HLA transgenic mice: application in reproducing idiosyncratic drug toxicity.

Various types of transgenic mice carrying either class I or II human leukocyte antigen (HLA) molecules are readily available, and reports describing their use in a variety of studies have been published for more than 30 years. Examples of their use include the discovery of HLA-specific antigens against viral infection as well as the reproduction of HLA-mediated autoimmune diseases for the development of therapeutic strategies. Recently, HLA transgenic mice have been used to reproduce HLA-mediated idiosyncratic drug toxicity (IDT), a rare and unpredictable adverse drug reaction that can result in death. For example, abacavir-induced IDT has successfully been reproduced in HLA-B*57:01 transgenic mice. Several reports using HLA transgenic mice for IDT have proven the utility of this concept for the evaluation of IDT using various HLA allele combinations and drugs. It has become apparent that such models may be a valuable tool to investigate the mechanisms underlying HLA-mediated IDT. This review summarizes the latest findings in the area of HLA transgenic mouse models and discusses the current challenges that must be overcome to maximize the potential of this unique animal model.

Mechanism-based integrated assay systems for the prediction of drug-induced liver injury.

Drug-induced liver injury (DILI) can cause hepatic failure and result in drug withdrawal from the market. It has host-related and compound-dependent mechanisms. Preclinical prediction of DILI risk is very challenging and safety assessments based on animals inadequately forecast human DILI risk. In contrast, human-derived in vitro cell culture-based models could improve DILI risk prediction accuracy. Here, we developed and validated an innovative method to assess DILI risk associated with various compounds. Fifty-four marketed and withdrawn drugs classified as DILI risks of "most concern", "less concern", and "no concern" were tested using a combination of four assays addressing mitochondrial injury, intrahepatic lipid accumulation, inhibition of bile canalicular network formation, and bile acid accumulation. Using the inhibitory potencies of the drugs evaluated in these in vitro tests, an algorithm with the highest available DILI risk prediction power was built by artificial neural network (ANN) analysis. It had an overall forecasting accuracy of 73%. We excluded the intrahepatic lipid accumulation assay to avoid overfitting. The accuracy of the algorithm in terms of predicting DILI risks was 62% when it was constructed by ANN but only 49% when it was built by the point-added scoring method. The final algorithm based on three assays made no DILI risk prediction errors such as "most concern " instead of "no concern" and vice-versa. Our mechanistic approach may accurately predict DILI risks associated with numerous candidate drugs.

Evaluation of immune-mediated idiosyncratic drug toxicity using chimeric HLA transgenic mice.

Immune-mediated idiosyncratic drug toxicity (IDT) is a rare adverse drug reaction, potentially resulting in death. Although genome-wide association studies suggest that the occurrence of immune-mediated IDT is strongly associated with specific human leukocyte antigen (HLA) allotypes, these associations have not yet been prospectively demonstrated. In this study, we focused on HLA-B*57:01 and abacavir (ABC)-induced immune-mediated IDT, and constructed transgenic mice carrying chimeric HLA-B*57:01 (B*57:01-Tg) to determine if this in vivo model may be useful for evaluating immune-mediated IDT. Local lymph node assay (LLNA) results demonstrated that percentages of BrdU+, IL-2+, and IFN-γ+ in CD8+ T cells of ABC (50 mg/kg/day)-applied B*57:01-Tg mice were significantly higher than those in littermates (LMs), resulting in the infiltration of inflammatory cells into the ear. These immune responses were not observed in B*57:03-Tg mice (negative control). Furthermore, oral administration of 1% (v/v) ABC significantly increased the percentage of CD44highCD62Llow CD8+ memory T cells in lymph nodes and spleen derived from B*57:01-Tg mice, but not in those from B*57:03-Tg mice and LMs. These results suggest that B*57:01-Tg mice potentially enable the reproduction and evaluation of HLA-B*57:01 and ABC-induced immune-mediated IDT.

Prediction of the Clinical Risk of Drug-Induced Cholestatic Liver Injury Using an In Vitro Sandwich Cultured Hepatocyte Assay.

