Allison J. Kwong

Downregulation of Hepatic Stellate Cell Activation by Retinol and Palmitate Mediated by Adipose Differentiation-Related Protein (ADRP)

Hepatic stellate cells (HSCs) store retinoids and triacylglycerols in cytoplasmic lipid droplets. Two prominent features of HSC activation in liver fibrosis are loss of lipid droplets along with increase of α‐smooth muscle actin (α‐SMA), but the link between these responses and HSC activation remains elusive. In non‐adipose cells, adipose differentiation‐related protein (ADRP) coats lipid droplets and regulates their formation and lipolysis; however its function in HSCs is unknown. Here, we observed, in human liver sections or primary HSC culture, ADRP localization to lipid droplets of HSCs, and reduced staining coincident with loss of lipid droplets in liver fibrosis and in culture‐activated HSCs, consistent with HSC activation. In the LX‐2 human immortalized HSCs, with scant lipid droplets and features of activated HSCs, we found that the upregulation of ADRP mRNA by palmitate is potentiated by retinol, accompanied by increased ADRP protein, generation of retinyl palmitate, and lipid droplet formation. ADRP induction also led to decreased expression of α‐SMA mRNA and its protein, while ADRP knockdown with small interfering RNA (siRNA) normalized α‐SMA expression. Furthermore, ADRP induction by retinol and palmitate resulted in decreased expression of collagen I and matrix metalloproteinase‐2 mRNA, fibrogenic genes associated with activated HSCs, while increasing matrix metalloproteinase‐1 mRNA; ADRP knockdown with siRNA reversed these changes. Tissue inhibitor of metalloproteinase‐1 was not affected. Thus, ADRP upregulation mediated by retinol and palmitate promotes downregulation of HSC activation and is functionally linked to the expression of fibrogenic genes. J. Cell. Physiol. 223:648–657, 2010.

Reduced hepatic stellate cell expression of kruppel-like factor 6 tumor suppressor isoforms amplifies fibrosis during acute and chronic rodent liver injury†‡

Kruppel‐like factor 6 (KLF6), a zinc finger transcription factor and tumor suppressor, is induced as an immediate‐early gene during hepatic stellate cell (HSC) activation. The paradoxical induction of a tumor suppressor in HSCs during proliferation led us to explore the biology of wildtype KLF6 (KLF6WT) and its antagonistic, alternatively spliced isoform KLF6SV1 in cultured HSCs and animal models. The animal models generated include a global heterozygous KLF6 mouse (Klf6+/−), and transgenic mice expressing either hKLF6WT or hKLF6SV1 under the control of the Collagen α2 (I) promoter to drive HSC‐specific gene expression following injury. The rat Klf6 transcript has multiple splice forms that are homologous to those of the human KLF6 gene. Following a transient increase, all rat Klf6 isoforms decreased in response to acute carbon tetrachloride (CCl4) liver injury and culture‐induced activation. After acute CCl4, Klf6+/− mice developed significantly increased fibrosis and enhanced fibrogenic messenger RNA (mRNA) and protein expression. In contrast, HSC‐specific transgenic mice overexpressing KLF6WT or KLF6SV1 developed significantly diminished fibrosis with reduced expression of fibrogenic genes. Chromatin IP and quantitative reverse‐transcription polymerase chain reaction in mouse HSCs overexpressing KLF6WT demonstrated KLF6WT binding to GC boxes in promoters of Colα1 (I), Colα2 (I), and beta‐platelet‐derived growth factor receptor (β‐Pdgfr) with reduced gene expression, consistent with transcriptional repression by KLF6. Stellate cells overexpressing either KLF6WT or KLF6SV1 were more susceptible to apoptotic stress based on poly (ADP‐ribose) polymerase (PARP) cleavage. Conclusion: KLF6 reduces fibrogenic activity of HSCs by way of two distinct mechanisms, direct transcriptional repression of target fibrogenic genes and increased apoptosis of activated HSCs. These results suggest that following its initial induction, sustained down‐regulation of KLF6 in liver injury may allow de‐repression of fibrogenic genes and decreased stellate cell clearance by inhibiting apoptosis. (HEPATOLOGY 2013)

