The sensing of essential amino acid deficiency in the anterior piriform cortex, that requires the uncharged tRNA/GCN2 pathway, is sensitive to wortmannin but not rapamycin.

Animals detect and reject their first essential/indispensable amino acid (IAA) deficient meal within 20min; this IAA sensing requires an intact anterior piriform cortex (APC). In the biochemical responses to IAA deficiency in the APC we have shown that: uncharged tRNA is the primary sensor; IAA transport is increased; and signaling, including the extracellular-regulated kinase (ERK1/2), is activated. The mammalian target of rapamycin (mTOR) is a potential AA sensor and is regulated by AA transport. Previously, the inhibitors, rapamycin for mTOR, wortmannin for phosphoinositide 3 kinase (PI3K) and PD98059 for ERK, each blocked the upregulation of the System A transporter in IAA depleted APC neurons. Here we injected these same inhibitors into the APC and measured intake of an IAA deficient diet. Rapamycin had no effect on the rejection of the IAA deficient diet, but wortmannin increased ERK activation and intake of the deficient diet before 40min and PD98059 acted after 40min to increase the second meal. While the specific wortmannin target involved in blocking the behavioral response remains unclear, we conclude that mTOR is dispensable for sensing IAA deficiency in the APC, and that ERK is associated with the secondary learned responses to IAA deficient diets.

Phosphorylation of Ca2+/calmodulin-dependent protein kinase type ii and the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (ampa) receptor in response to a threonine-devoid diet.

The anterior piriform cortex (APC) functions as a chemosensor for indispensable amino acid deficiency and responds to this deficiency with increased activity, as indicated by observations including averaged evoked-potentials and c-fos expression in the APC. Little is known of the intracellular signaling mechanisms that mediate this deficiency-related increase in neuronal excitability, but previous studies have shown effects on intracellular Ca2+ in deficient APC slices in vitro. In the present study we hypothesized that indispensable amino acid deficiency increases intraneuronal Ca2+, resulting in autophosphorylation of calcium/calmodulin-dependent protein kinase type II (CaMKII) in vivo. Results demonstrated that phosphorylation levels of CaMKII (pCaMKII) in APC neurons increased at 20 and 40 min after a single meal of threonine-devoid diet. Phosphorylation of the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptor subunit (GluR1) at the serine 831 (S831) site was modestly increased in the APC in response to a threonine-devoid meal. The GluR1 subunit also showed increased phosphorylation at the 845 (S845) site, suggesting additional signaling mechanisms. Although phosphorylation of CaMKII was sustained, phosphorylation of the GluR1 subunit returned to control levels by 40 min. These effects of amino acid deficiency did not occur throughout the brain as neither CaMKII nor GluR1 showed increased phosphorylation in the neocortex. These findings support the notion that calcium and glutamate signaling in the APC, but not throughout the brain, are triggered during early responses to amino acid deficiency. They also suggest that longer-term changes in APC neurons in response to such a deficiency may be mediated at least in part by CaMKII.

Peptide fragments released from Phe-caseinomacropeptide in vivo in the rat.

The aim of this study was to investigate the pharmacokinetics of bovine Phe-caseinomacropeptide (Phe-CMP) in the rat after oral administration. This polypeptide was monophosphorylated and mainly nonglycosylated: Phe-CMP-1P. During gastrointestinal digestion and absorption, Phe-CMP-1P was degraded. Intact Phe-CMP-1P and CMP-1P were rapidly released from the stomach. In contrast, partial hydrolysis by pancreatic enzymes was observed. In vitro hydrolysis by brush-border membrane vesicles also indicated that the peptide was degraded. In the blood, "CMP-immunoreactive material" appeared rapidly, reaching a maximum level of 5.5 microg/ml at 60 min.

Feeding patterns and meal microstructure during development of a taste aversion to a threonine devoid diet.

Food intake decreases and a conditioned taste aversion is induced when rats are fed a diet that is devoid of an indispensable amino acid. The purpose of this study was to characterize the meal patterns associated with (1) the onset of anorexia after the initial recognition of a threonine deficiency and (2) after the development of the conditioned taste aversion to this deficient diet. When rats ate the threonine-devoid diet for the first time, meal patterns were characterized by an increase in intermeal interval (IMI) between 3 and 6 h after food presentation, which was followed by a decrease in meal size and ingestion rate, between 6 and 12 h. Meal patterns on days 2 and 10 were associated with expression of the taste aversion, characterized by meals of smaller size, longer duration and by a reduction in ingestion rate, without variations in either IMI or meal frequency. Meals of the threonine-deficient group were composed of more frequent bouts, smaller size and shorter duration, with large within-meal pauses, which accounted for the reduced ingestion rate. This study presents the first analysis in terms of feeding patterns and meal microstructure of a conditioned taste aversion induced by a food rather than a toxin.