Drug-induced liver injury (DILI) is of concern to the pharmaceutical industry, and reliable preclinical screens are required. Previously, we established an in vitro bile acid-dependent hepatotoxicity assay that mimics cholestatic DILI in vivo. Here, we confirmed that this assay can predict cholestatic DILI in clinical situations by comparing in vitro cytotoxicity data with in vivo risk. For 38 drugs, the frequencies of abnormal increases in serum alkaline phosphatase (ALP), transaminases, gamma glutamyltranspeptidase (γGT), and bilirubin were collected from interview forms. Drugs with frequencies of serum marker increases higher than 1% were classified as high DILI risk compounds. In vitro cytotoxicity was assessed by monitoring lactate dehydrogenase release from rat and human sandwich-cultured hepatocytes (SCRHs and SCHHs) incubated with the test drugs (50 µM) for 24 hours in the absence or presence of a bile acids mixture. Receiver operating characteristic analyses gave optimal cutoff toxicity values of 19.5% and 9.2% for ALP and transaminases in SCRHs, respectively. Using this cutoff, high- and low-risk drugs were separated with 65.4-78.6% sensitivity and 66.7-79.2% specificity. Good separation was also achieved using SCHHs. In conclusion, cholestatic DILI risk can be successfully predicted using a sandwich-cultured hepatocyte-based assay.

Key determinants of the circulatory exposure of organic anions: differences in hepatic uptake between multidrug resistance-associated protein 2 (Mrp2)-deficient rats and wild-type rats.

1. Raloxifene-6-glucuronide (R6G) is a substrate of rat multidrug resistance-associated protein 2 (Mrp2), a transporter responsible for biliary excretion of organic anions. 2. Pharmacokinetic modeling of R6G in Eisai hyperbilirubinemic rats (EHBRs), hereditary Mrp2-deficient rats, and wild-type Sprague-Dawley rats (SDRs) indicated that reduction in not only biliary excretion but also hepatic uptake of R6G influenced low clearance in EHBRs. 3. An integration plot study demonstrated that the hepatic uptake of R6G was 66% lower in EHBRs than that in SDRs. A reduction was observed for the other Mrp2 substrate Valsartan (95% lower) but not for estradiol-17ß-glucuronide (E217ßG). This variation may be associated with the difference in substrate specificity of transporters and/or inhibition of hepatic uptake of organic anions by endogenous substances such as bilirubin glucuronides. 4. In conclusion, incidental alteration of the hepatic uptake of organic anions should be considered as an explanation of their enhanced systemic exposure in EHBRs.

HLA-B*57:01-dependent intracellular stress in keratinocytes triggers dermal hypersensitivity reactions to abacavir.

Specific human leukocyte antigen (HLA) polymorphisms combined with certain drug administration strongly correlate with skin eruption. Abacavir hypersensitivity (AHS), which is strongly associated with HLA-B*57:01, is one of the most representative examples. Conventionally, HLA transmits immunological signals via interactions with T cell receptors on the cell surface. This study focused on HLA-mediated intracellular reactions in keratinocytes that might determine the onset of skin immunotoxicity by drug treatments. Abacavir exposure resulted in keratinocytes expressing HLA-B*57:01 exhibiting endoplasmic reticulum (ER) stress responses, such as immediate calcium release into the cytosol and enhanced HSP70 expression. In contrast, keratinocytes expressing HLA-B*57:03 (closely related to HLA-B*57:01) did not show these changes. This indicated that HLA-B*57:01 has a specific intracellular response to abacavir in keratinocytes in the absence of lymphocytes. Furthermore, abacavir exposure in HLA-B*57:01-expressing keratinocytes elevated the expression of cytokines/chemokines such as interferon-γ, interleukin-1ß, and CCL27, and induced T lymphoblast migration. These effects were suppressed by ER stress relief using 4-phenylbutyrate (4-PB). HLA-B*57:01-transgenic mice also exhibited ER stress in epidermal areas following abacavir administration, and abacavir-induced skin toxicity was attenuated by the administration of 4-PB. Moreover, abacavir bound to HLA-B*57:01 within cells and its exposure led to HLA-B*57:01 protein aggregation and interaction with molecular chaperones in the ER of keratinocytes. Our results underscore the importance of HLA-mediated intracellular stress responses in understanding the onset of HLA-B*57:01-mediated AHS. We provide the possibility that the intracellular behavior of HLA is crucial for determining the onset of drug eruptions.

