HRD1 attenuates the high uptake of [ 18F]FDGin hepatocellular carcinoma PET imaging

Ai-Mei Li a,1, Xia-Wen Lin a,1, Jing-Tao Shen a, Min Li b, Qi-Huang Zheng c, Zheng-Yang Zhou a,*, Ming Shi a,*

Keywords:HMG-CoA reductase degradation 1;E3 ubiquitin ligase;Glucose transporter 1;Hepatocellular carcinoma;[18F]fluorodeoxyglucose positron emission tomography

Abstract
Introduction: Due to individual deviations in tumor tissue uptake, the role of [18F]fluorodeoxyglucose ([18F]FDG) positron emission tomography (PET) in hepatocellular carcinoma (HCC) diagnosis is limited. β-Hydroxy β- methylglutaryl-CoA reductase degradation 1 (HRD1) plays a key role in clearing misfolded proteins. This study is aimed to investigate the role and mechanism of HRD1 in [18F]FDG uptake for the diagnosis of HCC.

Methods: HRD1 expression level was detected using immunohistochemical (IHC) staining in 9 HCC patients. [18F] FDG PET/CT scans were conducted before treatment. [18F]FDG uptakes in HRD1 overexpressed and knockdown transgenic models were measured by γ-counter and microPET imaging. The GLUT1-HRD1 complex was exam- ined by co-immunoprecipitation and IHC assays. GLUT1 expression in different cell lines, xenograft models and HCC patients was evaluated by Western blot and IHC assays.
Results: HRD1 was highly expressed in the HCC tumors of patients with low [18F]FDG uptake, while the HRD1 ex- pression was obviously low in the higher [18F]FDG uptake group. Both in vitro and in vivo studies found that HRD1 signiicantly inhibited [18F]FDG uptake in HCC Huh7 cell lines and animal models. Furthermore, the co-location and interaction of HRD1 with GLUT1 were detected, and the results also indicate that HRD1 could induce the deg- radation of GLUT1 in vitro and in vivo.
Conclusion: HRD1 inhibits the high uptake of [18F]FDGin HCC tumor cells by inducing degradation of GLUT1, which leads to decreased diagnostic eficiency of [18F]FDG PET imaging for HCC.Advances in knowledge: This study suggests that HRD1 inhibits the high uptake of [18F]FDG in HCC tumor by in- ducing degradation of GLUT1.Implications for patient care: HCC diagnosis with [18F]FDG PET should be accompanied by determination of HRD1 expression, and patients with high tumor HRD1 expression might be unsuitable for [18F]FDG PET.

1. Introduction
Liver cancer is the ifth most common cancer in males and the ninth most common cancer in females worldwide, and hepatocellular carci- noma (HCC) accounts for most (70%–90%) of all primary liver cancers. The incidence of HCC is steadily on the rise and it is the second leading cause of death all over the world [1,2]. The ratio of male to female HCC patients is 2.43/1. It has a ive-year natural mortality rate of more than 95% and it affects more than 250,000 people in China annually. Chronic infection with hepatitis B virus (HBV) remains the dominant risk factor for liver cancer in China, where the prevalence of hepatitis B is more than 10% [3,4].Molecular imaging approaches, especially ultrasound (US), play a vital role in HCC screening and diagnosis [5]. Computed tomography (CT) and magnetic resonance imaging (MRI) are also used for conirming the diagnosis of HCC, such as triple-phase helical CT that im- proves the detection of HCC. MRI has about the same sensitivity to de- tect HCC as helical CT [6,7]. Based on high glucose metabolism of tumor cells, biomedical imaging technique positron emission tomogra- phy (PET) coupled with radiopharmaceutical [ 18F]fluorodeoxyglucose (FDG) is widely used for the diagnosis, surveillance, and prognostication of cancers [8]. However, the use of [18F]FDG PET is limited for detecting primary HCC because of the variable [18F]FDG uptakes in different HCC patients [9]. Nevertheless, [ 18F]FDG PET is a useful diagnostic tool for evaluating extra-hepatic metastasis [10]. Moreover, [ 18F]FDG PET has the potential to be used as an additional tool to assess biological behav- ior in HCCs.β-Hydroxy β-methylglutaryl-CoA (HMG-CoA) reductase degrada- tion (HRD) ligase is a membrane-anchored complex comprising at least ive distinct subunits. HRD1, an E3 ubiquitin ligase, was highly expressed in the rheumatoid synovium [11]. HRD1 mediates endoplasmic reticulum-associated degradation (ERAD),which detects misfolded proteins in the endoplasmic reticulum and targets them for destruction and degradation [12].