Elevated baseline triglyceride levels modulate effects of HMGCoA reductase inhibitors on plasma lipoproteins.

BACKGROUND: The response in levels of very-low-density (VLDL) and low-density (LDL) lipoproteins varies substantially among hyperlipidemic patients during treatment with HMGCoA reductase inhibitors. Apolipoprotein E genotype and gender are known to contribute to the regulation of steady state levels of plasma lipoproteins. This study explores the effect of these and other potential determinants of the response of VLDL and LDL to treatment with reductase inhibitors. METHODS: Using mixed linear statistical models, the response of lipoprotein lipid values was studied in 142 hyperlipidemic individuals who were treated with reductase inhibitors. Patients received one or more of the following drugs individually for a total of 623 treatment observations: lovastatin, pravastatin, simvastatin, or atorvastatin. For evaluation of the effects of treatment in the aggregate, actual doses were expressed as equivalent doses of atorvastatin, using factors based on random assignment comparisons in 16 reported studies. The analysis factors considered were apolipoprotein E genotype, baseline average triglycerides >170 mg/dL (vs less), and gender. RESULTS: Presence of an apo epsilon4 allele was associated with a trend toward greater reduction of triglyceride levels and a diminished ability of the reductase inhibitors to reduce LDL cholesterol levels. Gender had only minimal effect on the response of either LDL cholesterol or triglycerides. However, the effect of elevated baseline triglycerides on the response of both triglycerides and LDL cholesterol was striking and was exerted in opposite directions. The triglyceride-lowering effect of reductase inhibitors was greater in patients with initial triglyceride levels above 170 mg/dL (P=0.0001). The effect was even greater in patients with initial triglyceride levels over 250 mg/dL (P=0.015). Conversely, for LDL cholesterol levels, elevated baseline triglycerides were associated with a significantly decreased response to the drugs (P=0.0015). CONCLUSIONS: These findings indicate that baseline triglyceride levels are an important predictor of response of plasma lipoproteins to HMGCoA reductase inhibitors, perhaps reflecting fundamental differences in mechanism underlying the hyperlipidemic phenotype.

Acylation stimulating protein (ASP) acute effects on postprandial lipemia and food intake in rodents.

BACKGROUND: In vitro studies have shown that acylation stimulating protein (ASP) stimulates triglyceride (TG) synthesis and storage in adipocytes. We have previously demonstrated that intraperitoneal (i.p.) injection of ASP in C57BL/6J mice accelerated TG clearance following an orally-administered fat load as well as reducing postprandial glucose levels. RESULTS: In the present study, we first examined the effect of i.p. and intracerebroventricular (i.c.v.) injection of ASP on food intake in Sprague-Dawley rats. Intraperitoneal injection resulted in a short-term increase in food intake (maximum increase 29.3% within the first hour, P<0.025) decreasing thereafter as compared to vehicle alone. i.c.v. Administration of a comparable dose of ASP resulted in a similar but delayed increase in food intake with a maximum at 2-4 h, suggesting that the actions of ASP are peripherally mediated. However, there was no significant difference in 24 h food intake with either i.p. or i.c.v. injection. We also examined the effects of ASP on TG clearance in two obese mouse strains with different metabolic profiles: ob/ob (C57BL/6J-Lep(ob)) and db/db (C57BLKS/J-Lepr(db)). In a crossover design, the response to an oral fat load was determined with and without i.p. injection of exogenous ASP. In ob/ob mice, there was a 44% greater clearance of postprandial TG (area under the curve (AUC)=245+/-49 control vs 138+/-43 mg/dl h with ASP; P<0.05 by RM ANOVA). The db/db mice showed a greater response, with a 62% decrease in postprandial TG (AUC=4080+/-1489 control vs 1540+/-719 mg/dl h with ASP; P=0.004 by RM ANOVA). In addition there were decreases in postprandial glucose and non-esterified fatty acid (NEFA) levels in response to ASP. CONCLUSION: These results are the first to report that ASP can increase food intake in rats and also enhance postprandial TG clearance in obese animals. These data therefore support previous in vitro evidence pointing to ASP as a regulator of lipid metabolism.

