Mechanistic note

A remarkable mouse experiment has been sitting in the SREBP1 literature since 2011.

Mice were fed a high-fat diet for eight weeks to establish obesity and fatty liver. Only then was an antisense oligonucleotide administered for 14 days.

Total liver fat fell by approximately 69%.

Plasma triglycerides fell by approximately 61%.

The hepatic lipogenic machinery was strongly suppressed, and steatosis markedly improved.

The study was published as inhibition of SREBP1c.

But the reported antisense sequence targets the SREBP1a-specific transcript region.

That changes what the experiment means.

The sequence is the key

SREBP1a and SREBP1c arise from alternative first exons of the same Srebf1 gene.

The sequence reported by Frederico et al. contains:

AGGCGGCTCTGGAAC

Now compare the Srebf1a-specific primer used years later in the first hepatocyte-specific SREBP1a knockout study:

AGGCGGCTCTGGAACAGA

The same study used a completely different sequence to detect Srebf1c.

The isoform assignment is therefore not a semantic detail.

The 2011 oligonucleotide targets the SREBP1a-specific 5′ region, not the alternative first-exon region defining SREBP1c.

Read by sequence rather than label, Frederico et al. performed something considerably more interesting:

an acute SREBP1a-directed intervention after fatty liver had already developed.

And within 14 days, the hepatic phenotype changed dramatically.

A one-nucleotide inconsistency between the reported sense and antisense sequences remains a technical limitation. It does not change the identity of the targeted SREBP1a region.

SREBP1c going down does not change the target

Frederico et al. also reported a strong reduction in SREBP1c protein.

That does not restore the original isoform assignment.

SREBP1c expression is itself SREBP-dependent.

In primary human hepatocytes, we subsequently showed that isoform-specific knockdown of SREBP1a significantly reduced SREBP1c expression although SREBP1c itself was not targeted.

The same SREBP1a-specific knockdown significantly reduced major lipogenic target genes.

For most investigated SREBP target genes, simultaneous knockdown of both SREBP1 isoforms was not more effective than SREBP1a-specific knockdown alone.

Thus:

SREBP1a ↓ → SREBP1c ↓

is an experimentally observed consequence.

A decrease in SREBP1c after an SREBP1a-directed intervention therefore cannot identify SREBP1c as the primary target.

Acute function and chronic deletion are different questions

In 2022, the first hepatocyte-specific SREBP1a knockout study concluded:

“SREBP-1a does not contribute to hepatic lipogenesis.”

But that experiment examined chronic SREBP1a deficiency.

And the SREBP network did not remain unchanged.

Compensatory SREBP2 activity was observed. Under MCD feeding, Srebf1c and Srebf2 increased in SREBP1a-deficient liver, while nuclear SREBP1 and SREBP2 were maintained.

The SREBP1a-deficient liver was therefore not simply a normal liver minus SREBP1a.

The regulatory network had remodeled.

Direct hepatic DNL flux was not measured.

The knockout demonstrates that mouse liver can maintain much of its lipogenic machinery after chronic loss of hepatocellular SREBP1a.

It does not establish that an already operating lipogenic state is acutely independent of SREBP1a.

The distinction is fundamental:

Chronic deletion asks what the liver can become without SREBP1a.

Acute perturbation asks what the liver depends on now.

Frederico asked the second question.

And obtained a striking answer.

Frederico does not stand alone

A mechanistically independent experiment points in the same direction.

BF175 directly disrupts the interaction between the highly active SREBP1a transactivation domain and MED15-KIX.

This interaction was demonstrated biochemically, and BF175 strongly inhibited SREBP1a transactivation. SREBP1c and SREBP2 transactivation domains were inhibited with substantially lower efficiency.

Lipogenic output fell.

One intervention acts at the SREBP1a transcript.

The other acts at the high-potency SREBP1a transcriptional interface.

Different molecular perturbations.

Same direction.

The question has changed

In 2015, our human study was titled:

Human Sterol Regulatory Element-Binding Protein 1a Contributes Significantly to Hepatic Lipogenic Gene Expression.

That conclusion came from isoform-specific functional experiments in primary human hepatocytes.

Seven years later, chronic hepatocyte-specific deletion in mice was interpreted to mean that SREBP1a does not contribute to hepatic lipogenesis.

The acute evidence makes such a broad conclusion difficult to regard as settled.

Neither the Frederico study nor the later knockout directly quantified SREBP1a-dependent hepatic DNL flux.

The decisive experiment therefore remains to be done:

acute, adult-inducible, isoform-specific loss of SREBP1a followed immediately by direct measurement of hepatic DNL flux.

Until then, the question is not whether SREBP1a can contribute to hepatic lipogenesis.

It can.

The harder question is:

How much hepatic lipogenic function has been attributed to SREBP1c without experimentally separating it from SREBP1a?

One of the strongest acute mouse experiments relevant to that question may already have been published fifteen years ago.

It was called SREBP1c.

