- Email: [email protected]
Effect of lysine on afferent activity of the hepatic branch of the vagus nerve in normal and l -lysine-deficient rats
Physiology & Behavior 72 (2001) 685 ± 690
Effect of lysine on afferent activity of the hepatic branch of the vagus nerve in normal and L-lysine-deficient rats Kunio Toriia,*, Akira Niijimab a
Basic Research Laboratories, Central Research Laboratories, Ajinomoto Co., Inc., 1-1 Suzuki-cho, Kawasaki 210-8681, Japan b Department of Physiology, Niigata University School of Medicine, Niigata 951, Japan Received 18 October 2000; received in revised form 19 December 2000; accepted 29 December 2000
Abstract Amino acid deficiency was modeled by feeding rats a diet deficient in the essential L-amino acid, L-lysine (L-lys). There is a rapid anorectic response to such a diet, and a strong preference for L-lys develops during the deficiency. While the brain appears to trigger this preference, the peripheral pathways that inform the brain about the deficiency are not well understood. One possible information pathway may utilize an ``amino acid sensor'' in the hepatoportal region. In the present study, we measured in vivo neural activity in normal and L-lysdeficient rat. Compared to the normally fed controls, we found an approximately 100-fold increase in the firing sensitivity of the L-lys sensors in vagal afferent fibers from the hepatoportal region of the L-lys-deficient rats. Injection of 10 mM L-lys into the hepatoportal circulation, but not D-lysine (D-lys), evoked an increase in afferent activity. While L-lys deficiency enhanced the sensitivity of the L-lys sensors, the sensitivity due to other small amino acid sensors remained unchanged. Finally, we observed a time-dependent response of the lysine sensors to lysine deficiency. It required 3 ± 4 days of maintenance on the lysine-deficient diet for the sensitivity of the L-lys sensors to change. Taken together, these results provide additional data to support the existence of putative L-amino acid sensors in the hepatoportal circulation. Additionally, they describe several characteristics of the L-lys sensors and show that these sensors may contribute to the adaptation to dietary L-lys deficiency and to maintenance of L-amino acid homeostasis. D 2001 Elsevier Science Inc. All rights reserved. Keywords: Lysine; Amino acid; Deficiency; Hepatoportal sensors; Rat
1. Introduction Several studies report that feeding a diet deficient in an essential L-amino acid is followed by a change in taste preference to induce the selective intake of that amino acid, to the extent that the deficiency is alleviated. For example, L-lysine (L-lys), an essential amino acid, was most preferred during lysine deficiency in a choice paradigm over other solutions containing various L-amino acids [5,11]. It is known that the concentration of an essential amino acid L-lys in the plasma decreases a few hours after rats are exposed to a diet deficient in L-lys [5]. Brain circuits that include the anterior piriform cortex [3], several hypothalamic areas [1,8,9,12], and possibly the amygdala [3] trigger an anorectic response to an essential amino acid [3,10]
* Corresponding author. Tel.: +81-44-244-7183; fax: +81-44-2105893. E-mail address: [email protected] (K. Torii).
deficient diet. However, whether and how the L-lys deficiency is sensed in the periphery still remains unclear. It is feasible that a peripheral L-lys sensor [7] can detect L-lys in the circulating blood and alimentary organs. Neural information that is sent from such a sensor via vagal afferent fibers may be one of the pathways informing the brain about the level of peripherally circulating amino acids. The current report deals with sensory signals from hypothesized hepatoportal L-lys sensors in the normal and L-lys-deficient rats in vivo.
2. Methods Forty-two male Wistar rats weighing about 300 g were used. Twenty-one rats were fed a diet deficient in L-lys, while the others were fed a normal diet. To develop L-lys deficiency, rats were offered a diet deficient in L-lys (formulated as previously described [5]) for 7± 10 days. Afferent recordings from the vagus nerve were made in animals
0031-9384/01/$ ± see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 1 ) 0 0 4 2 6 - 7
686
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
Table 1 Effect of L-lys solution injection (10 mM, 0.1 ml ipv) on afferent discharge rates (impulses/5 s of five rats S.E.M.) in normally fed rats Rat (N)
Control
30 min
60 min
90 min
1 2 3 4 5 Mean
59.2 1.3 59.5 2.0 58.8 1.7 75.4 4.5 66.3 2.2 63.8 3.2
106.0 3.2* 75.8 3.8* 121.0 5.2* 80.6 3.6* 73.2 1.4* 91.3 9.4*
113.0 2.4* 77.6 3.0* 135.3 4.6* 119.4 8.0* 67.0 1.3 102.5 12.9*
117.3 2.1* 68.6 1.8* 80.0 2.6* 133.7 5.8* 83.6 2.9* 96.6 12.3*
* Denotes statistically significant increase, P < .05.
having consumed their respective diets for 7 ± 10 days. Food, but not water, was withdrawn at least 6 h before the beginning of each experiment. Animals were anesthetized with an intraperitoneal urethane injection (1 g/kg). An in vivo preparation was made by cannulating the portal vein. With a dissection microscope, the hepatic vagus was isolated and divided. An isolated nerve filament from the peripheral cut end of each rat was placed on a pair of silver wire recording electrodes to record afferent activity. The recording electrodes were immersed in a mixture of liquid paraffin and
petroleum jelly to prevent tissue dehydration. Afferent nerve activity was amplified in a condenser-coupled amplifier. All nerve activities were analyzed after conversion of raw data to standard pulses by a window discriminator, which separated discharges from background noise. Multiunits (about 5 ± 10 units) recordings were made. The techniques employed have been described in detail elsewhere [6]. To interpret the discharge rate data, a spike counter with a reset time of 5 s was used. For intraportal infusion, L-lys, D-lysine (D-lys), L-alanine and L-leucine (reagent grade; Ajinomoto, Japan) were used. All of these were individually dissolved in saline before use. Usually, one injection with duration of 30 s was made into the portal vein in each preparation, except the experiments presented in Figs. 1, 3 and 4) in which two or three successive injections were performed. The time of injection was after recording of stabilized afferent discharges for 30 min . The effect of injection of these amino acids was determined by comparing the mean number of impulses per 5 s in 10 successive 50-s impulse recordings. Values of afferent discharge rates were expressed as mean with standard error, and differences were evaluated with one-way analysis of variance (ANOVA, P < .05).