Increased [18F]FDG uptake, based on enhanced glucose metabolism in cancer cells, is a sensitive marker of tumor viability. [18F]FDG is intro- duced into tumor cells via glucose transporter proteins and converted into [ 18F]FDG-6-phosphate which is not a natural metabolic substrate. Increased [ 18F]FDG uptake in cancer cells is predominantly related to glucose transporters (GLUTs), especially GLUT1, and overexpression of GLUT1 results in increased glucose uptake in cancers [13]. Previous studies also indicate that HRD1-mediated ubiquitylation-dependent degradation is involved in regulating ion transportation across mem- brane and glucose metabolism [14,15]. Thus, we speculate that HRD1 may contribute to downregulation or even induce degradation of GLUT1, and subsequently attenuate the uptake of [ 18F]FDG in HCC tumor.Therefore, our present study was undertaken to investigate the mo- lecular mechanism for the different performances of tracer uptake in PET/CT imaging with [ 18F]FDG in HCCs. We hypothesized that HRD1 mediates the downregulated expression of GLUT1 protein, which results in the imaging deviations of [18F]FDG PET/CT for HCC diagnosis.

2. Materials and methods
2.1. Study design
This single-center retrospective-observational study was conducted in Nanjing Drum Tower Hospital in China. All the procedures performed involving human participants were in accordance with the ethical stan- dards of the Independent Ethics Committee of Nanjing University and the 1964 Helsinki Declaration and its later amendments (or comparable ethical standards). For this type of study, formal consent is not required. Between January 2008 and December 2012, 9 patients (8 male and 1 fe- male) conirmed HCC histologically using invasive hepatic biopsy underwent [ 18F]FDG PET/CT imaging (without any treatment), and their characteristics are summarized in Table 1. In these 9 patients, 6 were found infected with the hepatitis B virus. Nine cases of normal liver tissue (>5 cm from cancer tissue and identiied as normal liver tis- sue by hematoxylin and eosin (HE) staining) were used as a control group.

2.2.Immunohistochemical (IHC) staining
Slides were stained for HRD1 and GLUT1 at the Anatomic Pathology IHC Laboratory of Nanjing buy Taurine Drum Tower Hospital. Immunostaining was performed on Leica Bond III autostainers using Leica Bond ancillary re- agents and the REFINE polymer DAB detection system. Antibodies used included rabbit anti-HRD1 antibody (Ab170901, Abcam, 1/1000 dilution), rabbit anti-GLUT1 antibody (Sc7903, Santa Cruz, 1/1000), and mouse anti-GAPDH antibody (Beyotime Biotechnology, 1/5000). All staining was performed using the EliVisionTM super method.

2.3. [ 18F]FDG PET/CT imaging
All patients fasted for at least 6 h before [18F]FDG injection and their blood glucose levels were less than 150 mg/dL before radiotracer injec- tion. A dose of 10-15 mCi (370-555 MBq) of [18F]FDG was administered intravenously. Whole-body [ 18F]FDG PET/CT images were acquired from vertex to proximal thigh 45-60 min after [ 18F]FDG injection using GEMINI GXL 16 PET/CT system (Philips Medical Solutions, Netherlands). After the CT topogram, a spiral CT and subsequent PET scans with acquisition times of 2 min for each bed position, according to patient weight, were performed. After CT-based attenuation correc- tion, the PET images were reconstructed with an ordered-subset expec- tation maximization (OSEM) iterative reconstruction algorithm line of response (LOR) reconstruction. A circular region of interest was drawn manually around the hypermetabolic lesions on axial fused PET/CT im- ages. The maximum standardized uptake value (SUVmax), a clinical semi-quantitative index of [18F]FDG uptake in tissue, was calculated as the following formula: SUVmax = decay corrected selected region ac- tivity (CTAC) (mCi/mL)/(injected dose [mCi]/body weight [kg]).

2.4. Cell lines and validation
HRD1 overexpressed, HRD1 non-overexpressed, HRD1 shRNA, and HRD1 non-shRNA Huh7 cells were purchased from Novobio Scientiic (Shanghai, China). All human hepatic carcinoma cell lines (Huh7) were grown in Dulbecco’s minimal essential medium (DMEM) contain- ing 10% fetal bovine serum and 50 units/mL penicillin and 100 μg/mL streptomycin sulfate. Cells were incubated at 37 °C in a humidiied at- mosphere of 5% CO2. For the validation of different cell lines, HRD1 mRNA level was detected by real-time quantitative PCR and the proce- dure was consistent with the previous report [16].

2.5. In vitro [ 18F]FDG uptake studies
The uptake studies were performed according to a protocol already described [17]. For in vitro evaluation of [18F]FDG uptake, 2 × 106 Huh7 cells (HRD1 overexpressed cells, HRD1 non-overexpressed cells, HRD1 shRNA cells, and HRD1 non-shRNA cells) were plated in 60-mm culture dishes (Fisher Scientiic) and attached overnight in sugar-free DMEM without fetal bovine serum (FBS). The cells were treated with [ 18F]FDG at 37 kBq/mL for 0, 30, 60, and 120 min. Radioactivity was measured using a Cobra II auto γ-counter channeled for 0.908 MeV γ- rays (100%).

2.6. Co-immunoprecipitation and Western blot assays
The procedures for co-immunoprecipitation detection of the GLUT1- HRD1 complex andWestern blot assays were the same as described pre- viously [18,19]. Immunoprecipitation (IP) was done with anti-GLUT1 primary antibody, and Western blot analysis was done with anti- HRD1 and anti-GLUT1 antibodies.

2.7. Establishment of tumor-bearing mice model
All animal studies were performed in accordance with the guidelines of Animal Welfare ofice of Nanjing University for the humane use of animals, and all procedures were reviewed and approved by the Institu- tional Animal Care and Use Committee. For the xenograft model, four- week-old null mice (Charles River Laboratories, ratio of male to female, 3/2) were anesthetized using 1.5% inhaled isoflurane. Huh7 cells (1 × 107 ) in 50 μL of Dulbecco’s modiied Eagle medium containing 10% FBS were injected into the left front legs of the mice. The mice were used for microPET imaging studies when the tumor volume reached 100-300 mm3 (15-20 days after inoculation for both tumor models).

2.8. MicroPET imaging with mice
PET scans and image analysis were performed using an Inveon microPET scanner (Siemens Medical Solutions). Under isoflurane anes- thesia, each Huh7 tumor-bearing mouse was injected with 3.7 MBq (100 μCi) of [18F]FDG via a tail vein; 10-min static scans were acquired at 1 h after injection. For [ 18F]FDG PET scanning, mice were fasted for 8 h before the tracer injection and were maintained under isoflurane an- esthesia during the injection, accumulation, and scanning periods. The images were reconstructed using a two-dimensional OSEM algorithm, and no correction was applied for attenuation or scatter. For each microPET scan, regions of interest (ROIs) were drawn over the tumor using vendor software PMOD on decay-corrected whole-body coronal images. These values were then divided by the administered radioactiv- ity to obtain (assuming a tissue density of 1 g/mL) an image ROI-derived maximum percentage of injected dose per gram (%ID/g-max).

2.9. Statistical analysis
Statistical analysis was performed with SPSS 19.0 software. Statisti- cal analyses were done using either the analysis of variance (ANOVA) or Dunnett’s multiple comparison t-test. Data were expressed as mean± standard deviation (SD), or as mean ± standard error of the mean (SEM). Difference at P < 0.05 was considered to be statistically signiicant. 3. Results
3.1. PET/CT imaging and IHC results inpatients
The PET/CT results indicated that all 9 cases of primary lesions in the left or right lobe of the liver were solitary, irregularly shaped, with clear SUVmax of 2.5– 10.2. Patients were divided into two groups according to the SUVmax of [18F]FDG PET/CT. The cutoff point of the SUVmax in the receiver operation characteristic (ROC) curve was 7.0 (area under curve [AUC] of 0.669). Therefore, patients with SUVmax ≥7.0 were clas- siied as the high SUV group (n = 3), and those with SUVmax <7.0 as the low SUV group (n = 6). Representative PET/CT images of low and high [ 18F]FDG uptakes inpatients are shown in Fig. 1A and B, respec- tively. As shown in Fig. 1C, the IHC results indicated that the cytoplasm expression of HRD1 protein in the low uptake group was strongly posi- tive (+++), and HRD1 expression in the high uptake group was weakly positive (+), and the difference between the two groups was statistically signiicant (quantiied value of staining optical density; nor- malizeddata as the mean of high uptake group was1;low uptake group, 1.97 ± 0.23; high uptake group, 1.00 ± 0.24; P < 0.05, n = 3). 3.2. HRD1 inhibits [ 18F]FDG uptake in vitro and in vivo
Based on the above clinical results, we constructed four HCC tumor Huh7 cell lines, including HRD1 overexpressed, HRD1 non- overexpressed, HRD1 shRNA, and HRD1 non-shRNA groups, to uncover the underlying molecular mechanism for the attenuated [18F]FDG up- take in HCC tumors with up-regulated HRD1 expression. We conirmed the mRNA expression level of HRD1 in different Huh7 cell lines by RT- PCR. Using GAPDH mRNA as internal reference to make the quantitative analysis, the data “2-ΔΔT” of all sample wells were normalized with the minimal group mean being 1.00. As shown in Fig. 2A, the mRNA level of HRD1 in the HRD1 overexpressed group (21.07 ± 0.19) was signii- cantly higher than that of the HRD1 non-overexpressed group (2.50 ± 0.09) (P < 0.001), and it was lower in the HRD1 shRNA group (1.00 ± 0.11) than that of the HRD1 non-shRNA group (2.56 ± 0.08) (P < 0.001). Differences were also observed in the[18F]FDG uptake stud- Fig. 1. PET/CT and IHC images of HCC cases with differentiated [18F]FDG uptakes. (A) Representative PET/CT images of lower [18F]FDG uptake patients. (B) Representative PET/CT images of higher [18F]FDG uptake patients. (C) The expression of HRD1 protein in tumor of patients in low and high [18F]FDG uptake groups by IHC assay (10× scope). White arrows, indicating the location of HCC areas. Scale bar, 200 μm. Fig.2. Regulation of HRD1 genetic modification on [18F]FDG uptake in vitro and in vivo. (A) RT-PCR quantification of HRD1 mRNA expression in different transgenic Huh7 cell lines (normalized values of the data “2-ΔΔT”, with GAPDH mRNA as internal reference; mean ± SEM, n =10 per group; ***P < 0.001 vs. non-overexpressed group, ###P < 0.001 vs. non- shRNA group). (B) Quantification of in vitro cellular uptake of [18F]FDG in different transgenic cell lines (mean ± SEM, n = 6; **P < 0.01 vs. non-overexpressed group, ##P < 0.01 vs. non-shRNA group). (C) Quantification of [18F]FDG uptake in tumor of different transgenic Huh7 tumor xenograft models of mice (mean ± SEM, n = 6; **P < 0.01 vs. non- overexpressed group, ##P < 0.01 vs. non-shRNA group). (D) Representative microPET images in different transgenic Huh7 xenograft models of mice. Red arrow, indicating tumor area; scale bar, 1.0 cm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article in vitro between the four cell lines. The cell line with the highest [18F] FDG uptake (CPSavg) was the HRD1 shRNA group (CPSavg = 2130 ± 29), followed by the HRD1 non-overexpressed and HRD1 non-shRNA (CPSavg = 1623 ± 32 and 1530 ± 36, respectively) cell lines, and then the HRD1 overexpressed cell line (CPSavg = 898 ± 15) (as shown in Fig. 2B).Static microPET imaging was performed 60 min after the injection of [ 18F]FDG in the tumor-bearing mice. Using a visual analysis, we found that the tumor-to-background contrast was considerably higher in the HRD1 shRNA group than that of the HRD1 non-overexpressed, HRD1 non-shRNA, and HRD1 overexpressed groups, whereas tracer distribu- tion in other normal organs was nearly identical between the groups(Fig. 2C-D). Quantitative imaging analysis revealed signiicantly higher average SUV (SUVmean) in the tumors of the HRD1 shRNA group (2.20 ± 0.25) than the HRD1 non-shRNA group (1.59 ± 0.37, P < 0.01), and the SUVmean of the HRD1 overexpressed group (1.18 ± 0.19) was signiicantly lower than that of the HRD1 non- overexpressed group (1.71 ± 0.23, P < 0.01). These results strongly sug- gest that HRD1 inhibits the uptake of [ 18F]FDG in Huh7 tumor cells in vitro and in vivo. 3.3. HRD1 induces the degradation of GLUT1
To explore the molecular mechanism underlying GLUT1 and [ 18F] FDG PET imaging results, we investigated the expression of HRD1 and its interaction with GLUT1 in tumors. HRD1 mediates degradation of tumor-associated proteins and promotes hepatocellular carcinoma pro- gression [20], and previous studies also indicate that HRD1 may mediate the degradation of proteins related to glucose metabolism and transpor- tation [14,15]. [ 18F]FDG uptake depends on the distribution density of GLUT1 on the cell membrane, and GLUT1 is the most common trans- porter of [ 18F]FDG. We hypothesized that the interaction between HRD1 and GLUT1 may induce degradation and the downregulation of GLUT1 via ERAD. Co-immunoprecipitation results conirmed the inter- action of HRD1 and GLUT1 (Fig. 3A). As shown in Fig. 3B, the result of immunofluorescence study indicated that GLUT1 was mainly expressed in the membrane of Huh7 cells and that HRD1 was expressed in the membrane and cytoplasm of Huh7 cells. The merged IF images indicate the co-distribution and interaction of HRD1 and GLUT1 in Huh7 tumor cells.