Calorie Restriction Impairs Anti-Tumor Immune Responses in an Immunogenic Preclinical Cancer Model.

(1) Background: Although the important role of dietary energy intake in regulating both cancer progression and host immunity has been widely recognized, it remains unclear whether dietary calorie restriction (CR) has any impact on anti-tumor immune responses. (2) Methods: Using an immunogenic B16 melanoma cell expressing ovalbumin (B16-OVA), we examined the effect of the CR diet on B16-OVA tumor growth and host immune responses. To further test whether the CR diet affects the efficacy of cancer immunotherapy, we examined the effect of CR against anti-PD-1 monoclonal antibody (anti-PD-1 Ab) treatment. (3) Results: The CR diet significantly slowed down the tumor growth of B16-OVA without affecting both CD4+ and CD8+ T cell infiltration into the tumor. Although in vivo depletion of CD8+ T cells facilitated B16-OVA tumor growth in the control diet group, there was no significant change in the tumor growth in the CR diet group with or without CD8+ T cell-depletion. Anti-PD-1 Ab treatment lost its efficacy to suppress tumor growth along with the activation and metabolic shift of CD8+ T cells under CR condition. (4) Conclusions: Our present results suggest that a physical condition restricted in energy intake in cancer patients may impair CD8+ T cell immune surveillance and the efficacy of immunotherapy.

Drug-induced altered self-presentation increases tumor immunogenicity.

Anti-human immunodeficiency virus (HIV) drug abacavir (ABC) binds to the specific allele of human leukocyte antigen (HLA-B*57:01) and activates CD8+ T cells by presenting altered abnormal peptides. Here, we examined the effect of ABC-induced altered self-presentation by HLA-B*57:01 on immunogenicity of cancer cells and CD8+ T-cell-dependent anti-tumor immunity. We established human-mouse chimeric HLA-B*57:01-expressing tumor cell lines (B16F10 and 3LL) and tested the anti-tumor effect of ABC in vivo. ABC treatment inhibited the growth of HLA-B*57:01-expressing tumors by a CD8+ T-cell-dependent mechanism. ABC treatment induced CXCR3-dependent infiltration of CD8+ T cells into HLA-B*57:01-expressing tumors, and activated those tumor-infiltrating CD8+ T cells to proliferate and secrete IFN-γ. The activation of CD8+ T cells using drug-induced altered self-presentation may be a new strategy to increase tumor immunogenicity and improve the efficacy of immunotherapy.

Development of a Novel CD26-Targeted Chimeric Antigen Receptor T-Cell Therapy for CD26-Expressing T-Cell Malignancies.

Chimeric-antigen-receptor (CAR) T-cell therapy for CD19-expressing B-cell malignancies is already widely adopted in clinical practice. On the other hand, the development of CAR-T-cell therapy for T-cell malignancies is in its nascent stage. One of the potential targets is CD26, to which we have developed and evaluated the efficacy and safety of the humanized monoclonal antibody YS110. We generated second (CD28) and third (CD28/4-1BB) generation CD26-targeted CAR-T-cells (CD26-2G/3G) using YS110 as the single-chain variable fragment. When co-cultured with CD26-overexpressing target cells, CD26-2G/3G strongly expressed the activation marker CD69 and secreted IFNgamma. In vitro studies targeting the T-cell leukemia cell line HSB2 showed that CD26-2G/3G exhibited significant anti-leukemia effects with the secretion of granzymeB, TNFα, and IL-8, with 3G being superior to 2G. CD26-2G/3G was also highly effective against T-cell lymphoma cells derived from patients. In an in vivo mouse model in which a T-cell lymphoma cell line, KARPAS299, was transplanted subcutaneously, CD26-3G inhibited tumor growth, whereas 2G had no effect. Furthermore, in a systemic dissemination model in which HSB2 was administered intravenously, CD26-3G inhibited tumor growth more potently than 2G, resulting in greater survival benefit. The third-generation CD26-targeted CAR-T-cell therapy may be a promising treatment modality for T-cell malignancies.