Liver Fibrosis in Elderly Cadavers: Localization of Collagen Types I, III, and IV, α‐Smooth Muscle Actin, and Elastic Fibers

We have shown a high prevalence of liver fibrosis in elderly cadavers with diverse causes of death by Sirius red stain; however, the various collagen types in these samples have yet to be evaluated. To further characterize the histopathology of the fibrotic lesions in the livers of these elderly cadavers, this study used immunohistochemistry and histochemistry to identify the principal collagens produced in liver fibrosis, fibrogenic cells and elastic fibers. Collagen I and III immunoreactions were found to colocalize in collagen fibers of fibrotic central veins, perisinusoidal fibrotic foci, portal tract stroma, and fibrous septa. α‐Smooth muscle actin‐expressing perisinusoidal hepatic stellate cells (HSCs), as well as perivenular, portal, and septal myofibroblasts, were closely associated with collagen fibers, reflecting their fibrogenic functions. HSCs and myofibroblasts were also noted to express collagen IV, which may contribute to production of basal lamina‐like structures. In fibrotic livers, the sinusoidal lining showed variable immunostaining for collagen IV. Collagen IV immunostaining revealed vascular proliferation and atypical ductular reaction at the portal–septal parenchymal borders, as well as capillary‐like vessels in the lobular parenchyma. While elastic fibers were absent in the space of Disse, they were found to codistribute with collagens in portal tracts, fibrous septa and central veins. Our combined assessment of collagen types, HSCs, myofibroblasts, and elastic fibers is significant in understanding the histopathology of fibrosis in the aging liver. Anat Rec, 2012.

Hepatic Steatosis, Fibrosis, and Cancer in Elderly Cadavers

The incidence of hepatic steatosis, fibrosis, and cancer in the elderly population remains unknown. Human cadavers used in anatomy teaching, which come largely from older adults, may provide liver tissue for examining their pathologies. Livers were obtained from 68 cadavers (mean age 82.1 ± 10.4 years) with diverse causes of death in the Anatomy course at Mount Sinai School of Medicine. Paraffin sections were stained with hematoxylin and eosin and Sirius red and evaluated for steatosis, fibrosis, cancer, and lipofuscin. Tissue preservation was graded as good in 38.2% of the embalmed livers, fair in 36.7%, and poor in 25.0%. Steatosis was observed in 35.3% of the livers, central vein fibrosis in 49.2%, perisinusoidal fibrosis in 63.2%, portal tract (PT) fibrosis in 47.0%, septa formation in 44.1%, bridging fibrosis in 30.8%, and cirrhosis in 4.4%. One hepatocellular carcinoma (HCC) and six metastatic tumors were detected. Lobular inflammation occurred in 20.8% of the livers and PT inflammation in 38.8%. Nine livers showed minimal change. Lipofuscin was detected in 60.2% of the livers. Steatosis, fibrosis, and cancer are highly prevalent in elderly cadavers with diverse causes of death. The prevalence of steatosis and fibrosis is consistent with the data in patients with specific liver diseases. Steatosis alongside fibrosis can accelerate the progression of fibrosis to cirrhosis and ultimately HCC. Though not indicated as the primary cause of death, the liver injury may have compromised hepatic functions and enhanced disease susceptibility, thereby exacerbating the health conditions in this elderly population. Anat Rec, 2012.

Outcomes for liver transplant candidates listed with low model for end‐stage liver disease score

The Model for End‐Stage Liver Disease (MELD) score, which estimates mortality within 90 days, determines priority for liver transplantation (LT). However, longer‐term outcomes on the wait list for patients who are initially listed with low MELD scores are not well characterized. All adults listed for primary LT at a single, high‐volume center from 2005 to 2012 with an initial laboratory MELD score of 22 or lower were evaluated. Excluded were those patients listed with MELD exception points who underwent living donor liver transplantation (LDLT) or transplantation at another center, or who were removed from the wait list for nonmedical reasons. Outcomes and causes of death were identified by United Network for Organ Sharing, the National Death Index, and an electronic medical record review. Multivariate competing risk analysis evaluated predictors of death compared to deceased donor liver transplantation (DDLT); 893 patients were listed from 2005 to 2012. By the end of follow‐up, 27% had undergone DDLT, and 31% were removed from the wait list for death or clinical deterioration. In a competing risks assessment, only MELD score of 6‐9, older age, lower serum albumin, lower body mass index, and diabetes conferred an increased risk of wait‐list dropout compared to DDLT. Listing for simultaneous liver‐kidney transplantation was protective against wait‐list dropout. Of the patients included, 275 patients died or were delisted for being too sick; 87% of the identifiable causes of death were directly related to end‐stage liver disease or hepatocellular carcinoma. In conclusion, patients with low listing MELD scores remain at a significant risk for death due to liver‐related causes and may benefit from early access to transplantation, such as LDLT or acceptance of high‐risk donor livers. Predictors of death compared to transplantation may allow for early identification of patients who are at risk for wait‐list mortality. Liver Transpl 21:1403‐1409, 2015.

