No public transcriptome tests dehydroepiandrosterone at a physiological concentration: a systematic re-analysis of GEO

Abstract Background. Dehydroepiandrosterone (DHEA) is the most abundant circulating steroid in humans and the only major adrenal steroid whose level falls steeply and monotonically with age, from about 23 nM serum DHEA at age 20–30 to about 4.5 nM at age 70–80 in men. A receptor able to transduce that decline would need a half-maximal effective concentration inside roughly 5–50 nM. No such receptor has been established. Methods. All Gene Expression Omnibus series returned by a DHEA query were retrieved (n = 43), their treatment metadata curated from series-matrix headers, and the stated DHEA concentration compared with the physiological serum-DHEA window. The datasets with replication were then re-analysed: GSE3627 (LNCaP, 300 nM, 48 h), GSE109021 (LAPC4, 20 µM, 28 compounds across 3 days), GSE290297 (mouse Th17, 50 µM, n = 5 per stratum), with GSE56240, GSE56018, GSE103844 and the GSE56019 RORγt ChIP-seq used as reference sets. Results. Of the 43 series, 28 administer DHEA, 12 deposit a dose or concentration, and 5 state an in vitro concentration. None treats cells inside the physiological window. The lowest concentration in the public record is 300 nM (GSE3627), 13-fold above the young-adult serum level and 67-fold above the level at age 70–80; the median is 10 µM, 435-fold above the young-adult level. In androgen-receptor-positive cells the DHEA response is not separable from the androgen receptor response: in GSE3627 the variance of the DHEA-minus-DHT difference is −0.00102 (95% CI −0.00209 to +0.00007) against a DHT-response variance of +0.04445, and in GSE109021 DHEA behaves as a partial androgen-receptor antagonist (α = 0.631 ± 0.105; 62.1% of its response variance explained by the androgen-receptor axis) with no residual component above a potency-matched band (0.305 against 0.431 ± 0.126, rank 4 of 13, p = 0.769). In Th17 cells, where Ar transcript is effectively absent (0.134 TPM), 50 µM DHEA still moves 2230–4351 genes per stratum and suppresses the IL-17 effector module; the response projects onto a mouse Rorc -knockout signature (r = +0.422 whole signature; +0.721 restricted to the six effector genes; +0.189 for the remaining 54, p = 0.004) where an unrelated active compound, EGCG, does not (+0.028, +0.417, −0.161). Against RORγt ChIP-seq peaks, however, the DHEA effect does not reach direct targets: in an expression-matched test the Rorc knockout is displaced below its matched control (observed −0.0087 against +0.0085, p = 0.014) while DHEA is displaced in the opposite direction, indistinguishably from solvent. Conclusion. Every DHEA transcriptional effect in the public record was measured at 13- to 2174-fold above the physiological serum DHEA concentration. The concentration window in which the age-related decline could be transduced has never been assayed. The failure of these datasets to identify a DHEA receptor is therefore uninformative about its existence, and the screen specifies the experiment that would be informative.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-16
DOI
https://doi.org/10.21203/rs.3.rs-11725323/v1
Primary Topic
Estrogen and related hormone effects
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

No public transcriptome tests dehydroepiandrosterone at a physiological concentration: a systematic re-analysis of GEO

Kenji Ohe
Research Square
Estrogen and related hormone effects
preprint

No public transcriptome tests dehydroepiandrosterone at a physiological concentration: a systematic re-analysis of GEO

Kenji Ohe
preprint en

Abstract

Abstract Background. Dehydroepiandrosterone (DHEA) is the most abundant circulating steroid in humans and the only major adrenal steroid whose level falls steeply and monotonically with age, from about 23 nM serum DHEA at age 20–30 to about 4.5 nM at age 70–80 in men. A receptor able to transduce that decline would need a half-maximal effective concentration inside roughly 5–50 nM. No such receptor has been established. Methods. All Gene Expression Omnibus series returned by a DHEA query were retrieved (n = 43), their treatment metadata curated from series-matrix headers, and the stated DHEA concentration compared with the physiological serum-DHEA window. The datasets with replication were then re-analysed: GSE3627 (LNCaP, 300 nM, 48 h), GSE109021 (LAPC4, 20 µM, 28 compounds across 3 days), GSE290297 (mouse Th17, 50 µM, n = 5 per stratum), with GSE56240, GSE56018, GSE103844 and the GSE56019 RORγt ChIP-seq used as reference sets. Results. Of the 43 series, 28 administer DHEA, 12 deposit a dose or concentration, and 5 state an in vitro concentration. None treats cells inside the physiological window. The lowest concentration in the public record is 300 nM (GSE3627), 13-fold above the young-adult serum level and 67-fold above the level at age 70–80; the median is 10 µM, 435-fold above the young-adult level. In androgen-receptor-positive cells the DHEA response is not separable from the androgen receptor response: in GSE3627 the variance of the DHEA-minus-DHT difference is −0.00102 (95% CI −0.00209 to +0.00007) against a DHT-response variance of +0.04445, and in GSE109021 DHEA behaves as a partial androgen-receptor antagonist (α = 0.631 ± 0.105; 62.1% of its response variance explained by the androgen-receptor axis) with no residual component above a potency-matched band (0.305 against 0.431 ± 0.126, rank 4 of 13, p = 0.769). In Th17 cells, where Ar transcript is effectively absent (0.134 TPM), 50 µM DHEA still moves 2230–4351 genes per stratum and suppresses the IL-17 effector module; the response projects onto a mouse Rorc -knockout signature (r = +0.422 whole signature; +0.721 restricted to the six effector genes; +0.189 for the remaining 54, p = 0.004) where an unrelated active compound, EGCG, does not (+0.028, +0.417, −0.161). Against RORγt ChIP-seq peaks, however, the DHEA effect does not reach direct targets: in an expression-matched test the Rorc knockout is displaced below its matched control (observed −0.0087 against +0.0085, p = 0.014) while DHEA is displaced in the opposite direction, indistinguishably from solvent. Conclusion. Every DHEA transcriptional effect in the public record was measured at 13- to 2174-fold above the physiological serum DHEA concentration. The concentration window in which the age-related decline could be transduced has never been assayed. The failure of these datasets to identify a DHEA receptor is therefore uninformative about its existence, and the screen specifies the experiment that would be informative.

Research Square
Fukuoka University (JP)
Good health and well-being
Estrogen and related hormone effects
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon