Prior tonsillectomy is associated with an increased risk of esophageal adenocarcinoma

BACKGROUND

Esophageal cancer is a deadly cancer with 5-year survival <20%. Although multiple risk factors for esophageal adenocarcinoma (EAC) including obesity, GERD and smoking have been identified, these risk factors do not fully explain the rising incidence of EAC. In this study, we evaluated the association between prior history of tonsillectomy and EAC. Our goal was to determine whether tonsillectomies were more frequent in patients with EAC (cases) than in our thoracic surgery controls.

METHODS

Cases included 452 esophagectomy cases, including 396 with EAC and 56 who underwent esophagectomy for Barrett's esophagus (BE) with high grade dysplasia (HGD). 1,102 thoracic surgery patients with surgical indications other than dysplastic BE or esophageal cancer represented the controls for our analysis. The association of tonsillectomy and HGD/EAC were primarily evaluated by using univariate tests and then verified by logistic regression analysis. Baseline demographics, medical history, and thoracic surgery controls were compared by using χ2 tests or 95% CIs. Significant risk factors were considered as covariates in the multivariate models while evaluating the association between tonsillectomy and HGD/EAC. P-values or odds ratios were estimated with 95% confidence limits to identify significances which was more appropriate.

RESULTS

Tonsillectomy was more common in cases than controls and was found to have a significant association with esophageal cancer (19.9% vs. 12.7%; p-value = 0.0003). This significant association persisted after controlling for other known risk factors/covariates.

CONCLUSION

A prior history of tonsillectomy was significantly associated with HGD/EAC and may represent an independent risk factor for the development of EAC. However, the underlying biology driving this association remains unclear.

Plasmonic nanobiosensors for detection of microRNA cancer biomarkers in clinical samples

MicroRNAs (miRNAs) play an important role in the regulation of biological processes and have demonstrated great potential as biomarkers for the early detection of various diseases, including esophageal adenocarcinoma (EAC) and Barrett's esophagus (BE), the premalignant metaplasia associated with EAC. Herein, we demonstrate the direct detection of the esophageal cancer biomarker, miR-21, in RNA extracted from 17 endoscopic tissue biopsies using the nanophotonics technology our group has developed, termed the inverse molecular sentinel (iMS) nanobiosensor, with surface-enhanced Raman scattering (SERS) detection. The potential of this label-free, homogeneous biosensor for cancer diagnosis without the need for target amplification was demonstrated by discriminating esophageal cancer and Barrett's esophagus from normal tissue with notable diagnostic accuracy. This work establishes the potential of the iMS nanobiosensor for cancer diagnostics via miRNA detection in clinical samples without the need for target amplification, validating the potential of this assay as part of a new diagnostic strategy. Combining miRNA diagnostics with the nanophotonics technology will result in a paradigm shift in achieving a general molecular analysis tool that has widespread applicability for cancer research as well as detection of cancer. We anticipate further development of this technique for future use in point-of-care testing as an alternative to histopathological diagnosis as our method provides a quick result following RNA isolation, allowing for timely treatment.

A cancer rainbow mouse for visualizing the functional genomics of oncogenic clonal expansion

Field cancerization is a premalignant process marked by clones of oncogenic mutations spreading through the epithelium. The timescales of intestinal field cancerization can be variable and the mechanisms driving the rapid spread of oncogenic clones are unknown. Here we use a Cancer rainbow (Crainbow) modelling system for fluorescently barcoding somatic mutations and directly visualizing the clonal expansion and spread of oncogenes. Crainbow shows that mutations of ß-catenin (Ctnnb1) within the intestinal stem cell results in widespread expansion of oncogenes during perinatal development but not in adults. In contrast, mutations that extrinsically disrupt the stem cell microenvironment can spread in adult intestine without delay. We observe the rapid spread of premalignant clones in Crainbow mice expressing oncogenic Rspondin-3 (RSPO3), which occurs by increasing crypt fission and inhibiting crypt fixation. Crainbow modelling provides insight into how somatic mutations rapidly spread and a plausible mechanism for predetermining the intratumor heterogeneity found in colon cancers.

