Reply: To PMID 24027047.

Abstract

This work was supported in part by funding from the National Institutes of Health (R01 CA027607 and P01 AG034906) to ASL. Potential conflict of interest: Nothing to report. We thank Li et al. for their interest in our study on liver‐specific deletion of Grp94 and the Pten tumor‐suppressor gene using Cre‐recombinase driven by the albumin promoter (cPtenf/fGrp94f/f), which led to hyperproliferation of liver progenitor cells (LPCs), impaired cell adhesion, and accelerated development of both hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC).1 Given that deregulated expansion of LPCs has been shown to acquire liver tumor‐initiating ability in vivo,2 it is important to resolve whether GRP94 acts as a tumor suppressor in the liver. Thus, in a follow‐up study,3 we monitored the incidence of spontaneous liver tumor formation in liver‐specific Grp94 single‐knockout mice (cGrp94f/f). In contrast to wild‐type (WT) livers that were tumor free at least up to 2 years, cGrp94f/f livers at 15 months showed expanded LPCs and abnormal small nodules; and by 21 months, HCC development was evident, as well as ductal reactions.3 As the mice aged, in agreement with observations by us and others that gene knockout by the Alb‐Cre system may not be stably maintained,4cGrp94f/f livers were progressively repopulated by GRP94‐positive hepatocytes, associating with increased proliferation.3 Because HCC observed in these mice was mostly GRP94 positive, HCC development in aged cGrp94f/f mice is not the result of the lack of GRP94; rather, it is likely caused by some cell autonomous and/or nonautonomous events triggered by GRP94 deficiency, coupled with other carcinogenic events during the aging process. The issue remains of whether GRP94 overexpression, as evident in human HCC,6 is obligatory for HCC development. In cPtenf/fGrp94f/f tumors, we only observed partial repopulation of GRP94‐positive cells at 8‐9 months (Fig. 1A). Triple staining of liver sections with GRP94, HCC marker HepPar1, and CC marker panCK further revealed heterogeneous GRP94 expression from very low to WT level in cPtenf/fGrp94f/f HCC cells (Fig. 1B). Thus, though the pro‐oncogenic functions of GRP94 could confer growth and survival advantage to HCC cells, high GRP94 expression does not appear to be a requirement for HCC development. Notably, CC cells in cPtenf/fGrp94f/f livers remained mostly GRP94 negative (Fig. 1A,B); so, whether GRP94 is a suppressor for CC awaits resolution. Thus, whereas, on one hand, GRP94 maintains cell adhesion and stem cell quiescence, which are tumor‐suppressing factors, on the other hand, as a stress chaperone and regulator of pro‐oncogenic signaling pathways, it can also be supportive of tumorigenesis.6 The multifaceted role of GRP94 in cancer merits further investigation.Figure 1: Heterogeneous GRP94 expression in cPten f/f Grp94 f/f tumors at 8‐9 months. (A) Hematoxylin and eosin (H&E) and GRP94 immunohistochemical staining (brown) in WT livers as well as HCC and CC in cPten f/f Grp94 f/f livers. Black and white arrows indicate examples of GRP94‐positive and ‐negative cells, respectively. (B) Triple immunofluorescent staining with GRP94 (red), CC marker panCK (green), and HCC marker HepPar1 (blue) in liver of WT mice and tumors of cPten f/f Grp94 f/f mice. Nuclei were stained with 4',6‐diamidino‐2‐phenylindole (blue). Scale bars: 100 μm.