Fig. 1. Dose-related response of the afferent activity from putative L-lys sensors in the hepatoportal region of the normal rat. The upper trace represents the effect of 0.1 mM L-lys. There was no significant difference among the pre- and postinjective values. The lower trace presents neural activity after injections of L-lys at 1 and 10 mM. The discharge rates at 0, 30 and 60 min after injection of the 1 mM solution were 47.8 1.6, 72.1 2.4* and 48.3 1.3* (asterisk indicates significant increase, P < .05). The discharge rates at 0, 30 and 90 min after administration of 10 mM L-lys were 58.5 1.7, 121.0 5.2*, 135.3 4.6* and 80.0 2.6* (asterisk indicates significant increase, P < .05). Note that the horizontal axis in the lower trace is not linear rather is adjusted in order to compare the two abovedescribed administrations (1 and 10 mM).
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
3. Results Table 1 summarizes the mean discharge rate of five normal preparations before and 30, 60 and 90 min after an injection of 10 mM L-lys. There was a time-dependent increase in the firing rate after the L-lys injection. Not only mean values but also each individual preparation showed significant increase after injection. Fig. 1 shows the dose-dependent effects of injections of L-lys (0.1 mM, 1 mM, 10 mM; 0.1 ml each) into the portal vein on the afferent activity of the hepatic branch of the vagus nerve of a rat fed the normal diet for 7 days. As shown in the top trace, injection of 0.1 mM L-lys (0.1 ml) produced no effect on afferent activity. The discharge rates at 0, 30 and 60 min after injection of 0.1 mM were 56.2 3.1, 52.3 1.9 and 60.3 1.5 impulses/5 s, respectively. The lower trace and the graph present neural activity following injections of L-lys at 1 and 10 mM. The discharges immediately before, 30 and 60 min after injection of 1 mM solution were 47.8 1.6, 72.1 2.4* and 48.3 1.3* impulses/5 s, respectively (asterisk indicates significant increase, P < .05). The discharge rates at 0, 30, 60 and 90 min after administration of 10 mM solution were 58.5 1.7, 121.0 5.2*, 135.3 4.6* and 80.0 2.6*, impulses/5 s, respectively (asterisk indicates significant increase, P < .05; N = 6). Please note that the horizontal axis in the lower trace of Fig. 1 is not linear rather is adjusted in order to compare the two abovedescribed administrations (1 and 10 mM). Fig. 2 illustrates representative traces of hepatovagal afferent nerve activity on the effects of intraportal admin-
Fig. 2. Effect of intraportal injection of L-lys in the normal and L-lysdeficient rats. In the upper trace, injection of 0.01 mM (0.1 ml) L-lys in a normal rat resulted in the discharge rates at 0, 30, 60 and 90 min after injection of 79.3 1.8, 77.4 2.8, 74.8 3.3 and 80.2 2.9 impulses/5 s, respectively. There were no significant differences among these values. In the middle trace, injection of 0.01 mM (0.1 ml) L-lys into a L-lys-deficient rat resulted in discharge rates before (control) and 30, 60 and 90 min after injection of 61.2 3.3, 79.7 2.9*, 96.7 2.2* and 118.5 2.4* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). The bottom trace demonstrates the effect of intraportal injection of 0.001 mM L-lys into a L-lys-deficient rat. As shown in the trace, no remarkable changes in afferent activity were observed.
687
Table 2 Effect of L-lys solution injection (0.01 mM, 0.1 ml ipv) on afferent discharge rates (impulses/5 s of N rats S.E.M.) in L-lys-deficient and normally fed rats Rat (N) Normally 1 2 3 4 5 Mean
Control fed rats 53.8 2.3 79.3 1.8 57.2 4.1 55.5 2.1 56.1 1.4 60.4 4.8
L-lys-deficient rats 1 65.5 1.7 2 61.4 2.3 3 66.3 2.2 4 60.5 2.3 5 61.2 3.3 6 69.7 3.2 7 55.6 1.4 Mean 62.9 1.8
30 min
60 min
90 min
50.9 1.3 77.4 2.9 63.8 4.7 52.8 1.4 55.8 2.6 60.1 4.8
48.7 2.5 74.8 3.3 51.2 3.7 56.4 1.7 54.9 1.5 57.2 4.6
48.8 1.3 80.2 2.9 49.2 2.6 57.2 1.3 53.1 1.6 57.7 5.8
96.5 3.6* 79.4 2.7* 59.0 2.8 68.8 2.7 79.7 2.9* 82.0 1.6* 65.2 3.4* 75.9 4.7
193.6 7.4* 76.4 2.6* 94.1 2.7* 91.0 3.5* 96.7 2.2* 103.5 3.2* 64.3 4.4* 102.8 15.9*
158.4 6.5* 79.6 2.8* 117.7 4.6* 113.9 5.8* 118.5 2.4* 123.7 2.9* 67.1 2.7* 111.3 11.4*
(NS) (NS) (NS) (NS) (NS) (NS)
NS, nonsignificant difference. * Denotes statistically significant increase, P < .05.
