Choline deficiency-related multi-omics characteristics are susceptible factors for chemotherapy-induced thrombocytopenia.

The XELOX chemotherapy protocol that includes capecitabine and oxaliplatin is the routine treatment for colorectal cancer (CRC), but it can cause chemotherapy-related adverse events such as thrombocytopenia (TCP). To identify predictive biomarkers and clarify the mechanism of TCP susceptibility, we conducted integrative analysis using normal colorectal tissue (CRT), plasma, and urine samples collected before CRC patients received adjuvant XELOX chemotherapy. RNA-sequencing and DNA methylation arrays were performed on CRT samples, while liquid chromatography-mass spectrometry was performed on CRT, plasma, and urine samples. Differentially expressed features (DEFs) from each uni-omics analysis were then subjected to integrative analysis using Multi-Omics Factor Analysis (MOFA). Choline-deficiency in plasma and CRT was found as the most critical TCP-related feature. Based on bioinformatic analysis and literature research, we further concluded that choline-deficiency was the possible reason for most of the other TCP-related multi-omics DEFs, including metabolites representing reduced sphingolipid de novo synthesis and elevated solute carrier-mediated transmembrane transportation in CRT and plasma, DNA hypermethylation and elevated expression of genes involved in neuronal system genes. In terms of thrombocytopoiesis, these TCP-related DEFs may cause atypical maintenance and differentiation of megakaryocyte, resulting a suppressed ability of thrombocytopoiesis, making patients more susceptible to chemotherapy-induced TCP. At last, prediction models were developed and validated with reasonably good discrimination. The area under curves (AUCs) of training sets were all > 0.9, while validation sets had AUCs between 0.778 and 0.926. In conclusion, our results produced reliable marker systems for predicting TCP and promising target for developing precision treatment to prevent TCP.

A Comparison of the Gene Expression Profiles of Non-Alcoholic Fatty Liver Disease between Animal Models of a High-Fat Diet and Methionine-Choline-Deficient Diet.

Non-alcoholic fatty liver disease (NAFLD) embraces several forms of liver disorders involving fat disposition in hepatocytes ranging from simple steatosis to the severe stage, namely, non-alcoholic steatohepatitis (NASH). Recently, several experimental in vivo animal models for NAFLD/NASH have been established. However, no reproducible experimental animal model displays the full spectrum of pathophysiological, histological, molecular, and clinical features associated with human NAFLD/NASH progression. Although methionine-choline-deficient (MCD) diet and high-fat diet (HFD) models can mimic histological and metabolic abnormalities of human disease, respectively, the molecular signaling pathways are extremely important for understanding the pathogenesis of the disease. This review aimed to assess the differences in gene expression patterns and NAFLD/NASH progression pathways among the most common dietary animal models, i.e., HFD- and MCD diet-fed animals. Studies showed that the HFD and MCD diet could induce either up- or downregulation of the expression of genes and proteins that are involved in lipid metabolism, inflammation, oxidative stress, and fibrogenesis pathways. Interestingly, the MCD diet model could spontaneously develop liver fibrosis within two to four weeks and has significant effects on the expression of genes that encode proteins and enzymes involved in the liver fibrogenesis pathway. However, such effects in the HFD model were found to occur after 24 weeks with insulin resistance but appear to cause less severe fibrosis. In conclusion, assessing the abnormal gene expression patterns caused by different diet types provides valuable information regarding the molecular mechanisms of NAFLD/NASH and predicts the clinical progression of the disease. However, expression profiling studies concerning genetic variants involved in the development and progression of NAFLD/NASH should be conducted.

TGF-ß/YB-1/Atg7 axis promotes the proliferation of hepatic progenitor cells and liver fibrogenesis.

