- Email: [email protected]
Age-related decline of vasopressin mRNA in the bed nucleus of the stria terminalis
NeurobiologyofAging,Vol. 12, pp. 419--423. ©Pergamon Press plc, 1991. Printed in the U.S.A.
0197-4580/91 $3.00 + .00
Age-Related Decline of Vasopressin mRNA in the Bed Nucleus of the Stria Terminalis D O R C A S J. D O B I E , *l M A R G A R E T A. M I L L E R , * t J A N I C E H. U R B A N , : ~ M U R R A Y A. R A S K I N D * t A N D D A N I E L M. D O R S A * t : ~ §
*Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, WA 98195 ~-GRECC, Seattle Veterans' Affairs Medical Center, Seattle, WA 98108 ~:Department of Pharmacology, University of Washington, Seattle, WA 98195 §Department of Medicine, University of Washington, Seattle, WA 98195 R e c e i v e d 14 M a y 1990; A c c e p t e d 16 April 1991 DOBIE, D. J., M. A. MILLER, J. H. URBAN, M. A. RASKIND AND D. M. DORSA. Age-relateddecline of vasopressin mRNA in the bed nucleus of the stria terminalis. NEUROBIOL AGING 12(5) 419--423, 1991.--To determine whether aging influences arginine vasopressin (AVP) biosynthesis in the extrahypothalamic neurons of the bed nucleus of the stria terminalis (BNST), we used in situ hybridization and quantitative autoradiography to compare AVP mRNA in 3-month-old, 14-month-old, and 24-month-old male Fischer 344 rats. As AVP synthesis in the BNST has previously been shown to be steroid-dependent, plasma testosterone (T) was measured by radioimmunoassay. The 24-month-old animals had significantly fewer AVP-labelled cells than either the 3-month-old (p<0.01) or 14-month-old (p<0.05) animals. The cells that were present in the 24-month animals were less intensely labelled than in the other groups, as indicated by a significantly reduced number of grains per cell (p<0.01). Plasma T was also significantly lower in 24-month-old animals when compared with 3-month (p<0.01) or 14-month (p<0.05) groups. The results indicate that there is a marked age-related decline in vasopressin biosynthetic activity in neurons of the BNST. Vasopressin
Bed nucleus of the stria terminalis
Fischer 344 rats
ARGININE vasopressin (AVP) is a neurohypophyseal hormone that also appears to act as a CNS neuromodulator. In addition to the classical vasopressinergic projections to the neurohypophysis arising from the magnocellular neurons of the supraoptic (SON) and paraventricular (PVN) nuclei of the hypothalamus, AVP immunoreactive fibers and perikarya have been visualized in several extra-hypothalamic CNS sites. Immunoreactive perikarya have been found in the suprachiasmatic nucleus (SCN), the bed nucleus of the stria terminalis (BNST), the medial amygdala (MA), the locus coeruleus (LC), and accessory magnocellular groups. Immunoreactive fibers from these cell bodies, particularly from the BNST, project to a variety of extra-hypothalamic regions, including the septum (2,3). Vasopressinergic neurons in these regions have been implicated in a variety of functions that are disturbed in aging. An age-related decrease in the AVP content of the dorsal septum, organ vasculosum of the lamina terminalis, and LC, has been demonstrated by radioimmunoassay (7). Similarly, immunocytochemicai studies of senescent Brown-Norway rats have revealed a decrease in AVP immunoreactive fiber density in the MA, substantia nigra (SN), ventral hippocampus, LC, central grey, and nucleus ambiguus (8). Through the technique of in situ hybridization histochemistry, we have previously reported the detection of AVP messenger RNA (mRNA) in cells within
Testosterone
Aging
the BNST (18) and MA (23). In the current study, we use in situ hybridization histochemistry and quantitative autoradiography to examine age-related changes in the biosynthetic capacity of AVP neurons in the BNST. METHOD Three-month-old (n = 6), 14-month-old (n = 6), and 24-monthold (n = 6) virgin male Fisher 344 rats were obtained from the Harlan Sprague-Dawley National Institute of Aging breeding facility. The animals were housed for 21 days in filter-bonnetted cages and allowed ad lib access to commercial food and water. The rats were sacrificed by decapitation; their brains were removed and immediately frozen on dry ice. Twenty-micron thick serial brain sections were cut on a cryostat and thaw-mounted onto RNAse-free slides. The slides were stored at - 7 0 ° C . Trunk blood was collected for the measurement of plasma concentration of testosterone (T) by radioimmunoassay. Reagents for the assay were supplied by the WHO Matched Reagent Program. Details of this assay have been described previously (14). The minimum sensitivity of the assay was less than 10 pg/tube (0.1 ng/ml). The intraassay coefficient of variation was 8.4%. In situ hybridization was performed according to the protocol that we have previously described (17). Slides from all animals
~Requests for reprints should be addressed to Dorcas J. Dobie, M.D., RP-10, Department of Psychiatry, University of Washington, Seattle, WA 98195.