The enteric microbiota regulates jejunal Paneth cell number and function without impacting intestinal stem cells

Paneth cells (PCs) are epithelial cells found in the small intestine, next to intestinal stem cells (ISCs) at the base of the crypts. PCs secrete antimicrobial peptides (AMPs) that regulate the commensal gut microbiota. In contrast, little is known regarding how the enteric microbiota reciprocally influences PC function. In this study, we sought to characterize the impact of the enteric microbiota on PC biology in the mouse small intestine. This was done by first enumerating jejunal PCs in germ-free (GF) versus conventionally raised (CR) mice. We next evaluated the possible functional consequences of altered PC biology in these experimental groups by assessing epithelial proliferation, ISC numbers, and the production of AMPs. We found that PC numbers were significantly increased in CR versus GF mice; however, there were no differences in ISC numbers or cycling activity between groups. Of the AMPs assessed, only Reg3γ transcript expression was significantly increased in CR mice. Intriguingly, this increase was abrogated in cultured CR versus GF enteroids, and could not be re-induced with various bacterial ligands. Our findings demonstrate the enteric microbiota regulates PC function by increasing PC numbers and inducing Reg3γ expression, though the latter effect may not involve direct interactions between bacteria and the intestinal epithelium. In contrast, the enteric microbiota does not appear to regulate jejunal ISC census and proliferation. These are critical findings for investigators using GF mice and the enteroid system to study PC and ISC biology.

Ductular and proliferative response of esophageal submucosal glands in a porcine model of esophageal injury and repair

Esophageal injury is a risk factor for diseases such as Barrett's esophagus (BE) and esophageal adenocarcinoma. To improve understanding of signaling pathways associated with both normal and abnormal repair, animal models are needed. Traditional rodent models of esophageal repair are limited by the absence of esophageal submucosal glands (ESMGs), which are present in the human esophagus. Previously, we identified acinar ductal metaplasia in human ESMGs in association with both esophageal injury and cancer. In addition, the SOX9 transcription factor has been associated with generation of columnar epithelium and the pathogenesis of BE and is present in ESMGs. To test our hypothesis that ESMGs activate after esophageal injury with an increase in proliferation, generation of a ductal phenotype, and expression of SOX9, we developed a porcine model of esophageal injury and repair using radiofrequency ablation (RFA). The porcine esophagus contains ESMGs, and RFA produces a consistent and reproducible mucosal injury in the esophagus. Here we present a temporal assessment of this model of esophageal repair. Porcine esophagus was evaluated at 0, 6, 18, 24, 48, and 72 h and 5 and 7 days following RFA and compared with control uninjured esophagus. Following RFA, ESMGs demonstrated an increase in ductal phenotype, echoing our prior studies in humans. Proliferation increased in both squamous epithelium and ESMGs postinjury with a prominent population of SOX9-positive cells in ESMGs postinjury. This model promises to be useful in future experiments evaluating mechanisms of esophageal repair.NEW & NOTEWORTHY A novel porcine model of injury and repair using radiofrequency ablation has been developed, allowing for reproducible injury to the esophagus to study repair in an animal model with esophageal submucosal glands, a key anatomical feature and missing in rodent models but possibly harboring progenitor cells. There is a strong translational component to this porcine model given the anatomical and physiological similarities between pigs and humans.

Porcine Esophageal Submucosal Gland Culture Model Shows Capacity for Proliferation and Differentiation

BACKGROUND & AIMS

Although cells comprising esophageal submucosal glands (ESMGs) represent a potential progenitor cell niche, new models are needed to understand their capacity to proliferate and differentiate. By histologic appearance, ESMGs have been associated with both overlying normal squamous epithelium and columnar epithelium. Our aim was to assess ESMG proliferation and differentiation in a 3-dimensional culture model.

METHODS

We evaluated proliferation in human ESMGs from normal and diseased tissue by proliferating cell nuclear antigen immunohistochemistry. Next, we compared 5-ethynyl-2'-deoxyuridine labeling in porcine ESMGs in vivo before and after esophageal injury with a novel in vitro porcine organoid ESMG model. Microarray analysis of ESMGs in culture was compared with squamous epithelium and fresh ESMGs.