Hepatic fibrosis is characterized by excessive extracellular matrix deposition and ductular reactions, manifested as the expansion of hepatic progenitor cells (HPCs). We previously reported that the Y-box binding protein 1 (YB-1) in HPCs is involved in chronic liver injury. In this study, we constructed YB-1f/f Foxl1-Cre mice and investigated the role of YB-1 in HPC expansion in murine choline-deficient, ethionine-supplemented (CDE), and 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) models. Liver injury and fibrosis were measured using hematoxylin and eosin (HE), Masson, and Sirius Red staining. HPC proliferation was detected using EdU and immunofluorescence (IF). Autophagic flow was measured by mCherry-GFP-LC3B staining and transmission electron microscopy (TEM). YB-1 expression was measured by immunofluorescence and western blotting. CUT & Tag analysis, chromatin immunoprecipitation, and RT-PCR were performed to explore the regulation of autophagy-related protein 7 (Atg7) transcription by YB-1. Our results indicated that liver injury was accompanied by high expression of YB-1, proliferative HPCs, and activated autophagy in the CDE and DDC models. YB-1f/f Cre+/- mice displayed less liver injury and fibrosis than YB-1f/f Cre-/- mice in the CDE and DDC models. YB-1 promoted proliferation and autophagy of HPCs in vitro and in vivo. Transforming growth factor-ß (TGF-ß) induced YB-1 nuclear translocation and facilitated the proliferation and autophagy of HPCs. YB-1 nuclear translocation promoted the transcription of Atg7, which is essential for TGF-ß/YB-1 mediated HPCs expansion in vitro and in vivo. In summary, YB-1 nuclear translocation induced by TGF-ß in HPCs promotes the proliferation and autophagy of HPCs and Atg7 participates in YB-1-mediated HPC-expansion and liver fibrosis.

Inhibition of hyaluronan synthesis by 4-methylumbelliferone ameliorates non-alcoholic steatohepatitis in choline-deficient L-amino acid-defined diet-induced murine model.

Hyaluronan (HA) as a glycosaminoglycan can bind to cell-surface receptors, such as TLR4, to regulate inflammation, tissue injury, repair, and fibrosis. 4-methylumbelliferone (4-MU), an inhibitor of HA synthesis, is a drug used for the treatment of biliary spasms. Currently, therapeutic interventions are not available for non-alcoholic steatohepatitis (NASH). In this study, we investigated the effects of 4-MU on NASH using a choline-deficient amino acid (CDAA) diet model. CDAA diet-fed mice showed NASH characteristics, including hepatocyte injury, hepatic steatosis, inflammation, and fibrogenesis. 4-MU treatment significantly reduced hepatic lipid contents in CDAA diet-fed mice. 4-MU reversed CDAA diet-mediated inhibition of Ppara and induction of Srebf1 and Slc27a2. Analysis of serum ALT and AST levels revealed that 4-MU treatment protected against hepatocellular damage induced by CDAA diet feeding. TLR4 regulates low molecular weight-HA-induced chemokine expression in hepatocytes. In CDAA diet-fed, 4-MU-treated mice, the upregulated chemokine/cytokine expression, such as Cxcl1, Cxcl2, and Tnf was attenuated with the decrease of macrophage infiltration into the liver. Moreover, HA inhibition repressed CDAA diet-induced mRNA expression of fibrogenic genes, Notch1, and Hes1 in the liver. In conclusion, 4-MU treatment inhibited liver steatosis and steatohepatitis in a mouse model of NASH, implicating that 4-MU may have therapeutic potential for NASH.

Plasma lipoprotein profile from nonalcoholic steatohepatitis model rats.

We recently revealed that increases in particle sizes of very-low-density lipoproteins (VLDL) are highly correlated with the progression of nonalcoholic fatty liver disease (NAFLD)/nonalcoholic steatohepatitis (NASH), and VLDL particle size may be a minimally invasive indicator of these hepatic disorders. Methionine and choline-deficient (MCD) diet fed animals are usually used as a NASH model; however, the application of this minimally invasive biomarker in MCD diet fed animals remains unclear. In the present study, we measured the levels of liver disease markers and plasma lipoprotein profiles in MCD diet fed rats, and compared them with those of normal diet fed rats. Assessing lipoprotein profiles showed marked increases in VLDL particle sizes in MCD diet fed rats with pathologically and biochemically NASH-like features.

DSS colitis promotes tumorigenesis and fibrogenesis in a choline-deficient high-fat diet-induced NASH mouse model.