419
42O
DOBIE ET AL 100'
80 .J .M IM
ffl .1
80'
u
60 ¸
o r~ t~ .J ,J ~J
60"
,( =J
W D.
.... ... / / / / / / / / 1/11//// / / / / / / / / / / / / / / / / / / / / / i / / i l l / / i l l /111/111 / i / i / / / i
_z < 0c
40"
40" ////////
U.
0
20"
I
/ / / / / / / / /
~
z
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
0
/ / / / / / / / /
tu m
3
mos.
14 mos.
//////// / / / / / / / / /i/111// / i / / / / / / / / / / / / / / / / / / / i / / / / / / / / / i
~/////Z
0
20'
/ / / / / / / / IJJJJ/J/ / / / J / / / / / / / / / / / /
24 mos.
3 mos.
///j///.,I ,.//~-.,.
/ / A
~'/////A t / / / / / / ~ e.~.//.i/1~ ~.//j/llll ~.//////A
/.¢//...A /////'///1
14 mos.
2 4 mos.
AGE
FIG. 1. Number of cells in the bed nucleus of the stria terminalis (BNST) showing labelling for AVP mRNA following in situ hybridization (*p<0.05, **p<0.01, compared to 24-month animals). were run in a single assay and were distributed evenly throughout the assay. Sections were postfixed in 4% paraformaldehyde, rinsed in phosphate and triethanolamine buffers, treated with acetic anhydride, and delipidated and dehydrated through alcohols and chloroform. A 35S-labelled oligonucleotide probe was applied to slides in 45 Ixl aliquots at a concentration of 2.5 pmol/ml. This probe concentration was determined by performing a saturation curve. The slides were incubated overnight at 37°C, subjected to a series of washes in 1 x standard saline citrate (SSC), dehydrated, and coated with Kodak NTB-2 photographic emulsion. They were exposed for three weeks, developed, and stained with cresyl violet. The probe used was a 48-base oligonucleotide complementary to the mRNA coding for the last 16 amino acids of the glycopeptide portion of the vasopressin precursor. This portion of the vasopressin molecule does not overlap in sequence with oxytocin, hence allowing specific recognition of the vasopressin precursor molecule. We have previously verified the specificity of the signal obtained with this probe by RNAse pretreatment, by displacement with excess unlabelled probe, and by comparison with other methods of anatomical localization, including immunocytochemistry (18). Additionally, Young et al. detected homogenous bands of appropriate size on northern analysis of RNA extracted from vasopressinergic brain regions using this identical oligonucleotide sequence (24). The probe was end-labelled with 35S-dATP using terminal deoxyribonuceotidyl transferase, purified on a NENSORB column, and heat denatured. It was then diluted in a buffer containing 50% formamide, 10% dextran sulfate, 0.3 M NaC1, 10 mM Tris (ph 8.0), 1 mM EDTA (ethylene diaminetetraacetate), 1 x Denhardt's solution, 0.5 mg/ml yeast tRNA, and 10 mM dithiothreitol (DTT). The specific activity was estimated to be 6320 Ci/mmol (8.7 x 10s dpm/~g). Processed sections through the BNST of 3-month-old (n = 4), 14-month-old ( n = 4 ) , and 24-month-old animals ( n = 5 ) were anatomically matched at atlas levels 6360--6790 of the stereotaxic atlas of K6nig and Klippel (12). The sections were coded
FIG. 2, Number of labelled grains per cell in the BNST following in situ hybridization for AVP mRNA. (*p<0.01, 24-month animals compared to 3- or 14-month animals).
and read blindly using dark-field optics. The total number of labelled cells present in the BNST of one matched hemisection from each slide was recorded. A cell was considered to be labelled if its signal to background ratio was greater than 14:1. The number of grains per cell was measured using an automated image analysis system that has been developed by Dr. Donald Clifton, Department of Obstetrics and Gynecology, University of Washington (20). This system consists of an IBM-AT computer equipped with a video imaging board. Images were obtained by a Dage model 65 camera attached to an Olympus photomicroscope The images were acquired in dark-field using a 40 x objective. Background and signal grain density were calculated, as were the number of grains per cell. No more than ten labelled cells were grain-counted in each hemisection. An exhaustive discussion of the details of the methodology has been recently published (5).