istration of L-lys to a L-lys-deficient rat and to a normal rat. As shown in the top trace, injection of 0.01 mM (0.1 ml) L-lys caused no noticeable change in afferent activity in a normal rat. However, a gradual significant increase in afferent activity was observed following administration of 0.01 mM (0.1 ml) L-lys to a rat fed the L-lys-deficient diet for 7 days (middle trace). The discharge rates immediately before injection, 30, 60 and 90 min after injection were 61.2 3.3, 79.7 2.9*, 96.7 2.2* and 118.5 2.4*, impulses/5 s, respectively (asterisk indicates significant increase, P < .05). The bottom trace demonstrates the effect of intraportal injection of 0.001 mM L-lys to the L-lysdeficient rat (the same rat as above). As shown in this lower trace, no significant changes in afferent activity were observed. The discharge rates immediately before, 30, 60 and 90 min after injection were 59.1 3.0, 59.0 2.7, 56.2 3.9 and 52.0 2.3 impulses/5 s, respectively (N = 7). Table 2 summarize the mean discharge rates of vagal hepatic afferents following intraportal injection of 0.01 mM (0.1 ml) L-lys in normal (N = 5) and L-lys-deficient (N = 7) rats. In rats fed a normal diet, there were no significant differences in afferent activity observed after injection, not only for the mean values but also for each individual preparation. However, in rats fed the L-lys-deficient diet for 7 days, not only the mean values but also each individual preparation showed significant increases in activity at 60 and 90 min after injection of 0.01 mM L-lys. The upper trace of Fig. 3 shows the effect of intraportal injection of 0.01 mM D-lys (0.1 ml) on the afferent activity of the hepatic branch of the vagus nerve in a rat fed the normal diet. The discharge rates immediately before, 30, 60 and 90 min after injection were 67.3 3.0, 60.9 3.2 and 66.9 2.2 impulses/5 s, respectively. The middle trace
688
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
Fig. 3. Effect of intraportal injections of D-lys and L-lys on the mean afferent discharge rate of vagal hepatic afferents in the normal and L-lysdeficient rat. The upper trace of shows the effect of intraportal injection of 0.01 mM D-lys (0.1 ml) on the afferent activity of the hepatic branch of the vagus nerve in a rat fed the normal diet. The middle trace shows the effect of 10 mM D-lys (0.1 ml). The bottom trace presents the effect of intraportal injection of 0.01 mM (0.1 ml) D-lys and 0.01 mM (0.1 ml) L-lys to a L-lysdeficient rat. For quantitative data, see Results.
shows the effect of 10 mM D-lys (0.1 ml). The discharge rates just before and 30, 60, 90 and 120 min after injection were 66.2 2.8, 91.0 2.7*, 187.3 4.4*, 216.9 7.3* and 210.0 5.1* impulses/5 s, respectively (asterisk denotes significant difference, P < .05). These data suggest that the sensitivity of a D-lys sensor operates at approximately the same level as that of the L-lys sensor. The bottom trace presents the effect of intraportal injection of 0.01 mM (0.1 ml) D-lys and 0.01 mM (0.1 ml) L-lys to a L-lys-deficient rat. The discharge rates immediately before, 30 and 60 min after injection of D-lys were 79.6 3.4, 76.9 3.5 and 72.6 2.3
impulses/5 s, respectively (N = 4). The discharge rates immediately before, 30, 60, 90 and 120 min after injection of L -lys were 72.6 2.3, 76.3 2.3 and 85.4 4.7, 98.6 4.5* and 100.6 3.3* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). Upper trace of the Fig. 4 demonstrates the effects of an injection of 0.01 mM L-alanine and L-lys on afferent activity of a L-lys-deficient rat. Afferent discharge rates before, 30, 60 and 90 min after injection of L-alanine (0.1 ml) were 68.0 2.6, 62.4 2.4, 72.6 2.7 and 66.4 2.9 impulses/5 s, respectively. Discharge rate immediately before, 30, 60 and 90 min after application of L -lys were 69.7 3.2, 82.0 1.6*, 103.5 3.2* and 123.7 2.9* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). Lower trace of Fig. 4 shows the effects of an injection of 0.01 mM (0.1 ml) L-alanine, L-leucine and L-lys in L-lys in a deficient rat. The discharge rates immediately before, 30 and 60 min after injection of L -alanine were 59.4 1.9, 54.0 3.0 and 58.6 3.1 impulses/5 s, respectively; those of L-leucine were 58.6 3.1, 58.4 3.2 and 60.1 1.3 impulses/5 s, respectively. The discharge rates before, 30, 60 and 90 min after injection of 0.01 mM L-lys were 55.6 1.4, 65.3 3.4*, 64.3 4.4* and 67.1 2.7* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). To observe the change in sensitivity of L-lys sensors in the hepatoportal region over a 12-day period in 12 rats (one rat each day) offered the L-lys-deficient diet, the change in sensitivity (the least effective concentration, mM) was plotted against time (day; Fig. 5). As shown in Fig. 5, on the third day after changing animals from a normal diet to a L-lys-deficient diet, the sensitivity index increased from 1 to 0.1 mM. By the fourth day, the sensitivity further increased to 0.01 mM. On the seventh day, food was switched back
Fig. 4. Effect of intraportal injection of L-alanine, L-leucine and L-lys on the vagal hepatic afferents in L-lys-deficient rat. Intraportal administrations of 0.01 mM (0.1 ml) L-alanine and L-leucine were without effect, although administration of the same amount of L-lys resulted in a significant increase in afferent activity. For quantitative data, see Results.