We examined the expression of HRD1 and GLUT1 by Western blot assay in four HRD1 genetically modiied Huh7 cell lines in vitro. Com- pared with the HRD1 non-overexpressed group, the expression of HRD1 in the HRD1 overexpressed group increased to 1.78-fold (P < 0.05), while the expression of GLUT1 decreased by 52.0% (P < 0.05). Compared with the HRD1 non-shRNA group, the expression of HRD1 in the HRD1 shRNA group decreased by 42.2% (P < 0.05), and the expression of GLUT1 in the HRD1 shRNA group increased to 1.84- fold (P < 0.05) (Fig. 4A-C).To examine the effect of HRD1 on GLUT1 expression in vivo, we established different types of Huh7 xenograft tumor models of mice using four HRD1 genetically modiied Huh7 cell lines. Western blotting results veriied that the protein expression manners of HRD1 and GLUT1 proteins in the animal tumor models were consistent with the in vitro research results. Compared with the HRD1 non-overexpressed group, expression of HRD1 in the HRD1 overexpressed group increased to 2.31-fold (P < 0.05), while the GLUT1 expression decreased by 62.6% (P < 0.05). When compared with the HRD1 non-shRNA group, expres- sion of HRD1 in the HRD1 shRNA group decreased by 69.3% (P < 0.05), while the expression of GLUT1 in the HRD1 shRNA group increased to 1.52-fold of that (P < 0.05) (Fig. 4D-F).In addition, in the study with tumor tissues of the HCC patients, IHC results showed that the protein expression of GLUT1 was up-regulated in the tumors with high [ 18F]FDG uptakes. The GLUT1 expression in the high uptake group (staining optical density, 3.29 ± 0.55)was almost 2.3-fold higher than that of the low uptake group (1.00 ± 0.41, P < 0.05) (Fig. 4G). Collectively, the above results of this study showed that HRD1 can induce the downregulation of GLUT1 in HCC tumors, which leads to decreased [18F]FDG uptake in HCC tumor cells and discounted diagnos- tic eficiency of [18F]FDG PET for HCC tumors. 4. Discussion
In the present study, we demonstrated that the expression of GLUT1 increased in the tumors of HCC patients with higher [ 18F]FDG uptake while the expression of HRD1 decreased, and this was closely related to the tumor [ 18F]FDG uptake performances in PET imaging. Genetic overexpression of HRD1 decreased the [ 18F]FDG uptake in vitro and in vivo. In contrast, genetic deletion of HRD1 led to the opposite effect. Furthermore, genetic overexpression of HRD1 induced downregulation of GLUT1 in Huh7 tumor cells and animal models.HCC is a major global health problem and its incidence is on the rise [21]. The detection of increased [18F]FDG uptake by PET is based on the enhanced glucose metabolism of tumor cells [22]. However, the accu- racy of [ 18F]FDG PET imaging in the diagnosis of patients with HCC is about 50% because of the differentiated [ 18F]FDGuptakes in different HCC patients [9,23]. In this study, we irstly conirmed the variation of [ 18F]FDGuptakein HCC patients. Then, we classiied these patients as high or low uptake group by setting the cutoff point in order to investi- gate the underlying mechanism.
HRD1, a typical E3 ubiquitin ligase, is implicated in screen media ERAD [16]. Huh7 is a well-differentiated HCC cell line that was originally taken from an HCC patient. We constructed HRD1 overexpressed and HRD1 shRNA Huh7 cell lines to investigate the role of HRD1 in [18F]FDG uptake in tu- mors.