References

Frederico, M.J.S. et al. Short-term inhibition of SREBP-1c expression reverses diet-induced non-alcoholic fatty liver disease in mice. Scandinavian Journal of Gastroenterology 46, 1381–1388 (2011). https://doi.org/10.3109/00365521.2011.613945

Zhao, X. et al. Inhibition of SREBP transcriptional activity by a boron-containing compound improves lipid homeostasis in diet-induced obesity. Diabetes 63, 2464–2473 (2014). https://doi.org/10.2337/db13-0835

Bitter, A. et al. Human sterol regulatory element-binding protein 1a contributes significantly to hepatic lipogenic gene expression. Cellular Physiology and Biochemistry 35, 803–815 (2015). https://doi.org/10.1159/000369739

Araki, M. et al. Hepatocyte- or macrophage-specific SREBP-1a deficiency in mice exacerbates methionine- and choline-deficient diet-induced nonalcoholic fatty liver disease. American Journal of Physiology-Gastrointestinal and Liver Physiology 323, G627–G639 (2022). https://doi.org/10.1152/ajpgi.00090.2022

A new Science Advances study identifies TRIM29-dependent degradation as a new level of AKR1B10 control in experimental MASH. The finding adds post-translational regulation to a functional AKR1B10–ACC lipogenic connection demonstrated in human hepatic cells more than a decade ago.

AKR1B10 can now be controlled at the level of protein stability.

That is new.

Its functional connection to hepatic lipogenesis is not.

Ma Z. et al. identify TRIM29 as a direct AKR1B10-binding regulator that promotes its ubiquitination and proteasomal degradation.

In human samples, AKR1B10 was elevated in diseased liver tissue and in serum from patients with MASLD compared with controls. In mice fed a high-fat/high-cholesterol diet, genetic loss of Akr1b10 attenuated steatosis, inflammation and fibrosis.

The study then moved from expression to protein turnover.

Hepatic overexpression of TRIM29 reduced AKR1B10 protein and improved multiple features of experimental MASH. Notably, hepatic Akr1b10 mRNA remained unchanged.

The authors further report that coptisine enhanced the TRIM29–AKR1B10 interaction and promoted AKR1B10 degradation. In hepatic cell models, AKR1B10 protein declined without a corresponding decrease in AKR1B10 mRNA.

The separation between transcript and protein is particularly informative.

In 2008, Ma J. et al. demonstrated in breast cancer cells that AKR1B10 interacts with and stabilizes acetyl-CoA carboxylase-α, ACCα, thereby promoting fatty-acid synthesis.

Seven years later, its functional relevance was extended into human liver biology.

Bitter et al. showed that PXR knockdown increased AKR1B10 mRNA and protein in human hepatic cells. The increase was reproduced after PXR knockdown in primary human hepatocytes.

In low-PXR hepatic cells, ACC protein was approximately 1.6-fold higher despite reduced ACACA mRNA. Silencing AKR1B10 reduced de novo lipogenesis by approximately 20%. When ACC was inhibited with TOFA, the AKR1B10-dependent difference in lipogenesis was essentially lost.

The same study extended the connection to human disease. In histologically classified liver samples, median AKR1B10 mRNA expression was approximately threefold higher in moderate-to-severe NASH, while hepatic PXR protein was reduced by more than half compared with non-NASH liver.

By 2015, AKR1B10 had therefore already been functionally linked to ACC-dependent de novo lipogenesis in human hepatic cells and associated with human steatohepatitis.

That chronology matters.

The new Science Advances paper cites the original 2008 study establishing AKR1B10-mediated stabilization of ACCα. An important hepatic intermediate in that chronology, however, is absent from its reference trail: the 2015 study had already connected AKR1B10 to ACC-dependent de novo lipogenesis in human hepatic cells, reproduced AKR1B10 regulation in primary human hepatocytes, and demonstrated increased AKR1B10 in human NASH.

This distinction also defines what is genuinely new in 2026.

The new work does not establish the hepatic functional relevance of AKR1B10 from scratch. It adds something different: regulation of AKR1B10 itself through TRIM29-dependent protein degradation, together with genetic evidence that loss of Akr1b10 attenuates experimental MASH.

The boundaries remain important. The human findings in the new study establish association, whereas the causal MASH evidence derives from mouse models and the detailed degradation mechanism from cellular and biochemical experiments. PXR was not examined, and the study therefore does not test the previously described PXR–AKR1B10 relationship.

Nor does it establish a complete TRIM29–AKR1B10–ACC–DNL pathway in MASH.

What it establishes is a new control point: AKR1B10 protein stability.

Earlier work established stabilization of ACCα by AKR1B10 and functionally linked AKR1B10 to ACC-dependent de novo lipogenesis in human hepatic cells. The 2026 study moves the regulatory question one level upstream: to the stability of AKR1B10 itself.

The degradation mechanism is new.

The hepatic story had already begun.

References

Ma, Z. et al. TRIM29-dependent AKR1B10 degradation ameliorates MASH. Science Advances 12, eaeb9309 (2026). https://doi.org/10.1126/sciadv.aeb9309

Bitter, A. et al. Pregnane X receptor activation and silencing promote steatosis of human hepatic cells by distinct lipogenic mechanisms. Archives of Toxicology 89, 2089-2103 (2015). https://doi.org/10.1007/s00204-014-1348-x

Ma, J. et al. Aldo-keto reductase family 1 B10 affects fatty acid synthesis by regulating the stability of acetyl-CoA carboxylase-alpha in breast cancer cells. Journal of Biological Chemistry 283, 3418-3423 (2008). https://doi.org/10.1074/jbc.M707650200

Elevated G6PD expression coexisted with higher NADP⁺/NADPH and GSSG/GSH ratios under prolonged injury. The result shows why metabolic compensation is not necessarily the same as redox recovery.