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
Fig. 5. Change in sensitivity of putative L-lys sensors in the hepatoportal region before, during and after supplying L-lys-deficient food. To observe the change in sensitivity of L-lys sensors in hepatoportal region in 12 rats over 12 days (one observation corresponds to one rat), the change in sensitivity (the least effective concentration, mM) was plotted against time (day).
from the L-lys-deficient diet to the normal diet. On the eighth day (1 day after the diet change), the sensitivity remained at the same level (0.01 mM). On the second and third days after switching back to the normal diet (i.e., Days 9 and 10), the sensitivity decreased to 0.1 mM and by the fourth day after the return to the normal diet, the sensitivity of the hepatoportal region to L-lys returned to the control level (1 mM). 4. Discussion This study supports the existence of essential amino acid sensors in the hepatoportal region [7]. The putative sensors are thought to relay information on the concentration of a particular amino acid in the portal venous blood to the brain through the afferent fibers in the hepatic branch of the vagus nerve. Data of this study are consistent with the existence of an amino acid sensor(s) for the essential amino acid, L-lys. Nerve activity was increased by L-lys in both rats fed a normal diet and those fed a L-lys-deficient diet. However, in the rats fed the normal diet, the lowest effective concentration of L-lys to evoke a neural response was estimated to be near 1 mM, while in those fed the L-lys-deficient diet for 7 ± 10 days, the lowest effective concentration was estimated to be near 0.01 mM, a 100-fold difference. Based on our previous work [5,10,11], we have used 1 week to induce the L-lys deficiency. The present data, however, indicate that even 4 days are enough to significantly increase the peripheral L-lys sensor sensitivity. Recognition of such a diet involves two phases, the first one is relatively immediate behavioral effect and the second one is gradually increased acceptance of a corrected diet. As the sensitization of the vagal sensory mechanisms described in this report required up to 4 days, it is probable that the sensory vagal mechan-
689
isms participate in the second phase of responses to an amino acid-deficient diet. While the L-lys-deficient rats were sensitive to 0.01 mM levels of L-lys, they were insensitive to the same concentrations of L-alanine, L-leucine and D-lys. The fact that sensitivity to L-lys increased in the L-lys-deficient rat, while sensitivity to D-lys did not, suggests the presence of an enantiomeric receptor (or other chiral process) for this peripheral activity. The recognition of an amino acid deficiency is functional only after amino acids are absorbed from the small intestine into the portal vein. This absorption mechanism may play important roles in maintaining homeostasis of each nutrient, and the portal vein would seem to be an appropriate area for recognizing an amino acid deficiency. The increase in vagal afferent activity specifically towards L-lys in the deficient animals may accelerate selective intake behavior for L-lys in food to aid recovery from the L-lys-deficient state. These findings are in line with the recent report by Dixon et al. [2] who, by the means of subdiaphragmatic vagotomy, reported the involvement of vagal branches in the recognition of an amino acid-imbalanced diet. Previous studies have shown that when a L-lys-deficient diet was offered to rats, an anorexic response was observed [3,4,8,10]. This finding suggests that unpleasant symptoms evoked by the decrease of L-lys in the blood, the alimentary organs and the brain could either enhance the sensitivity of the digestive tract for L-lys (due to the augmented recognition of dietary L-lys) or enhance selective intake of a solution containing L-lys [10]. Overwhelming evidence suggests that both phases of the recognition of amino acid deficiency are brain mediated. The central mechanisms are not well understood, except that local injection of the missing amino acid into the anterior piriform cortex eliminates the hypophagia on deficient diet [3]. Also, hypothalamic areas were found instrumental in the recognition of an amino acid deficiency [1,8,9,12]. It appears that the brain reacts very rapidly to amino acid dietary deficiency (for a recent review, see Ref. [3]). Thus, the presently described sensory vagal mechanisms definitively do not provide initial afferent signals about the missing amino acid. Those could be possibly provided by rapid changes of plasma levels of a missing amino acid [5]. In summary, these data document that the afferent vagus nerve innervating the portal vein reacts to a L-lys-deficient diet and may play an important role in maintaining the homeostasis of systemic L-lys, thus preventing malnutrition and anorexia.
References [1] Blevins JE, Dixon KD, Hernandez EJ, Barrett JA, Gietzen DW. Effect of threonine injections in the lateral hypothalamus on intake of amino acid imbalanced diets in rats. Brain Res 2000;879:65 ± 72. [2] Dixon KD, Williams FE, Wiggins RL, Pavellka J, Lucente J, Bel-
690
[3] [4] [5] [6] [7]
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690 tzen DW, Gietzen DW. Differential effects of selective vagotomy and tropisetron in aminoiprivic feeding. Am J Physiol 2000;279: R997 ± 1009. Gietzen DW, Erecius LF, Rogers QR. Neurochemical changes after imbalanced diets suggest a brain circuit mediating anorectic responses to amino acid deficiency in rats. J Nutr 1998;128:771 ± 81. Markison S, Gietzen DW, Spector AC. Essential amino acid deficiency enhances long-term intake but not short-term licking of the required nutrient. J Nutr 1998;129:1604 ± 12. Mori M, Kawada T, Ono T, Torii K. Taste preference, protein nutrition and L-amino acid homeostasis in male Sprague ± Dawley rats. Physiol Behav 1991;49:987 ± 96. Niijima A. Glucose-sensitive afferent nerve fibres in the hepatic branch of the guinea pig. J Physiol 1982;332:315 ± 23. Niijima A, Meguid MM. An electrophysiological study on amino acid sensors in the hepatoportal system in the rat. Obes Res 1995;3: 741S ± 5S.
[8] Smriga M, Mori M, Torii K. Circadian release of hypothalamic norepinephrine in rats in vivo is depressed during early L-lys deficiency. J Nutr 2000;130:1641 ± 3. [9] Smriga M, Murakami H, Mori M, Torii K. Effects of L-lysine deficient diet on the hypothalamic interstitial NE and diet-induced thermogenesis in rats in vivo. BioFactors 2000;12:137 ± 42. [10] Torii K, Yokawa T, Tabuchi E, Hawkins RL, Mori M, Kondoh T, Ono T. Recognition of deficient nutrient intake in the brain of rats with the Llysine deficiency monitored by functional magnetic resonance imaging, electrophysiologically and behaviorally. Amino Acids 1996;10:73 ± 81. [11] Torii K. Central mechanisms of umami taste perception and effect of dietary protein on the preference for amino acids and sodium chloride in rats. Food Rev Int 1998;2:273 ± 308. [12] Wang CX, Yang H, Perrot JC, Gietzen DW. Inhibition of norepinephrine release in the rat ventromedial hypothalamic nucleus in essential amino acid deficiency. Neurosci Lett 1998;259:53 ± 5.