Molecular mechanisms in the brain involved in the anorexia of branched-chain amino acid deficiency.

The anterior piriform cortex (APC) of the rat is thought to be the site of indispensable amino acid (IAA) chemosensation in the brain. The branched-chain amino acids, including leucine, are among the IAA that are recognized in the APC. The behavioral outcome of IAA deficiency is an anorectic response. The specific transduction mechanisms by which IAA deficiency and repletion activate the APC are not fully understood, but clearly phosphorylation of proteins, increases in intracellular calcium, and expression of the immediate early gene c-fos, which are among the earliest events occurring after the initial drop in the concentration of the limiting IAA, cause stimulation in the APC. Subsequently, several neurotransmitter systems, including those for norepinephrine, GABA, serotonin, dopamine and nitric oxide, are activated in the APC of rats that have consumed an IAA-imbalanced diet. These systems appear to modulate the output cells from the APC, glutamatergic pyramidal cells that send neural signals to activate subsequent relays in the brain. Ultimately, the feeding circuits of the brain carry out the anorectic response. Continued consumption of a diet containing an IAA imbalance causes a conditioned taste aversion to the diet in all animals that have been studied. Such learning involves synaptic reorganization, requiring both degradation and synthesis of protein, along with alterations in genomic activity.

Chronic marginal iron intakes during early development in mice result in persistent changes in dopamine metabolism and myelin composition.

Marginal iron (Fe) deficiency is prevalent in children worldwide, yet the behavioral and biochemical effects of chronic marginal Fe intakes during early development are not well characterized. Using a murine model, previous work in our laboratory demonstrated persistent behavioral disturbances as a consequence of marginal Fe intakes during early development. In the present study, Swiss-Webster mice fed a control Fe diet (75 microgram Fe/g diet, n = 13 litters) or marginal Fe diet (14 microgram Fe/g diet, n = 16 litters) during gestation and through postnatal day (PND) 75 were killed on PND 75 for assessment of tissue mineral concentrations, dopamine metabolism, myelin fatty acid composition, and c- and m-aconitase activities. In addition, these outcomes were assessed in a group of offspring (n = 13 litters) fed a marginal Fe diet during gestation and lactation and then fed a control diet from PND 21-75. Marginal Fe mice demonstrated significant differences in brain iron concentrations, dopamine metabolism and myelin fatty acid composition relative to control mice; however, no difference in c- or m-aconitase activity was demonstrated in the brain. The postnatal consumption of Fe-adequate diets among marginal Fe offspring did not fully reverse all of the observed biochemical disturbances. This study demonstrates that chronic marginal Fe intakes during early development can result in significant changes in brain biochemistry. The persistence of some of these biochemical changes after postnatal Fe supplementation suggests that they are an irreversible consequence of developmental Fe restriction.

Effects of threonine injections in the lateral hypothalamus on intake of amino acid imbalanced diets in rats.

Previous work from this laboratory suggests that animals decrease their intake of an amino acid imbalanced diet (IMB), due in part to a drop in the concentration of the dietary limiting amino (DLAA) in the anterior piriform cortex (APC). Administration of the DLAA, but not of a non-limiting amino acid into the APC, blocks the anorectic response to IMB. To our knowledge, the effects of DLAA injections on intake of a diet devoid of the DLAA (DEV), have not been examined in areas outside the APC. We hypothesized that the LH is a potential chemosensory area for DLAA. Our objectives were: (1) to determine whether injections of the DLAA threonine into the lateral hypothalamus (LH) alter intake of a threonine-devoid diet (DEV); and (2) to examine the dose-response effects of threonine injections into the LH on intake of threonine-corrected diet (COR). Administration of threonine into the LH stimulated DEV intake during the first 6 h at the 0.25 and 1-nmol doses by approximately 26 and 24%, respectively. Threonine (0.25, 2.5 nmol) did not alter COR intake at any time during the first 12 h. Our results suggest that: (1) the LH, along with the APC, likely acts as a chemosensory brain area for indispensable amino acids; and (2) both the APC and LH are part of a circuit that is involved in the short term anorectic response to amino acid imbalanced diets.

Differential effects of selective vagotomy and tropisetron in aminoprivic feeding.