The liver increased a central component of its NADPH-generating machinery—and still became more oxidized.

That is the striking pattern in a study by Ou-Yang et al.

In mice with hepatocyte-specific loss of nuclear receptor corepressor 1, NCoR1, G6PD protein expression was elevated. Through the oxidative pentose phosphate pathway, G6PD connects glucose metabolism to cytosolic NADPH generation.

Yet four weeks after a single injection of the genotoxic hepatocarcinogen diethylnitrosamine, DEN, administered at 15 days of age, the NADP⁺/NADPH and GSSG/GSH ratios were also higher. Both redox systems shifted toward oxidation, while mitochondrial reactive oxygen species increased, hepatic ATP declined, and oxidative damage and apoptosis rose.

The liver responded.

But response was not the same as recovery.

The same genetic model shows how context and persistence can change the outcome. Loss of NCoR1 increased Fasn and Acc2 expression, hepatic lipid accumulation and the broader de novo fatty-acid synthesis program.

After partial hepatectomy, this lipogenic program supported regeneration. Hepatocyte proliferation increased, liver mass recovered earlier, and ACC2 knockdown or treatment with orlistat reduced the regenerative advantage.

The acute response carried a measurable redox cost. During early regeneration, the NADP⁺/NADPH ratio and mitochondrial reactive oxygen species were elevated. By 48 hours, however, the differences between NCoR1-deficient and control livers had largely narrowed.

The disturbance was real, but temporally contained.

Four weeks after DEN, the pattern no longer resolved in the same way. Lipogenic proteins and G6PD remained elevated, while the NADP⁺/NADPH and GSSG/GSH ratios remained shifted toward oxidation. Mitochondrial reactive oxygen species, energy loss and cellular damage accompanied the persistent metabolic response.

Orlistat attenuated several of these lipid, oxidative and apoptotic changes, supporting a contribution of enhanced fatty-acid synthesis to the phenotype.

This is more informative than a simple model of NADPH depletion.

The data define a systems-level pattern rather than a direct map of NADPH flux. G6PD expression documents induction of a central component of the oxidative pentose phosphate pathway but does not quantify pathway flux or NADPH production. Whole-liver NADP⁺/NADPH measurements capture the integrated redox outcome rather than the distribution of reducing equivalents among individual enzymes or intracellular compartments.

A central component of a major NADPH-generating pathway was induced. Lipogenic demand persisted. The measured NADP and glutathione redox states nevertheless became more oxidized.

This pattern is consistent with NADPH allocation pressure.

Here, NADPH allocation pressure describes a state in which a generating response is induced, yet simultaneous reductive demands persist beyond the capacity of compensation to restore the previous redox state.

Production, availability and recovery are not interchangeable concepts.

Independent work in PXR-null mice adds a complementary perspective. Following partial hepatectomy, loss of the xenobiotic and endobiotic sensor PXR blunted transient hepatic lipid accumulation and impaired normal liver regeneration, with changes extending to genes involved in lipid biosynthesis. The NCoR1 phenotype was not shown to be PXR-mediated; the convergence lies at the level of regulated lipid remodeling as part of the adaptive response to tissue loss.

In the DEN model, increased oxidative damage and apoptosis were ultimately associated with fewer tumors, consistent with enhanced elimination of damaged or preneoplastic cells. This model-specific outcome should not be transferred directly to human metabolic liver disease.

The broader principle is more fundamental.

Transient lipogenic activity supported regeneration, and the associated redox disturbance receded.

Under prolonged injury, a persistent lipogenic program was accompanied by an NADPH-generating response, but redox balance was no longer restored.

The liver did not fail to respond.

Its response failed to bring the liver back.

References

Ou-Yang, Q. et al. Distinct role of nuclear receptor corepressor 1 regulated de novo fatty acids synthesis in liver regeneration and hepatocarcinogenesis in mice. Hepatology 67, 1071-1087 (2018). https://doi.org/10.1002/hep.29562

Dai, G. et al. Pregnane X receptor is essential for normal progression of liver regeneration. Hepatology 47, 1277-1287 (2008). https://doi.org/10.1002/hep.22129

A recent Nature Metabolism study reconstructs human MASLD progression as a continuous molecular trajectory rather than as a set of static histological stages. Within this trajectory-based framework, SREBF1 appears as a dynamic regulatory node. The authors place SREBF1 at the interface between lipogenic regulation, early stress responses and hepatocellular injury, citing Bitter et al. 2015 at this point. This context is notable for human SREBP1a biology. The 2015 study showed that human SREBP1a contributes significantly to hepatic lipogenic gene expression. In the new trajectory-based framework, this earlier human SREBP1a work is connected to a systems-level view of fatty liver disease progression, where metabolic regulation, cellular stress and disease-state transitions are analysed as part of a continuous molecular process. The relevance is therefore not the citation alone. It is the position of the citation. Human hepatic SREBP1a biology appears at the conceptual interface that is increasingly shaping the field: dynamic regulatory states, progression trajectories and stress-linked metabolic remodelling.