Effect of lysine on afferent activity of the hepatic branch of the vagus nerve in normal and L-lysine-deficient rats Kunio Toriia,*, Akira Niijimab a
Basic Research Laboratories, Central Research Laboratories, Ajinomoto Co., Inc., 1-1 Suzuki-cho, Kawasaki 210-8681, Japan b Department of Physiology, Niigata University School of Medicine, Niigata 951, Japan Received 18 October 2000; received in revised form 19 December 2000; accepted 29 December 2000
Abstract Amino acid deficiency was modeled by feeding rats a diet deficient in the essential L-amino acid, L-lysine (L-lys). There is a rapid anorectic response to such a diet, and a strong preference for L-lys develops during the deficiency. While the brain appears to trigger this preference, the peripheral pathways that inform the brain about the deficiency are not well understood. One possible information pathway may utilize an ``amino acid sensor'' in the hepatoportal region. In the present study, we measured in vivo neural activity in normal and L-lysdeficient rat. Compared to the normally fed controls, we found an approximately 100-fold increase in the firing sensitivity of the L-lys sensors in vagal afferent fibers from the hepatoportal region of the L-lys-deficient rats. Injection of 10 mM L-lys into the hepatoportal circulation, but not D-lysine (D-lys), evoked an increase in afferent activity. While L-lys deficiency enhanced the sensitivity of the L-lys sensors, the sensitivity due to other small amino acid sensors remained unchanged. Finally, we observed a time-dependent response of the lysine sensors to lysine deficiency. It required 3 ± 4 days of maintenance on the lysine-deficient diet for the sensitivity of the L-lys sensors to change. Taken together, these results provide additional data to support the existence of putative L-amino acid sensors in the hepatoportal circulation. Additionally, they describe several characteristics of the L-lys sensors and show that these sensors may contribute to the adaptation to dietary L-lys deficiency and to maintenance of L-amino acid homeostasis. D 2001 Elsevier Science Inc. All rights reserved. Keywords: Lysine; Amino acid; Deficiency; Hepatoportal sensors; Rat
1. Introduction Several studies report that feeding a diet deficient in an essential L-amino acid is followed by a change in taste preference to induce the selective intake of that amino acid, to the extent that the deficiency is alleviated. For example, L-lysine (L-lys), an essential amino acid, was most preferred during lysine deficiency in a choice paradigm over other solutions containing various L-amino acids [5,11]. It is known that the concentration of an essential amino acid L-lys in the plasma decreases a few hours after rats are exposed to a diet deficient in L-lys [5]. Brain circuits that include the anterior piriform cortex [3], several hypothalamic areas [1,8,9,12], and possibly the amygdala [3] trigger an anorectic response to an essential amino acid [3,10]
* Corresponding author. Tel.: +81-44-244-7183; fax: +81-44-2105893. E-mail address: [email protected] (K. Torii).
deficient diet. However, whether and how the L-lys deficiency is sensed in the periphery still remains unclear. It is feasible that a peripheral L-lys sensor [7] can detect L-lys in the circulating blood and alimentary organs. Neural information that is sent from such a sensor via vagal afferent fibers may be one of the pathways informing the brain about the level of peripherally circulating amino acids. The current report deals with sensory signals from hypothesized hepatoportal L-lys sensors in the normal and L-lys-deficient rats in vivo.
2. Methods Forty-two male Wistar rats weighing about 300 g were used. Twenty-one rats were fed a diet deficient in L-lys, while the others were fed a normal diet. To develop L-lys deficiency, rats were offered a diet deficient in L-lys (formulated as previously described [5]) for 7± 10 days. Afferent recordings from the vagus nerve were made in animals
0031-9384/01/$ ± see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 0 0 3 1 - 9 3 8 4 ( 0 1 ) 0 0 4 2 6 - 7
686
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
Table 1 Effect of L-lys solution injection (10 mM, 0.1 ml ipv) on afferent discharge rates (impulses/5 s of five rats S.E.M.) in normally fed rats Rat (N)
Control
30 min
60 min
90 min
1 2 3 4 5 Mean
59.2 1.3 59.5 2.0 58.8 1.7 75.4 4.5 66.3 2.2 63.8 3.2
106.0 3.2* 75.8 3.8* 121.0 5.2* 80.6 3.6* 73.2 1.4* 91.3 9.4*
113.0 2.4* 77.6 3.0* 135.3 4.6* 119.4 8.0* 67.0 1.3 102.5 12.9*
117.3 2.1* 68.6 1.8* 80.0 2.6* 133.7 5.8* 83.6 2.9* 96.6 12.3*
* Denotes statistically significant increase, P < .05.
having consumed their respective diets for 7 ± 10 days. Food, but not water, was withdrawn at least 6 h before the beginning of each experiment. Animals were anesthetized with an intraperitoneal urethane injection (1 g/kg). An in vivo preparation was made by cannulating the portal vein. With a dissection microscope, the hepatic vagus was isolated and divided. An isolated nerve filament from the peripheral cut end of each rat was placed on a pair of silver wire recording electrodes to record afferent activity. The recording electrodes were immersed in a mixture of liquid paraffin and
petroleum jelly to prevent tissue dehydration. Afferent nerve activity was amplified in a condenser-coupled amplifier. All nerve activities were analyzed after conversion of raw data to standard pulses by a window discriminator, which separated discharges from background noise. Multiunits (about 5 ± 10 units) recordings were made. The techniques employed have been described in detail elsewhere [6]. To interpret the discharge rate data, a spike counter with a reset time of 5 s was used. For intraportal infusion, L-lys, D-lysine (D-lys), L-alanine and L-leucine (reagent grade; Ajinomoto, Japan) were used. All of these were individually dissolved in saline before use. Usually, one injection with duration of 30 s was made into the portal vein in each preparation, except the experiments presented in Figs. 1, 3 and 4) in which two or three successive injections were performed. The time of injection was after recording of stabilized afferent discharges for 30 min . The effect of injection of these amino acids was determined by comparing the mean number of impulses per 5 s in 10 successive 50-s impulse recordings. Values of afferent discharge rates were expressed as mean with standard error, and differences were evaluated with one-way analysis of variance (ANOVA, P < .05).
