, encouraging correlations in the best hands, real disease associations at group level — and no validated individual test yet. Good doctors use these tools the way scientists use instruments still being calibrated: carefully, comparatively, and without pretending the calibration is finished. (Evidence: moderate for the associations; not established for individual diagnosis.)

MeScreen: the test that doesn't test your mitochondria

MeScreen markets itself as "the world's first and only at-home mitochondrial efficiency test." It costs $699, needs no prescription, and works from a finger-prick at home. Here is the part the marketing does not emphasise: at no point does anyone measure your mitochondria.

Follow the specimen. You prick your finger and dry drops of blood on a proprietary card that separates serum. The card is mailed to the lab. There, your serum is reconstituted and applied to a proprietary living reporter cell line — the company's own cells, grown in the lab. Those cells' mitochondria are then measured: oxygen consumption and acidification (respirometry), reactive oxygen species, and the structure of their mitochondrial network, imaged with fluorescent-protein tags, at baseline and under stress. Machine learning converts the readouts into 11 scores — the product page says 12, an inconsistency the vendor's own pages do not resolve — grouped into Energy Profile, Mito ROS, Mito Network, and Energy Balance (including Glycolytic, Aerobic, a Q-Score, and Energetic), plus an aggregate score called the mescore, a supplement and lifestyle plan, and a short consultation.

The company calls this "adoptive transfer": the idea is that your blood's metabolic environment — its hormones, metabolites, inflammatory signals — transfers onto the reporter cells, so their mitochondrial behaviour becomes a mirror of your internal state. As a concept, this is a legitimate bioassay design; drug screening works this way. The peer-reviewed ancestor of the method (Dzotsi and colleagues, Scientific Reports, 2025, with MeScreen's scientific advisor as a co-author) applied 1% patient plasma for one hour to C2C12 mouse muscle cells and measured the response in 20 patients with POTS. Most mitochondrial stress parameters did not change; one (the mitochondrial ATP production rate) rose slightly. That is the published foundation: a real paper, a clever idea, a mostly null result, in mouse cells.

For the commercial test itself, the validation cupboard is bare. The only peer-reviewed paper using MeScreen is a 30-day, single-arm pilot of a wearable light-therapy device (10 athletes, 10 people with chronic fatigue, funded by the device maker, with a company employee as an author). Its abstract calls the assay "validated" but cites no validation study — because none could be found: no analytical validation, no test–retest reliability, no reference intervals, no comparison against muscle biopsy or standard respirometry for MeScreen itself. The claim repeated in company materials that the system was "validated at UC San Diego against 60+ FDA-approved compounds" has no located publication behind it. The reporter cell line's identity, the serum dose and exposure time used commercially, and the scoring weights are not disclosed.

The people are real. The company is Verséa Health/Discovery of Tampa, Florida; its chief scientific advisor is Hemal Patel, a tenured professor and vice-chair at UC San Diego's Department of Anesthesiology and a VA research scientist, whose laboratory genuinely pioneered blood-cell mitochondrial profiling, including work connected to NASA's Twins Study. Note the exact shape of the "NASA-developed" claim: the science lineage runs through astronaut research Patel participated in; NASA did not develop or endorse this test. And the vendor's own disclaimer, on its own product page, states that MeScreen "is for research and informational use... and has not been reviewed by the U.S. Food and Drug Administration." Its CLIA laboratory status could not be verified from its public pages. Even friendly expert commentary carries the caveat: Theodoros Kelesidis of UT Southwestern calls such tests "promising tools" that "need validation in clinical studies," and David Perlmutter warns the results are "relative information, patterns and trends, not an absolute diagnosis."

The distance problem deserves emphasis, because it is the crux. A PBMC respirometry test measures your cells' mitochondria — arguably the wrong tissue, but at least yours. MeScreen measures a cell line's mitochondria responding to your serum. Every step in that chain — finger-prick drying, ambient shipping, serum reconstitution, dose, exposure time, the reporter cells' own state — is a place where your biology can be amplified, distorted, or lost, and none of those steps has published error bars. Interesting research instrument; unproven consumer diagnostic. (Evidence: vendor-only for the commercial test — grade D for clinical claims, with a genuinely novel bioassay concept underneath.)

Nanolive: the microscope that watches mitochondria live — and claims nothing about you

The third entry in this comparison is a category error on purpose. Nanolive's Smart Mitochondrial Assay LIVE is not a blood test, not a home kit, and not something a patient will ever take. It is software for a research microscope, sold to pharmaceutical companies and university laboratories. It earns its place here as the control case: a mitochondrial measurement technology whose marketing claims are exactly the size of its peer-reviewed evidence.

The technology: holotomography. Nanolive's 3D Cell Explorer splits a low-power 520 nm laser into an object beam, which passes through living cells on a rotating arm that sweeps the illumination through 360 degrees (100 angular snapshots), and a reference beam. The interference pattern — a hologram — encodes how much each structure in the cell slowed the light down: its refractive index. Software reconstructs a 3D refractive-index map at roughly 200 nm lateral and 400 nm axial resolution, across 96 optical slices spanning 30 µm of depth. Refractive index is proportional to dry-mass concentration (the non-water content of a structure), so the map is not just a picture: it is a quantitative mass measurement of every organelle, with no dyes, no fluorescent tags, and laser power low enough (about 20 mW/cm²) that cells can be filmed alive for days without phototoxic damage.