References

Kamzolas, I. et al. A data-driven framework reconstructs the molecular continuum of human MASLD progression. Nature Metabolism (2026). https://doi.org/10.1038/s42255-026-01543-7

Bitter, A. et al. Human sterol regulatory element-binding protein 1a contributes significantly to hepatic lipogenic gene expression. Cellular Physiology and Biochemistry 35, 803-815 (2015). https://doi.org/10.1159/000369739

SREBP1a is back at the center of the PXR story.

More than ten years ago, our 2015 Archives of Toxicology study identified SREBP1a induction as a key mechanism linking PXR activation to the SREBP1-dependent lipogenic pathway in human hepatic cells.

A new Archives of Toxicology study now provides independent in vivo support for that axis.

The key signal is clear:

Atorvastatin induced hepatic Srebf1a in Pxr+/+ mice.
In Pxr−/− mice, this response disappeared.

That is the genetic point.

PCN adds the biological weight. As the prototypical mouse PXR agonist, PCN also induced Srebf1a in wild-type liver and activated nuclear SREBP1.

Together, these findings place SREBP1a within the PXR-shaped hepatic transcriptional state space.

This is not just another lipid gene response.

It is independent in vivo support for a mechanism that has been central to srebp1a.com from the beginning: the PXR–SREBP1a axis.

The timing is remarkable.

Recent work has already pushed AKR1B10 beyond marker status and strengthened it as a functional lipogenic stress-state node. Now, SREBP1a also receives independent support as a second molecular anchor of a detoxification–lipogenic hepatic state.

This does not prove Detoxification State Fixation as a whole.

But it changes the weight of interpretation.

SREBP1a is no longer "only" an early human hepatic cell observation.

More than ten years later, it is independently supported in vivo — downstream of PXR.

References

Nabil, H. et al. Atorvastatin regulates hepatic transcriptome PXR dependently but distinct from pregnenolone 16α-carbonitrile and does not induce PXR-mediated liver steatosis. Archives of Toxicology 100, 1443-1463 (2026). https://doi.org/10.1007/s00204-025-04280-0

Bitter, A. et al. Pregnane X receptor activation and silencing promote steatosis of human hepatic cells by distinct lipogenic mechanisms. Archives of Toxicology 89, 2089-2103 (2015). https://doi.org/10.1007/s00204-014-1348-x

AKR1B10 is rising.

CYP2C19 is falling.

That may be more than a biomarker pattern.

It may be a sign that detoxification is changing direction.

The previous note argued that AKR1B10 has moved beyond passive marker status in NAFLD. The next question is where this AKR1B10 signal sits within disease progression. The AKR1B10–CYP2C19 divergence reported by Liu et al. provides a sharper frame: AKR1B10 does not rise in isolation. It rises while a classical CYP drug-metabolizing output falls.

A recent integrative multi-omics study by Liu et al. mapped the progressive disease landscape of metabolic dysfunction-associated steatotic liver disease and identified four recurrent progression-associated candidates: AKR1B10, COL1A2, SPP1 and CYP2C19. Among them, AKR1B10 and CYP2C19 are particularly interesting because both were predominantly localized to hepatocytes and both were functionally interrogated in hepatocyte models.

The direction of change was striking.

AKR1B10 increased during disease progression.

CYP2C19 decreased.

At first glance, this could be read simply as one marker going up and another going down. But that would miss the deeper biological signal. AKR1B10 and CYP2C19 do not represent random hepatocyte proteins. They point into different regions of hepatic detoxification-state biology.

AKR1B10 is an NADPH-dependent aldo-keto reductase involved in carbonyl and aldehyde handling, retinoid metabolism, lipid-associated stress biology and redox adaptation. CYP2C19, by contrast, represents one classical cytochrome P450 drug-metabolizing output. The divergence between AKR1B10 and CYP2C19 therefore does not look like a simple “detoxification up” or “detoxification down” event.

It looks like redistribution.

AKR1B10 is not a passive marker

The importance of AKR1B10 has been increasing steadily.

In fatty liver disease, AKR1B10 has repeatedly appeared as a disease-associated hepatic signal. More recent work has moved it beyond marker status. Yang et al. showed that berberine directly targets AKR1B10 protein and that AKR1B10 perturbation influences lipid and glucose metabolic outputs in experimental NAFLD models. This does not make berberine the central point. The important point is more general: AKR1B10 can be touched, and touching it can alter metabolic disease-relevant outputs.

A recent pathway-level review further strengthens this interpretation. It places AKR1B10 at the interface of carbonyl detoxification, retinoid metabolism, redox control, ACCα-linked lipogenesis, DAG/PKC/ERK signaling, inflammatory signaling and hepatocellular carcinoma biology. In that framework, AKR1B10 is not merely a diagnostic label attached to injured liver tissue. It is a metabolic-redox-detoxification node.

This matters for fatty liver disease progression.

