- Email: [email protected]
Mammillary polydipsia and diabetes insipidus: A study of the rhythmicity of water intake
Physiology & Behavior, Vol. 45, pp. 911-915. ©Pergamon Press plc., 1989. Printed in the U.S.A.
0031-9384/89 $3.00 + .00
Mammillary Polydipsia and Diabetes Insipidus: A Study of the Rhythmicity of Water Intake I ALBERTO MORALES, 2 INMACULADA CUBERO AND AMADEO PUERTO
Area de Psicobiologfa, Departamento de Psicologfa Experimental y Fisiologfa del Comportamiento Universidad de Granada, Spain R e c e i v e d 30 M a r c h 1988
MORALES, A., I. CUBERO AND A. PUERTO. Mammillarypolydipsia and diabetes insipidus: A study of the rhythmicity of water intake. PHYSIOL BEHAV 45(5) 911-915, 1989.--Rats with polydipsia induced by electrolytic mammillary lesions show a normal daily rhythmicity of water intake compared with sham lesioned animals, when kept in a 12:12 hours light-dark cycle of illumination. Water is mainly consumed during the dark phase (approximately 80-90% of the total amount). On the other hand, rats with centrally induced diabetes insipidus by means of electrolytic lesions in the median eminence show a clear-cut alteration in this rhythmicity, drinking only 67% of the total amount during the dark phase. This effect could be due to the continuous necessity of these animals to drink water in order to maintain fluid homeostasis, and is not related to food rhythmicity alterations. Taken together, and on the basis of the daily rhythmicity of water intake, these results suggest that mammillary polydipsia may be different from that observed when diabetes insipidus is present. Polydipsia Diabetes insipidus Median eminence
Mammillary area
Circadian rhythms
THE production of electrolytic lesions of the mammillary area in the rat is an experimental procedure which induces polydipsia, as initially described by Santacana et al. (19), through as yet unexplained mechanisms. Although in their original study these authors brought up the possibility that such polydipsia may result from a process of diabetes insipidus (DI), they observed that lesioned animals were able to regulate urine excretion in response to water deprivation. Subsequent histological examinations moreover failed to reveal any damage to the infundibulo-neurohypophyseal system. Nevertheless, later studies showed that although polydipsia induced by mammillary area lesions (mammillary polydipsia) possessed a well-defined primary component [as described by Santacana et al., see (19)], water urinary loss was also significant, and to a certain extent could also be considered a causal factor of the polydipsia under study (14). This polyuric effect once again raised the possibility that diabetes might be the cause of the polydipsia observed in these studies. In our laboratory, electrical stimulation of the supraoptic nucleus has been shown to significantly reduce water intake in mammillary lesioned rats without affecting other behaviors (e.g., food intake). These results suggested that the hypothalamus-neurohypophyseal tract had remained intact, thus implying normal vasopressin secretion (11). Mammillary rats also showed a differential response to IP admin-
Water intake rhythmicity
istration of hypertonic saline in relation to other dipsogenic treatments. No such differential response was recorded in rats with centrally induced DI, and this seems to point to a different biological basis for each type of polydipsia (12,13). In the present study, aimed at characterizing and differentiating mammillary polydipsia on the basis of behavioral indices, the daily rhythmicity of water intake was analyzed under a light-dark period of 12 hr (LD 12:12). Earlier investigations have confirmed experimentally that drinking rhythmicity can be modified by certain brain lesions which are also capable of altering internal fluid dynamics, such as occurs following lesions of the nucleus medianus (5,6) and the zona incerta (8). The results obtained with mammillary rats were compared with those recorded in a group of animals in which DI had been centrally induced by lesions of the median eminence (ME), a large area located ventromedially in the hypothalamus. Lesions of the ME have been shown to cause DI (16,17) possibly as a consequence of damaging the hypothalamusneurohypophyseal tract. METHOD
Subjects Male Wistar rats weighing 250-350 g were used in this
~Part of the results of the mammillary group were presented in an absuact-like form at the NATO Advanced Research Workshop on The Physiology of Thirst and Sodium Appetite, held July 12-20, 1984, in Camerino, Italy [see (12)]. 2Requests for reprints should be addressed to Dr. Alberto Morales, Laboratorio de Psicobiologia, Psicologia, Universidad de Granada, Campus Cartuja, E-18071 Granada, Spain.
911
912
MORALES, CUBERO AND PUERTO
experiment. The rats were housed in individual cages with food (Sandersmus, Spain) and tap water ad lib unless otherwise noted. Room temperature was kept about 22°C with lights on from 8:30 a . m . to 8:30 p.m.
100-
J
Surgical Procedure
75-
The animals were anesthetized with sodium pentothal (Thiopentone sodium, IP 50 mg/kg of body weight, Lab. Abbott, Spain) and placed in a stereotaxic frame (David Kopf Instruments). The experimental animals received bilateral electrolytic lesions in the mammillary area (19) through a stainless steel monopolar electrode isolated throughout (INSL-X) except for the distal 0.5 mm. A 1.2 mA DC current was delivered for 20 seconds by a DCML-5 lesion maker (Grass Instruments). Stereotaxic coordinates were obtained from the Albe-Fessard and associates atlas (1). Control animals received a standard sham lesion. On the other hand, median eminence lesioned rats received a similar procedure. In this case, however, stereotaxic coordinates were obtained from the De Groot atlas (4), according to Roll's procedure (16), and a 1.5 mA DC current was delivered for 15 seconds. Control animals received a standard sham lesion.