RESULTS

Marked proliferation was observed in human ESMGs of diseased tissue. This activated ESMG state was recapitulated after esophageal injury in an in vivo porcine model, ESMGs assumed a ductal appearance with increased proliferation compared with control. Isolated and cultured porcine ESMGs produced buds with actively cycling cells and passaged to form epidermal growth factor-dependent spheroids. These spheroids were highly proliferative and were passaged multiple times. Two phenotypes of spheroids were identified: solid squamous (P63+) and hollow/ductal (cytokeratin 7+). Microarray analysis showed spheroids to be distinct from parent ESMGs and enriched for columnar transcripts.

CONCLUSIONS

Our results suggest that the activated ESMG state, seen in both human disease and our porcine model, may provide a source of cells to repopulate damaged epithelium in a normal manner (squamous) or abnormally (columnar epithelium). This culture model will allow the evaluation of factors that drive ESMGs in the regeneration of injured epithelium. The raw microarray data have been uploaded to the National Center for Biotechnology Information Gene Expression Omnibus (accession number: GSE100543).

Intestinal Enteroendocrine Lineage Cells Possess Homeostatic and Injury-Inducible Stem Cell Activity

Several cell populations have been reported to possess intestinal stem cell (ISC) activity during homeostasis and injury-induced regeneration. Here, we explored inter-relationships between putative mouse ISC populations by comparative RNA-sequencing (RNA-seq). The transcriptomes of multiple cycling ISC populations closely resembled Lgr5+ ISCs, the most well-defined ISC pool, but Bmi1-GFP+ cells were distinct and enriched for enteroendocrine (EE) markers, including Prox1. Prox1-GFP+ cells exhibited sustained clonogenic growth in vitro, and lineage-tracing of Prox1+ cells revealed long-lived clones during homeostasis and after radiation-induced injury in vivo. Single-cell mRNA-seq revealed two subsets of Prox1-GFP+ cells, one of which resembled mature EE cells while the other displayed low-level EE gene expression but co-expressed tuft cell markers, Lgr5 and Ascl2, reminiscent of label-retaining secretory progenitors. Our data suggest that the EE lineage, including mature EE cells, comprises a reservoir of homeostatic and injury-inducible ISCs, extending our understanding of cellular plasticity and stemness.

GI stem cells - new insights into roles in physiology and pathophysiology

This overview gives a brief historical summary of key discoveries regarding stem cells of the small intestine. The current concept is that there are two pools of intestinal stem cells (ISCs): an actively cycling pool that is marked by Lgr5, is relatively homogeneous and is responsible for daily turnover of the epithelium; and a slowly cycling or quiescent pool that functions as reserve ISCs. The latter pool appears to be quite heterogeneous and may include partially differentiated epithelial lineages that can reacquire stem cell characteristics following injury to the intestine. Markers and methods of isolation for active and quiescent ISC populations are described as well as the numerous important advances that have been made in approaches to the in vitro culture of ISCs and crypts. Factors regulating ISC biology are briefly summarized and both known and unknown aspects of the ISC niche are discussed. Although most of our current knowledge regarding ISC physiology and pathophysiology has come from studies with mice, recent work with human tissue highlights the potential translational applications arising from this field of research. Many of these topics are further elaborated in the following articles.

Tissue underlying the intestinal epithelium elicits proliferation of intestinal stem cells following cytotoxic damage

The goals of this study were to document the proliferative response of intestinal stem cells (ISCs) during regeneration after damage from doxorubicin (DXR), and to characterize the signals responsible for ISC activation. To this end, jejuni from DXR-treated mice were harvested for histology, assessment of ISC numbers and proliferation by flow cytometry, crypt culture, and RNA analyses. Histology showed that crypt depth and width were increased 4 days after DXR. At this time point, flow cytometry on tissue collected 1 h after EdU administration revealed increased numbers of CD24(lo)UEA(-) ISCs and increased percentage of ISCs cycling. In culture, crypts harvested from DXR-treated mice were equally proliferative as those of control mice. Addition of subepithelial intestinal tissue (SET) collected 4 days after DXR elicited increased budding (1.4 ± 0.3 vs. 5.1 ± 1.0 buds per enteroid). Microarray analysis of SET collected 4 days after DXR revealed 1030 differentially expressed transcripts. Cross-comparison of Gene Ontology terms considered relevant to ISC activation pointed to 10 candidate genes. Of these, the epidermal growth factor (EGF) family member amphiregulin and the BMP antagonist chordin-like 2 were chosen for further study. In crypt culture, amphiregulin alone did not elicit significant budding, but amphiregulin in combination with BMP antagonism showed marked synergism (yielding 6.3 ± 0.5 buds per enteroid). These data suggest a critical role for underlying tissue in regulating ISC behavior after damage, and point to synergism between amphiregulin and chordin-like 2 as factors which may account for activation of ISCs in the regenerative phase.