If AKR1B10 rises together with lipogenic, inflammatory and oxidative stress programs, the signal is not just “AKR1B10 is present.” The signal may be that hepatocytes are entering a different mode of stress handling: one in which aldehyde detoxification, retinoid routing, NADPH use, carbonyl stress buffering and lipogenic output become increasingly connected.

AKR1B10 should not be read as intrinsically pathological. In a transient stress context, its carbonyl-detoxifying, retinaldehyde-reducing and redox-buffering functions may be protective. A transient lipogenic output may also be adaptive, for example by expanding lipid-buffering capacity for lipophilic stress mediators and by supporting membrane and lipid requirements during repair and regeneration. The disease-relevant question is whether this AKR1B10-linked stress-handling mode resolves — or whether it becomes persistently co-maintained with lipogenic output.

That is precisely the type of biology that a state-based framework should notice.

The CYP2C19 decrease changes the interpretation

The decrease in CYP2C19 makes the pattern sharper.

If AKR1B10 alone increased, one could describe it as a stress marker, a lipogenic contributor, or a disease-associated enzyme. But when AKR1B10 rises while CYP2C19 falls, the interpretation changes.

This is not simply detoxification becoming stronger.

It is not simply detoxification failing.

It is a directional reorganization of hepatocyte output.

One arm associated with redox-retinoid-carbonyl-lipogenic stress handling rises. One classical CYP drug-metabolizing output falls. In Liu et al., AKR1B10 knockdown and CYP2C19 overexpression both reduced lipid droplet accumulation and intracellular triglyceride levels in FFA-treated hepatocytes, with the combined intervention showing the strongest effect. The same perturbations also reduced oxidative stress and inflammatory readouts.

That functional direction is important.

It suggests that the AKR1B10/CYP2C19 pattern is not only descriptive. It may be linked to the hepatocyte state itself.

Not global CYP loss

There is an important caveat.

The fall of CYP2C19 must not be misunderstood as global cytochrome P450 collapse.

The CYP system is not a single output. Some CYP-linked routes may remain active or even increase under steatohepatitic stress, including lipid-oxidation and omega-oxidation pathways such as those involving CYP2E1 and CYP4A. These routes are often discussed in relation to microsomal oxidative stress, lipid peroxidation and fatty-acid stress metabolism.

Therefore, the key observation is not global CYP loss.

The key observation is selective output redistribution.

CYP2C19 decreases as one classical drug-metabolizing CYP output within the progression landscape. Other CYP-linked stress routes may behave differently. That is not a weakness of the interpretation. It is the point.

Fatty liver disease progression may not switch hepatic detoxification simply on or off. It may redistribute detoxification-state outputs across different enzymatic arms.

PXR-connected, but not simple PXR activation

This pattern is also PXR-relevant.

Pregnane X receptor biology has long been linked to hepatic xenobiotic metabolism, but it is not restricted to a simple drug-detoxification switch. In human hepatic cells, PXR perturbation has been connected to steatotic lipogenic output and to AKR1B10-associated mechanisms. Notably, PXR activation and PXR silencing can both promote steatosis, but through distinct lipogenic routes.

That matters here.

The AKR1B10/CYP2C19 divergence should not be reduced to simple PXR activation or repression. It is more consistent with a PXR-connected imbalance in detoxification-state biology: selected stress-adaptive outputs rise, while selected classical CYP outputs fall.

This is exactly where simple on/off language becomes insufficient.

The pattern asks for state language.

NADPH allocation, not simple energy failure

There is another layer.

AKR1B10 is NADPH-dependent. Lipogenesis is NADPH-demanding. Carbonyl detoxification, antioxidant defense and lipid peroxide handling also draw on redox capacity. The relevant question is therefore not whether the liver can generate NADPH. The relevant question is how NADPH-dependent outputs are chronically allocated across detoxification, lipogenesis, carbonyl handling, retinoid routing and antioxidant defense.

In a transient adaptive state, this allocation may be useful.

In a fixed state, the same allocation may become self-maintaining.

That is where detoxification-state biology begins to overlap with disease persistence.

A DSF-compatible reading

Detoxification State Fixation (DSF) proposes that progressive fatty liver disease may involve fixation of an originally adaptive hepatic detoxification-lipogenic rescue state beyond its useful window. In that view, pathology does not arise because detoxification is simply active. It arises when a protective mode loses reversibility and becomes partly self-maintaining.

The AKR1B10/CYP2C19 pattern fits this logic remarkably well.

AKR1B10 rises as a redox-retinoid-carbonyl-lipogenic stress node.

CYP2C19 falls as a classical CYP drug-metabolizing output.

Other CYP-linked stress routes may remain active or increase.

The result is not uniform detoxification activation.

It is not uniform detoxification failure.

It is output-selective disequilibrium.

That is the important conceptual step.

A hepatocyte under progressive fatty liver disease pressure may not simply detoxify more or less. It may detoxify differently. It may allocate redox capacity, NADPH use, lipid handling and carbonyl stress buffering into a new pattern. If that pattern becomes persistent, it may help stabilize the disease state itself.

Why this matters

The Liu et al. study does not prove DSF.

It does something more specific and, in some ways, more useful: it provides a modern progression-associated pattern that requires better language than “marker up” and “marker down.”