Statistical Analysis Variables are expressed as m e a n - S E M . The testosterone levels, the number of labelled cells, and the grains per cell were compared among the three age groups using one-way analysis of variance. When ANOVA revealed significance, Duncan's multiple range test was used to assess the magnitude of the significance of the differences between groups. RESULTS There were significantly fewer labelled cells in the BNST of the 24-month-old animals (35.4_+ 11.2) than there were in the BNST of the 3-month-old group (77.5 _+6.5; p<0.01) or the 14month-old group (67.8___4.8; p<0.05) (Fig. 1). Similarly, the 24-month-old animals demonstrated significantly fewer grains per cell (36.65-+3.3) than did the 3-month-old animals (56.25 -+ 3.4; p<0.01) or the 14-month-old animals (53.03 _ 2.6; p<0.01) (Fig. 2). Plasma T levels were markedly lower in 24-
FACING PAGE FIG. 3. Dark-field photomicrographs of hybridized sections through the BNST at atlas level 6670. Twenty-four-month-old rats (Ax 37.2) show fewer and less intensely labelled cells than do 3-month-old rats (B x 37.2).
AGE-RELATED DECLINE OF VASOPRESSIN mRNA
421
422
DOBIE ET AL.
month-old animals (0.156---0.045 ng/ml) than in either 3-monthold (1.103---0.243 ng/ml; p<0.01) or 14-month-old animals (0.905__+0.276 ng/ml; p<0.05). Figure 3 presents high power dark-field photomicrographs of representative hybridized sections through the BNST of 3-month-old and 24-month-old animals; 24-month-old animals show fewer and less intensely labelled cells. There were no statistically significant differences between 3-month and 14-month-old animals in any of the parameters measured. DISCUSSION The results of this experiment indicate that aged Fischer 344 rats show not only a reduced number of AVP-synthesizing cells in the BNST, but also a diminished capacity for synthesizing AVP within the remaining active cells. These neurons and their projections to the septum have been implicated in a variety of CNS functions that are disturbed in aging, including memory, social behaviors, conditioned taste aversion, antipyresis, and regulation of blood pressure. The ability of centrally administered vasopressin to prevent the extinction of conditioned avoidance responses, prevent or reverse amnesia, and enhance performance in positively rewarded behavior has been repeatedly demonstrated in increasingly sophisticated studies in rodents (13,22). It is plausible to consider the possibility that the vasopressin deficiency in extra-hypothalamic brain regions contributes to disturbances in a variety of CNS functions in aged animals. Aged animals provide an interesting model for studying the effects of low levels of androgens on AVP synthesis (10,19). Our laboratory has previously shown that AVP synthesis in the BNST of young rats is a sexually dimorphic, T-dependent phenomenon (16). The finding of a decrease in AVP synthetic activity in the BNST of inherently androgen-deficient aged animals is consistent with gonadal steroid regulation of these neurons. Whether neurons in the BNST of aged animals are still capable of responding to T remains to be explored. Studies are underway to further evaluate the mechanisms by which T modulates AVP biosynthetic activity in senescence. That the interaction between vasopressin and gonadal hormones may be relevant to an understanding of CNS changes in aging is suggested by behavioral studies. For example, aged rats show impairment on conditioned taste aversion (CTA) paradigms; this impairment can be reversed by vasopressin (6). The underlying neurobiology of this observation has not yet been elucidated. If vasopressin is important in normal memory processes, it is possible that the age-related impairment in performance on the CTA paradigm may be due to a decreased synthesis of AVP in cell groups such as the BNST that project to the septum. Moreover, as CTA has been shown by our group and others to be an androgen-dependent phenomenon (1,4), the age-
