Hepatocyte levels of CD73 contribute to Mallory-Denk body formation


  • Potential conflict of interest: Nothing to report.

To the Editor:

Snider et al.[1] reported low levels of CD73 in hepatitis C and nonalcoholic steatohepatitis (NASH) livers, where Mallory-Denk bodies (MDBs) formed. In a mouse model (C57BL), low levels of CD73 developed when fed 3,5-diethoxycarbonyl-1,4-dihydrocolidine (DDC) and MDBs formed. Thus, both human and mouse livers with reduced levels of CD73 correlated with increased MDB formation. Paradoxically, Snider et al.[1] observed that CD73−/− mice fed DDC failed to form MDBs. They offer no mechanism to explain this paradox. “If low levels of CD73 increase MDB formation, isn't it logical to expect that, in the absence of CD73 in CD73−/− mice, they would form even more MDBs?” They do show that some of the CD73−/− mice fed DDC formed a lower amount of high molecular weight (HMW) proteins and ubiquitinated HMW proteins (their fig. 4D). The normal K8 blot, on the other hand, was greatly increased by DDC feeding in the CD73−/− mice. This did not occur in the wild-type mice where MDBs were formed. These paradoxical observations are hard to reconcile. One explanation could be that low levels of CD73 somehow causes a decrease in 26S proteasome activity. However, the absence of CD73 does the opposite, that is, it prevents the loss of 26s proteasome activity. MDBs form when a shift occurs in the formation of the 26s proteasome. This shift is to the formation of the immunoproteasome.[2, 3] The shift to form the immunoproteasome is the result of the response of interferon (IFN) sequence response element on the UbD (FAT10 in humans) promoter stimulated by IFNγ and tumor necrosis factor alpha (TNFα).[4] UbD is markedly induced in the DDC mouse model of MBD formation.[5] SAMe prevents this DDC induction of UbD, MDB formation, and the induction of immunoproteasome formation.[3, 5] UbD−/− mice fed DDC cannot form MDBs.[6]

  • Samuel French, M.D.

  • Harbor-UCLA Medical Center

  • Torrance, CA