AKR1B10 rising while CYP2C19 falls is a hepatocyte-centered clue. It suggests that fatty liver disease progression may involve directional remodeling of detoxification-related outputs. When combined with evidence that AKR1B10 is functionally linked to lipid metabolism, redox stress, retinoid metabolism and ACCα-associated lipogenic biology, the signal becomes difficult to dismiss as passive.

This is why AKR1B10 matters.

This is why CYP2C19 matters.

And this is why the divergence between them may matter even more than either marker alone.

Detoxification does not disappear.

It changes direction.

Related framework

Detoxification State Fixation (DSF)

Related note

AKR1B10 moves beyond marker status in NAFLD

References

Liu, K. et al. Integrative Multi-Omics Analysis Elucidates the Progressive Disease Landscape and Reveals Dynamic Protein Biomarkers for MASLD Surveillance. The FASEB Journal 40, e71998 (2026). https://doi.org/10.1096/fj.202601011R

Yang, S. et al. Berberine directly targets AKR1B10 protein to modulate lipid and glucose metabolism disorders in NAFLD. Journal of Ethnopharmacology 332, 118354 (2024). https://doi.org/10.1016/j.jep.2024.118354

Wang, C. et al. The pathway network of aldo-keto reductase 1B10: a new perspective on gene-targeted therapy. npj Gut and Liver 3, 20 (2026). https://doi.org/10.1038/s44355-026-00064-0

Bitter, A. et al. Pregnane X receptor activation and silencing promote steatosis of human hepatic cells by distinct lipogenic mechanisms. Archives of Toxicology 89, 2089-2103 (2015). https://doi.org/10.1007/s00204-014-1348-x

Lin, X.-L. et al. Nicotinate-curcumin improves NASH by inhibiting the AKR1B10/ACCα-mediated triglyceride synthesis. Lipids in Health and Disease 23, 201 (2024). https://doi.org/10.1186/s12944-024-02162-5

Ma, J. et al. Aldo-keto reductase family 1 B10 affects fatty acid synthesis by regulating the stability of acetyl-CoA carboxylase-alpha in breast cancer cells. Journal of Biological Chemistry 283, 3418-3423 (2008). https://doi.org/10.1074/jbc.M707650200

AKR1B10 is no longer only a marker.

It can be touched.

A recent study by Yang et al., published in the Journal of Ethnopharmacology, identifies AKR1B10 as a direct berberine-binding target in experimental NAFLD models.

The important point is not berberine as a supplement.

The important point is AKR1B10.

In this study, AKR1B10 is chemically engaged, enzymatically inhibited, genetically perturbed and functionally connected to lipid and glucose metabolic outputs.

That changes the weight of the argument.

From marker to functional node

AKR1B10 has often been read as a disease-associated marker in fatty liver disease, steatohepatitis, fibrosis and liver cancer risk.

That view is becoming too small.

AKR1B10 is a NADPH-dependent aldo-keto reductase. It belongs to carbonyl, aldehyde and retinoid-related stress-handling biology.

In a diseased liver, that matters.

AKR1B10 does not only mark stress.

It may help shape the metabolic state that follows from stress.

That is why recent AKR1B10 work is important. Earlier studies linked AKR1B10 to ACCα stability and the lipogenic machinery that drives de novo lipogenesis and triglyceride synthesis. More recent NASH work placed the AKR1B10/ACCα axis directly within hepatic triglyceride accumulation.

Yang et al. now add another layer.

AKR1B10 appears pharmacologically addressable.

What Yang et al. show

Yang et al. used high-fat diet-fed mice and oleic acid-treated HepG2 cells as experimental NAFLD models.

Berberine improved several metabolic readouts, including hepatic steatosis, triglyceride accumulation, glucose-related parameters and insulin resistance-associated changes.

But the key point is target engagement.

Using a berberine-derived probe and click-chemistry proteomics, the authors identified AKR1B10 among candidate berberine-binding proteins.

They then supported AKR1B10 engagement with additional approaches, including co-localization, molecular docking, SPR, DARTS and CETSA.

Berberine also inhibited AKR1B10 enzymatic activity.

This makes the result stronger than an expression observation.

AKR1B10 is not only increased in disease-associated settings.

It is directly engaged by a small molecule.

The strongest signal is dependence

The most important part of the study is not the target list.

It is the perturbation logic.

In oleic acid-treated HepG2 cells, AKR1B10 knockdown itself increased glucose consumption and reduced triglyceride content.

After AKR1B10 knockdown, berberine produced little or no additional effect on these readouts.

Pharmacological AKR1B10 inhibition supported the same dependency pattern.

AKR1B10 overexpression moved the system the other way: glucose consumption decreased and triglyceride content increased. Berberine counteracted this overexpression phenotype.

The in vivo data point in the same direction. In high-fat diet-fed mice, AAV-mediated AKR1B10 knockdown weakened the additional metabolic effects of berberine.

This is best read as functional dependence.

When AKR1B10 is already reduced or blocked, there is less room for berberine to act.

That is what one expects from a relevant metabolic control point.

Why this matters

A marker can describe disease.

A functional node can help explain how a disease state is maintained.

That is the step AKR1B10 is beginning to take.

Yang et al. do not show that AKR1B10 alone drives NAFLD.