DPP8 Selective Inhibitor Tominostat as a Novel and Broad-Spectrum Anticancer Agent against Hematological Malignancies.

DPP8/9 inhibition induces either pyroptotic or apoptotic cell death in hematological malignancies. We previously reported that treatment with the DPP8/9 inhibitor 1G244 resulted in apoptotic cell death in myeloma, and our current study further evaluates the mechanism of action of 1G244 in different blood cancer cell lines. Specifically, 1G244 inhibited DPP9 to induce GSDMD-mediated-pyroptosis at low concentrations and inhibited DPP8 to cause caspase-3-mediated-apoptosis at high concentrations. HCK expression is necessary to induce susceptibility to pyroptosis but does not participate in the induction of apoptosis. To further characterize this DPP8-dependent broad-spectrum apoptosis induction effect, we evaluated the potential antineoplastic role for an analog of 1G244 with higher DPP8 selectivity, tominostat (also known as 12 m). In vitro studies demonstrated that the cytotoxic effect of 1G244 at high concentrations was enhanced in tominostat. Meanwhile, in vivo work showed tominostat exhibited antitumor activity that was more effective on a cell line sensitive to 1G244, and at higher doses, it was also effective on a cell line resistant to 1G244. Importantly, the weight loss morbidity associated with increasing doses of 1G244 was not observed with tominostat. These results suggest the possible development of novel drugs with antineoplastic activity against selected hematological malignancies by refining and increasing the DPP8 selectivity of tominostat.

Regulation of the immune tolerance system determines the susceptibility to HLA-mediated abacavir-induced skin toxicity.

Idiosyncratic drug toxicity (IDT) associated with specific human leukocyte antigen (HLA) allotype is a rare and unpredictable life-threatening adverse drug reaction for which prospective mechanistic studies in humans are difficult. Here, we show the importance of immune tolerance for IDT onset and determine whether it is susceptible to a common IDT, HLA-B*57:01-mediated abacavir (ABC)-induced hypersensitivity (AHS), using CD4+ T cell-depleted programmed death-1 receptor (PD-1)-deficient HLA-B*57:01 transgenic mice (B*57:01-Tg/PD-1-/-). Although AHS is not observed in B*57:01-Tg mice, ABC treatment increases the proportion of cytokine- and cytolytic granule-secreting effector memory CD8+ T cells in CD4+ T cell-depleted B*57:01-Tg/PD-1-/- mice, thereby inducing skin toxicity with CD8+ T cell infiltration, mimicking AHS. Our results demonstrate that individual differences in the immune tolerance system, including PD-1highCD8+ T cells and regulatory CD4+ T cells, may affect the susceptibility of humans to HLA-mediated IDT in humans.

The PD1 inhibitory pathway and mature dendritic cells contribute to abacavir hypersensitivity in human leukocyte antigen transgenic PD1 knockout mice.

Based on recent genome-wide association studies, abacavir-induced hypersensitivity is highly associated with human leukocyte antigen (HLA)-B*57:01 allele. However, the underlying mechanism of this occurrence is unclear. To investigate the underlying mechanism, we developed HLA-B*57:01 transgenic mice and found that application of abacavir could cause CD8 T cell activation with elevation in PD1 expression; however, severe skin hypersensitivity was not observed. To eliminate the immunosuppressive effect of PD1, HLA-B*57:01 transgenic/PD1 knockout (01Tg/PD1) mice were generated by mating HLA-B*57:01 transgenic mice and PD1 knockout mice. Thereafter, 01Tg/PD1 mice were treated with abacavir. Similar to the above results, severe skin hypersensitivity was not observed. Therefore, we treated 01Tg/PD1 mice with an anti-CD4 antibody to deplete CD4 T cells, followed by abacavir topically and orally. Severe abacavir-induced skin hypersensitivity was observed in 01Tg/PD1 mice after depletion of CD4 T cells, in addition to significant CD8 T cell activation and dendritic cell maturation. Taken together, we succeeded in reproducing severe skin hypersensitivity in a mouse model. And we found that through the combined depletion of PD1 and CD4 T cells, CD8 T cells could be activated and could proceed to clonal proliferation, which is promoted by mature dendritic cells, thereby eventually inducing severe skin hypersensitivity.