Nonalcoholic steatohepatitis (NASH) patients progress to liver cirrhosis and even hepatocellular carcinoma (HCC). Several lines of evidence indicate that accumulation of lipopolysaccharide (LPS) and disruption of gut microbiota play contributory roles in HCC. Moreover, in a dextran sodium sulfate (DSS)-induced colitis model in mice, a high-fat diet increases portal LPS level and promotes hepatic inflammation and fibrosis. However, this diet-induced NASH model requires at least 50 weeks for carcinogenesis. In this study, we sought to determine whether increased intestinal permeability would aggravate liver inflammation and fibrosis and accelerate tumorigenesis in a diet-induced NASH model. Mice were fed a choline-deficient high-fat (CDHF) diet for 4 or 12 weeks. The DSS group was fed CDHF and intermittently received 1% DSS in the drinking water. Exposure to DSS promoted mucosal changes such as crypt loss and increased the number of inflammatory cells in the colon. In the DSS group, portal LPS levels were elevated at 4 weeks, and the proportions of Clostridium cluster XI in the fecal microbiota were elevated. In addition, levels of serum transaminase, number of lobular inflammatory cells, F4/80 staining-positive area, and levels of inflammatory cytokines were all elevated in the DSS group. Liver histology in the DSS group revealed severe fibrosis at 12 weeks. Liver tumors were detected in the DSS group at 12 weeks, but not in the other groups. Thus, DSS administration promoted liver tumors in a CDHF diet-induced NASH mouse over the short term, suggesting that the induction of intestinal inflammation and gut disruption of microbiota in NASH promote hepatic tumorigenesis.

Prenatal choline deficiency does not enhance hippocampal vulnerability after kainic acid-induced seizures in adulthood.

Choline is a vital nutrient needed during early development for both humans and rodents. Severe dietary choline deficiency during pregnancy leads to birth defects, while more limited deficiency during mid- to late pregnancy causes deficits in hippocampal plasticity in adult rodent offspring that are accompanied by cognitive deficits only when task demands are high. Because prenatal choline supplementation confers neuroprotection of the adult hippocampus against a variety of neural insults and aids memory, we hypothesized that prenatal choline deficiency may enhance vulnerability to neural injury. To examine this, adult offspring of rat dams either fed a control diet (CON) or one deficient in choline (DEF) during embryonic days 12-17 were given multiple injections (i.p.) of saline (control) or kainic acid to induce seizures and were euthanized 16 days later. Perhaps somewhat surprisingly, DEF rats were not more susceptible to seizure induction and showed similar levels of seizure-induced hippocampal histopathology, GAD expression loss, upregulated hippocampal GFAP and growth factor expression, and increased dentate cell and neuronal proliferation as that seen in CON rats. Although prenatal choline deficiency compromises adult hippocampal plasticity in the intact brain, it does not appear to exacerbate the neuropathological response to seizures in the adult hippocampus at least shortly after excitotoxic injury.

Perinatal choline deficiency produces abnormal sensory inhibition in Sprague-Dawley rats.

Adequate choline levels in rodents during gestation have been shown to be critical to several functions, including certain learning and memory functions, when tested at adulthood. Choline is a selective agonist for the alpha7 nicotinic receptor which appears in development before acetylcholine is present. Normal sensory inhibition is dependent, in part, upon sufficient numbers of this receptor in the hippocampus. The present study assessed sensory inhibition in Sprague-Dawley rats gestated on normal (1.1 g/kg), deficient (0 g/kg) or supplemented (5 g/kg) choline in the maternal diet during the critical period for cholinergic cell development (E12-18). Rats gestated on deficient choline showed abnormal sensory inhibition when tested at adulthood, while rats gestated on normal or supplemented choline showed normal sensory inhibition. Assessment of hippocampal alpha-bungarotoxin to visualize nicotinic alpha7 receptors revealed no difference between the gestational choline levels. These data suggest that attention to maternal choline levels for human pregnancy may be important to the normal functioning of the offspring.

Prenatal choline deficiency increases choline transporter expression in the septum and hippocampus during postnatal development and in adulthood in rats.

