3D illustration of a caffeine molecule with clearly defined atoms and bonds. Scientific visualization for educational, medical, chemical, pharmaceutical or research purposes.
Omar Lopez-Rincon/Unsplash.com
Japanese researchers have reported a compound that stops hepatitis B virus (HBV) and hepatitis D virus (HDV) from entering cultured human liver cells by weakening the partnership between two receptors the viruses need, NTCP and EGFR. In the team’s preprint, the compound left NTCP’s normal job of moving bile acids into the cell undisturbed. Kyoto University announced the work on Nov. 13 and says it appears in PNAS (volume 123, issue 46).
That matters because the only FDA-approved hepatitis D entry blocker, bulevirtide (Hepcludex), works by binding NTCP itself, and every participant who received it in the pivotal MYR301 trial had elevated serum bile salts, which peaked by week 8 and resolved after treatment stopped. A way to stop viral entry without touching that transporter would be a different kind of entry blocker, but this compound has so far been tested only in lab-grown cells.
The Oxy229 figures below come from the team’s February 2026 preprint, which the published PNAS paper may have revised, so the published numbers could differ. Kyoto’s announcement lists no funding sources and quotes no researchers. This report draws on Kyoto’s announcement and the preprint rather than the full PNAS paper.
Koichi Watashi, a virologist at Japan’s National Institute of Infectious Diseases, is the paper’s last-listed author. Norimichi Nomura of Kyoto University’s Graduate School of Biostudies is a co-author. Other contributors come from Kitasato University, the National Institute of Advanced Industrial Science and Technology (AIST) and Max BioPharma, a privately held, preclinical-stage company in Santa Monica, California, built around proprietary oxysterol compounds.
How Does Hepatitis B Virus Enter Liver Cells Through NTCP and EGFR?
HBV first sticks to heparan sulfate proteoglycans on the liver-cell surface, then the preS1 segment of its large surface protein binds NTCP, a bile-acid transporter encoded by the SLC10A1 gene, and the virus is taken in by endocytosis. HDV, a satellite virus that can replicate only where HBV is present, enters through the same transporter. Researchers identified NTCP as that receptor in 2012.
NTCP alone is not enough. A 2019 PNAS paper from the same Watashi group found that EGFR, the epidermal growth factor receptor, binds NTCP and helps carry the virus-bound complex inside: knocking down EGFR left attachment intact but cut internalization. The same study showed that an NTCP mutant, G144,148A, lost EGFR binding but kept preS1 binding. That separability is the biological premise for going after the partnership instead of the transporter.
Oxy229 behaves the way that premise predicts. It did not significantly change viral attachment to cells, but the share of cells with internalized preS1, a fluorescently tracked piece of the virus’s binding segment, fell from about 70% to about 5%, while EGFR itself was still internalized. The authors read this as the virus-NTCP complex coming apart from the traveling EGFR. They have not shown whether Oxy229 binds NTCP or EGFR, and they describe direct binding to one of the two as likely.
Bead-Based Screen of 1,000 Compounds Singled Out Oxysterol Oxy229
To find molecules that pry the pair apart, the team built a cell-free AlphaScreen assay for NTCP-EGFR binding. AlphaScreen attaches one protein to a donor bead and the other to an acceptor bead; laser light makes the donor bead release singlet oxygen, which diffuses only a short distance, so the acceptor bead emits light only when the proteins hold the beads close together. A compound that breaks the contact dims the glow. In this assay the NTCP-EGFR signal ran 21 times above one negative control and 24 times above another.
The team screened about 1,000 compounds at 100 µM, and 40 cut the signal below half of control levels; a counter-screen then removed compounds that also dimmed an interaction-independent signal, which flags false positives. Oxy229, a semi-synthetic oxysterol built on a cholesterol backbone, was the compound the team pursued. Oxysterols are oxidized cholesterol derivatives.
Oxy229 Left Bile-Acid Uptake and EGFR Signaling Untouched, Unlike Bulevirtide and Gefitinib
At active concentrations, Oxy229 did not significantly change uptake of radiolabeled taurocholic acid, a bile acid; bulevirtide, tested in the preprint under its earlier name Myrcludex B, did block it, and gefitinib did not. Oxy229 also left phosphorylation of EGFR, Akt and ERK alone, whereas gefitinib nearly shut those signals down, and Oxy229 did not significantly affect entry of an NTCP-independent control virus or show cytotoxicity at the concentrations tested.