In vitro bile acid-dependent hepatocyte toxicity assay system using human induced pluripotent stem cell-derived hepatocytes: Current status and disadvantages to overcome.

Cholestatic drug-induced liver injury (DILI) is a type of hepatotoxicity. Its underlying mechanisms are dysfunction of bile salt export pump (BSEP) and multidrug resistance-associated protein 2/3/4 (MRP2/3/4), which play major roles in bile acid (BA) excretion into the bile canaliculi and blood, resulting in accumulation of BAs in hepatocytes. The sandwich-cultured hepatocyte (SCH) model can simultaneously analyze hepatic uptake and biliary excretion. Therefore, we investigated whether sandwich-cultured human induced pluripotent stem cell (iPS cell)-derived hepatocytes (SCHiHs) are suitable for evaluating cholestatic DILI. Fluorescent N-(24-[7-(4-N,N-dimethylaminosulfonyl-2,1,3-benzoxadiazole)]amino-3α,7α,12α-trihydroxy-27-nor-5ß-cholestan-26-oyl)-2'-aminoethanesulfonate (tauro-nor-THCA-24-DBD, a BSEP substrate) was accumulated in bile canaliculi, which supports the presence of a functional bile canaliculi lumen. MRP2 was highly expressed in the Western blot analysis, whereas the mRNA expression of BSEP was hardly detectable. MRP3/4 mRNA levels were maintained. Of the 22 compounds known to cause DILI with BAs, 7 showed significant cytotoxicity. Most high-risk drugs were detected using the developed SCHiH system. However, a shortcoming was the considerably low expression level of BSEP, which prevented the detection of some relevant drugs whose risks should be detected in primary human hepatocytes.

An Animal Model of Abacavir-Induced HLA-Mediated Liver Injury.

Genome-wide association studies indicate that several idiosyncratic adverse drug reactions are highly associated with specific human leukocyte antigen (HLA) alleles. For instance, abacavir, a human immunodeficiency virus reverse transcriptase inhibitor, induces multiorgan toxicity exclusively in patients carrying the HLA-B*57:01 allele. However, the underlying mechanism is unclear due to a lack of appropriate animal models. Previously, we developed HLA-B*57:01 transgenic mice and found that topical application of abacavir to the ears induced proliferation of CD8+ lymphocytes in local lymph nodes. Here, we attempted to reproduce abacavir-induced liver injury in these mice. However, oral administration of abacavir alone to HLA-B*57:01 transgenic mice did not increase levels of the liver injury marker alanine aminotransferase. Considering the importance of innate immune activation in mouse liver, we treated mice with CpG oligodeoxynucleotide, a toll-like receptor 9 agonist, plus abacavir. This resulted in a marked increase in alanine aminotransferase, pathological changes in liver, increased numbers of activated CD8+ T cells, and tissue infiltration by immune cells exclusively in HLA-B*57:01 transgenic mice. These results indicate that CpG oligodeoxynucleotide-induced inflammatory reactions and/or innate immune activation are necessary for abacavir-induced HLA-mediated liver injury characterized by infiltration of CD8+ T cells. Thus, we developed the first mouse model of HLA-mediated abacavir-induced idiosyncratic liver injury. Further investigation will show that the proposed HLA-mediated liver injury model can be applied to other combinations of drugs and HLA types, thereby improving drug development and contributing to the development of personalized medicine.

Identification of Bile Acids Responsible for Inhibiting the Bile Salt Export Pump, Leading to Bile Acid Accumulation and Cell Toxicity in Rat Hepatocytes.