Supplementation of maternal diet with the essential nutrient, choline, during the second half of pregnancy in rats causes long-lasting improvements in spatial memory in the offspring and protects them from the memory decline characteristic of old age. In contrast, prenatal choline deficiency is associated with poor performance in certain cognitive tasks. The mechanism by which choline influences learning and memory remains unclear; however, it may involve changes to the hippocampal cholinergic system. Previously, we showed that the hippocampi of prenatally [embryonic days (E) 11-17] choline-deficient animals have increased synthesis of acetylcholine (ACh) from choline transported by the high-affinity choline transporter (CHT) and reduced ACh content relative to the control and to the E11-17 choline-supplemented rats. In the current study, we found that, during postnatal period [postnatal days (P) 18-480], prenatal choline deficiency increased the expression of CHT mRNA in the septum and CHT mRNA and protein levels in the hippocampus and altered the pattern of CHT immunoreactivity in the dentate gyrus. CHT immunoreactivity was more prominent in the inner molecular layer in prenatally choline-deficient rats compared to controls and prenatally choline-supplemented animals. In addition, in all groups, we observed a population of hilar interneurons that were CHT-immunoreactive. These neurons are the likely source of the hippocampal CHT mRNA as their number correlated with the levels of this mRNA. The abundance of hippocampal CHT mRNA rose between P1 and P24 and then declined reaching 60% of the P1 value by P90. These data show that prenatal availability of choline alters its own metabolism (i.e., CHT expression). While the upregulated CHT expression during the period of prenatal choline deficiency may be considered as a compensatory mechanism that could enhance ACh synthesis when choline supply is low, the persistent upregulation of CHT expression subsequent to the brief period of prenatal deprivation of choline in utero might be beneficial during choline deficiency in adulthood.

Choline cannot be replaced by propanolamine in mice.

Choline is an important nutrient for humans and animals. Animals obtain choline from the diet and from the catabolism of phosphatidylcholine made by phosphatidylethanolamine N-methyltransferase (PEMT). The unique model of complete choline deprivation is Pemt(-/-) mice that are fed a choline-deficient diet. This model, therefore, can be used for the examination of choline substitutes in mammalian systems. Recently, propanolamine was found to be a replacement for choline in yeast. Thus, we tested to see whether or not choline can be replaced by propanolamine in mice. Mice were fed a choline-deficient diet and supplemented with either methionine, 2-amino-propanol, 2-amino-isopropanol and 3-amino-propanol. We were unable to detect the formation of any of the possible phosphatidylpropanolamines. Moreover, none of them prevented liver damage, reduction of hepatic phosphatidylcholine levels or fatty liver induced in choline-deficient-Pemt(-/-) mice. These results suggest that choline in mice cannot be replaced by any of the three propanolamine derivatives.

Review of the carcinogenic activity of diethanolamine and evidence of choline deficiency as a plausible mode of action.

Diethanolamine (DEA) is a chemical used widely in a number of industries and is present in many consumer products. Studies by the National Toxicology Program (NTP) have indicated that lifetime dermal exposure to DEA increased the incidence and multiplicity of liver tumors in mice, but not in rats. In addition, DEA was not carcinogenic when tested in the Tg.Ac transgenic mouse model. Short-term genotoxicity tests have yielded negative results. In view of these apparent inconsistencies, we have critically evaluated the NTP studies and other data relevant to assessing the carcinogenic potential of DEA. The available data indicate that DEA induces mouse liver tumors by a non-genotoxic mode of action that involves its ability to cause choline deficiency. The following experimental evidence supports this hypothesis. DEA decreased the hepatic choline metabolites and S-adenosylmethionine levels in mice, similar to those observed in choline-deficient mice. In contrast, DEA had no effect in the rat, a species in which it was not carcinogenic at a maximum tolerated dose level. In addition, a consistent dose-effect relationship had been established between choline deficiency and carcinogenic activity since all DEA dosages that induced tumors in the NTP studies were also shown to cause choline deficiency. DEA decreased phosphatidylcholine synthesis by blocking the cellular uptake of choline in vitro, but these events did not occur in the presence of excess choline. Finally, DEA induced transformation in the Syrian hamster embryo cells, increased S-phase DNA synthesis in mouse hepatocytes, and decreased gap junctional intracellular communication in primary cultured mouse and rat hepatocytes, but all these events were prevented with choline supplementation. Since choline is an essential nutrient in mammals, this mode of action is qualitatively applicable to humans. However, there are marked species differences in susceptibility to choline deficiency, with rats and mice being far more susceptible than other mammalian species including humans. These differences are attributed to quantitative differences in the enzyme kinetics controlling choline metabolism. The fact that DEA was carcinogenic in mice but not in rats also has important implications for human risk assessment. DEA has been shown to be less readily absorbed across rat and human skin than mouse skin. Since a no observed effect level for DEA-induced choline deficiency in mice has been established to be 10 mg/kg/d, this indicates that there is a critical level of DEA that must be attained in order to affect choline homeostasis. The lack of a carcinogenic response in rats suggests that exposure to DEA did not reach this critical level. Since rodents are far more sensitive to choline deficiency than humans, it can be concluded that the hepatocarcinogenic effect of DEA in mice is not predictive of similar susceptibility in humans.

