Integrated immune, hormonal, and transcriptomic profiling reveals sex-specific dysregulation in long COVID patients with ME/CFS, 2025, Shahbaz et al.

[Chandelier]

https://www.cell.com/cell-reports-medicine/fulltext/S2666-3791(25)00522-1

Spoiler: Authors

Shima Shahbaz1 ∙ Mohammed Osman2,3,4,7 ∙ Hussain Syed5 ∙ Andrew Mason5 ∙ Rhonda J. Rosychuk6 ∙ Jan Willem Cohen Tervaert2 ∙ Shokrollah Elahi

Highlights​

• Female long COVID (LC) patients show heightened immune activation and inflammation
• Disrupted sex hormone levels differ between male and female LC patients
• Neuroinflammatory gene signatures may explain cognitive symptoms in females
• Sex-specific biomarkers suggest need for tailored LC therapies

Summary​

Long COVID (LC) manifests with sex-specific differences, particularly in those with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS).

Our study reveals that female LC patients (LCF) with ME/CFS show a shift toward myelopoiesis, reduced lymphocytes, increased neutrophils/monocytes, and depleted regulatory T cells—suggesting persistent immune activation.
Elevated CD71+ erythroid cells and disrupted erythropoiesis contribute to fatigue and tissue damage in LCF. Cytokine profiling indicates a stronger pro-inflammatory response in LCF compared to males (LCM), along with markers of gut barrier dysfunction.
Hormonal analysis shows reduced testosterone in LCF and estradiol in LCM. Transcriptomic data reveal neuroinflammatory signatures in LCF, potentially explaining cognitive symptoms.
We also identify biomarkers that distinguish LCF from LCM and correlate with sex-specific clinical symptoms.

Overall, LC with ME/CFS is characterized by sex-specific immune, hormonal, and transcriptional alterations, with females exhibiting more severe inflammation. These insights underscore the need for sex-tailored interventions, including consideration of hormone replacement therapy.

Graphical abstract​

 

[Chandelier]

Women are three times more likely than men to get severe long COVID: Here’s why

U of A research published in Cell Reports Medicine reveals the underlying mechanisms that explain why women are three times more likely than men to have severe long COVID.

www.ualberta.ca

 

[Eleanor]

Importantly, most of the upregulated genes were associated with neuronal injury response and cognitive dysfunction. The most upregulated gene in LCF was the zinc-finger-containing gene (ZNF469; >6.7-fold), which is involved in neuronal differentiation34,35(Figure S6). Another significant upregulated gene, bone morphogenetic protein and retinoic acid (RA) inducible neural-specific protein 2 (BRINP2, >5.8-fold) (Figure S6), is associated with neurological development and disorders.36 The third most highly upregulated gene, forebrain embryonic zinc finger (FEZF2, >5.5-fold), is essential for deep-layer neuron development and function.

HOXC12, another upregulated gene (>5.1-fold), is involved in hematopoietic stem cell (HSC) and hematopoietic stem and progenitor cell (HSPC) renewal and differentiation.39,40 Similarly, zinc-finger homeobox 3 (ZFHX3; >4.8-fold), which is involved in neuronal differentiation34,35 and extracellular matrix (ECM) remodeling,41 was upregulated in LCFs (Figure S6). Reln (>4-fold), an extracellular matrix protein (Reelin) influencing synaptic plasticity and cognitive function,42,43 which has been reported to be elevated at the gene and protein levels in LC patients,17 was upregulated in LCFs

None of the top 20 genes in LCF overlapped with the top 20 genes in LCM. Notably, our comparative RNA-seq analysis revealed that the top 20 upregulated genes in LCM were enriched for pathways related to persistent innate immune activation, interferon signaling, mitochondrial stress, and inflammation. In contrast, the top 20 upregulated genes in LCF were associated with neuroinflammation and cognitive dysfunction, suggesting both qualitative and quantitative differences in transcriptomic activation between the sexes.

 

[forestglip]

Previous paper from many of the same authors has a very large overlap in upregulated genes.

From thread paper:

The most upregulated gene in LCF was the zinc-finger-containing gene (ZNF469; >6.7-fold), which is involved in neuronal differentiation34,35(Figure S6). Another significant upregulated gene, bone morphogenetic protein and retinoic acid (RA) inducible neural-specific protein 2 (BRINP2, >5.8-fold) (Figure S6), is associated with neurological development and disorders.36 The third most highly upregulated gene, forebrain embryonic zinc finger (FEZF2, >5.5-fold), is essential for deep-layer neuron development and function.37,38 HOXC12, another upregulated gene (>5.1-fold), is involved in hematopoietic stem cell (HSC) and hematopoietic stem and progenitor cell (HSPC) renewal and differentiation.39,40 Similarly, zinc-finger homeobox 3 (ZFHX3; >4.8-fold), which is involved in neuronal differentiation34,35 and extracellular matrix (ECM) remodeling,41 was upregulated in LCFs (Figure S6). Reln (>4-fold), an extracellular matrix protein (Reelin) influencing synaptic plasticity and cognitive function,42,43 which has been reported to be elevated at the gene and protein levels in LC patients,17 was upregulated in LCFs (Figures 6I and S6A).