Fig. 1. Dose-related response of the afferent activity from putative L-lys sensors in the hepatoportal region of the normal rat. The upper trace represents the effect of 0.1 mM L-lys. There was no significant difference among the pre- and postinjective values. The lower trace presents neural activity after injections of L-lys at 1 and 10 mM. The discharge rates at 0, 30 and 60 min after injection of the 1 mM solution were 47.8 1.6, 72.1 2.4* and 48.3 1.3* (asterisk indicates significant increase, P < .05). The discharge rates at 0, 30 and 90 min after administration of 10 mM L-lys were 58.5 1.7, 121.0 5.2*, 135.3 4.6* and 80.0 2.6* (asterisk indicates significant increase, P < .05). Note that the horizontal axis in the lower trace is not linear rather is adjusted in order to compare the two abovedescribed administrations (1 and 10 mM).
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
3. Results Table 1 summarizes the mean discharge rate of five normal preparations before and 30, 60 and 90 min after an injection of 10 mM L-lys. There was a time-dependent increase in the firing rate after the L-lys injection. Not only mean values but also each individual preparation showed significant increase after injection. Fig. 1 shows the dose-dependent effects of injections of L-lys (0.1 mM, 1 mM, 10 mM; 0.1 ml each) into the portal vein on the afferent activity of the hepatic branch of the vagus nerve of a rat fed the normal diet for 7 days. As shown in the top trace, injection of 0.1 mM L-lys (0.1 ml) produced no effect on afferent activity. The discharge rates at 0, 30 and 60 min after injection of 0.1 mM were 56.2 3.1, 52.3 1.9 and 60.3 1.5 impulses/5 s, respectively. The lower trace and the graph present neural activity following injections of L-lys at 1 and 10 mM. The discharges immediately before, 30 and 60 min after injection of 1 mM solution were 47.8 1.6, 72.1 2.4* and 48.3 1.3* impulses/5 s, respectively (asterisk indicates significant increase, P < .05). The discharge rates at 0, 30, 60 and 90 min after administration of 10 mM solution were 58.5 1.7, 121.0 5.2*, 135.3 4.6* and 80.0 2.6*, impulses/5 s, respectively (asterisk indicates significant increase, P < .05; N = 6). Please note that the horizontal axis in the lower trace of Fig. 1 is not linear rather is adjusted in order to compare the two abovedescribed administrations (1 and 10 mM). Fig. 2 illustrates representative traces of hepatovagal afferent nerve activity on the effects of intraportal admin-
Fig. 2. Effect of intraportal injection of L-lys in the normal and L-lysdeficient rats. In the upper trace, injection of 0.01 mM (0.1 ml) L-lys in a normal rat resulted in the discharge rates at 0, 30, 60 and 90 min after injection of 79.3 1.8, 77.4 2.8, 74.8 3.3 and 80.2 2.9 impulses/5 s, respectively. There were no significant differences among these values. In the middle trace, injection of 0.01 mM (0.1 ml) L-lys into a L-lys-deficient rat resulted in discharge rates before (control) and 30, 60 and 90 min after injection of 61.2 3.3, 79.7 2.9*, 96.7 2.2* and 118.5 2.4* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). The bottom trace demonstrates the effect of intraportal injection of 0.001 mM L-lys into a L-lys-deficient rat. As shown in the trace, no remarkable changes in afferent activity were observed.
687
Table 2 Effect of L-lys solution injection (0.01 mM, 0.1 ml ipv) on afferent discharge rates (impulses/5 s of N rats S.E.M.) in L-lys-deficient and normally fed rats Rat (N) Normally 1 2 3 4 5 Mean
Control fed rats 53.8 2.3 79.3 1.8 57.2 4.1 55.5 2.1 56.1 1.4 60.4 4.8
L-lys-deficient rats 1 65.5 1.7 2 61.4 2.3 3 66.3 2.2 4 60.5 2.3 5 61.2 3.3 6 69.7 3.2 7 55.6 1.4 Mean 62.9 1.8
30 min
60 min
90 min
50.9 1.3 77.4 2.9 63.8 4.7 52.8 1.4 55.8 2.6 60.1 4.8
48.7 2.5 74.8 3.3 51.2 3.7 56.4 1.7 54.9 1.5 57.2 4.6
48.8 1.3 80.2 2.9 49.2 2.6 57.2 1.3 53.1 1.6 57.7 5.8
96.5 3.6* 79.4 2.7* 59.0 2.8 68.8 2.7 79.7 2.9* 82.0 1.6* 65.2 3.4* 75.9 4.7
193.6 7.4* 76.4 2.6* 94.1 2.7* 91.0 3.5* 96.7 2.2* 103.5 3.2* 64.3 4.4* 102.8 15.9*
158.4 6.5* 79.6 2.8* 117.7 4.6* 113.9 5.8* 118.5 2.4* 123.7 2.9* 67.1 2.7* 111.3 11.4*
(NS) (NS) (NS) (NS) (NS) (NS)
NS, nonsignificant difference. * Denotes statistically significant increase, P < .05.