They do not show that berberine treats Detoxification State Fixation.

They do not test the full DSF architecture.

But they do show something important:

AKR1B10 is perturbable.

That matters for a disease-state framework in which stress handling, redox chemistry, lipid remodeling, de novo lipogenesis and lipogenic output may become linked in a self-maintaining hepatic state.

AKR1B10 sits close to that interface.

It connects detoxification-state biology with lipid metabolism.

That is why AKR1B10 should not be read only as a marker of where fatty liver disease has been.

It may be one of the places where the diseased hepatic state can be touched.

A broader pattern is emerging

AKR1B10 may not be standing alone.

A broader progression-level pattern is now emerging.

That pattern deserves its own discussion.

For now, the message is already clear:

AKR1B10 has moved beyond marker status in NAFLD.

Related framework:

Detoxification State Fixation (DSF)

References

Yang, S. et al. Berberine directly targets AKR1B10 protein to modulate lipid and glucose metabolism disorders in NAFLD. Journal of Ethnopharmacology 332, 118354 (2024). https://doi.org/10.1016/j.jep.2024.118354

Lin, X.-L. et al. Nicotinate-curcumin improves NASH by inhibiting the AKR1B10/ACCα-mediated triglyceride synthesis. Lipids in Health and Disease 23, 201 (2024). https://doi.org/10.1186/s12944-024-02162-5

Ma, J. et al. Aldo-keto reductase family 1 B10 affects fatty acid synthesis by regulating the stability of acetyl-CoA carboxylase-alpha in breast cancer cells. Journal of Biological Chemistry 283, 3418-3423 (2008). https://doi.org/10.1074/jbc.M707650200

Bitter, A. et al. Pregnane X receptor activation and silencing promote steatosis of human hepatic cells by distinct lipogenic mechanisms. Archives of Toxicology 89, 2089-2103 (2015). https://doi.org/10.1007/s00204-014-1348-x

Related note:
AKR1B10 as a lipogenic control node in NASH

Mechanistic note

AKR1B10 has often been discussed as a disease-associated marker in steatohepatitis and liver injury. However, an increasingly important question is whether AKR1B10 is only a marker of disease state, or whether it can actively contribute to the persistence of a lipogenic hepatic state.

A 2024 study by Lin et al. adds a particularly relevant functional layer to this question. The authors investigated nicotinate-curcumin in experimental NASH models and focused on the AKR1B10/ACCα pathway. In a high-fat/high-fructose rat model and in Ox-LDL/high-glucose-stressed HepG2 cells, they report increased AKR1B10 and ACCα together with changes in Malonyl-CoA, free fatty acids and triglycerides. Treatment with nicotinate-curcumin reduced AKR1B10/ACCα signaling and was associated with lower Malonyl-CoA, FFA and TG levels.

This supports the view that AKR1B10 is more than a passive NASH-associated readout. ACCα catalyzes the formation of Malonyl-CoA, a central building block for fatty acid synthesis. If AKR1B10 helps to maintain ACCα-dependent lipid synthesis, it may participate in stabilizing the metabolic output of the diseased hepatocyte.

This connects back to my earlier PXR/steatosis work in human hepatic cells. In 2015, we reported that ligand-dependent PXR activation and reduced PXR signaling can both promote steatosis, but through distinct mechanisms. PXR activation induced SREBP1a and lipogenic SREBP1 target genes, whereas PXR knockdown increased AKR1B10 and an ACC-dependent branch of de novo lipogenesis. In histologically classified human NASH liver samples, PXR protein was reduced, while AKR1B10, SREBP1a and lipogenic target genes were increased.

The work by Lin et al. therefore does not stand isolated. It strengthens a mechanistic line in which AKR1B10 links stress-associated hepatic remodeling to ACCα, Malonyl-CoA, FFA and triglyceride synthesis.

This is also relevant for the Detoxification State Fixation (DSF) framework. In the DSF model, AKR1B10 is proposed as one of the molecular anchors that may help stabilize an originally adaptive detoxification-lipogenic hepatic state beyond its normal resolution window. The important point is not that AKR1B10 alone “causes” NASH. Rather, AKR1B10 may help maintain one part of a self-reinforcing state: persistent lipogenic output under disease-shaped stress conditions.

From this perspective, AKR1B10 becomes more than a biomarker. It becomes a candidate state-stabilizing node.

This does not prove DSF as a whole. But it provides independent functional support for one of its proposed molecular anchors: AKR1B10 as a lipogenic control node connecting detoxification-associated stress biology with persistent hepatic lipid synthesis.

Related framework:
Detoxification State Fixation (DSF)
https://srebp1a.com/state-fixation/

References

Lin, X.-L. et al. Nicotinate-curcumin improves NASH by inhibiting the AKR1B10/ACCα-mediated triglyceride synthesis. Lipids in Health and Disease 23, 201 (2024). https://doi.org/10.1186/s12944-024-02162-5

Bitter, A. et al. Pregnane X receptor activation and silencing promote steatosis of human hepatic cells by distinct lipogenic mechanisms. Archives of Toxicology 89, 2089-2103 (2015). https://doi.org/10.1007/s00204-014-1348-x

In biomedical research, accurate nomenclature is not a cosmetic detail. It determines whether relevant biology remains visible to researchers, databases, search engines, and automated text-mining systems.