Mouse Paneth cell antimicrobial function is independent of Nod2

OBJECTIVE

Although polymorphisms of the NOD2 gene predispose to the development of ileal Crohn's disease, the precise mechanisms of this increased susceptibility remain unclear. Previous work has shown that transcript expression of the Paneth cell (PC) antimicrobial peptides (AMPs) α-defensin 4 and α-defensin-related sequence 10 are selectively decreased in Nod2(-/-) mice. However, the specific mouse background used in this previous study is unclear. In light of recent evidence suggesting that mouse strain strongly influences PC antimicrobial activity, we sought to characterise PC AMP function in commercially available Nod2(-/-) mice on a C57BL/6 (B6) background. Specifically, we hypothesised that Nod2(-/-) B6 mice would display reduced AMP expression and activity.

DESIGN

Wild-type (WT) and Nod2(-/-) B6 ileal AMP expression was assessed via real-time PCR, acid urea polyacrylamide gel electrophoresis and mass spectrometry. PCs were enumerated using flow cytometry. Functionally, α-defensin bactericidal activity was evaluated using a gel-overlay antimicrobial assay. Faecal microbial composition was determined using 454-sequencing of the bacterial 16S gene in cohoused WT and Nod2(-/-) littermates.

RESULTS

WT and Nod2(-/-) B6 mice displayed similar PC AMP expression patterns, equivalent α-defensin profiles, and identical antimicrobial activity against commensal and pathogenic bacterial strains. Furthermore, minimal differences in gut microbial composition were detected between the two cohoused, littermate mouse groups.

CONCLUSIONS

Our data reveal that Nod2 does not directly regulate PC antimicrobial activity in B6 mice. Moreover, we demonstrate that previously reported Nod2-dependent influences on gut microbial composition may be overcome by environmental factors, such as cohousing with WT littermates.

Side population sorting separates subfractions of cycling and non-cycling intestinal stem cells

We report here that side population (SP) sorting allows for the simultaneous isolation of two intestinal stem cell (ISC) subsets from wild-type (WT) mice which are phenotypically different and represent cycling and non-cycling pools of cells. Following 5-ethynyl-2'-deoxyuridine (EdU) injection, in the upper side population (USP) the percentage of EdU+ was 36% showing this fraction to be highly proliferative. In the lower side population (LSP), only 0.4% of cells were EdU+, indicating this fraction to be predominantly non-cycling. Using Lgr5-EGFP mice, we show that Lgr5-EGFP(hi) cells, representing actively cycling ISCs, are essentially exclusive to the USP. In contrast, using histone 2B-YFP mice, SP analysis revealed YFP label retaining cells (LRCs) in both the USP and the LSP. Correspondingly, evaluation of the SP fractions for mRNA markers by qRT-PCR showed that the USP was enriched in transcripts associated with both quiescent and active ISCs. In contrast, the LSP expressed mRNA markers of quiescent ISCs while being de-enriched for those of the active ISC. Both the USP and LSP are capable of generating enteroids in culture which include the four intestinal lineages. We conclude that sorting of USP and LSP fractions represents a novel isolation of cycling and non-cycling ISCs from WT mice.

Intestinal stem cells remain viable after prolonged tissue storage

Intestinal stem cells (ISCs) are responsible for renewal of the epithelium both during normal homeostasis and following injury. As such, they have significant therapeutic potential. However, whether ISCs can survive tissue storage is unknown. We hypothesize that, although the majority of epithelial cells might die, ISCs would remain viable for at least 24 h at 4 °C. To explore this hypothesis, jejuna of C57Bl6/J or Lgr5-LacZ mice were removed and either processed immediately or placed in phosphate-buffered saline at 4 °C. Delayed isolation of epithelium was performed after 24, 30, or 48 h storage. At the light microscope level, despite extensive apoptosis of villus epithelial cells, small intestinal crypts remained morphologically intact for 30 h and ISCs were identifiable via Lgr5-LacZ positivity. Electron microscopy showed that ISCs retained high integrity for 24 h. When assessed by flow cytometry, ISCs were more resistant to degeneration than the rest of the epithelium, including neighboring Paneth cells, with higher viability across all time points. Cultured isolated crypts showed no loss of capacity to form complex enteroids after 24 h tissue storage, with efficiencies after 7 days of culture remaining above 80 %. By 30 h storage, efficiencies declined but budding capability was retained. We conclude that, with delay in isolation, ISCs remain viable and retain their proliferative capacity. In contrast, the remainder of the epithelium, including the Paneth cells, exhibits degeneration and programmed cell death. If these findings are recapitulated in human tissue, storage at 4 °C might offer a valuable temporal window for the harvesting of crypts or ISCs for therapeutic application.