What it measures. An AI segmentation pipeline — an extra-trees pixel classifier for mitochondria in the published work, with U-Net networks for larger structures — outlines every mitochondrion and reports more than 30 metrics per mitochondrion and per cell: count, length, radius, dry mass, dry-mass density, branch number, endpoints, and junctions, from which it derives the dynamics biologists actually argue about: fission and fusion balance, swelling, fragmentation, network complexity, mitophagy and biogenesis, mitochondrial mass and content. Watch the classic positive control: add antimycin A, a complex III inhibitor, and within minutes the network swells into rings — morphology announcing a functional collapse in real time.

What it does not measure matters as much. No oxygen consumption. No ATP production rate. No membrane potential. No reactive oxygen species. Holotomography sees form and mass, and infers function from them. A fragmented network usually means stressed mitochondria; it does not tell you which complex failed or how much ATP the cell is making. Nanolive's own positioning respects that boundary: the assay is sold for drug discovery, mechanism-of-action studies, and investigative toxicology — "detect stress earlier, understand mechanism, compare compounds."

The evidence is published and specific. The method's foundations are in PLoS Biology (Sandoz and colleagues, 2019), where refractive-index signatures of mitochondria were validated against fluorescence imaging and used to track dry-mass dynamics through the cell cycle. The direct ancestor of the Smart Mitochondrial Assay is a 2024 Nature Communications paper from Institut Pasteur and Nanolive researchers, which used the platform to quantify how SARS-CoV-2 infection remodels mitochondria and other organelles across hundreds of cells. A 2026 study validating the imaging against MitoTracker dye through mitosis found one-to-one correspondence. When the company says the platform appears in "450+ publications," that figure is platform-wide, not assay-specific — a small inflation worth noting precisely because everything else checks out. Pricing is quote-only; there is no patient price because there are no patients.

Why include a research microscope in a consumer testing guide? Because Nanolive demonstrates the order of operations the consumer tests skipped: publish the method, validate it against the incumbent technique in the open literature, sell it for exactly what it was validated to do. (Evidence: strong for its stated research purpose — the only entry in this series whose claims and evidence match.)

The directness ladder: all of them, ranked by what is actually measured

Lay every approach in this series on one ladder, from the most direct measurement of a patient's mitochondrial function to the least:

1. Muscle or fibroblast functional testing (specialist centres): the patient's own tissue, enzyme assays and respirometry — most direct, invasive, validated for diagnosing mitochondrial disease alongside genetics.
2. Genetics (first-line for disease): measures the cause, not current function; finds a definite answer in roughly half to two-thirds of suspected cases.
3. PBMC/platelet respirometry and BHI-style scores (BioVis, AONM/MMD): the patient's own blood cells — direct for those cells, an imperfect and unproven proxy for muscle, brain, and heart, with real group-level science behind it.
4. MeScreen: the patient's serum acting on a laboratory cell line's mitochondria — a bioassay of your blood's metabolic environment, two steps removed from your mitochondria, with an unpublished error structure.
5. ProgeneDx GENIE ($720 plus shipping, per the first article): blood mRNA abundance for 188 genes — three steps removed from actual energy flux, research-use-only by its own disclaimer.
6. Nanolive Smart Mitochondrial Assay: off the ladder entirely — direct, quantitative imaging of cultured cells a researcher chooses to stage, with no patient in the loop at all.

Every approach on the ladder shares one elegant idea: stress reveals phenotype. Push a system — with an uncoupler in a respirometer, with serum on reporter cells, with a virus in an imaging dish, with antimycin A — and the reserve capacity and failure modes show themselves. The idea is good science. What separates the rungs is not the idea; it is whether anyone has published how wrong the measurement can be, in whom, and compared with what.

The bottom line

Doctors reach for blood-based mitochondrial testing for defensible reasons: the biology is plausible, the best studies show real correlations with tissue, the alternative is often no objective measure at all, and a repeatable blood draw beats a repeat biopsy. That is the preliminary science, and it deserves respect — as preliminary science. MeScreen wraps the most indirect consumer measurement in the series in the most confident marketing, at $699, with its own research-use disclaimer in the fine print. Nanolive, which no patient can buy, quietly shows the standard: method published, validation published, claims sized to fit.

So the questions from the first article stand, with one added for this round: Whose mitochondria are being measured — and how many steps sit between them and me? The fewer the steps, and the more published error bars on each one, the more a result deserves to steer your care.

Where the evidence stands

• Moderate: Blood-cell respirometry correlates with muscle respiration and ATP production in the best studies (r around 0.6 in older adults) and shows group-level disease signals — real research science, not yet a validated individual test.
• Vendor-only (grade D): MeScreen, a $699 at-home test, measures a laboratory cell line’s mitochondria responding to your serum; no analytical validation, test–retest data, or reference intervals for the commercial test could be found.
• Strong (for its stated purpose): Nanolive’s Smart Mitochondrial Assay is research-microscope software whose method and validation are published; it makes no claims about patients and is the one entry here whose claims match its evidence.