Experimental Procedure MammiUary lesion. The animals were assigned at random to the experimental mammillary group (EMG) or to the control mammillary group (CMG) before surgery. Those subjects which did not develop the previously described polydipsia were excluded (19). The resulting EMG consisted of 6 polydipsic animals and the CMG of 5 normodipsic animals. Water intake in both groups was measured during the five days before surgery and during a similar period after 15 days recovery from operation. Thereafter, the consumption of water and food available ad lib was measured in both groups in 12 hourly periods beginning with the onset of light or darkness, for 3 days and the results obtained in darkness were analyzed and expressed as percentages of the total daily consumption. Median eminence lesion. The subjects forming the experimental median eminence group (EMEG) and the control median eminence group (CMEG) (seven and five, respectively) were chosen and studied in the same way as the mammillary group. On the other hand, and on the basis of the data obtained (see below), rats in the EMEG were food deprived throughout the 12 hr lights on period, with food available ad lib during the dark cycle. Water was always available ad lib, and drinking was recorded as described above. Histology On completion of testing, the rats were killed with a sodium pentothal overdose and then perfused intracardially with saline followed by 10% buffered formaldehyde solution. The brains were removed and stored in formaldehyde until lamination.
Statistical Analysis Values reported are means ---S.D. Comparisons between groups were analyzed for statistical significance using the unpaired Student's t-test. Data from the same group were analyzed using now a paired Student's t-test. RESULTS
As expected, electrolytic lesions in the mammillary area induced polydipsia: an average of 89.2---22.5 ml as opposed to
,< IZ uJ (J 50-
25-
E MG EMEG WATER
EMG
EMEG FOOD
FIG. 1. Percentage of water and food consumedduring the dark phase by the experimental mammillary group (EMG) and by the experimental median eminence group (EMEG). Values reported are means---S.D. (~'k~-)p<0.005. 38.3---8.0 ml by the control animals (t= 4.3, p<0.005). The EMG rats consumed 80.8 __-4.4% of their daily total water intake in the 12 hours of the dark cycle and the CMG rats 85.8--.4.0% (t= 1.7, n.s.). That is to say, the rhythmicity of the polydipsic intake was not affected by the mammillary lesion. On the other hand, electrolytic lesions of the ME led to a marked polydipsia with intake about 200 ml daily on postoperative day seven. Thereafter, water intake fell progressively (possibly as a result of compensatory mechanisms of unknown origin), stabilizing at a mean volume of 135.9---47.8 ml in the EMEG as compared to 25.5 ± 7.3 ml in the CMEG (t = 4.7, p<0.001). The lesion also altered the daily rhythmicity or water intake, experimental animals consuming 67.2 ± 7.2% of their total intake during the nocturnal phase in comparison to 89.4± 3.9% in the control group (t = 5.9, p<0.001). Although ME lesioned animals drank greater amounts of water than mammillary lesioned rats (135.9 ---47.8 ml vs. 89.2 ± 22.5 ml respectively), the difference fell short of significance (t=2.0, n.s.). However, the percentage of water consumed during the dark phase differed markedly, reaching 6 7 . 2 - 7.2% in the EMEG and up to 80.8±4.4% in the EMG (t=3.6, p<0.005) (Fig. 1). Moreover, upon analyzing the data individually, some mammillary rats were noted to drink more daily water than some diabetic rats, a fact which was nevertheless insufficient to offset the difference in percentage nocturnal intake observed between the two groups. This finding seems to be independent of food intake (Table 1), and was unchanged in rats from both groups which consumed similar amounts of water (Table 2). No significant differences were seen in mean food intake between the EMG and EMEG, which consumed 30.3---6.8 g and 34.0± 10.9 g respectively (t=0.6, n.s.). Percentage food intake (during the nocturnal phase) likewise presented no significant difference, at 71.6±5.5% for the EMG and 65.0± 12.7% for the EMEG (t= 1.1, n.s.) (Fig. 1). Finally, EMEG rats changed their rhythmicity of water intake in response to the 12 hr period of food deprivation. During the nocturnal phase, the percentage of total water consumed with food available rose from 67.2±7.2% to 77.7±4.8% (t=5.3, p<0.005).
M A M M I L L A R Y POLYDIPSIA AND RHYTHMICITY
TABLE 1 WATER AND FOOD INTAKE CHARACTERISTICS OF TWO RATS, MAMMILLARY AND MEDIAN EMINENCE LESIONED (DIABETIC) RESPECTIVELY
Rat
Total Water Intake (nil)
Total Food Intake (g)
Water Intake Dark Phase (%)
Food Intake Light Phase (%)
Mammillary Diabetic
124.8 -+ 9.9 108.3 ± 4.2
31.3 - 1.5 35.3 ± 1.2
80.0 - 3.0 68.6 ± 1.7
30.2 ± 2.9 34.8 - 4.9
Values reported are means - S.D. (for details, see text).