CD24 can be used to isolate Lgr5+ putative colonic epithelial stem cells in mice

A growing body of evidence has implicated CD24, a cell-surface protein, as a marker of colorectal cancer stem cells and target for antitumor therapy, although its presence in normal colonic epithelium has not been fully characterized. Previously, our group showed that CD24-based cell sorting can be used to isolate a fraction of murine small intestinal epithelial cells enriched in actively cycling stem cells. Similarly, we hypothesized that CD24-based isolation of colonic epithelial cells would generate a fraction enriched in actively cycling colonic epithelial stem cells (CESCs). Immunohistochemistry performed on mouse colonic tissue showed CD24 expression in the bottom half of proximal colon crypts and the crypt base in the distal colon. This pattern of distribution was similar to enhanced green fluorescent protein (EGFP) expression in Lgr5-EGFP mice. Areas expressing CD24 contained actively proliferating cells as determined by ethynyl deoxyuridine (EdU) incorporation, with a distinct difference between the proximal colon, where EdU-labeled cells were most frequent in the midcrypt, and the distal colon, where they were primarily at the crypt base. Flow cytometric analyses of single epithelial cells, identified by epithelial cell adhesion molecule (EpCAM) positivity, from mouse colon revealed an actively cycling CD24(+) fraction that contained the majority of Lgr5-EGFP(+) putative CESCs. Transcript analysis by quantitative RT-PCR confirmed enrichment of active CESC markers [leucine-rich-repeat-containing G protein-coupled receptor 5 (Lgr5), ephrin type B receptor 2 (EphB2), and CD166] in the CD24(+)EpCAM(+) fraction but also showed enrichment of quiescent CESC markers [leucine-rich repeats and immunoglobin domains (Lrig), doublecortin and calmodulin kinase-like 1 (DCAMKL-1), and murine telomerase reverse transcriptase (mTert)]. We conclude that CD24-based sorting in wild-type mice isolates a colonic epithelial fraction highly enriched in actively cycling and quiescent putative CESCs. Furthermore, the presence of CD24 expression in normal colonic epithelium may have important implications for the use of anti-CD24-based colorectal cancer therapies.

Mouse background strain profoundly influences Paneth cell function and intestinal microbial composition

BACKGROUND

Increasing evidence supports the central role of Paneth cells in maintaining intestinal host-microbial homeostasis. However, the direct impact of host genotype on Paneth cell function remains unclear. Here, we characterize key differences in Paneth cell function and intestinal microbial composition in two widely utilized, genetically distinct mouse strains (C57BL/6 and 129/SvEv). In doing so, we demonstrate critical influences of host genotype on Paneth cell activity and the enteric microbiota.

METHODOLOGY AND PRINCIPAL FINDINGS

Paneth cell numbers were determined by flow cytometry. Antimicrobial peptide (AMP) expression was evaluated using quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR), acid urea-polyacrylamide gel electrophoresis, and mass spectrometry. Effects of mouse background on microbial composition were assessed by reciprocal colonization of germ-free mice from both background strains, followed by compositional analysis of resultant gut bacterial communities using terminal restriction fragment length polymorphism analysis and 16 S qPCR. Our results revealed that 129/SvEv mice possessed fewer Paneth cells and a divergent AMP profile relative to C57BL/6 counterparts. Novel 129/SvEv á-defensin peptides were identified, including Defa2/18v, Defa11, Defa16, and Defa18. Host genotype profoundly affected the global profile of the intestinal microbiota, while both source and host factors were found to influence specific bacterial groups. Interestingly, ileal α-defensins from 129/SvEv mice displayed attenuated antimicrobial activity against pro-inflammatory E. coli strains, a bacterial species found to be expanded in these animals.