Evaluation of a Resident-Led Project to Decrease Phlebotomy Rates in the Hospital: Think Twice, Stick Once

LESS IS MORE Evaluation of a Resident-Led Project to Decrease Phlebotomy Rates in the Hospital: Think Twice, Stick Once Excessive inpatient laboratory testing leads to unnecessary health care cost and exposes patients to uncomfortable blood draws, false-positive results, and hospital-acquired anemia.1,2 Many strategies have sought to decrease laboratory testing in the hospital,3 but most do not focus on decreasing needlesticks for patients. In July 2014, we launched the resident-led Think Twice, Stick Once program on the Internal Medicine teaching service at the University of California, San Francisco (UCSF) Medical Center to reduce the number of phlebotomies per patient per day by at least 5%.

Improved posttransplant mortality after share 35 for liver transplantation

The Share 35 policy was implemented in June 2013 to improve equity in access to liver transplantation (LT) between patients with fulminant liver failure and those with cirrhosis and severe hepatic decompensation. The aim of this study was to assess post‐LT outcomes after Share 35. Relevant donor, procurement, and recipient data were extracted from the Organ Procurement and Transplantation Network/United Network for Organ Sharing database. All adult deceased donor LTs from January 1, 2010, to March 31, 2016, were included in the analysis. One‐year patient survival before and after Share 35 was assessed by multivariable Cox proportional hazards analysis, with adjustment for variables known to affect graft survival. Of 34,975 adult LT recipients, 16,472 (47.1%) were transplanted after the implementation of Share 35, of whom 4,599 (27.9%) had a Model for End‐Stage Liver Disease (MELD) score ≥35. One‐year patient survival improved from 83.9% to 88.4% after Share 35 (P < 0.01) for patients with MELD ≥35. There was no significant impact on survival of patients with MELD <35 (P = 0.69). Quality of donor organs, as measured by a donor risk index without the regional share component, improved for patients with MELD ≥35 (P < 0.01) and worsened for patients with lower MELD (P < 0.01). In multivariable Cox regression analysis, Share 35 was associated with improved 1‐year patient survival (hazard ratio, 0.69; 95% confidence interval, 0.60‐0.80) in recipients with MELD ≥35. Conclusion: Share 35 has had a positive impact on survival after transplantation in patients with MELD ≥35, without a reciprocal detriment in patients with lower acuity; this was in part a result of more favorable donor–recipient matching. (Hepatology 2018;67:273‐281).

Decreasing mortality and disease severity in hepatitis C patients awaiting liver transplantation in the United States

Hepatitis C virus (HCV) infection has been the leading indication for liver transplantation (LT) in the United States. Since 2013, interferon‐free antiviral therapy has led to sustained virological response in many LT candidates. We compared the wait‐list mortality of HCV patients with that of patients with other chronic liver diseases. Data for primary LT candidates were obtained from the Organ Procurement and Transplantation Network database. Adult wait‐list registrants were divided into 3 cohorts: cohort 1 included patients on the waiting list as of January 1, 2004; cohort 2 as of January 1, 2009; and cohort 3 as of January 1, 2014. The primary outcome was wait‐list mortality, and the secondary outcome was the rate of change in Model for End‐Stage Liver Disease (MELD). Multivariate Cox proportional hazards analysis was performed to evaluate 12‐month wait‐list mortality. The cohorts included 7627 LT candidates with HCV and 13,748 patients without HCV. Compared with cohort 2, HCV patients in cohort 3 had a 21% lower risk of death (hazard ratio [HR], 0.79; 95% confidence interval [CI], 0.67‐0.93). Among patients with non‐HCV liver disease, no difference in mortality was seen between cohorts 2 and 3 (HR, 0.97; 95% CI, 0.86‐1.09). Among HCV patients, the mean rate of change in MELD decreased from 2.35 per year for cohort 2 to 1.90 per year for cohort 3, compared with 1.90 and 1.66 in cohorts 2 and 3, respectively, among non‐HCV patients. In this population‐based study, wait‐list mortality and progression of disease severity decreased in recent HCV patients for whom direct‐acting antiviral agents were available. Liver Transplantation 24 735–743 2018 AASLD.

