8.3 Molecular Mechanisms of Target Organ Toxicity: Hepatotoxicity, Nephrotoxicity & Neurotoxicity
Key Takeaways
- CYP2E1-rich, relatively hypoxic centrilobular zone 3 explains many bioactivation hepatotoxicants; BSEP/ABCB11 inhibition is a bile-salt cholestatic mechanism distinct from zonal necrosis.
- Proximal-tubule organic anion and cation transporters concentrate many nephrotoxicants; male-rat α2u-globulin hyaline-droplet nephropathy has limited human relevance.
- Neurotoxicity is classified first as myelin, axon, or neurotransmitter; strychnine is a glycine-receptor antagonist that removes spinal inhibition.
- Systems toxicology (II.6) translates a molecular initiating event to cellular, organ, and organism key events; omics without function is not an adverse outcome.
- rasH2, p53-heterozygous, and obese or diabetic models have defined strengths and human-disease gaps (II.8); mixture terms are additivity, synergism, potentiation, and antagonism (II.7 C); mechanism plus exposure informs regulation of endocrine disruptors, PBTs, nanoparticles, and tobacco or nicotine (II.7 A).
From a molecular interaction to an organ and a decision
Handbook II.6 asks you to translate results across sub-cellular, cellular, tissue, individual, species, and population levels using systems toxicology. II.7 asks you to apply mechanistic results to outcomes, prevention, and risk, including endocrine disruptors, microbeads, tobacco/nicotine, persistent bioaccumulative toxins (PBTs), and nanoparticles, and to name mixture interactions. II.8 asks you to use disease and genetically engineered models without confusing them with human disease. Independent OpenExamPrep material in this section works three classic target organs—liver, kidney, and nervous system—then practices that translation. Full adverse outcome pathway (AOP) formalisms (weight of evidence, essentiality, quantitative key-event relationships) belong in chapter 10. Here you need the mechanism-translation sketch: molecular initiating event → cellular key event → organ key event → organism adverse outcome.
This section is not an ABT product and does not claim official approval, review, or partnership with ABT.
Hepatotoxicity: zonation and bile-salt transporters
The hepatic acinus is oxygen- and enzyme-zoned. Zone 3 (centrilobular) hepatocytes are relatively hypoxic, CYP2E1- and many CYP3A-rich, and lower in GSH than periportal zone 1. That is why carbon tetrachloride, chloroform, and acetaminophen (NAPQI after CYP2E1/CYP3A/CYP1A2) produce centrilobular necrosis rather than a random hepatitis. Low zone-3 oxygen tension also favors the reductive dehalogenation of CCl4 taught in section 8.1. Zone 1 injury (allyl alcohol; some iron-overload patterns) is the contrast case: oxygen-rich, GSH-rich, different enzymes.
Cholestatic injury is often a transporter story, not a CYP-zonation story. The canalicular bile salt export pump (BSEP, ABCB11) is the ATP-dependent efflux pump for taurine- and glycine-conjugated bile acids. Human genetic BSEP failure (PFIC2) shows what intracellular bile-salt detergent does to hepatocytes when the pump is missing—physiology that II.1 D wants you to connect to drug inhibition. Chemicals that inhibit or down-regulate BSEP (troglitazone, bosentan, some cyclosporine exposures) let bile salts accumulate. The clinicopathologic face is cholestasis with ALP, GGT, and bilirubin more than a pure ALT leak. MRP2 (ABCC2) and NTCP (SLC10A1) complete the uptake/efflux set; a chemical can inhibit BSEP and also block compensatory basolateral efflux, worsening the intracellular load. Do not call every high ALP “zone-3 ROS.”
Nephrotoxicity: proximal-tubule transporters and α2u-globulin
The proximal tubule (S1–S3) reabsorbs the filtered load and runs organic anion transporters (OAT1/SLC22A6, OAT3/SLC22A8) on the basolateral membrane plus apical efflux. Those transporters concentrate β-lactam antibiotics, many antivirals, ochratoxin A, and conjugated metabolites inside the tubule cell—an uptake molecular initiating event, not a slogan that “the kidney is sensitive.” Organic cation transporter 2 (OCT2) similarly concentrates cisplatin; cimetidine can compete at OCT2. Injury markers and outer-stripe (S3) preference follow blood flow, hypoxia, and that transporter map.