Diethanolamine induces hepatic choline deficiency in mice.

The purpose of the present experiments was to test the hypothesis that diethanolamine (DEA), an alkanolamine shown to be hepatocarcinogenic in mice, induces hepatic choline deficiency and to determine whether altered choline homeostasis was causally related to the carcinogenic outcome. To examine this hypothesis, the biochemical and histopathological changes in male B6C3F1 mice made choline deficient by dietary deprivation were first determined. Phosphocholine (PCho), the intracellular storage form of choline was severely depleted, decreasing to about 20% of control values with 2 weeks of dietary choline deficiency. Other metabolites, including choline, glycerophosphocholine (GPC), and phosphatidylcholine (PC) also decreased. Hepatic concentrations of S-adenosylmethionine (SAM) decreased, whereas levels of S-adenosylhomocysteine (SAH) increased. Despite these biochemical changes, fatty liver, which is often associated with choline deficiency, was not observed in the mice. The dose response, reversibility, and strain-dependence of the effects of DEA on choline metabolites were studied. B6C3F1 mice were dosed dermally with DEA (0, 10, 20, 40, 80, and 160 mg/kg) for 4 weeks (5 days/week). Control animals received either no treatment or dermal application of 95% ethanol (1.8 ml/kg). PCho was most sensitive to DEA treatment, decreasing at dosages of 20 mg/kg and higher and reaching a maximum 50% depletion at 160 mg/kg/day. GPC, choline, and PC also decreased in a dose-dependent manner. At 80 and 160 mg/kg/day, SAM levels decreased while SAH levels increased in liver. A no-observed effect level (NOEL) for DEA-induced changes in choline homeostasis was 10 mg/kg/day. Choline metabolites, SAM and SAH returned to control levels in mice dosed at 160 mg/kg for 4 weeks and allowed a 2-week recovery period prior to necropsy. In a manner similar to dietary choline deficiency, no fatty change was observed in the liver of DEA-treated mice. In C57BL/6 mice, DEA treatment (160 mg/kg) also decreased PCho concentrations, without affecting hepatic SAM levels, suggesting that strain-specific differences in intracellular methyl group regulation may influence carcinogenic outcome with DEA treatment. Finally, in addition to the direct effects of DEA on choline homeostasis, dermal application of 95% ethanol for 4 weeks decreased hepatic betaine levels, suggesting that the use of ethanol as a vehicle for dermal application of DEA may exacerbate or confound the biochemical actions of DEA alone. Collectively, the results demonstrate that DEA treatment causes a spectrum of biochemical changes consistent with choline deficiency in mice and demonstrate a clear dose concordance between DEA-induced choline deficiency and hepatocarcinogenic outcome.

Potential mechanisms of tumorigenic action of diethanolamine in mice.

Diethanolamine (DEA), a secondary amine found in a number of consumer products, reportedly induces liver tumors in mice. In an attempt to define the tumorigenic mechanism of DEA, N-nitrosodiethanolamine (NDELA) formation in vivo and development of choline deficiency were examined in mice. DEA was administered with or without supplemental sodium nitrite to B6C3F1 mice via dermal application (with or without access to the application site) or via oral gavage for 2 weeks. Blood levels of DEA reflected the dosing method used; oral greater than dermal with access greater than dermal without access. No NDELA was observed in the urine, blood or gastric contents of any group of treated mice. Choline, phosphocholine and glycerophosphocholine were decreased

Inhibitory effects of N,N'-diphenyl-p-phenylenediamine on the early stage of the enhanced hepatocarcinogenesis caused by coadministration of ethionine and a choline-deficient L-amino acid-defined diet in rats.

