# Evidence behind selected mitochondrial microprotein reports

ProteinIQ literature audit, checked September 19, 2026. The accompanying `audit.csv` contains ten selected sequences from five discovery papers. We transcribed reported lengths, canonical gene regions, detection methods, origin controls, and limitations. These are our structured summaries of published experiments, not new experimental data or a revised mitochondrial gene annotation.

## Scope and interpretation

The question is whether reported additional peptides justify replacing the standard count of 13 mitochondrial protein-coding genes. We selected humanin, MOTS-c, SHLP1–6, SHMOOSE, and MTALTND4 because they illustrate overlapping RNA/protein coding regions and different kinds of experimental evidence. This is a bounded audit of their original reports, not an exhaustive catalogue or a systematic review of all subsequent replication studies. Gau, CYTB187AA, and other candidates are outside this selected set. Ten selected sequences must not be interpreted as ten universally accepted additional genes.

We recorded sequence length as reported, without predicting ORFs or recalculating translated sequences. Locations identify overlapping canonical regions rather than normalized nucleotide coordinates. A detection method is evidence reported in the selected paper, not an independent validation by ProteinIQ. No absence of evidence in an original report establishes that evidence is absent from all later literature.

Antibody signals, peptide identification by mass spectrometry, and biological effects of an added synthetic peptide answer different questions. Cells depleted of mtDNA provide an origin control, but depletion can also indirectly alter nuclear expression. Nuclear mitochondrial DNA segments and related nuclear genes can complicate source assignment. We therefore report the actual controls rather than assign numerical confidence scores.

## Sources and reproduction

Read the indicated Results sections and figure/table locators alongside each CSV row. All article URLs below were accessed on the audit date.

1. Hashimoto Y, Niikura T, Tajima H, et al. (2001). *A rescue factor abolishing neuronal cell death by a wide spectrum of familial Alzheimer's disease genes and Abeta.* PNAS. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC33469/), DOI: 10.1073/pnas.101133498. The selected original experiment establishes activity of the 24-residue sequence. The MT-RNR2 overlap and nuclear-homologue caveat are also documented in Cobb et al.'s Results.
2. Lee C, Zeng J, Drew BG, et al. (2015). *The mitochondrial-derived peptide MOTS-c promotes metabolic homeostasis and reduces obesity and insulin resistance.* Cell Metabolism. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC4350682/), DOI: 10.1016/j.cmet.2015.02.009. Results and Figure 1 distinguish mitochondrial DNA origin from the standard code needed for the reported peptide.
3. Cobb LJ, Lee C, Xiao J, et al. (2016). *Naturally occurring mitochondrial-derived peptides are age-dependent regulators of apoptosis, insulin sensitivity, and inflammatory markers.* Aging. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC4925829/), DOI: 10.18632/aging.100943. Lengths were read from Figure 1A, including SHLP1 = 24 and SHLP5 = 24. Protein and origin evidence comes from the Results accompanying Figure 1. SHLP5 lacked a specific antibody; SHLP2/3 transcript amplification did not distinguish mitochondrial from nuclear origins. The article describes the supplementary transcript experiments; we did not reanalyse their raw data.
4. Miller B, Kim SJ, Mehta HH, et al. (online 2022; issue 2023). *Mitochondrial DNA variation in Alzheimer's disease reveals a unique microprotein called SHMOOSE.* Molecular Psychiatry. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC10027624/), DOI: 10.1038/s41380-022-01769-3. Figure 2 reports endogenous detection. The Discussion leaves the translation compartment open and describes the antibody epitope as residues 32–58. Kienzle et al.'s Introduction specifies the overlap with serine tRNA, leucine tRNA, and ND5; the 58-residue length is also tabulated in [Mitochondrial Microproteins from Discovery to Function](https://pmc.ncbi.nlm.nih.gov/articles/PMC11794013/). We checked the [2023 correction](https://doi.org/10.1038/s41380-023-01956-w): it corrects attribution and the D47N amino-acid description. We make no disease-risk inference from the association experiments.
5. Kienzle L, Bettinazzi S, Choquette T, et al. (2023). *A small protein coded within the mitochondrial canonical gene nd4 regulates mitochondrial bioenergetics.* BMC Biology. [Full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC10193809/), DOI: 10.1186/s12915-023-01609-y. Table 1 and Results describe the 99-residue candidate, origin controls, and immunoprecipitation/MS evidence. The length uses the authors' stop-codon assumption, and is not a new ProteinIQ translation result.

## Tissue-data update in the accompanying article

The separate copy-number section uses published results from Rath SP, Gupta R, Todres E, et al. (2024), [Mitochondrial genome copy number variation across tissues in mice and humans](https://pmc.ncbi.nlm.nih.gov/articles/PMC11331085/), DOI: 10.1073/pnas.2402291121. We did not recompute its tissue statistics or combine its observations with Frahm et al. (2005).

Rath et al. report 52 tissues, 952 donors, approximately 50-fold variation in tissue medians, and more than 200-fold interindividual variation within a tissue. The denominator is a diploid nuclear genome equivalent. The abstract says 10,499 samples; Results and Figure 1A say 10,449 donor-tissue pairs. We retain this unresolved discrepancy here and omit the exact sample total from the article rather than silently choose one. It does not affect the agreement on the tissue and donor counts.