Both total subdiaphragmatic vagotomy (TVAGX) and serotonin(3) receptor blockade with tropisetron or ondansetron attenuate amino acid-imbalanced diet (Imb) anorexia. Total vagotomy is less effective than tropisetron in reducing Imb-induced anorexia and also blunts the tropisetron effect. With the use of electrocautery at the subdiaphragmatic level of the vagus, we severed the ventral and dorsal trunks as well as the hepatic, ventral gastric, dorsal gastric, celiac, and accessory celiac branches separately or in combination to determine which vagal branches or associated structures may be involved in these responses. Rats were prefed a low-protein diet. On the first experimental day, tropisetron or saline was given intraperitoneally 1 h before presentation of Imb. Cuts including the ventral branch, i.e., TVAGX, ventral vagotomy (above the hepatic branch), and hepatic + gastric vagotomies (but not hepatic branch cuts alone) caused the highest (P < 0.05) Imb intake on day 1 with or without tropisetron. The responses to tropisetron were not affected significantly. On days 2-8, groups having vagotomies that included the hepatic branch recovered faster than sham-treated animals. Because the hepatic and gastric branches together account for most of the vagal innervation to the proximal duodenum, this area may be important in the initial responses, whereas structures served by the hepatic branch alone apparently act in the later adaptation to Imb.

Basolateral and central amygdaloid lesions leave aversion to dietary amino acid imbalance intact.

Expression of c-fos is increased in the central amygdaloid nucleus (CE) of rats ingesting a diet with a severely imbalanced essential amino acid profile (IMB), at a time associated with development of a conditioned taste aversion (CTA). The CE and the basolateral amygdaloid nucleus (BL) both are reported to be involved in the development of CTA. Large amygdaloid lesions involving CE and BL mitigate the normal decrease in intake of IMB; this treatment also impairs CTA to a flavor cue associated with gastrointestinal discomfort. To differentiate their potential roles in aversive responses to IMB, we electrolytically lesioned CE and BL separately. Neither lesion attenuated IMB-induced anorexia, or prevented the avoidance of flavored solutions previously paired with IMB. In contrast, after saccharin-LiCl pairing, CE-lesioned animals showed attenuated CTA to saccharin solution in a two-bottle test. We conclude that neither the CE nor the BL is essential for the reduction of IMB intake, or for CTA associated with IMB. Furthermore, these results suggest that the aversive consequences of IMB intake do not involve gastrointestinal malaise-evoked neurotransmission involving the CE.

Essential amino acids affect interstitial dopamine metabolites in the anterior piriform cortex of rats.

The anterior piriform cortex (APC) is essential for the anorectic reactions to an amino acid-imbalanced diet, and it also responds to repletion of the limiting amino acid. In the present study, we examine the dynamic changes of the interstitial dopamine metabolites in the APC following feeding of either an amino acid-corrected or -imbalanced diet. Microdialysates, collected from the APC, were analyzed using HPLC with electrochemical detection. The concentrations were 19.7 +/- 4.8 microg/L for 3, 4-dyhydroxyphenylacetic acid and 25.1 +/- 4.4 microg/L for homovanillic acid, respectively, in the baseline dialysates. After diet treatments, no significant changes occurred in 3, 4-dyhydroxyphenylacetic acid in the corrected (n = 7) or imbalanced (n = 9) groups vs. the basal group (n = 7). However, after feeding the threonine-corrected diet, the concentration of homovanillic acid was significantly less (P < 0.01) than after the basal and imbalanced diets. The homovanillic acid level in the corrected group was already significantly lower than in the basal group by 20 min (P < 0.05), and reached its lowest level at 70 min (P < 0.05). The concentrations of homovanillic acid in the corrected group remained at this low level until the end of the experiment. The present results introduce the idea that the dopaminergic system is involved in the feeding responses to essential amino acid repletion.

Serotonergic blockade in the treatment of the cancer anorexia-cachexia syndrome.