related impairment on this conditioned avoidance paradigm could be attributed to the influence of low levels of circulating T on AVP synthesizing cells in the BNST of aged animals. Our observations are consistent with previously cited reports of decreased vasopressinergic immunoreactive fiber density in extra-hypothalamic brain regions in senescent rats (8) and with a report that this decrease in fiber density can be reversed by administration of T (9). However, the findings obtained by immunocytochemistry are difficult to interpret. The decrease in immunoreactivity of a neuron could be explained by decreased peptide synthesis, increased peptide degradation, or increased peptide release. Moreover, changes in the activity of neuronal systems may be detected more immediately by measuring mRNA than by measuring peptide immunoreactivity. We have recently shown that, after castration, AVP gene expression in the BNST as measured by in situ hybridization declines much more rapidly than does AVP immunoreactivity (15). Thus it appears that in situ hybridization is a more direct method than immunocytochemistry to assess the synthetic activity of the neurons in this extra-hypothalamic vasopressin system. Although in situ hybridization remains a semiquantitative technique, the number of grains per cell has been demonstrated to be proportional to the mRNA content of a cell by both Northern analysis (11,21) and solution hybridization (5). The exact mathematical nature of the relationship is unclear. A standardization process that would allow the number of silver grains over a cell to be related to the absolute number of mRNA copies has not been developed. Nonetheless, it is possible to quantify relative changes in gene expression using this technique, and so to estimate differences in vasopressin biosynthetic activity between groups in the same assay. We have used in situ hybridization to show that AVP synthesis in the BNST is diminished in aged Fischer 344 male rats. This is consistent with reports of a decrease in vasopressinergic fiber density in areas that project from the BNST in senescent Brown-Norway rats (8) and extends the observation to a strain of rat which has been more consistently used for studies of aging. In situ hybridization provides an advantage over immunocytochemical reports in that it allows the observation of a decrease in vasopressin content to be more convincingly explained as being due to a decrease in AVP synthesis. That these extra-hypothalamic neurons have been shown to be steroid sensitive in young animals suggests that this system may be a useful model in which to investigate the regulation of AVP gene expression in aging. ACKNOWLEDGEMENTS We would like to thank Ms. Cheryl Refsdal, Ms. Kimberly Donnell, and Mr. Carl Sikkema for their invaluable technical assistance. This work was supported by the Department of Veterans' Affairs, by NIH grants AG8419 and N.S. 20311, and by NARSAD.
REFERENCES 1. Brot, M. D.; Berstein, I. L.; Dorsa, D. M. Chronic central administration of vasopressin antagonist hastens extinction of conditioned taste aversion in rats. Soc. Neurosci. Abstr. 15:1069; 1989. 2. Buijs, R. M.; Swaab, D. F.; Dogterom, J.; Van Leeuwen, F. W. Intra and extra hypothalamic vasopressin and oxytocin pathways in the rat. Cell Tissue Res. 186:423-433; 1978. 3. Caffe, R.; Van Leeuwen, F. W. Vasopressin-immunoreactive cells in the dorsomedial hypothalamic region, medial amygdaloid nucleus, and locus coeruleus of the rat. Cell Tissue Res. 233:23-33; 1983. 4. Chambers, K. C. Hormonal influences on sexual dimorphism in rate of extinction of a conditioned taste aversion in rats. J. Comp. Physiol. Psychol. 90(9):851-856; 1976.
5. Chowen, J. A.; Steiner, R. A.; Clifton, D. K. Semiquantitative analysis of cellular somatostatin mRNA levels by in situ hybridization histochemistry. Methods Neurosci.; in press. 6. Cooper, R. L.; McNamara, M. C.; Thompson, W. U. Vasopressin and conditioned flavor aversion in aged rats. Neurobiol. Aging 1:53-57; 1980. 7. Dorsa, D. M.; Bottemiller, L. A. Age-related changes of vasopressin content in microdissected areas of rat brain. Brain Res. 242: 151-156; 1982. 8. Fliers, E.; deVries, G. J.; Swaab, D. F. Changes with aging in the vasopressin and oxytocin innervation of the rat brain. Brain Res. 348:1-8; 1985. 9. Goudsmit, E.; Fliers, E.; Swaab, D. F. Testosterone supplementa-
AGE-RELATED DECLINE OF VASOPRESSIN mRNA
10. 11.
12.
13. 14.
15.
16.
17.