913
Histological analysis confm'ned that the mammillary lesion had affected mainly the mammillary bodies, including the medial (pars medialis and pars lateralis) and lateral mammillary nuclei, as well as the most posterior zone of the same. In some cases the lesion involved a very limited portion of the posterior edge of the posterior hypothalamic nucleus, although more anterior portions of the hypothalamus (e.g., the median eminence) were never affected. The median eminence lesions on the other hand destroyed the most ventral region of the ventromedial hypothalamus. As seen in Figs. 2 and 3, the lesion, owing to its markedly medial location, affected the lateral walls of the third ventricle, including a large part of the nucleus arcuatus. The lateral extension of the lesion was variable, at times involving the most medial region of the ventromedial hypothalamic nucleus (VMH). The suprachiasmatic nucleus (SCN), located well forward of the site of the lesion, was undamaged in all rats.
TABLE 2
DISCUSSION
WATER INTAKE AND NOCTURNAL RHYTHMICITY OF TWO RATS, MAMMILLARY AND MEDIAN EMINENCE LESIONED (DIABETIC) RESPECTIVELY
As shown in the results, electrolytic lesions of the mammiUary area have no effect on the daily rhythmicity of water intake. Despite polydipsia, the percentage of water intake during the dark phase is similar (around 80% of the total water intake) in lesioned rats and in controls. These findings contrast with the data obtained in the ME lesioned group, where rhythmicity was affected by surgery. It thus appears that in principle, water intake throughout the day is distributed differently in rats with mammillary polydipsia and rats with centrally induced DI, and that these two groups can be clearly distinguished on the basis of this behavioral characteristic.
Rat
Total Water Intake (ml)
Water Intake Dark Phase (%)
Mammillary Diabetic
82.7 - 1.2 87.7 ± 5.5
84.0 _ 3.0 69.8 ± 3.7
Values reported are means ± S.D. (for details, see text).
FIG. 2. Photomicrograph of an electrolytic lesion of the median eminence. The most ventral zone of the ventromedial hypothalamus was affected by the lesion (see arrow).
914
MORALES, CUBERO AND PUERTO
vmh
ot
FIG. 3. Schematic representation of the median eminence electrolytic lesion (dark). Eventually, the lesions can extend to the most medial part of the VMH nucleus (shaded). Abbreviations: aha: anterior hypolhalamic area; cpu: nucleus caudate-putamen; fx: fornix; gp: globus pallidus; ot: optic tract; pvh: hypothalamic paraventricular nucleus; re: nucleus reuniens; v: ventricle; vmh: hypothalamic ventromedial nucleus. The findings in the DI group however deserve to be examined in greater detail. It is possible that the change in percentage water intake during the dark phase simply reflected the greater need to drink in these animals. This notion however seems unlikely, as water consumption reached extremely high levels (around 200 ml daily) approximately seven days postsurgery, although dark intake comprised only about 60% of the total (data not shown). Polydipsia stablized at around 15 days postlesion, when mean intake had fallen to 135 ml per day as noted in the results. Under these conditions and despite the reduction just noted, night time drinking remained at around 67%, a value significantly different from that recorded in mammillary and control rats. In addition, a study of the individual data in Tables 1 and 2 reveals that some mammillary rats drank more water than some diabetic rats, although the differences in nocturnal distribution of water intake were uninfluenced by this fact; hence this cannot be interpreted as a consequence of a greater food intake during the day in diabetic rats. Moreover, the differences noted in nocturnal rhythmicity could also be seen when mammillary and diabetic rats consumed similar amounts of water. Furthermore, the rhythmicity of water intake in diabetic rats was susceptible to experimental manipulation, e.g., food deprivation during the 12 hr light period. Under this condition, the percentage of nocturnal water intake rose significantly to make up 77% of the total. This implies that the percentage of water consumed under ad lib conditions during this period (67%) cannot be taken as a possible "ceiling effect." To sum up, the data suggest that the alteration recorded in daily rhythmicity of water intake in rats with centrally induced DI is not conditioned by the amount of water consumed, but rather by the nature of the hormonal alteration itself. Similar findings have been described in rats with genetically determined DI (Brattleboro rats), which also present changes in rhythmicity owing not so much to a loss of circadian control over drinking, which is unaffected (7) as to the need to drink in order to maintain water balance, due to the continuous urinary loss (10,15). Given that the ME lesion may affect the most medial part of the VMH nucleus, the changes in drinking rhythmicity may be traced indirectly to an alteration in feeding behavior, with which this nucleus is closely related [see (18) for review]. This variable does not however appear to be essential, as mean food intake in diabetic rats was no greater than in mammillary lesioned animals [it should
be recalled that the mammillary lesion never invaded the ventromedial hypothalamus, nor were alterations in food intake recorded in connection with these lesions, see (19)]. The lack of significant differences between the diabetic and mammillary group in nocturnal distribution of food intake further supports this conclusion. Experimental studies in other laboratories have demonstrated that only massive lesions to the VMH nucleus were capable of causing changes in the circadian rhythmicity of certain variables such as feeding and locomotor activity (9,18). Such alterations usually arise as a result of the destruction of efferent fibers from the SCN (9); only part of these fibers cross the ventromediai hypothalamus near the VMH nucleus (2). In the present series of experiments, only the most medial part of this nucleus is damaged, although some efferent fibers from the SCN (left intact in all animals studied) may have been affected. Nevertheless, this is insufficient to explain the present findings, as born out by a) the absence of hyperphagic effects, as would be expected following a lesion to the VMH nucleus, b) the lack of differences in the nocturnal rhythmicity of food intake between diabetic and mammillary rats, and c) the susceptibility of drinking rhythmicity in diabetic rats to experimental manipulation, a phenomenon which may well not have occurred if the efferent system of the SCN had been damaged seriously. To sum up, the results brought to light in this article suggest that the two types of polydipsia investigated differ in origin, as suggested by previous studies carded out in our laboratory (13). The alteration in the daily rhythmicity of water intake in rats with centrally induced DI appears not to be a secondary effect related to a possible change in feeding patterns, nor to reflect a general disruption of circadian rhythms, but rather to result from the continuous need to drink in order to maintain fluid homeostasis. Although our findings apparently do not support a diabetic origin for mammillary polydipsia, further studies will be needed to identify the ultimate cause of this behavior. ACKNOWLEDGEMENTS
Our thanks to David Butler and Karen Shashok for their help in the English version of this text. We are indebted to Professor Neil McClaren, of the English Department of the University of Granada, who supervised the final form of the manuscript.