istration of L-lys to a L-lys-deficient rat and to a normal rat. As shown in the top trace, injection of 0.01 mM (0.1 ml) L-lys caused no noticeable change in afferent activity in a normal rat. However, a gradual significant increase in afferent activity was observed following administration of 0.01 mM (0.1 ml) L-lys to a rat fed the L-lys-deficient diet for 7 days (middle trace). The discharge rates immediately before injection, 30, 60 and 90 min after injection were 61.2 3.3, 79.7 2.9*, 96.7 2.2* and 118.5 2.4*, impulses/5 s, respectively (asterisk indicates significant increase, P < .05). The bottom trace demonstrates the effect of intraportal injection of 0.001 mM L-lys to the L-lysdeficient rat (the same rat as above). As shown in this lower trace, no significant changes in afferent activity were observed. The discharge rates immediately before, 30, 60 and 90 min after injection were 59.1 3.0, 59.0 2.7, 56.2 3.9 and 52.0 2.3 impulses/5 s, respectively (N = 7). Table 2 summarize the mean discharge rates of vagal hepatic afferents following intraportal injection of 0.01 mM (0.1 ml) L-lys in normal (N = 5) and L-lys-deficient (N = 7) rats. In rats fed a normal diet, there were no significant differences in afferent activity observed after injection, not only for the mean values but also for each individual preparation. However, in rats fed the L-lys-deficient diet for 7 days, not only the mean values but also each individual preparation showed significant increases in activity at 60 and 90 min after injection of 0.01 mM L-lys. The upper trace of Fig. 3 shows the effect of intraportal injection of 0.01 mM D-lys (0.1 ml) on the afferent activity of the hepatic branch of the vagus nerve in a rat fed the normal diet. The discharge rates immediately before, 30, 60 and 90 min after injection were 67.3 3.0, 60.9 3.2 and 66.9 2.2 impulses/5 s, respectively. The middle trace
688
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
Fig. 3. Effect of intraportal injections of D-lys and L-lys on the mean afferent discharge rate of vagal hepatic afferents in the normal and L-lysdeficient rat. The upper trace of shows the effect of intraportal injection of 0.01 mM D-lys (0.1 ml) on the afferent activity of the hepatic branch of the vagus nerve in a rat fed the normal diet. The middle trace shows the effect of 10 mM D-lys (0.1 ml). The bottom trace presents the effect of intraportal injection of 0.01 mM (0.1 ml) D-lys and 0.01 mM (0.1 ml) L-lys to a L-lysdeficient rat. For quantitative data, see Results.
shows the effect of 10 mM D-lys (0.1 ml). The discharge rates just before and 30, 60, 90 and 120 min after injection were 66.2 2.8, 91.0 2.7*, 187.3 4.4*, 216.9 7.3* and 210.0 5.1* impulses/5 s, respectively (asterisk denotes significant difference, P < .05). These data suggest that the sensitivity of a D-lys sensor operates at approximately the same level as that of the L-lys sensor. The bottom trace presents the effect of intraportal injection of 0.01 mM (0.1 ml) D-lys and 0.01 mM (0.1 ml) L-lys to a L-lys-deficient rat. The discharge rates immediately before, 30 and 60 min after injection of D-lys were 79.6 3.4, 76.9 3.5 and 72.6 2.3
impulses/5 s, respectively (N = 4). The discharge rates immediately before, 30, 60, 90 and 120 min after injection of L -lys were 72.6 2.3, 76.3 2.3 and 85.4 4.7, 98.6 4.5* and 100.6 3.3* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). Upper trace of the Fig. 4 demonstrates the effects of an injection of 0.01 mM L-alanine and L-lys on afferent activity of a L-lys-deficient rat. Afferent discharge rates before, 30, 60 and 90 min after injection of L-alanine (0.1 ml) were 68.0 2.6, 62.4 2.4, 72.6 2.7 and 66.4 2.9 impulses/5 s, respectively. Discharge rate immediately before, 30, 60 and 90 min after application of L -lys were 69.7 3.2, 82.0 1.6*, 103.5 3.2* and 123.7 2.9* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). Lower trace of Fig. 4 shows the effects of an injection of 0.01 mM (0.1 ml) L-alanine, L-leucine and L-lys in L-lys in a deficient rat. The discharge rates immediately before, 30 and 60 min after injection of L -alanine were 59.4 1.9, 54.0 3.0 and 58.6 3.1 impulses/5 s, respectively; those of L-leucine were 58.6 3.1, 58.4 3.2 and 60.1 1.3 impulses/5 s, respectively. The discharge rates before, 30, 60 and 90 min after injection of 0.01 mM L-lys were 55.6 1.4, 65.3 3.4*, 64.3 4.4* and 67.1 2.7* impulses/5 s, respectively (asterisk denotes significant increase, P < .05). To observe the change in sensitivity of L-lys sensors in the hepatoportal region over a 12-day period in 12 rats (one rat each day) offered the L-lys-deficient diet, the change in sensitivity (the least effective concentration, mM) was plotted against time (day; Fig. 5). As shown in Fig. 5, on the third day after changing animals from a normal diet to a L-lys-deficient diet, the sensitivity index increased from 1 to 0.1 mM. By the fourth day, the sensitivity further increased to 0.01 mM. On the seventh day, food was switched back
Fig. 4. Effect of intraportal injection of L-alanine, L-leucine and L-lys on the vagal hepatic afferents in L-lys-deficient rat. Intraportal administrations of 0.01 mM (0.1 ml) L-alanine and L-leucine were without effect, although administration of the same amount of L-lys resulted in a significant increase in afferent activity. For quantitative data, see Results.
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690
Fig. 5. Change in sensitivity of putative L-lys sensors in the hepatoportal region before, during and after supplying L-lys-deficient food. To observe the change in sensitivity of L-lys sensors in hepatoportal region in 12 rats over 12 days (one observation corresponds to one rat), the change in sensitivity (the least effective concentration, mM) was plotted against time (day).