Other upregulated genes include Ski family transcriptional corepressor 2 (SKOR2; >3.7-fold), linked to cerebellar Purkinje cells (PCs)44,45; RPL17-C18orf32 (>3.4-fold), a ribosomal protein associated with Alzheimer disease46; and Meis homeobox 2 (MEIS2; >3.2-fold), associated with neuronal dysfunction and intellectual impairment,47,48,49 which were also highly upregulated in LCFs versus RFs (Figure 6I).

Upregulation of olfactory receptors and neuronal-associated genes highlights complex immune and neuronal dysregulation in Long COVID patients, 2025, Shahbaz et al

Key findings include the upregulation of genes involved in immune dysregulation and neuronal development, such as Fezf2, BRINP2, HOXC12, MEIS2, ZFHX3, and RELN.

Similarly, Zinc-finger homeobox 3 (ZFHX3; > 4.5fold) and Zin-finger containing gene (ZFN469; > 4 fold), [...] were upregulated in LC patients (Fig. 3A).

 

[arnoble]

Am I the only one skimming this and seeing they found lots of things, but little pathological insight?

 

[ME/CFS Science Blog]

Should we have a closer look at reelin? Both the blood marker and gene expression were increased in patients. This team has reported this before in LC.

Here's the wikipedia description:

Reelin, encoded by the RELN gene,[5] is a large secreted extracellular matrix glycoprotein that helps regulate processes of neuronal migration and positioning in the developing brain by controlling cell–cell interactions.

Besides this important role in early development, reelin continues to work in the adult brain.[6] It modulates synaptic plasticity by enhancing the induction and maintenance of long-term potentiation.[7][8] It also stimulates dendrite and dendritic spine development in the hippocampus,[9][10] and regulates the continuing migration of neuroblasts generated in adult neurogenesis sites of the subventricular and subgranular zones. It is found not only in the brain but also in the liver, thyroid gland, adrenal gland, fallopian tube, breast and in comparatively lower levels across a range of anatomical regions.[11]

Reelin has been suggested to be implicated in pathogenesis of several brain diseases. The expression of the protein has been found to be significantly lower in schizophrenia and psychotic bipolar disorder,[12] but the cause of this observation remains uncertain, as studies show that psychotropic medication itself affects reelin expression. Moreover, epigenetic hypotheses aimed at explaining the changed levels of reelin expression[13] are controversial.[14][15] Total lack of reelin causes a form of lissencephaly. Reelin may also play a role in Alzheimer's disease,[16] temporal lobe epilepsy and autism.

 

[ME/CFS Science Blog]

This team has reported this before in LC.

Here's the link to that previous study:
Upregulation of olfactory receptors and neuronal-associated genes highlights complex immune and neuronal dysregulation in Long COVID patients - ScienceDirect

Looks like in other disorders such as schizophrenia, reeling has mostly been found to be decreased (while it was increased in LC studies):
Reelin - Long Covid, Depression, and Alzheimer's

 

[Chandelier]

Here's the link to that previous study:

Discussion on S4ME

 

[Jonathan Edwards]

Should we have a closer look at reelin?

Yes.

 

[Kitty]

I searched on 'elevated levels of reelin protein', and got:

In cancer

• Negative outcomes: High Reelin levels can be detrimental in several cancers, including multiple myeloma, lung cancer, and prostate cancer, where it is associated with disease progression, resistance to therapy, and poor patient outcomes.
• Protective effects: In some cases, such as triple-negative breast cancer, higher Reelin levels are linked to a reduction in cancer cell migration and invasion.

In neurological disorders

• Neurodegenerative diseases: Increased levels of the 180-kDa Reelin fragment are seen in the cerebrospinal fluid of patients with frontotemporal dementia, progressive supranuclear palsy, and Alzheimer's disease.
• Autism: Some studies show elevated plasma Reelin levels in children with autism, particularly in males.
• Schizophrenia: High Reelin levels have been observed in first-episode schizophrenia patients.
• Other conditions: High Reelin levels have also been observed in other neuropsychiatric disorders, such as bipolar disorder, and in conditions like epilepsy.

In other conditions

• Liver disease: Reelin is increased in the liver tissue of patients with conditions like HCV-related liver fibrosis and cirrhosis, where it is linked to the progression of liver fibrosis.
• Autoimmune diseases: Reelin levels were found to be increased in the synovial fluid of patients with rheumatoid arthritis.

 

[forestglip]

This team has reported this before in LC.

I think it's the same data (or at least part of the data is the same), just split by sex, which would explain why the top upregulated genes are almost identical.

Very similar cohort numbers, and identical age summaries:

Upregulation of olfactory receptors and neuronal-associated genes highlights complex immune and neuronal dysregulation in Long COVID patients

For bulk RNAseq, we randomly selected 19 (15 females & 4 males) LC patients with ME/CFS and 17 (14 females & 5 males) controls. Demographic analysis confirmed that the cohorts were comparable in age (median 47.65 ± 10:65 for LC and 48.15 ± 9.1 for controls)

Integrated immune, hormonal, and transcriptomic profiling reveals sex-specific dysregulation in long COVID patients with ME/CFS

We randomly selected 18 patients (15 LCF and 3 LCM) and 17 control individuals (12 RF and 5 RM) for bulk RNA-seq, with a comparable age range (median 47.65 ± 10:65 for LC and 48.15 ± 9.1 for R).

Edit: Same typo of a colon instead of a period in ± 10:65 for the age range in both papers.