MAMMILLARY POLYDIPSIA AND RHYTHMICITY
915
REFERENCES 1. Albe-Fessard, D.; Stutinskym, F.; Libouban, S. Atlas ster6otaxique du dienc6phale du rat blanc. Paris: Editions du CNRS; 1969. 2. Berk, M. L.; Finkelstein, J. A. An autoradiographic determination of the efferent projections of the suprachiasmatic nucleus of the hypothalamus. Brain Res. 226:1-13; 1981. 3. Brodal, A. Neurological anatomy in relation to clinical medicine. New York: Oxford University Press; 1981. 4. De Groot, J. The rat brain in stereotaxic coordinates. Trans. R. Netherland Acad. Sci. 52:1--40; 1959. 5. Gardiner, T. W.; Stricker, E. M. Hyperdipsia in rats after electrolytic lesions of nucleus medianus. Am. J. Physiol. 248:R214-R223; 1985. 6. Gardiner, T. W.; Stricker, E. M. Impaired drinking responses of rats with lesions of nucleus medianus: circadian dependence. Am. J. Physiol. 248:R224-R230; 1985. 7. Grobleski, T. A.; Ndfiez, A. A.; Gold, R. M. Circadian rhythms in vasopressin deficient rats. Brain Res. Bull. 6:125-130; 1981. 8. Grossman, S. P. A reassessment of brain mechanisms that control thirst. Neurosci. Biobehav. Rev. 8:95-104; 1984. 9. Inouye, S. T. Does the VMH contain a self-sustained circadian oscillator associated with periodic feedings? Brain Res. 279:53-63; 1983. 10. Kutscher, C. L.; Wright, W. A. Drinking characteristics of normal rats and rats heterozygous and homozygous for diabetes insipidus. Physiol. Behav. 18:833-839; 1977. 11. Morales, A.; Puerto, A. Efecto de la estimulaci6n el6ctrica del nticleo
12.
13.
14. 15.
16. 17.
18. 19.
supra6ptico sobre la polidipsia mamilar. Arch. Neurobiol. 51:154158; 1988. Morales, A.; Puerto, A. Posterior hypothalamic polydipsia: differential effects to several dipsogenic treatments. In: de Caro, G.; Epstein, A. N.; Massi, M., eds. The physiology of thirst and sodium appetite. New York: Plenum Press; 1986:77-81. Morales, A.; Puerto, A. A study of the response to several dipsogenic treatments in rats with mammilllary polydipsia and with centrally induced diabetes insipidus. Behav. Brain Res. 31:69-74; 1988. Morales, A.; Puerto, A. Un estudio del car~icter primario/secundario de la polidipsia inducida por lesi6n electrolftica de los cuerpos mamilares. Rev. Esp. Fisiol., in press; 1989. Peterson, G. M.; Watkins, W. B.; Moore, R. Y. The suprachiasmatic hypothalamic nuclei of the rat. VI. Vasopressin neurons and circadian rhythmicity. Behav. Neural Biol. 29:236-245; 1980. Rolls, B. J. Drinking by rats after irritative lesions in the hypothalamus. Physiol. Behav. 5:1385-1393; 1970. Smith, R. W.; McCann, S. M. Increased and decreased water intake in the rat with hypothalamic lesions. Am. J. Physiol. 203:366-370; 1962. Stoynev, A. G.; Ikonomov, O. C. Circadian regulation of feeding in rats: suprachiasmatic versus ventromedial hypothalamic nuclei. Appetite 9:217-229; 1987. Tejedor del Real, M. P.; Santacana-Altimiras, M. P.; Alvarez-Pel"iez, R. Alteraci6n en la toma de agua despu6s de la lesi6n de los cuerpos mamilares. Rev. Esp. Fisiol. 28:95-102; 1972.