from the L-lys-deficient diet to the normal diet. On the eighth day (1 day after the diet change), the sensitivity remained at the same level (0.01 mM). On the second and third days after switching back to the normal diet (i.e., Days 9 and 10), the sensitivity decreased to 0.1 mM and by the fourth day after the return to the normal diet, the sensitivity of the hepatoportal region to L-lys returned to the control level (1 mM). 4. Discussion This study supports the existence of essential amino acid sensors in the hepatoportal region [7]. The putative sensors are thought to relay information on the concentration of a particular amino acid in the portal venous blood to the brain through the afferent fibers in the hepatic branch of the vagus nerve. Data of this study are consistent with the existence of an amino acid sensor(s) for the essential amino acid, L-lys. Nerve activity was increased by L-lys in both rats fed a normal diet and those fed a L-lys-deficient diet. However, in the rats fed the normal diet, the lowest effective concentration of L-lys to evoke a neural response was estimated to be near 1 mM, while in those fed the L-lys-deficient diet for 7 ± 10 days, the lowest effective concentration was estimated to be near 0.01 mM, a 100-fold difference. Based on our previous work [5,10,11], we have used 1 week to induce the L-lys deficiency. The present data, however, indicate that even 4 days are enough to significantly increase the peripheral L-lys sensor sensitivity. Recognition of such a diet involves two phases, the first one is relatively immediate behavioral effect and the second one is gradually increased acceptance of a corrected diet. As the sensitization of the vagal sensory mechanisms described in this report required up to 4 days, it is probable that the sensory vagal mechan-
689
isms participate in the second phase of responses to an amino acid-deficient diet. While the L-lys-deficient rats were sensitive to 0.01 mM levels of L-lys, they were insensitive to the same concentrations of L-alanine, L-leucine and D-lys. The fact that sensitivity to L-lys increased in the L-lys-deficient rat, while sensitivity to D-lys did not, suggests the presence of an enantiomeric receptor (or other chiral process) for this peripheral activity. The recognition of an amino acid deficiency is functional only after amino acids are absorbed from the small intestine into the portal vein. This absorption mechanism may play important roles in maintaining homeostasis of each nutrient, and the portal vein would seem to be an appropriate area for recognizing an amino acid deficiency. The increase in vagal afferent activity specifically towards L-lys in the deficient animals may accelerate selective intake behavior for L-lys in food to aid recovery from the L-lys-deficient state. These findings are in line with the recent report by Dixon et al. [2] who, by the means of subdiaphragmatic vagotomy, reported the involvement of vagal branches in the recognition of an amino acid-imbalanced diet. Previous studies have shown that when a L-lys-deficient diet was offered to rats, an anorexic response was observed [3,4,8,10]. This finding suggests that unpleasant symptoms evoked by the decrease of L-lys in the blood, the alimentary organs and the brain could either enhance the sensitivity of the digestive tract for L-lys (due to the augmented recognition of dietary L-lys) or enhance selective intake of a solution containing L-lys [10]. Overwhelming evidence suggests that both phases of the recognition of amino acid deficiency are brain mediated. The central mechanisms are not well understood, except that local injection of the missing amino acid into the anterior piriform cortex eliminates the hypophagia on deficient diet [3]. Also, hypothalamic areas were found instrumental in the recognition of an amino acid deficiency [1,8,9,12]. It appears that the brain reacts very rapidly to amino acid dietary deficiency (for a recent review, see Ref. [3]). Thus, the presently described sensory vagal mechanisms definitively do not provide initial afferent signals about the missing amino acid. Those could be possibly provided by rapid changes of plasma levels of a missing amino acid [5]. In summary, these data document that the afferent vagus nerve innervating the portal vein reacts to a L-lys-deficient diet and may play an important role in maintaining the homeostasis of systemic L-lys, thus preventing malnutrition and anorexia.
References [1] Blevins JE, Dixon KD, Hernandez EJ, Barrett JA, Gietzen DW. Effect of threonine injections in the lateral hypothalamus on intake of amino acid imbalanced diets in rats. Brain Res 2000;879:65 ± 72. [2] Dixon KD, Williams FE, Wiggins RL, Pavellka J, Lucente J, Bel-
690
[3] [4] [5] [6] [7]
K. Torii, A. Niijima / Physiology & Behavior 72 (2001) 685±690 tzen DW, Gietzen DW. Differential effects of selective vagotomy and tropisetron in aminoiprivic feeding. Am J Physiol 2000;279: R997 ± 1009. Gietzen DW, Erecius LF, Rogers QR. Neurochemical changes after imbalanced diets suggest a brain circuit mediating anorectic responses to amino acid deficiency in rats. J Nutr 1998;128:771 ± 81. Markison S, Gietzen DW, Spector AC. Essential amino acid deficiency enhances long-term intake but not short-term licking of the required nutrient. J Nutr 1998;129:1604 ± 12. Mori M, Kawada T, Ono T, Torii K. Taste preference, protein nutrition and L-amino acid homeostasis in male Sprague ± Dawley rats. Physiol Behav 1991;49:987 ± 96. Niijima A. Glucose-sensitive afferent nerve fibres in the hepatic branch of the guinea pig. J Physiol 1982;332:315 ± 23. Niijima A, Meguid MM. An electrophysiological study on amino acid sensors in the hepatoportal system in the rat. Obes Res 1995;3: 741S ± 5S.
[8] Smriga M, Mori M, Torii K. Circadian release of hypothalamic norepinephrine in rats in vivo is depressed during early L-lys deficiency. J Nutr 2000;130:1641 ± 3. [9] Smriga M, Murakami H, Mori M, Torii K. Effects of L-lysine deficient diet on the hypothalamic interstitial NE and diet-induced thermogenesis in rats in vivo. BioFactors 2000;12:137 ± 42. [10] Torii K, Yokawa T, Tabuchi E, Hawkins RL, Mori M, Kondoh T, Ono T. Recognition of deficient nutrient intake in the brain of rats with the Llysine deficiency monitored by functional magnetic resonance imaging, electrophysiologically and behaviorally. Amino Acids 1996;10:73 ± 81. [11] Torii K. Central mechanisms of umami taste perception and effect of dietary protein on the preference for amino acids and sodium chloride in rats. Food Rev Int 1998;2:273 ± 308. [12] Wang CX, Yang H, Perrot JC, Gietzen DW. Inhibition of norepinephrine release in the rat ventromedial hypothalamic nucleus in essential amino acid deficiency. Neurosci Lett 1998;259:53 ± 5.