Use of hepatitis C viraemic organs in kidney transplantation: a need to hit the pause button?

Innovative solutions are needed to overcome the global disparity in patients awaiting kidney transplantation versus donor organs available. A new study reports a promising new strategy of transplanting kidneys from HCV-infected donors into HCV-uninfected recipients and treating their HCV with direct-acting antivirals post-transplant — recipients achieved HCV cure with excellent one-year kidney allograft function.

Artificial neural networks and liver transplantation: Are we ready for self‐driving cars?

The persistent shortage of donor organs has motivated efforts to predict posttransplant outcome, maximize survival benefit, and encourage the appropriate allocation of donor organs. Machine learning, a branch of statistics and computer science that has revolutionized the analysis of large and complex data sets, and its application to medicine has accelerated in recent years. Not only have machine-learning algorithms been used to develop the technology for Google Home, Siri, and self-driving cars, they have also been used to predict hospitalization in patients with heart failure, remission in patients with inflammatory bowel disease, and graft failure after solid organ transplantation. In this issue of Liver Transplantation, Ayll on et al. applied an artificial neural network (ANN), a type of machine-learning algorithm, to a cohort of 858 liver transplantation (LT) recipients from 2002 to 2010 at King’s College Hospital (KCH) using 55 donor, procurement, and recipient variables. This ANN had been previously used to develop a model in a Spanish multicenter cohort (Model for Allocation of Donor and Recipient in Espa~ na [MADR-E]) of 1003 LT recipients with an area under the curve (AUC), or c-statistic, of 0.82 for 3-month graft failure. The KCH model was developed using the same ANN architecture (although the predictive model was different) used for the MADR-E study, which is designed to maximize the accuracy of graft survival by the correct classification rate (CCR) and the sensitivity of graft failure by minimum sensitivity (MS). In the KCH cohort, the ANN resulted in an AUC of 0.94 for 3month graft failure, superior to other published models. This result implies that for 2 randomly drawn grafts, 1 of graft survival and 1 of graft failure, the model would correctly assign higher 3-month graft risk to the failed graft 94% of the time—compared with AUCs of 0.73 for donor Model for End-Stage Liver Disease (DMELD), 0.82 for survival outcomes following liver transplantation (SOFT), and 0.84 for balance of risk (BAR), in this cohort. The ANN was also trained to predict 1-year graft failure and resulted in an AUC of 0.82, also superior to previously developed scores (AUCs of 0.63 for D-MELD, 0.65 for SOFT, and 0.71 for BAR). Variables reported to have the highest weight in the KCH model included pretransplant status, Model for End-Stage Liver Disease (MELD) at transplant, time on waiting list, etiology of liver disease, cold ischemia time, and donor hypertension, cause of death, and aspartate aminotransferase (AST) levels. The authors illustrated examples in which organ acceptance can be guided using the probability of graft survival and nonsurvival in combination with the MELD of a potential recipient. A donor-recipient match is chosen if their predicted 3-month graft survival by the CCR model is >3% of a competing match or if their predicted graft failure by the MS model is >5%—these cutoffs represent approximately 1 standard deviation in the Abbreviations: ANN, artificial neural network; AST, aspartate aminotransferase; AUC, area under the curve; BAR, balance of risk; CCR, correct classification rate; D-MELD, donor Model for EndStage Liver Disease; KCH, King’s College Hospital; LT, liver transplantation; MADR-E, Model for Allocation of Donor and Recipient in Espa~ na; MELD, Model for End-Stage Liver Disease; MS, minimum sensitivity; SOFT, survival outcomes following liver transplantation.