0031-9384/89 $3.00 + .00
Mammillary Polydipsia and Diabetes Insipidus: A Study of the Rhythmicity of Water Intake I ALBERTO MORALES, 2 INMACULADA CUBERO AND AMADEO PUERTO
Area de Psicobiologfa, Departamento de Psicologfa Experimental y Fisiologfa del Comportamiento Universidad de Granada, Spain R e c e i v e d 30 M a r c h 1988
MORALES, A., I. CUBERO AND A. PUERTO. Mammillarypolydipsia and diabetes insipidus: A study of the rhythmicity of water intake. PHYSIOL BEHAV 45(5) 911-915, 1989.--Rats with polydipsia induced by electrolytic mammillary lesions show a normal daily rhythmicity of water intake compared with sham lesioned animals, when kept in a 12:12 hours light-dark cycle of illumination. Water is mainly consumed during the dark phase (approximately 80-90% of the total amount). On the other hand, rats with centrally induced diabetes insipidus by means of electrolytic lesions in the median eminence show a clear-cut alteration in this rhythmicity, drinking only 67% of the total amount during the dark phase. This effect could be due to the continuous necessity of these animals to drink water in order to maintain fluid homeostasis, and is not related to food rhythmicity alterations. Taken together, and on the basis of the daily rhythmicity of water intake, these results suggest that mammillary polydipsia may be different from that observed when diabetes insipidus is present. Polydipsia Diabetes insipidus Median eminence
Mammillary area
Circadian rhythms
THE production of electrolytic lesions of the mammillary area in the rat is an experimental procedure which induces polydipsia, as initially described by Santacana et al. (19), through as yet unexplained mechanisms. Although in their original study these authors brought up the possibility that such polydipsia may result from a process of diabetes insipidus (DI), they observed that lesioned animals were able to regulate urine excretion in response to water deprivation. Subsequent histological examinations moreover failed to reveal any damage to the infundibulo-neurohypophyseal system. Nevertheless, later studies showed that although polydipsia induced by mammillary area lesions (mammillary polydipsia) possessed a well-defined primary component [as described by Santacana et al., see (19)], water urinary loss was also significant, and to a certain extent could also be considered a causal factor of the polydipsia under study (14). This polyuric effect once again raised the possibility that diabetes might be the cause of the polydipsia observed in these studies. In our laboratory, electrical stimulation of the supraoptic nucleus has been shown to significantly reduce water intake in mammillary lesioned rats without affecting other behaviors (e.g., food intake). These results suggested that the hypothalamus-neurohypophyseal tract had remained intact, thus implying normal vasopressin secretion (11). Mammillary rats also showed a differential response to IP admin-
Water intake rhythmicity
istration of hypertonic saline in relation to other dipsogenic treatments. No such differential response was recorded in rats with centrally induced DI, and this seems to point to a different biological basis for each type of polydipsia (12,13). In the present study, aimed at characterizing and differentiating mammillary polydipsia on the basis of behavioral indices, the daily rhythmicity of water intake was analyzed under a light-dark period of 12 hr (LD 12:12). Earlier investigations have confirmed experimentally that drinking rhythmicity can be modified by certain brain lesions which are also capable of altering internal fluid dynamics, such as occurs following lesions of the nucleus medianus (5,6) and the zona incerta (8). The results obtained with mammillary rats were compared with those recorded in a group of animals in which DI had been centrally induced by lesions of the median eminence (ME), a large area located ventromedially in the hypothalamus. Lesions of the ME have been shown to cause DI (16,17) possibly as a consequence of damaging the hypothalamusneurohypophyseal tract. METHOD
Subjects Male Wistar rats weighing 250-350 g were used in this
~Part of the results of the mammillary group were presented in an absuact-like form at the NATO Advanced Research Workshop on The Physiology of Thirst and Sodium Appetite, held July 12-20, 1984, in Camerino, Italy [see (12)]. 2Requests for reprints should be addressed to Dr. Alberto Morales, Laboratorio de Psicobiologia, Psicologia, Universidad de Granada, Campus Cartuja, E-18071 Granada, Spain.
911
912
MORALES, CUBERO AND PUERTO
experiment. The rats were housed in individual cages with food (Sandersmus, Spain) and tap water ad lib unless otherwise noted. Room temperature was kept about 22°C with lights on from 8:30 a . m . to 8:30 p.m.
100-
J
Surgical Procedure
75-
The animals were anesthetized with sodium pentothal (Thiopentone sodium, IP 50 mg/kg of body weight, Lab. Abbott, Spain) and placed in a stereotaxic frame (David Kopf Instruments). The experimental animals received bilateral electrolytic lesions in the mammillary area (19) through a stainless steel monopolar electrode isolated throughout (INSL-X) except for the distal 0.5 mm. A 1.2 mA DC current was delivered for 20 seconds by a DCML-5 lesion maker (Grass Instruments). Stereotaxic coordinates were obtained from the Albe-Fessard and associates atlas (1). Control animals received a standard sham lesion. On the other hand, median eminence lesioned rats received a similar procedure. In this case, however, stereotaxic coordinates were obtained from the De Groot atlas (4), according to Roll's procedure (16), and a 1.5 mA DC current was delivered for 15 seconds. Control animals received a standard sham lesion.