α2u-globulin nephropathy is the human-relevance trap. Adult male rats synthesize a large amount of hepatic α2u-globulin under androgen control. Some hydrocarbons and their metabolites (d-limonene 1,2-oxide, 2,2,4-trimethylpentane metabolites, decalin, isophorone) bind that protein. The complex resists lysosomal degradation, appears as hyaline droplets in proximal tubules, and can drive cytotoxicity, regeneration, and, in chronic studies, a male-rat renal-tumor promotion pathway. Female rats, male mice, and humans do not run this androgen-driven protein load. Humans have related lipocalins at much lower abundance; they do not reproduce the syndrome. For this mode of action, human relevance is limited. Do not promote a male-rat hyaline-droplet chemical to a default human renal carcinogen on α2u evidence alone, and do not ignore a second mechanism the same molecule might have.
Neurotoxicity: myelin, axon, and neurotransmitter
Classify nervous-system injury by the structure first.
Myelin / Schwann cell or oligodendrocyte. Vacuolation, stripping, slowed conduction. Teaching examples include hexachlorophene-type white-matter edema and tellurium-induced peripheral myelin injury. The cell that fails first is the myelinating cell, not the axon cylinder.
Axon (axonopathy). Distal dying-back after cytoskeletal or energy failure. Acrylamide is a standard distal axonopathy example. Some organophosphate delayed neuropathies travel through neuropathy target esterase aging rather than through acute acetylcholinesterase crisis. The soma may look intact early.
Neurotransmitter / receptor. Too much or too little synaptic inhibition or excitation without a requirement for structural myelin loss. Strychnine is a competitive glycine-receptor antagonist (section 8.2): loss of spinal inhibition, tetanic convulsions, individual death from respiratory failure. Organophosphate acetylcholinesterase inhibition is a different transmitter story (acetylcholine excess). Tetrodotoxin is a voltage-gated sodium-channel pore block—still channel-level, not myelin. Mixing these three is how every tremor becomes undifferentiated “neurotoxicity.”
Systems toxicology: translating across levels (II.6)
II.6 A–B are the craft of moving a result from a simple system to a more complex one without pretending they are equal. A mitochondrial ROS assay is a sub-cellular observation. A hepatocyte GSH drop is cellular. Centrilobular necrosis is tissue/organ. Jaundice and death are individual. A mouse-to-human CYP2E1 difference is species. A change in liver-related mortality in an exposed town is population. Each step needs a method: enzyme kinetics, histology, PBPK, epidemiology—not a leap.
II.6 C reminds you that anatomical and biochemical differences change magnitude: a male-rat α2u load, a dog’s corticosteroid ALP isoform, a human cardiomyocyte’s low catalase. II.6 D is honesty about computational network tools: they organize omics and suggest nodes; they do not replace an apical adverse outcome. II.6 E again: omics relate to organism effects only through those key events. The four-box sketch below is the translation you should be able to fill with named biology for a bioactivated solvent. Chapter 10 will add Bradford Hill-style essentiality and quantitative key-event relationships.
Disease and genetically engineered models (II.8)
rasH2 (c-Ha-ras transgenic mouse) is a 26-week alternative carcinogenicity model used in ICH S1B weight-of-evidence packages. It can detect many genotoxic and some nongenotoxic rodent carcinogens faster than a two-year mouse study. It is not a miniature human carcinoma. Ras-pathway bias and mouse background tumors are limitations (II.8 B).
p53+/− (heterozygous Trp53) 26-week assays are generally more sensitive to genotoxic carcinogens that benefit from reduced p53 dosage. They can miss some receptor-mediated, nongenotoxic rodent carcinogens that need a full lifetime and an intact tumor-suppressor context. A negative p53+/− study is not “not a human carcinogen.”
Obese and diabetic models (ob/ob, db/db, Zucker diabetic fatty, diet-induced obesity) increase hepatic fat, alter CYP expression, and can unmask drug-induced liver injury that a lean Wistar would miss. The leptin-mutation models are not polygenic human type 2 diabetes. Use them to test a mechanistic hypothesis about steatosis or insulin resistance as a susceptibility factor (II.3 disease-state), and state the gap.
II.8 A also covers building an alternative model from a mechanism (a humanized CYP2E1 mouse after you already know CYP2E1 is the initiating enzyme). That is the opposite of hoping a random disease model will reveal the mode of action.
Mixtures: additivity, synergism, potentiation, antagonism (II.7 C)
Additivity means the combined effect equals the sum predicted by dose addition (similar mode of action; concentration addition) or response addition (independent modes). Naming which null model you used is the skill.
Synergism means the combined effect is greater than that additive prediction, and both agents can be toxic on their own.
Potentiation means chemical A is not toxic alone at the dose used but increases the toxicity of B. Classic teaching: isopropanol potentiates carbon tetrachloride hepatotoxicity. Ethanol induction of CYP2E1 increasing CCl4 injury is a related susceptibility phenotype that often gets discussed in the same breath.
Antagonism means the combined effect is less than additive. Functional (opposing physiology), chemical (chelation), dispositional (induced clearance), or receptor (a blocker) are the usual subtypes.