Fig. 3. Co-immunoprecipitation and immunofluorescence results of GLUT1 and HRD1 in Huh7 cell lines. (A) Representative co-immunoprecipitation blots indicating the interaction of HRD1 and GLUT1 in Huh7 cells. (B) Representative immunofluorescence images indicating the co-location of HRD1 and GLUT1 expressed in Huh7 cells. Red, GLUT1; green, HRD1. White arrows, indicating the cellular membrane distribution of GLUT1. 40× scope; scale bar, 10 μm. (For interpretation of the references to colour in this igure legend, the reader is referred to the web version of this article.)

Fig. 4. Genetic manipulation of HRD1 regulates GLUT1 expression in cellular and animal HCC models. (A) Representative Western blots indicating the negative correlation between the expression of HRD1 and GLUT1 in different transgenic Huh7 cell lines. (B) Quantiication of the protein density of GLUT1 in each group. N = 6; *P < 0.05 vs. HRD1 overexpressed group, #P < 0.05 vs. non-shRNA group. (C) Quantiication of the protein density of HRD1 in each group. N = 6; *P < 0.05 vs. HRD1 overexpressed group, #P < 0.05 vs. non-shRNA group. (D) Representative Western blots indicating the negative correlation between the expression of HRD1 and GLUT1 in different Huh7 tumor xenograft mice models. (E) Quantiication of the protein density of GLUT1 in each group in vivo. N = 6; *P < 0.05 vs. HRD1 overexpressed group, #P < 0.05 vs. non-shRNA group. (F) Quantiication of the protein density of HRD1 in each group in vivo. N = 6; *P < 0.05 vs. HRD1 overexpressed group, #P < 0.05 vs. non-shRNA group. (G) The expression of GLUT1 in tumor of HCC patients in different [18F]FDG uptake groups by immunofluorescence assay. Green, GLUT1 protein; blue, DAPI (4′,6-diamidino-2-phenylindole). 40× scope; scale bar, 10 μm. (For interpretation of the references to colour in this igure legend, the reader is referred to the web version of this article.) ated with lower [18F]FDG uptake in vitro and in vivo. Studies on the relation of ubiquitin ligase with [ 18F]FDG PET behavior are lacking in the current literature. The mechanism underlying the regulation of HRD1 expression in tumors is still not well deined. Several studies have indi- cated that inflammatory cytokines increase the expression of HRD1[24]. We speculated that viral infection, such as HBV, induces the secretion of inflammatory cytokines which increase the expression of HRD1, and it may also potentially explain the association between cancer and chronic inflammation. However, this hypothesis needs to be explored in the future. Glucose transportation mainly relies on sodium ion/glucose co- transporter (SGLT) and GLUT family members. Numerous studies have demonstrated that [ 18F]FDG is transported through the cell membrane via glucose transporter proteins [24,25]. A few investigations have found that the expression of GLUT1 in tumors is influenced by several factors, such as the p53 and the extracellular signal-regulated kinase (ERK) signaling pathways [26,27]. GLUT1 is a highly conserved uniporter protein that can be degraded by ubiquitin ligases [28]. GLUT1 is the most common glucose transporter protein in human sub- jects and it is overexpressed in many types of tumor cells [24]. Consis- tent with the results obtained in previous studies, our IHC results conirmed the remarkable membrane expression of GLUT1 in tumor tissue of HCC patients. By both PET imaging and IF assay, we found that [18F]FDG uptake in the tumors of HCC patients is positively corre- lated with their GLUT1 expression, which is consistent with previous reports. 5. Conclusion
In this study, we testiied the association of [18F]FDG uptake and HRD1 expression in Infection ecology Huh7 tumor cells and the tumor tissues of HCC pa- tients and animal models, and our results support the negative correla- tion between the expression of HRD1 and GLUT1 in vitro and in vivo. We found that HRD1 and GLUT1 are co-expressed in the same HCC cells by means of co-immunoprecipitation and immunofluorescence assays. These indings indicate that HRD1 may play a critical role in the degra- dation and downregulation of GLUT1 protein and this can result in the attenuated [18F]FDG uptake in PET imaging of HCC tumors.