Experimental Procedure MammiUary lesion. The animals were assigned at random to the experimental mammillary group (EMG) or to the control mammillary group (CMG) before surgery. Those subjects which did not develop the previously described polydipsia were excluded (19). The resulting EMG consisted of 6 polydipsic animals and the CMG of 5 normodipsic animals. Water intake in both groups was measured during the five days before surgery and during a similar period after 15 days recovery from operation. Thereafter, the consumption of water and food available ad lib was measured in both groups in 12 hourly periods beginning with the onset of light or darkness, for 3 days and the results obtained in darkness were analyzed and expressed as percentages of the total daily consumption. Median eminence lesion. The subjects forming the experimental median eminence group (EMEG) and the control median eminence group (CMEG) (seven and five, respectively) were chosen and studied in the same way as the mammillary group. On the other hand, and on the basis of the data obtained (see below), rats in the EMEG were food deprived throughout the 12 hr lights on period, with food available ad lib during the dark cycle. Water was always available ad lib, and drinking was recorded as described above. Histology On completion of testing, the rats were killed with a sodium pentothal overdose and then perfused intracardially with saline followed by 10% buffered formaldehyde solution. The brains were removed and stored in formaldehyde until lamination.
Statistical Analysis Values reported are means ---S.D. Comparisons between groups were analyzed for statistical significance using the unpaired Student's t-test. Data from the same group were analyzed using now a paired Student's t-test. RESULTS
As expected, electrolytic lesions in the mammillary area induced polydipsia: an average of 89.2---22.5 ml as opposed to
,< IZ uJ (J 50-
25-
E MG EMEG WATER
EMG
EMEG FOOD
FIG. 1. Percentage of water and food consumedduring the dark phase by the experimental mammillary group (EMG) and by the experimental median eminence group (EMEG). Values reported are means---S.D. (~'k~-)p<0.005. 38.3---8.0 ml by the control animals (t= 4.3, p<0.005). The EMG rats consumed 80.8 __-4.4% of their daily total water intake in the 12 hours of the dark cycle and the CMG rats 85.8--.4.0% (t= 1.7, n.s.). That is to say, the rhythmicity of the polydipsic intake was not affected by the mammillary lesion. On the other hand, electrolytic lesions of the ME led to a marked polydipsia with intake about 200 ml daily on postoperative day seven. Thereafter, water intake fell progressively (possibly as a result of compensatory mechanisms of unknown origin), stabilizing at a mean volume of 135.9---47.8 ml in the EMEG as compared to 25.5 ± 7.3 ml in the CMEG (t = 4.7, p<0.001). The lesion also altered the daily rhythmicity or water intake, experimental animals consuming 67.2 ± 7.2% of their total intake during the nocturnal phase in comparison to 89.4± 3.9% in the control group (t = 5.9, p<0.001). Although ME lesioned animals drank greater amounts of water than mammillary lesioned rats (135.9 ---47.8 ml vs. 89.2 ± 22.5 ml respectively), the difference fell short of significance (t=2.0, n.s.). However, the percentage of water consumed during the dark phase differed markedly, reaching 6 7 . 2 - 7.2% in the EMEG and up to 80.8±4.4% in the EMG (t=3.6, p<0.005) (Fig. 1). Moreover, upon analyzing the data individually, some mammillary rats were noted to drink more daily water than some diabetic rats, a fact which was nevertheless insufficient to offset the difference in percentage nocturnal intake observed between the two groups. This finding seems to be independent of food intake (Table 1), and was unchanged in rats from both groups which consumed similar amounts of water (Table 2). No significant differences were seen in mean food intake between the EMG and EMEG, which consumed 30.3---6.8 g and 34.0± 10.9 g respectively (t=0.6, n.s.). Percentage food intake (during the nocturnal phase) likewise presented no significant difference, at 71.6±5.5% for the EMG and 65.0± 12.7% for the EMEG (t= 1.1, n.s.) (Fig. 1). Finally, EMEG rats changed their rhythmicity of water intake in response to the 12 hr period of food deprivation. During the nocturnal phase, the percentage of total water consumed with food available rose from 67.2±7.2% to 77.7±4.8% (t=5.3, p<0.005).
M A M M I L L A R Y POLYDIPSIA AND RHYTHMICITY
TABLE 1 WATER AND FOOD INTAKE CHARACTERISTICS OF TWO RATS, MAMMILLARY AND MEDIAN EMINENCE LESIONED (DIABETIC) RESPECTIVELY
Rat
Total Water Intake (nil)
Total Food Intake (g)
Water Intake Dark Phase (%)
Food Intake Light Phase (%)
Mammillary Diabetic
124.8 -+ 9.9 108.3 ± 4.2
31.3 - 1.5 35.3 ± 1.2
80.0 - 3.0 68.6 ± 1.7
30.2 ± 2.9 34.8 - 4.9
Values reported are means - S.D. (for details, see text).