BACKGROUND: Imbalanced amino acid diets in animals rapidly produce anorexia and weight loss. Blockade of type 3 serotonergic receptors (5HT(3)) can ameliorate anorexia in this animal model. Imbalanced plasma amino acid levels also have been documented in both animal models and human patients with cancer cachexia. Therefore a trial of the 5HT(3) receptor antagonist, ondansetron, was undertaken in the treatment of patients with cancer cachexia. METHODS: Patients with metastatic cancer who were not undergoing chemotherapy or radiotherapy and who had lost >5% of their body weight were eligible. Baseline physical examination; weight; anthropometric studies; levels of retinol binding protein, albumin, and prealbumin; and skin testing for anergy were obtained. The ability to enjoy food was assessed utilizing a seven-point hedonic category scale for specific foods. Therapy was comprised of oral ondansetron, 8 mg twice a day. RESULTS: Twenty-seven patients were enrolled; all were evaluable for toxicity and 20 patients were evaluable for response. Toxicity of ondansetron was minimal. Patients demonstrated significant weight loss prior to disease entry (mean baseline weight of 76.9 kg vs. 72. 1 kg; P < 0.000002). Patients continued to lose weight on study (Week 0: 72.5 kg vs. Week 4: 71.4 kg; P = 0.027); in addition, there was significant deterioration of midarm circumference and hand grip strength, all of which indicated worsening nutritional status. However, a significant improvement in food enjoyment was noted (P = 0.04). CONCLUSIONS: Although it apparently improved the ability of patients to enjoy food, the blockade of 5HT(3) receptors failed to prevent weight loss in patients with cancer cachexia or alter laboratory parameters of protein nutrition.

Essential amino acid deficiency enhances long-term intake but not short-term licking of the required nutrient.

Rats can adjust their nutrient intake in response to nutritional deficiency. This phenomenon has been described extensively for sodium deficiency, whereas other nutrient deficiencies have not been explored thoroughly. Essential amino acid (EAA) deficiency represents a relevant model to describe adaptive changes in behavior resulting from deficiency. The purpose of these experiments was to examine more closely the behavioral responses that occur as a result of lysine (LYS) and threonine (THR) deficiency. Licking to LYS, THR, glycine and distilled water during 10-s trials was measured in control (CON) and EAA-deficient rats. Licking tests were conducted both before and after 23-h intake tests. Although EAA-deficient rats did not show increased licking to the deficient EAA in any of the brief-access tests, in all cases, they did initiate significantly more overall trials than did CON. The EAA-deficient rats also had elevated intake of the deficient EAA in long-duration tests. These findings suggest that LYS or THR deficiency does not emulate the behavioral properties of sodium deficiency in that it does not result in enhanced immediate licking responses to the limiting EAA in brief-access tests. Nevertheless, an appetite is expressed to the relevant EAA in a long-term intake test.

Effects of dorsomedial hypothalamic nuclei lesions on intake of an imbalanced amino acid diet.

Within 3 h of ingesting an imbalanced amino acid diet (Imb), rats show attenuated intake, which can be ameliorated by prior administration of the serotonin receptor antagonist tropisetron (Trop). Earlier work in which the dorsomedial hypothalamic nucleus (DMN) was electrolytically lesioned (DMNL) determined that this structure plays a role in the early detection of and subsequent adaptation to Imb. However, that study did not address whether cell bodies in the DMN, fibers of passage, or both were involved in the DMNL response to Imb. In the present investigation in experiment 1, rats were given electrolytic DMNL or a sham operation (Sham). The rats were injected with saline (Sal) or Trop just before introduction of Imb. By 3 h Sal-DMNL rats consumed more Imb than did the Sal-Sham rats; intake was normal by 12 h. Trop enhanced Imb intake, with Trop and DMNL being additive. By day 4 the DMNL rats were eating and gaining weight less than were Sham rats. In experiment 2, DMN cell bodies were destroyed by ibotenic acid (Ibo). Sal-injected Ibo-lesioned and Sham rats showed similar food intake depression on Imb; Trop similarly increased Imb intake in both groups. By day 4 both Ibo-L rats were eating and gaining weight less than were Sham rats. In experiment 3, groups of rats were given knife cuts posterior, lateral, ventral, dorsal, or anterior to the DMN. During the first 3 h of consuming Imb, all cuts except posterior enhanced the intake of Imb. Over the next 24 h the anterior cut group continued to eat more Imb than did the Sham rats. In experiment 4 DMNL rats were given novel diets; the DMNL rats did not display a neophilic response. The data suggest that fiber tracts that pass through the DMN may be involved in the early detection of Imb. DMN cell bodies, or fibers of passage, are not involved in the Trop effect. Finally, DMN cell bodies are necessary for proper long-term adaptation to Imb.

Increased intracellular calcium in rat anterior piriform cortex in response to threonine after threonine deprivation.

