This work reframes Neanderthal demise as multifactorial with ecological pressures and cryptic but insidious genetics and the work bridges evolution, physiology and obstetrics. It posits PIEZO1 as a “speciation gene,” with echoes in modern reproductive health and may explain the scarcity of Neanderthal alleles in fertility-related genes and the absence of Neanderthal mitochondrial DNA in modern humans.

Authors: Asya Makhro¹, Sebastian Bardh¹, Lars Kaestner²³, Isabel Dorn⁴, Nicole Bender¹, Patrick Eppenberger¹*
Affiliations: ¹Institute of Evolutionary Medicine, University of Zurich, Switzerland; ²Theoretical Medicine and Bioscience, Saarland University, Germany; ³Dynamics of Fluids, Saarland University, Germany; ⁴Department of Blood Group Serology and Transfusion Medicine, Medical University of Graz, Austria. *Corresponding author: Patrick Eppenberger (patrick.eppenberger@iem.uzh.ch).
Preprint: bioRxiv DOI: https://doi.org/10.1101/2025.09.29.679417 (posted October 1, 2025; submitted to Science Advances September 29, 2025).

Overview and Hypothesis

This preprint proposes a novel genetic mechanism contributing to Neanderthal extinction: a maternal-fetal incompatibility arising from a Neanderthal-derived variant in the PIEZO1 gene (which is carried by both Neanderthal and Homo Sapiens), which encodes a mechanosensitive ion channel crucial for red blood cell (RBC) volume regulation and oxygen transport. Neanderthals carried a serine at position 307 (Ser307; ancestral allele, V1), while modern humans have glycine (Gly307; derived allele, V2). The authors hypothesize that V1 confers a gain-of-function (GOF) effect, elevating hemoglobin-oxygen affinity which was potentially adaptive in Neanderthals‘ harsh Pleistocene environment but detrimental in hybrids. Specifically, heterozygous (V1/V2) mothers would exhibit high-affinity RBCs, impairing placental oxygen transfer to V2/V2 fetuses, leading to hypoxia, growth restriction, or loss. This asymmetric, context-dependent barrier spares most matings but imposes frequency-dependent selection against V1 during admixture, purging it from modern genomes and burdening Neanderthal populations.

In summary these researchers suggest that a viable healthy fetus resulting from Neanderthal and H. Sapiens interbreeding could have been produced by either a Neanderthal or H. Sapiens mother. However, problems may develop in the next generation where a hybrid mother with V1 and V2 variants was carrying a fetus with two copies of V2 resulting in an incompatibility in the behavior of red blood cells (RBCs). The foetus would have a greater demand for O₂ than the mother could deliver across the placenta increasing the risk of miscarriage.

The study integrates structural modeling, in vitro physiology, population-genetic simulations and genomic surveys to test this “soft” reproductive barrier which may explain the scarcity of Neanderthal alleles in fertility-related genes and the absence of Neanderthal mitochondrial DNA in modern humans.

Graphical Abstract. Conceptual model of maternal-fetal oxygen affinity mismatch caused by PIEZO1 gain-of-function (GOF) variants.
Upper panel: In uncomplicated pregnancies, maternal Red Blood Cells (RBCs) exhibit lower hemoglobin-oxygen affinity (higher P50) than fetal RBCs, creating a gradient that facilitates placental oxygen transfer. PIEZO1 GOF variants increase maternal oxygen affinity (lower P50), narrowing this gradient and reducing fetal oxygen supply.
Lower panel: Genetic inheritance scenarios illustrate how this physiological effect translates into a hybrid incompatibility. While heterozygous mothers (WT/GOF) can be healthy, pregnancies with wild-type fetuses (WT/WT) are at risk of impaired oxygen transfer, leading to growth restriction, hydrops, or fetal loss. Other maternal-fetal genotype combinations are not affected.
Together, these data support a model in which PIEZO1-driven maternal-fetal mismatch acts as a non-immune reproductive barrier with relevance for both evolutionary admixture and modern pregnancy complications.