913
Histological analysis confm'ned that the mammillary lesion had affected mainly the mammillary bodies, including the medial (pars medialis and pars lateralis) and lateral mammillary nuclei, as well as the most posterior zone of the same. In some cases the lesion involved a very limited portion of the posterior edge of the posterior hypothalamic nucleus, although more anterior portions of the hypothalamus (e.g., the median eminence) were never affected. The median eminence lesions on the other hand destroyed the most ventral region of the ventromedial hypothalamus. As seen in Figs. 2 and 3, the lesion, owing to its markedly medial location, affected the lateral walls of the third ventricle, including a large part of the nucleus arcuatus. The lateral extension of the lesion was variable, at times involving the most medial region of the ventromedial hypothalamic nucleus (VMH). The suprachiasmatic nucleus (SCN), located well forward of the site of the lesion, was undamaged in all rats.
TABLE 2
DISCUSSION
WATER INTAKE AND NOCTURNAL RHYTHMICITY OF TWO RATS, MAMMILLARY AND MEDIAN EMINENCE LESIONED (DIABETIC) RESPECTIVELY
As shown in the results, electrolytic lesions of the mammiUary area have no effect on the daily rhythmicity of water intake. Despite polydipsia, the percentage of water intake during the dark phase is similar (around 80% of the total water intake) in lesioned rats and in controls. These findings contrast with the data obtained in the ME lesioned group, where rhythmicity was affected by surgery. It thus appears that in principle, water intake throughout the day is distributed differently in rats with mammillary polydipsia and rats with centrally induced DI, and that these two groups can be clearly distinguished on the basis of this behavioral characteristic.
Rat
Total Water Intake (ml)
Water Intake Dark Phase (%)
Mammillary Diabetic
82.7 - 1.2 87.7 ± 5.5
84.0 _ 3.0 69.8 ± 3.7
Values reported are means ± S.D. (for details, see text).
FIG. 2. Photomicrograph of an electrolytic lesion of the median eminence. The most ventral zone of the ventromedial hypothalamus was affected by the lesion (see arrow).
914
MORALES, CUBERO AND PUERTO
vmh
ot
FIG. 3. Schematic representation of the median eminence electrolytic lesion (dark). Eventually, the lesions can extend to the most medial part of the VMH nucleus (shaded). Abbreviations: aha: anterior hypolhalamic area; cpu: nucleus caudate-putamen; fx: fornix; gp: globus pallidus; ot: optic tract; pvh: hypothalamic paraventricular nucleus; re: nucleus reuniens; v: ventricle; vmh: hypothalamic ventromedial nucleus. The findings in the DI group however deserve to be examined in greater detail. It is possible that the change in percentage water intake during the dark phase simply reflected the greater need to drink in these animals. This notion however seems unlikely, as water consumption reached extremely high levels (around 200 ml daily) approximately seven days postsurgery, although dark intake comprised only about 60% of the total (data not shown). Polydipsia stablized at around 15 days postlesion, when mean intake had fallen to 135 ml per day as noted in the results. Under these conditions and despite the reduction just noted, night time drinking remained at around 67%, a value significantly different from that recorded in mammillary and control rats. In addition, a study of the individual data in Tables 1 and 2 reveals that some mammillary rats drank more water than some diabetic rats, although the differences in nocturnal distribution of water intake were uninfluenced by this fact; hence this cannot be interpreted as a consequence of a greater food intake during the day in diabetic rats. Moreover, the differences noted in nocturnal rhythmicity could also be seen when mammillary and diabetic rats consumed similar amounts of water. Furthermore, the rhythmicity of water intake in diabetic rats was susceptible to experimental manipulation, e.g., food deprivation during the 12 hr light period. Under this condition, the percentage of nocturnal water intake rose significantly to make up 77% of the total. This implies that the percentage of water consumed under ad lib conditions during this period (67%) cannot be taken as a possible "ceiling effect." To sum up, the data suggest that the alteration recorded in daily rhythmicity of water intake in rats with centrally induced DI is not conditioned by the amount of water consumed, but rather by the nature of the hormonal alteration itself. Similar findings have been described in rats with genetically determined DI (Brattleboro rats), which also present changes in rhythmicity owing not so much to a loss of circadian control over drinking, which is unaffected (7) as to the need to drink in order to maintain water balance, due to the continuous urinary loss (10,15). Given that the ME lesion may affect the most medial part of the VMH nucleus, the changes in drinking rhythmicity may be traced indirectly to an alteration in feeding behavior, with which this nucleus is closely related [see (18) for review]. This variable does not however appear to be essential, as mean food intake in diabetic rats was no greater than in mammillary lesioned animals [it should
be recalled that the mammillary lesion never invaded the ventromedial hypothalamus, nor were alterations in food intake recorded in connection with these lesions, see (19)]. The lack of significant differences between the diabetic and mammillary group in nocturnal distribution of food intake further supports this conclusion. Experimental studies in other laboratories have demonstrated that only massive lesions to the VMH nucleus were capable of causing changes in the circadian rhythmicity of certain variables such as feeding and locomotor activity (9,18). Such alterations usually arise as a result of the destruction of efferent fibers from the SCN (9); only part of these fibers cross the ventromediai hypothalamus near the VMH nucleus (2). In the present series of experiments, only the most medial part of this nucleus is damaged, although some efferent fibers from the SCN (left intact in all animals studied) may have been affected. Nevertheless, this is insufficient to explain the present findings, as born out by a) the absence of hyperphagic effects, as would be expected following a lesion to the VMH nucleus, b) the lack of differences in the nocturnal rhythmicity of food intake between diabetic and mammillary rats, and c) the susceptibility of drinking rhythmicity in diabetic rats to experimental manipulation, a phenomenon which may well not have occurred if the efferent system of the SCN had been damaged seriously. To sum up, the results brought to light in this article suggest that the two types of polydipsia investigated differ in origin, as suggested by previous studies carded out in our laboratory (13). The alteration in the daily rhythmicity of water intake in rats with centrally induced DI appears not to be a secondary effect related to a possible change in feeding patterns, nor to reflect a general disruption of circadian rhythms, but rather to result from the continuous need to drink in order to maintain fluid homeostasis. Although our findings apparently do not support a diabetic origin for mammillary polydipsia, further studies will be needed to identify the ultimate cause of this behavior. ACKNOWLEDGEMENTS
Our thanks to David Butler and Karen Shashok for their help in the English version of this text. We are indebted to Professor Neil McClaren, of the English Department of the University of Granada, who supervised the final form of the manuscript.