Do not call every “worse than I expected” synergism until you have stated the additive null.
Applying mode of action to regulation, conceptually (II.7 A)
You are not asked to recite every statute. You are asked how mechanism plus exposure changes a control decision.
Endocrine disruptors. An estrogen-receptor, androgen-receptor, or thyroid-pathway initiating event is why receptor-binding and steroidogenesis screens exist and why some uses are restricted even when apical cancer bioassays are messy.
PBTs. Persistence, bioaccumulation, and toxicity drive listing (Stockholm Convention; TSCA PBT). The T is a mode-of-action problem (AhR for some dioxin-like compounds; neuronal calcium and mitochondria for methylmercury). P and B are fate problems; a beautiful cellular mechanism does not make a chemical less persistent.
Nanoparticles. Primary particles, agglomerates, dissolution (ion release versus particle), surface reactivity, and frustrated phagocytosis ROS in macrophages are the mechanistic vocabulary. Regulation (TSCA; REACH nano provisions) still treats many nanoforms as poorly predicted by the dissolved ion alone.
Tobacco / nicotine. Nicotine’s nicotinic acetylcholine receptor pharmacology explains addiction and some cardiovascular effects; nitrosamines and PAHs explain much of the cancer chemistry. Product regulation uses both exposure reduction and mechanistic toxicology.
Microbeads. The handbook lists them with these products. The dominant problem is physical persistence and sorption of other chemicals in aquatic systems—an exposure and fate control (bans on rinse-off microbeads)—not a CYP2E1 story.
II.7 B and II.7 D close the loop: in vitro and in vivo mechanistic data populate the translation sketch, and cellular and biochemical mechanisms are how you evaluate risk to humans, animals, and the environment—without claiming a binding assay is already a population adverse outcome.
| Organ or decision | Initiating event or uptake | Cellular key event | Organ key event | Human-relevance note |
|---|---|---|---|---|
| Zone-3 hepatocyte | CYP2E1 bioactivation and ROS | Lipid peroxidation and MPT | Centrilobular necrosis, ALT leak | Depends on human CYP2E1 and GSH |
| Cholestasis | BSEP inhibition | Bile-salt retention, mitochondrial injury | Cholestatic chemistry, biliary reaction | Transporter homology is good; kinetics differ |
| Proximal tubule | OAT or OCT concentrative uptake | ATP failure, cell death | Acute tubular injury | Human OATs exist; substrate-specific |
| Male-rat hyaline droplets | α2u-globulin binding | Lysosomal overload | Tubular injury; possible male-rat tumors | Limited human relevance for this mode of action |
| Myelin vs axon vs synapse | Cell-specific target | Demyelination, dying-back, or receptor block | Neuropathy or acute convulsant death | Map the structure before the statute |
| rasH2 or p53+/− | Engineered pathway bias | Altered tumor latency | 26-week tumor call | Strength for screening; not human identity |
| Mixture potentiation | A changes B’s activation or defense | More of B’s cellular key event | More organ injury from B | State the additive null first |
Scenario
A hydrocarbon produces male-rat hyaline droplets, no female-rat renal tumors, and no human-protein binding analog. The cautious reading for that mode of action is limited human renal-tumor relevance. The same dossier must still ask whether the chemical is also a PBT, an endocrine-active substance, or a nanoparticle surface catalyst—those are separate II.7 questions.
A second file: a nanoparticle plus endotoxin contamination produces a greater-than-additive macrophage IL-1β response, and both materials can activate innate immunity alone. That is synergism. Isopropanol plus a CCl4 dose that is not hepatotoxic alone becomes necrotic: potentiation. A computational network that lights an Nrf2 node without necrosis or dysfunction remains a hypothesis, not an organism adverse outcome.
Traps
- Treating α2u-globulin male-rat tumors as default human renal carcinogens.
- Calling BSEP cholestasis “centrilobular CYP2E1 necrosis.”
- Using rasH2 or p53+/− as interchangeable with human cancer or with each other.
- Skipping the additive model before calling synergism.
- Equating an omics node or a receptor screen with an organism or population adverse outcome.
- Treating a microbead ban as if it were a Fenton-chemistry decision.
A hydrocarbon produces hyaline droplets in male-rat proximal tubules, no equivalent droplets in females or mice, and no comparable androgen-driven protein load in humans. Which human-relevance statement is correct for this mode of action?
Which statement correctly names mixture interaction terms used in handbook II.7 C?
A 26-week rasH2 study is positive, a 26-week p53-heterozygous study is negative for a receptor-mediated rodent carcinogen, and an ob/ob mouse shows more steatotic liver injury than a lean control. Which model interpretation matches handbook II.8?