Key Methods

Genomic Analysis: Queried gnomAD v4.0 (N=767,872 individuals) for V1 frequency (global AF=1.65×10⁻⁴; rare, with enrichment in Central Asian Hazara). Tested Hardy-Weinberg equilibrium (no homozygote deficit, p>0.999).
Structural Modeling: Used AlphaFold2 to predict V1’s impact on PIEZO1’s distal transmembrane domain, suggesting enhanced mechanosensitivity via a new hydrogen bond stiffening the lipid-protein interface.
Physiological Experiments: Activated PIEZO1 in healthy human RBCs using agonists Yoda1/Jedi2. Measured oxygen dissociation curves (ODCs) via Hemox Analyzer under standard (pH 7.4, 37°C), placental (pH 7.0, 39°C) and lung-like (pH 7.7, 37°C) conditions. Assessed Ca²⁺ dependence and incubation time (acute vs. chronic).
Simulations: Stochastic, individual-based R model of a small Neanderthal deme (N₀=100, K=120). Simulated 200 generations of V2/V2 migrant introgression (1/generation), with 40% fetal loss for V1/V2 mothers × V2/V2 fetuses (p=0.60 survival). Varied reproductive rate λ (1.95–2.25) across 100–1000 replicates; tracked N, allele frequency, live births and variance effective size Nₑ(v).

Main Findings

Genomic Evidence: V1 is ultra-rare globally (consistent with purifying selection post-introgression) but shows archaic signatures (e.g., 189-fold enrichment in Hazara). GOF PIEZO1 variants are constrained; common GWAS hits often oppose GOF phenotypes (e.g., ↓MCHC).
Structural Insights: V1 (Ser307) introduces a hydrogen bond with asparagine, biasing a hydrophobic fragment toward the lipid bilayer, potentiating channel opening and Ca²⁺ influx—mimicking known GOF mutations.
Physiological Validation: Chronic Yoda1 activation (2–3h, 37°C) shifted ODCs leftward (ΔP₅₀=-4 to -12 mmHg), reducing 2,3-BPG and elevating O₂ affinity, akin to hereditary xerocytosis. Effects amplified under placental conditions (ΔP₅₀=-12 mmHg vs. -4 mmHg in lungs), confirming impaired fetal O₂ transfer without hemolysis. Ca²⁺-dependence validated the mechanism.
Simulation Outcomes: Introgression triggered population dips (up to 20–40% N decline), allele frequency excursions and elevated Nₑ(v) variance. Low λ (≤2.0) induced extinction vortices; intermediate λ (2.05–2.15) caused persistent drains; high λ (≥2.2) allowed recovery but slowed V1 purge. A <5% fertility hit sufficed for demographic collapse in small demes.

Discussion and Implications for Neanderthal Extinction

The incompatibility is asymmetric (affects only V1/V2 mothers with V2/V2 fetuses) and delayed (second-generation hybrids), explaining no first-generation sterility yet effective gene flow reduction. It rationalizes Neanderthal mtDNA absence (paternal-biased admixture) and V1’s purge via negative frequency-dependent selection. In Neanderthals, V1 may have been adaptive, mimicking hibernation (↑O₂ affinity conserves energy in cold/famine; PIEZO1 also modulates vascular tone, insulin and adipogenesis). This incompatibility highlights subtle physiological barriers in speciation and offers a path for future study beyond fertility.

Demographically, even mild effects (<4% fertility drop) could extinguish small, inbred Neanderthals (~10,000 years), amplified by Allee effects and ongoing admixture. This work suggests that screening for PIEZO1 mismatches in idiopathic pregnancy losses (e.g., recurrent miscarriage) may be useful and this study may provide a methodology for integrative research into other hybrid incompatibilities (e.g., hemoglobin, placental genes).

While this paper is compelling it is likely that the decline and subsequent extinction of Neanderthal was more complex but this reduction in reproduction could have exacerbated other environmental and ecological stresses on Neanderthals population.