The anterior piriform cortex (APC) may serve as the chemosensor for amino acid (AA) deficiency in rats. To investigate the mechanism by which the APC recognizes a limiting indispensable AA (IAA), we examined changes in intracellular calcium ([Ca2+]i) in APC slices after culture in medium with or without threonine (Thr) or lysine (Lys). The addition of 1 or 10 mM Thr to slices previously incubated in Thr-devoid medium resulted in a significant and sustained increase in [Ca2+]i compared to control slices; an effect not seen when isoleucine, another IAA, was added. Similar results were seen when lysine, but not threonine, was added to slices incubated in lysine-devoid medium. The rise in [Ca2+]i resulting from the addition of the limiting IAA to deficient slices may be linked to enhanced activity of the appropriate AA transporter. This is suggested by preliminary findings that serine, a small neutral AA that uses the same transporter as threonine, gave rise to an enhanced response in the Thr-deficient slice.

CCK(A) and 5-HT3 receptors interact in anorectic responses to amino acid deficiency.

Serotonin3 (5-HT3) receptors in the periphery mediate anorectic responses to the amino acid deficiency, which occurs after eating amino acid-imbalanced diets (IMB). However, other neurochemical systems, notably cholecystokinin (CCK), are known to affect food intake. We pretreated rats systemically with tropisetron, a 5-HT3 receptor antagonist, alone and combined with antagonists of CCK(A) and CCK(B) receptors, and measured intake of an IMB. Devazepide, a CCK(A) receptor antagonist, appeared to interact with tropisetron in the anorectic responses to IMB, blunting the usual remediation of IMB anorexia by tropisetron. The CCK(B) receptor antagonist, L-365, 260, increased intake of both IMB and an amino acid-balanced basal diet (BAS) and did not interact with tropisetron. Our data suggest that activation of CCK(A) receptors is interactive with 5-HT3 receptor activity in mediating IMB anorexia in the aminoprivic feeding model.

DA1 receptor activity opposes anorectic responses to amino acid-imbalanced diets.

The serotonin3 (5-HT3) receptor plays an important role in the aminoprivic feeding model. Other neurochemical systems, including cholecystokinin (CCK) and dopamine (DA), are known to affect food intake. We pretreated rats systemically with tropisetron, a 5-HT3 receptor antagonist, alone and combined with antagonists of DA1 and DA2 receptors, and measured intake of an amino acid-imbalanced diet (IMB). As expected, tropisetron significantly increased intake of IMB. SCH-23390, a DA1 antagonist, increased IMB anorexia. When combined with tropisetron, DA2 antagonism with eticlopride reduced short-term intake of both the basal diet (BAS) and IMB. In the IMB model, specificity of 5-HT3-DA2 interactions, and of 5-HT3-CCK(A) interactions from previous studies, prompted investigation of CCK(A)-DA2 interactions; there appeared to be none. SKF-38393, a DA1 agonist, combined with the CCK(A) receptor antagonist, devazepide, increased BAS and tended to increase IMB intake. Thus, CCK(A)-DA1 interactions were not specific for IMB. These data suggest that DA1 receptor activity opposes IMB anorexia, possibly via an interaction with the 5-HT3 receptor.

Lysine deficiency alters diet selection without depressing food intake in rats.

Under states of protein deficiency, the dietary limiting amino acid, rather than protein content, can act as the dietary stimulus to control diet selection. If fact, threonine-deficient rats will alter their diet selection patterns solely on the basis of very small changes (0.009 g/100 g) in the dietary threonine concentration. In these studies, we assessed whether lysine-deficient rats will also alter their diet selection patterns on the basis of small changes in dietary Lys concentration. In all experiments, growing rats were adapted to diets in which the protein fraction (purified amino acids or wheat gluten) was limiting in Lys. They were then given a choice between the adaptation diet (AD) diet and a slightly more deficient diet. Rats that were adapted to a Lys-deficient diet (0.25 g Lys/100 g) selected their AD over diets containing as little as 0.01% less Lys (P < 0.01) within 5 d. To determine how deficient rats must be before they alter their selection patterns, rats were adapted to diets containing various levels of Lys, i.e., 2 levels below the requirement for growth and 2 levels above the requirement for growth, but below the requirement for maximal nitrogen retention. Only rats adapted to diets containing Lys below their requirement for growth selected their AD over a diet containing 0.05% less Lys (P < 0.005). Finally, to determine whether rats will alter their selection to whole protein-based diets, rats were adapted to 25% wheat gluten diets supplemented with 0.03-0.21% Lys. Rats selected the AD over a diet containing as little as 0.09% less supplemental Lys by d 4 of the trial (P < 0.05). We conclude that rats are sensitive to changes as small as 0.01% in dietary Lys concentration, but that sensitivity requires prior adaptation to Lys-deficient diets.