MAMMILLARY POLYDIPSIA AND RHYTHMICITY
915
REFERENCES 1. Albe-Fessard, D.; Stutinskym, F.; Libouban, S. Atlas ster6otaxique du dienc6phale du rat blanc. Paris: Editions du CNRS; 1969. 2. Berk, M. L.; Finkelstein, J. A. An autoradiographic determination of the efferent projections of the suprachiasmatic nucleus of the hypothalamus. Brain Res. 226:1-13; 1981. 3. Brodal, A. Neurological anatomy in relation to clinical medicine. New York: Oxford University Press; 1981. 4. De Groot, J. The rat brain in stereotaxic coordinates. Trans. R. Netherland Acad. Sci. 52:1--40; 1959. 5. Gardiner, T. W.; Stricker, E. M. Hyperdipsia in rats after electrolytic lesions of nucleus medianus. Am. J. Physiol. 248:R214-R223; 1985. 6. Gardiner, T. W.; Stricker, E. M. Impaired drinking responses of rats with lesions of nucleus medianus: circadian dependence. Am. J. Physiol. 248:R224-R230; 1985. 7. Grobleski, T. A.; Ndfiez, A. A.; Gold, R. M. Circadian rhythms in vasopressin deficient rats. Brain Res. Bull. 6:125-130; 1981. 8. Grossman, S. P. A reassessment of brain mechanisms that control thirst. Neurosci. Biobehav. Rev. 8:95-104; 1984. 9. Inouye, S. T. Does the VMH contain a self-sustained circadian oscillator associated with periodic feedings? Brain Res. 279:53-63; 1983. 10. Kutscher, C. L.; Wright, W. A. Drinking characteristics of normal rats and rats heterozygous and homozygous for diabetes insipidus. Physiol. Behav. 18:833-839; 1977. 11. Morales, A.; Puerto, A. Efecto de la estimulaci6n el6ctrica del nticleo
12.
13.
14. 15.
16. 17.
18. 19.
supra6ptico sobre la polidipsia mamilar. Arch. Neurobiol. 51:154158; 1988. Morales, A.; Puerto, A. Posterior hypothalamic polydipsia: differential effects to several dipsogenic treatments. In: de Caro, G.; Epstein, A. N.; Massi, M., eds. The physiology of thirst and sodium appetite. New York: Plenum Press; 1986:77-81. Morales, A.; Puerto, A. A study of the response to several dipsogenic treatments in rats with mammilllary polydipsia and with centrally induced diabetes insipidus. Behav. Brain Res. 31:69-74; 1988. Morales, A.; Puerto, A. Un estudio del car~icter primario/secundario de la polidipsia inducida por lesi6n electrolftica de los cuerpos mamilares. Rev. Esp. Fisiol., in press; 1989. Peterson, G. M.; Watkins, W. B.; Moore, R. Y. The suprachiasmatic hypothalamic nuclei of the rat. VI. Vasopressin neurons and circadian rhythmicity. Behav. Neural Biol. 29:236-245; 1980. Rolls, B. J. Drinking by rats after irritative lesions in the hypothalamus. Physiol. Behav. 5:1385-1393; 1970. Smith, R. W.; McCann, S. M. Increased and decreased water intake in the rat with hypothalamic lesions. Am. J. Physiol. 203:366-370; 1962. Stoynev, A. G.; Ikonomov, O. C. Circadian regulation of feeding in rats: suprachiasmatic versus ventromedial hypothalamic nuclei. Appetite 9:217-229; 1987. Tejedor del Real, M. P.; Santacana-Altimiras, M. P.; Alvarez-Pel"iez, R. Alteraci6n en la toma de agua despu6s de la lesi6n de los cuerpos mamilares. Rev. Esp. Fisiol. 28:95-102; 1972.