That contrast targets a known cost of NTCP-directed drugs. The FDA label records the clinical consequence: elevated serum bile salts in every participant who received Hepcludex in MYR301.
The weight of that cost is unsettled. The label states that bile-salt elevations resolved after Hepcludex was stopped, that their size did not correlate with the severity of pruritus (itching), and that pruritus occurred in 14% of recipients, described as Grade 1 or 2 and self-limited. Sparing transport is therefore a design hypothesis with a plausible rationale, not a demonstrated clinical advantage.
Gefitinib illustrates the other trade-off. It is an EGFR kinase inhibitor, a cancer drug that blocks the receptor’s signaling broadly. The 2019 paper found gefitinib inhibited HBV infection in cultured cells. Disrupting the NTCP-EGFR contact without silencing EGFR’s kinase signaling is the alternative that Oxy229 tests in cells.
Oxy229 Did Not Slow Virus Production in Already-Infected Cells, Unlike Entecavir
In a cell line that already produces HBV, Oxy229 did not significantly change viral DNA output, while the standard antiviral entecavir did, and Oxy229 did not stop the virus from attaching to cells. That fits an entry-only mechanism, and it marks a limit: HBV persists in infected liver-cell nuclei as covalently closed circular DNA (cccDNA), the template for viral RNA, which an entry blocker cannot remove. Gene-editing approaches aimed at cccDNA itself have reached Phase 1 testing.
Potency, Binding Target and Stability Remain Open Questions for Oxy229
Half-maximal inhibition of HBV infection took 23.2 µM in HepG2-hNTCP-C4 cells and 13.0 µM in primary human hepatocytes, and the authors plan to optimize activity. Oxy283, a derivative, reached 3.23 µM, but the preprint reports that figure only in primary hepatocytes. The preprint notes that natural oxysterols tend to be weakly active and metabolically unstable, and it describes the semi-synthetic derivatives as designed to improve both.
HDV was tested in a single cell model and inhibition was dose-dependent, but no HDV potency figure is reported. The preprint lists pharmacokinetic properties and in vivo infection assays as work still to be done.
WHO Counts 240 Million Chronic Hepatitis B Cases, Fewer Than 5% Treated
WHO’s 2026 Global Hepatitis Report estimated 240 million people living with chronic hepatitis B in 2024, with fewer than 5% receiving treatment and about 1.1 million hepatitis B deaths that year.
For hepatitis D, the FDA granted accelerated approval to Gilead’s Hepcludex on May 22, 2026, the first U.S. therapy for chronic HDV. The approval rests on HDV RNA decline and ALT normalization, the label states that clinical improvement has not been established, and a boxed warning says stopping treatment may cause severe flares of HDV and HBV.
What the Kyoto-led work shows, so far, is narrow: in cultured cells, a small molecule can stop HBV and HDV internalization by disturbing NTCP-EGFR contact while bile-acid transport and EGFR signaling carry on. It does not show that Oxy229 works in animals, binds a defined target, clears infected cells, or delivers a patient benefit from sparing transport. Kyoto’s announcement says the authors expect the findings to support future drug design and development, and it gives no clinical timeline.
Frequently Asked QuestionsWhat does the NTCP receptor normally do in the liver?
NTCP, encoded by the SLC10A1 gene, is a sodium-dependent transporter in the membrane of liver cells that pulls conjugated bile salts out of the blood, and it also moves some steroid hormones, thyroid hormones and foreign compounds. HBV and HDV use it as an entry door, which is why drugs aimed at it can disturb bile-acid handling.
Does sparing bile-acid transport matter for patients?
That is not yet known. The elevated bile salts seen with bulevirtide were reversible after treatment stopped, so whether avoiding them improves safety or adherence can only be tested in clinical trials of a compound that has not yet been tested in animals.
Could EGFR cancer drugs such as gefitinib treat hepatitis B instead?
Only in cultured cells so far. Gefitinib inhibited HBV infection in the 2019 PNAS work, which suggested that EGFR-targeting cancer drugs deserve study against HBV and HDV. Gefitinib blocks EGFR’s kinase signaling broadly, which is the effect Oxy229 avoided in cell tests.
When could an Oxy229-based drug reach patients?
No timeline has been announced. Kyoto’s announcement says the results should support future drug design, the preprint lists pharmacokinetic and in vivo studies as steps still ahead, and the authors say they plan to optimize antiviral activity.
AloJapan.com