Inhibition of bile salt export pump (BSEP) causes hepatic accumulation of toxic bile acid (BA), leading to hepatocyte death. We reported a sandwich-cultured hepatocyte (SCH)-based model that can estimate potential cholestatic compounds by assessing their ability to induce hepatotoxicity in combination with a titrated amount of human 12 BA species. However, there is little information about the specific BAs responsible for hepatotoxicity, when BSEP is inhibited. This study measured the accumulation of each BA in rat SCHs in the presence of 10 µM cyclosporine A (CsA), which only inhibits BSEP, and 50 µM CsA, which further inhibits basolateral BA efflux transporters. The accumulation of all BAs (not significant for deoxycholic acid [DCA]) was observed in the presence of 10 µM CsA. In particular, 3 BAs (chenodeoxycholic acid [CDCA], DCA, and glyco-DCA [GDCA]) showed increased toxicity in the presence of 10 µM CsA, whereas the other BAs did not. In addition to these BAs, taurolithocholic acid, glyco-CDCA, and glycocholic acid showed increased toxicity in the presence of 50 µM CsA, but additional accumulation of these BAs could not be observed. These results indicate the inhibiting BSEP results in the accumulation of CDCA, GDCA, and partially DCA, thereby resulting in hepatotoxicity.

Establishment of a Drug-Induced, Bile Acid-Dependent Hepatotoxicity Model Using HepaRG Cells.

Bile acid (BA) retention within hepatocytes is an underlying mechanism of cholestatic drug-induced liver injury (DILI). We previously developed an assay using sandwich-cultured human hepatocytes (SCHHs) to evaluate drug-induced hepatocyte toxicity accompanying intracellular BA accumulation. However, due to shortcomings commonly associated with the use of primary human hepatocytes (e.g., limited availability, lot-to-lot variability, and high cost), we examined if the human hepatic stem cell line, HepaRG, might also be applicable to our assay system. Consequently, mRNA expression levels of human BA efflux and uptake transporters were lower in HepaRG cells than in SCHHs but higher than in HepG2 human hepatoma cells. Nevertheless, HepaRG cells and SCHHs showed similar toxicity responses to 22 selected drugs, including cyclosporine A (CsA). CsA (10 µM) was cytotoxic toward HepaRG cells in the presence of BAs and also reduced the biliary efflux rate of [(3)H]taurocholic acid from 38.5% to 19.2%. Therefore, HepaRG cells are useful for the evaluation of BA-dependent drug toxicity caused by biliary BA efflux inhibition. Regardless, the prediction accuracy for cholestatic DILI risk was poor for HepaRG cells versus SCHHs, suggesting that our DILI model system requires further improvements to increase the utility of HepaRG cells as a preclinical screening tool.

Basal efflux of bile acids contributes to drug-induced bile acid-dependent hepatocyte toxicity in rat sandwich-cultured hepatocytes.

The bile salt export pump (BSEP or Bsep) functions as an apical transporter to eliminate bile acids (BAs) from hepatocytes into the bile. BSEP or Bsep inhibitors engender BA retention, suggested as an underlying mechanism of cholestatic drug-induced liver injury. We previously reported a method to evaluate BSEP-mediated BA-dependent hepatocyte toxicity by using sandwich-cultured hepatocytes (SCHs). However, basal efflux transporters, including multidrug resistance-associated proteins (MRP or Mrp) 3 and 4, also participate in BA efflux. This study examined the contribution of basal efflux transporters to BA-dependent hepatocyte toxicity in rat SCHs. The apical efflux of [(3)H]taurocholic acid (TC) was potently inhibited by 10 µM cyclosporine A (CsA), with later inhibition of basal [(3)H]TC efflux, while MK571 simultaneously inhibited both apical and basal [(3)H]TC efflux. CsA-induced BA-dependent hepatocyte toxicity was 30% at most at 10 µM CsA and ∼60% at 50 µM, while MK571 exacerbated hepatocyte toxicity at concentrations of ≥50 µM. Quinidine inhibited only basal [(3)H]TC efflux and showed BA-dependent hepatocyte toxicity in rat SCHs. Hence, inhibition of basal efflux transporters as well as Bsep may precipitate BA-dependent hepatocyte toxicity in rat SCHs.