Effects of N,N'-diphenyl-p-phenylenediamine (DPPD), an antioxidant, on liver carcinogenesis caused by a choline-deficient L-amino acid-defined (CDAA) diet containing ethionine were studied in Fischer 344 rats. Male animals, 6 weeks old, were fed a CDAA diet, a choline-supplemented L-amino acid-defined (CSAA) diet or a CDAA diet containing 0.05% ethionine with or without 0.2% DPPD. Histological changes and lesions positive for gamma-glutamyltransferase (GGT) were analyzed 12 weeks after the beginning of the experiment. The levels of 8-hydroxyguanine (8-OHGua) in DNA and 2-thiobarbituric acid-reacting substances (TBARS) were measured as the parameters for cellular oxidative damage after 4 and 11 days of treatment. Expression of c-myc and c-Ha-ras was also investigated in relation to cell proliferation after 2, 4, 8 and 11 days. Histologically, development of diffuse fatty liver observed in rats fed a CDAA diet was inhibited, while massive oval cell proliferation and cholangiofibrosis resulted from the addition of ethionine with/without DPPD. The sizes but not numbers of GGT-positive lesions seen in the liver of rats fed a CDAA diet were increased and the levels of 8-OHGua formation and TBARS generation were also increased by the ethionine supplement. Both numbers and sizes of GGT-positive lesions were decreased and the level of TBARS, but not 8-OHGua, was decreased by adding DPPD. The increased expression of c-myc and c-Ha-ras detected in the liver of rats fed a CDAA diet was further increased by addition of ethionine and again reduced by DPPD. These results indicate that an antioxidant DPPD can inhibit the early stage of enhanced hepatocarcinogenesis caused by coadministration of ethionine and a CDAA diet, by blocking cellular oxidative damage as well as c-myc and c-Ha-ras expression.

Incomplete reversibility of an experimentally induced hypocholinergic state: biochemical and physiological, but not behavioral, recovery.

In previous reports, we described the experimental development of a hypocholinergic state in rats following the total replacement of dietary choline by an artificial isostere, N-aminodeanol (NADe). NADe shares most of the physicochemical and biochemical characteristics of choline (Ch) but is utilized less efficiently in pathways leading to the formation of both acetylcholine and phospholipids. This experimental model mimics many of the features of human degenrative dementias. We now discuss the behavioral and physiological effects of restoring a normal diet after the hypocholinergic state has become well established. The procedure by which that state was induced has been described in detail in earlier publications. After replacing Ch in the diets of weanling rats for 270 days, NADe replaced 70-85% of the phospholipid-bound Ch in plasma, brain, and peripheral tissue. When dietary NADe was removed and Ch was restored in the diet, NADe disappeared and plasma and erythrocyte (RBC) choline levels returned to normal within 30-60 days. Quinuclidinyl benzilate (QNB) binding showed that muscarinic receptors continued to be depressed in animals remaining on the NADe diet, but returned to control levels in the reversal group. There were no differences in cholinesterase activity among the three treatments. Choline acetyltransferase activity returned to control levels, while continuing to be lower in the NADe animals. Liver lipids were elevated in the latter and not significantly different in the control and reversal groups. Among physiological functions, body weight increased more rapidly in the reversal group than in animals continuing on the NADe diet. Brain weights of the reversal animals were significantly greater than those of animals not reversed, but less than controls. Core body temperatures did not differ from controls at any time during the reversal period. Behaviorally, nociceptive thresholds indicative of sensory-reflexive and sensory-perceptual responses remained significantly below normal, that is, a hyperalgesic state. Reversal animals also remained hyperactive and displayed memory significantly poorer than those on the normal diet, that is, no improvement over animals continuing on NADe. In general, the results suggest that behavioral losses induced by NADe reflect persisting changes in the CNS, despite essentially complete recovery of biochemical parameters. The changes may be morphological or be associated with adaptive changes in other neurochemical events in the CNS.

Chronic toxicity of methotrexate in rats: partial to complete projection of the liver by choline: Brief communication.

Methotrexate (MTX) inhibits the enzyme dihydrofolate reductase, which in turn limits the body's ability to perform transmethylation reactions. This study examined the hypothesis that the consequent deficiency of an important methylated compound, choline, may have contributed to the MTX-induced fatty change in the liver of W rats. Groups of rats were given MTX alone or MTX plus choline in varying dose combinations. All groups but one receiving the combined treatment showed a significantly lower triglyceride concentration in their livers and much less visible hepatocytic fat on histologic examination than did those given MTX alone. The protective effect of choline on the liver was dose related, the unaffected group having received a very small amount. Growth rate, survival, and hematopoietic depression due to MTX were unaltered by choline administration.