7.4 Phase II Biotransformation: Conjugation Pathways, Bioactivation & Detoxification

Conjugation as the fork between excretion and covalent binding

Phase I often creates a nucleophilic or electrophilic handle. Phase II attaches a cofactor that usually makes the product more polar and a substrate for MRP2, OATs, or biliary/renal excretion. The same chemistry can instead yield a better electrophile (sulfate esters of N-hydroxy aromatic amines; some acyl glucuronides; epoxides that escape GST). Independent OpenExamPrep teaching in this Domain II.2 section covers the major human conjugation paths, glutathione as a depletable defense, three textbook bioactivations (CCl4, aflatoxin B1, benzo[a]pyrene), and conjugation capacity defects in cats and in Crigler-Najjar disease. It is not an ABT product and does not claim official approval, review, or partnership with ABT.

Keep the fork in mind for every parent: Phase I reactive intermediate → Phase II conjugate and excretion, versus covalent binding to protein or DNA when the intermediate outruns conjugation.

UGT, SULT, GST, NAT, amino acids, methylation

UDP-glucuronosyltransferases (UGTs) in the ER transfer glucuronic acid from UDP-glucuronic acid to phenols, alcohols, carboxylic acids, and amines. Products are organic anions exported by MRP2 and related pumps. UGT1A1 glucuronidates bilirubin. UGT1A6 glucuronidates many simple phenols and a large share of acetaminophen in humans. Acyl glucuronides of some carboxylic acids can rearrange and bind protein—an uncommon but real bioactivation.

Sulfotransferases (SULTs) in cytosol transfer sulfonate from PAPS. Sulfation is often high affinity, low capacity: it saturates before glucuronidation at overdose. SULT can detoxify phenols or bioactivate N-hydroxy arylamines by forming sulfate esters that leave a nitrenium ion.

Glutathione-S-transferases (GSTs) conjugate glutathione (GSH) to electrophiles (epoxides, NAPQI, many unsaturated carbonyls). Products are processed to mercapturic acids and excreted. GSTM1-null genotypes are common and are discussed as susceptibility modifiers for some epoxide-forming carcinogens. GSH is a finite pool: synthesis via γ-glutamylcysteine ligase can be outrun; N-acetylcysteine supplies cysteine.

N-acetyltransferases (NAT1, NAT2) transfer acetyl from acetyl-CoA to aromatic amines and hydrazines. NAT2 slow acetylators accumulate isoniazid and some arylamines. Dogs remain poor acetylators (7.3); humans are not.

Amino-acid conjugation (often mitochondrial) activates a carboxylic acid as an acyl-CoA, then links glycine (benzoate → hippurate), taurine, or glutamine. It is a detoxification path for some aromatic acids and a competitor for CoA.

Methylation uses S-adenosylmethionine (SAM): COMT, TPMT (thiopurines—another polymorphism), PNMT, AS3MT (arsenic). Methylation does not always increase water solubility; it often reduces receptor activity. Arsenic methylation is a mixed toxicologic story (some methylated species remain highly toxic).

Acetaminophen: NAPQI, GSH, and centrilobular necrosis

At therapeutic doses, most acetaminophen is glucuronidated (UGT) or sulfated (SULT) and excreted in urine. A minority is oxidized by CYP2E1, with contributions from CYP1A2 and CYP3A4, to N-acetyl-p-benzoquinone imine (NAPQI). GST conjugates NAPQI to GSH. When the dose is large, sulfate capacity saturates, more parent reaches CYPs, GSH falls, and NAPQI covalently binds cysteines on hepatocellular proteins, injuring mitochondria. Necrosis is centrilobular (zone 3), where CYP2E1 is richest and oxygen tension favors some bioactivation patterns. N-acetylcysteine works if given while hepatocytes can still synthesize GSH and before massive necrosis. Ethanol induction of CYP2E1 plus fasting (less GSH) worsens the risk after the ethanol is gone; simultaneous ethanol can compete at CYP2E1. Cats skip much of the UGT safety net (below), so the same milligram-per-kilogram is a different internal-dose problem.

Three bioactivation archetypes

Carbon tetrachloride. CYP2E1 catalyzes reductive dehalogenation to the trichloromethyl radical (•CCl3), which adds oxygen to form trichloromethylperoxy radical. The radicals drive lipid peroxidation, break calcium homeostasis, and kill centrilobular hepatocytes. Phenobarbital or ethanol induction of CYPs can worsen CCl4 injury; hypoxia in zone 3 favors the reductive path. GSH and other antioxidants modulate, but this is a radical story more than a stable epoxide story.

Aflatoxin B1. Hepatic CYP3A4 and CYP1A2 form AFB1-8,9-epoxide. The exo-epoxide binds N7-guanine, producing a promutagenic adduct implicated in hepatocellular carcinoma, especially with chronic hepatitis B. GST conjugation of the epoxide is a major detoxification; epoxide hydrolase also contributes. Species and human GST differences help explain why trout, rats, and humans do not share identical AFB1 potency.

Benzo[a]pyrene. CYP1A1/1B1 (AhR-inducible) form an epoxide; microsomal epoxide hydrolase yields the 7,8-dihydrodiol; a second CYP oxidation creates the bay-region 7,8-diol-9,10-epoxide (BPDE). BPDE adducts DNA (notably N2-guanine). GST and further conjugation compete with adduction. Note the double-edged hydrolase: the diol is on the path to the ultimate electrophile, not always a dead-end detox.

Cats, UGT1A6, and Crigler-Najjar as capacity lessons

Domestic cats carry UGT1A6 as a pseudogene and express a narrower UGT repertoire, so they glucuronidate many phenolic drugs poorly (acetaminophen, some other phenols; propofol handling is also unfavorable). More parent is shunted to oxidation and to pathways that produce methemoglobin and hepatic injury. This is a conjugation-capacity species difference, not a proof that cats lack CYP2E1.

Crigler-Najjar syndrome is a human UGT1A1 deficiency: unconjugated bilirubin cannot be glucuronidated efficiently, producing severe jaundice and kernicterus risk in the complete (type I) form. Gilbert syndrome is a milder UGT1A1 promoter variant. Neither disease means the patient is a cat. The shared lesson is: if the conjugating enzyme is missing, a normally “detox” substrate becomes a toxic parent. Use Crigler-Najjar to remember UGT1A1/bilirubin; use feline UGT1A6 to remember phenols and acetaminophen.

Realistic scenario

A 24-hour mouse APAP study shows centrilobular necrosis, collapsed hepatic GSH, and APAP-protein adducts. Co-administration of NAC 1 hour after APAP reduces adducts and necrosis; a CYP2E1 inducer (chronic ethanol, then a dry interval) worsens them. That package is NAPQI outrunning GST, not NAT2 acetylation and not BSEP cholestasis. A second file: dietary AFB1 in a species with low hepatic GST toward the 8,9-epoxide produces high N7-guanine adducts—bioactivation minus conjugation. A third: a cat that received a “human pediatric” acetaminophen dose; calling it a CYP2D6 polymorphism case misses the UGT1A6 species defect.

Traps

• Treating every Phase II product as automatically less toxic than parent.
• Forgetting that epoxide hydrolase can sit on the path to BPDE.
• Equating Crigler-Najjar (UGT1A1/bilirubin) with feline UGT1A6 (phenols).
• Calling CCl4 injury an aflatoxin-style DNA-epoxide story.
• Using dog acetylation deficiency to explain cat acetaminophen poisoning.