The history of human SREBP1a illustrates this point. SREBP1a and SREBP1c are not interchangeable names. They are distinct SREBF1-derived isoforms with different biological and experimental implications. Nevertheless, for many years, human database entries made SREBP1c highly visible while SREBP1a was much harder to find.

The historical oversight

Until early 2016, the official alias structure for the human SREBF1 gene did not properly reflect SREBP1a. Historical database entries listed SREBP1, bHLHd1, and SREBP-1c, but SREBP1a was missing from the visible alias field.

Figure 1 | Historical NCBI Gene entry for human SREBF1, archived in July 2015, showing SREBP1, bHLHd1, and SREBP-1c as aliases, while SREBP1a was not listed.

This was more than a naming inconvenience. If a human protein isoform is missing from alias structures, database searches, literature searches, and automated semantic tools may fail to connect that protein to its gene.

The 2016 correction

In February 2016, I contacted NCBI and HGNC regarding the absence of SREBP1a from the human SREBF1 alias listings. Shortly thereafter, SREBP1a was added, restoring a direct database link between the human SREBP1a protein and the SREBF1 gene.

Figure 2 | NCBI RefSeq curator confirmation from February 1, 2016 documenting that SREBP1a was added as an additional alias to the human SREBF1 gene record (GeneID: 6720).

 Figure 3 | HGNC curator confirmation from February 3, 2016 documenting that SREBP1a was added as an additional alias to the human SREBF1 gene record.

This correction mattered because SREBP1a is not a synonym for SREBP1c. Both arise from SREBF1, but they represent distinct isoform biology. Making SREBP1a visible in official alias systems improved the connection between human experimental protein research and genomic database structure.

Current status

In recent years, database displays have changed. Some resources no longer show all transcript- or isoform-specific names in the most prominent summary fields. This can make rapid searches less straightforward.

HGNC remains especially important because it provides the official curated framework for human gene nomenclature. For SREBF1, this means that SREBP1a remains connected to the human SREBF1 gene in an official nomenclature context.

Figure 4 | HGNC symbol report for human SREBF1, accessed in June 2026, showing SREBP1a as a listed alias symbol and thereby confirming continued visibility of the isoform in an official gene nomenclature resource.

Conclusion

The 2016 correction helped preserve an important semantic link: human SREBP1a belongs visibly to the SREBF1 gene context.

That may sound like a small database detail. It is not. When nomenclature is incomplete, biology becomes harder to find. When nomenclature is corrected, human protein isoform research becomes more searchable, more transparent, and easier to integrate across literature, databases, and automated discovery tools.

For human SREBP1a research, this link matters.

Related: Detoxification State Fixation (DSF)

Please use the Internet Archive...

Jung et al. Role of the AMPK/SREBP-1 pathway in the development of orotic acid-induced fatty liver. J Lipid Res. 2011 Sep;52(9):1617-25.

 »Whereas decreased phosphorylation of AMPK and modulation of a series of downstream events leading to fatty acid synthesis and lipogenesis were observed in rat hepatocytes and human hepatoma cell lines, mouse hepatocytes were resistant to OA with regard to the AMPK/SREBP-1-dependent lipogenic pathway. Similar results were observed in animal studies using SD rats and C57BL/6 mice. These two species showed completely different responses to OA in terms of AMPK/SREBP-1 signaling and development of steatosis, indicating that not only pharmacokinetic but also pharmacodynamic factors participate in determining interspecies differences.«

Ohno et al. Transgenic Mice Overexpressing SREBP-1a in Male ob/ob Mice Exhibit Lipodystrophy and Exacerbate Insulin Resistance. Endocrinology. 2018 Jun 1;159(6):2308-2323.

What is going on?

Q?

Future proves Past

Be strong (anons)

HMGCR-dependent SREBP1a maturation (simplified scheme). 1a, mature SREBP1a protein; ER, endoplasmic reticulum. (see section »PXR versus LXR«)

The dogma of the SREBP1 protein:

»Sterols inhibit the cleavage of the precursor, and the mature nuclear form is rapidly catabolized, thereby reducing transcription.« (see ref 1, ref 2, ref 3)

However, with all due respect, I would like to mention that the results of dozens of scientific articles contradict and/or do not support this dogma. The literature analysis on srebp1a.com and the results of the following mentioned article reveal it.

Hwang et al. Contribution of Accelerated Degradation to Feedback Regulation of 3-Hydroxy-3-methylglutaryl Coenzyme A Reductase and Cholesterol Metabolism in the Liver. J Biol Chem. 2016 Jun 24;291(26):13479-94.

Before the literature analysis on srebp1a.com and the analysis of the article mentioned above, please consider the following two points:

1. HMGCR is the rate-limiting enzyme of sterol biosynthesis.

2. HMGCR protein/activity induces the maturation of SREBP1.

Alarming predictions:

Estes et al. Modeling the epidemic of nonalcoholic fatty liver disease demonstrates an exponential increase in burden of disease. Hepatology. 2017 Aug 12. doi: 10.1002/hep.29466. [Epub ahead of print]

Abstract