CONCLUSIONS AND SIGNIFICANCE

This work establishes the important impact of host genotype on Paneth cell function and the composition of the intestinal microbiota. It further identifies specific AMP and microbial alterations in two commonly used inbred mouse strains that have varying susceptibilities to a variety of disorders, ranging from obesity to intestinal inflammation. This will be critical for future studies utilizing these murine backgrounds to study the effects of Paneth cells and the intestinal microbiota on host health and disease.

Sorting mouse jejunal epithelial cells with CD24 yields a population with characteristics of intestinal stem cells

Intestinal stem cells (ISCs) have been studied for more than three decades; however, their isolation has remained a challenge. We hypothesized that, just as for stem cells of other tissues, one or more membrane markers would allow positive selection of ISCs by antibody-based sorting. To explore this hypothesis, microarray data of putative ISC fractions generated by side population sorting and laser capture microdissection were subjected to bioinformatic analysis to identify common membrane antigens. The microarray comparison suggested CD24 as a candidate surface marker, and immunohistochemistry showed expression of CD24 in epithelial cells of crypt bases. Flow cytometry of jejunal epithelial preparations revealed a CD24(+) CD45(-) fraction comprising ∼1% of the cells. Analysis with epithelial cell adhesion molecule and CD31 confirmed that the cell preparations were epithelial and without endothelial contamination. Cycling cells identified by prior injection with 5-ethynyl-2'-deoxyuridine were found predominantly in the CD24(lo) subfraction. Transcript analysis by real-time RT-PCR showed this subfraction to be enriched in the ISC markers leucine-rich-repeat-containing G-protein-coupled receptor 5 (40-fold) and Bmi1 (5-fold), but also enriched in lysozyme (10-fold). Flow cytometry with anti-lysozyme antibodies demonstrated that Paneth cells comprise ∼30% of the CD24(lo) subfraction. Additional flow analyses with leucine-rich-repeat-containing G-protein-coupled receptor 5-enhanced green fluorescent protein (EGFP) epithelium demonstrated colocalization of EGFP(hi) and CD24(lo). In contrast, CD24 cells were negative for the quiescent ISC marker doublecortin and CaM kinase-like-1. Culture of CD24(lo) cells in Matrigel generated organoid structures, which included all four epithelial lineages, thus giving functional evidence for the presence of ISCs. We conclude that the CD24(lo) fraction of jejunal epithelium is highly enriched with cycling ISCs. This isolation method should be useful to many investigators in the field to advance both the basic understanding of ISC biology and the therapeutic applications of ISCs.

Rosiglitazone attenuates suppression of RXRalpha-dependent gene expression in inflamed liver

BACKGROUND/AIMS

A recently determined target of lipopolysaccharide (LPS) and cytokine signaling in liver is the central Type II nuclear receptor (NR) heterodimer partner, retinoid X receptor alpha (RXRalpha). We sought to determine if Rosiglitazone (Rosi), a peroxisome proliferator activated receptor gamma (PPARgamma) agonist with anti-inflammatory properties, can attenuate LPS and cytokine-induced molecular suppression of RXRalpha-regulated genes.

METHODS

In vivo, mice were gavage-fed Rosi for 3 days, prior to intraperitoneal injection of LPS, followed by harvest of liver and serum. In vitro, HepG2 cells were treated with IL-1beta, +/- short-term Rosi pretreatment. RNA was analyzed by quantitative RT-PCR, while nuclear and cytoplasmic proteins were analyzed by immunoblotting and gel shifts.

RESULTS

Rosi attenuated LPS-mediated suppression of RNA levels of several Type II NR-regulated genes, including bile acid transporters and the major drug metabolizing enzyme, Cyp3a11, without affecting cytokine expression, suggesting a novel, direct anti-inflammatory effect in hepatocytes. Rosi suppressed the inflammation-induced nuclear export of RXRalpha, in both LPS-injected mice and IL-1beta-treated HepG2 cells, leading to maintenance of nuclear RXRalpha levels and heterodimer binding activity.

CONCLUSIONS

Rosi directly attenuates the suppressive effects of inflammation-induced cell signaling on nuclear RXRalpha levels in liver.