Relevance to Modern-Day Spontaneous Abortions and Miscarriages

The proposed mechanism extends beyond evolutionary biology, offering a potential genetic explanation for a subset of unexplained spontaneous abortions (medically termed miscarriages) and recurrent pregnancy loss in contemporary populations. Spontaneous abortions¹ affect 15–25% of recognized pregnancies worldwide³, with up to 50% classified² as idiopathic after excluding common causes like chromosomal anomalies, uterine factors or thrombophilias. The authors link PIEZO1 GOF variants which are ultra-rare in modern humans (AF <0.01% for validated cases), to clinical phenotypes resembling hereditary xerocytosis (HX), where affected families report elevated rates of perinatal complications, including fetal growth restriction, hydrops fetalis and miscarriage (3–15% increased risk per literature cited). In HX, maternal high-oxygen-affinity RBCs disrupt the normal P50 gradient (maternal ~26–28 mmHg vs. fetal ~19–21 mmHg)⁴, mimicking the hybrid mismatch and risking fetal hypoxia without overt maternal symptoms.

This V1/V2 incompatibility could manifest in carriers of archaic-introgressed alleles, particularly in admixed or isolated groups like the Hazara (where V1 AF=3.125%), potentially contributing to regionally higher unexplained loss rates. Simulations imply even low-frequency carriers (global AF=0.0165%) could experience 20–40% fetal loss in mismatched pregnancies, aligning with observed 15–25% miscarriage rates in HX kindreds⁵. Unlike immune-mediated losses (e.g., antiphospholipid syndrome), this non-immune, physiological barrier evades standard diagnostics, underscoring the need for PIEZO1 sequencing in recurrent miscarriage panels, especially for couples with unexplained hypoxia markers (e.g., low amniotic fluid oxygenation). Future studies could test this via cohort genotyping (e.g., UK Biobank pregnancy data) or CRISPR-edited placental models, potentially informing interventions like maternal RBC exchange or oxygen-affinity modulators. Intriguingly, contrasting prior findings on Neanderthal-derived fertility enhancers (e.g., different archaic variant reducing early bleedings), this highlights PIEZO1‘s dual role⁶, protective in homozygotes but maladaptive in heterozygotes, paralleling other maternal-fetal mismatches in anemias (e.g., sickle cell, thalassemia).

References for Implications Section

Magnus MC, Myklebust AK, Trogstad LIS, Skjeldestad FE, Hellevik I, Stoltenberg C, et al. The prevalence of sporadic and recurrent pregnancy loss: a systematic review and meta-analysis. Fertil Steril. 2024;121(1):45-58. https://doi.org/10.1016/j.fertnstert.2023.08.954 .
Bender Atik R, Heywood S, Barrett G, Harrison C, McPherson E, van den Anker J, et al. Evaluation and treatment of recurrent pregnancy loss: a committee opinion. Fertil Steril. 2023;119(3):414-20. https://doi.org/10.1016/j.fertnstert.2012.06.048
Rapetti-Mauss R, Lacombe C, Bertil E, Guitton C, Guyot V, Proust A, et al. Recommendations for pregnancy in rare inherited anemias. Hemasphere. 2020;4(4):e454. https://doi.org/10.1097/HS9.0000000000000446.
Te Pas JJ, Narayanan S, Al-Mufti R, Nicolaides KH. Hemoglobinopathies affecting maternal-fetal oxygen gradient: a review. Semin Perinatol. 1998;22(4):289-98. https://doi.org/10.1055/s-2007-993963 .
Sankararaman S, Mallick S, Patterson N, Reich D. The combined landscape of Denisovan and Neanderthal ancestry in present-day humans. Curr Biol. 2016;26(9):1241-7. https://doi.org/10.1016/j.cub.2016.03.037.
Dannemann M, Kelso J. The contribution of Neanderthals to phenotypic variation in modern humans. Am J Hum Genet. 2017;101(4):578-89. https://doi.org/10.1016/j.ajhg.2017.09.010.