tion restores vasopressin innervation in the senescent rat brain. Brain Res. 473:306-313; 1988. Kaler, L. W.; Neaves, W. B. The androgen status of the aging male rat. Endocrinology 108:712-719; 1981. Kelsey, J. E.; Watson, S. J.; Burke, S.; Akil, H.; Roberts, J. L. Characterization of proopiomelanocortin mRNA detected by in situ hybridization. J. Neurosci. 6(1):38; 1986. KSnig, J. F. R.; Klippel, R. A. The rat brain: A stereotaxic atlas of the forebrain and lower parts of the brain stem. New York: Krieger; 1963. Koob, G. F.; Bloom, F. E. Behavioral effects of neuropeptides: endorphins and vasopressin. Annu. Rev. Physiol. 44:571-582; 1982. Matsumoto, A. M.; Paulsen, C. A.; Hopper, B. R.; Rebar, R. W.; Bremner, W. J. Human chorionic gonadotropin and testicular function: stimulation of testosterone, testosterone precursors, and sperm production despite high estradiol levels. J. Clin. Endocrinol. Metab. 56:730; 1983. Miller, M. A.; Dorsa, D. M.; DeVries. G. J. Decline of vasopressin mRNA and immunoreactivity in the BNST following castration: a time course study. Soc. Neurosci. Abstr. 16:661; 1990. Miller, M. A.; Urban, J. H.; Dorsa, D. M. Steroid dependency of vasopressin neurons in the bed nucleus of the stria terminalis by in situ hybridization. Endocrinology 125:2335-2340; 1989. Miller, M. A.; Urban, J. H.; Dorsa, D. M. Quantification of mRNA in discrete cell groups of brain by in situ hybridization histochemis-
423
try. Methods Neurosci. 1:164-181; 1989. 18. Miller, M. A.; Zoeller, R. T.; Dorsa, D. M. Detection of vasopressin messenger RNA in cells within the bed nucleus of the stria terminalis by in situ hybridization histochemistry. Neurosci. Lett. 94:264-268; 1988. 19. Ravid, R.; Fliers, E.; Swaab, D. F.; Zurcher, C. Changes in vasopressin and testosterone in the senescent Brown-Norway (BN/BiRij) rat. Gerontology 33:87-98; 1987. 20. Rogers, K. V.; Vician, L.; Steiner, R. A.; Clifton, D. A. The effect of hypophysectomy and growth hormone administration on preprosomatostain mRNA in the periventricular nucleus of the rat hypothalamus. Endocrinology 122:586-591; 1988. 21. Sherman, T. G.; Akil, H.; Watson, S. J. Vasopressin mRNA expression: a northern and in situ hybridization analysis. In: Schrier, R., ed. Vasopressin. New York: Raven Press; 1985:475--483. 22. Strupp, B. J.; Levitsky, D. A. A mnenomic role for vasopressin: The evidence for and against. Neurosci. Biobehav. Rev. 9:399-411; 1985. 23. Urban, J. H.; Miller, M. A.; Drake, C. A.; Dorsa, D. M. Detection of vasopressin mRNA in cells of medial amygdala but not locus coeruleus by in situ hybridization. J. Chem. Neuroanat. 3:27783; 1990. 24. Young, W. S., III; Bonner, T. I.; Brann, M. R. Mesencephalic dopamine neurons regulate the expression of neuropeptide m/RNAs in the rat forebrain. Proc. Natl. Acad. Sci. USA 83:9827-9831; 1986.
0197-4580/91 $3.00 + .00
Age-Related Decline of Vasopressin mRNA in the Bed Nucleus of the Stria Terminalis D O R C A S J. D O B I E , *l M A R G A R E T A. M I L L E R , * t J A N I C E H. U R B A N , : ~ M U R R A Y A. R A S K I N D * t A N D D A N I E L M. D O R S A * t : ~ §
*Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, WA 98195 ~-GRECC, Seattle Veterans' Affairs Medical Center, Seattle, WA 98108 ~:Department of Pharmacology, University of Washington, Seattle, WA 98195 §Department of Medicine, University of Washington, Seattle, WA 98195 R e c e i v e d 14 M a y 1990; A c c e p t e d 16 April 1991 DOBIE, D. J., M. A. MILLER, J. H. URBAN, M. A. RASKIND AND D. M. DORSA. Age-relateddecline of vasopressin mRNA in the bed nucleus of the stria terminalis. NEUROBIOL AGING 12(5) 419--423, 1991.--To determine whether aging influences arginine vasopressin (AVP) biosynthesis in the extrahypothalamic neurons of the bed nucleus of the stria terminalis (BNST), we used in situ hybridization and quantitative autoradiography to compare AVP mRNA in 3-month-old, 14-month-old, and 24-month-old male Fischer 344 rats. As AVP synthesis in the BNST has previously been shown to be steroid-dependent, plasma testosterone (T) was measured by radioimmunoassay. The 24-month-old animals had significantly fewer AVP-labelled cells than either the 3-month-old (p<0.01) or 14-month-old (p<0.05) animals. The cells that were present in the 24-month animals were less intensely labelled than in the other groups, as indicated by a significantly reduced number of grains per cell (p<0.01). Plasma T was also significantly lower in 24-month-old animals when compared with 3-month (p<0.01) or 14-month (p<0.05) groups. The results indicate that there is a marked age-related decline in vasopressin biosynthetic activity in neurons of the BNST. Vasopressin
Bed nucleus of the stria terminalis
Fischer 344 rats
ARGININE vasopressin (AVP) is a neurohypophyseal hormone that also appears to act as a CNS neuromodulator. In addition to the classical vasopressinergic projections to the neurohypophysis arising from the magnocellular neurons of the supraoptic (SON) and paraventricular (PVN) nuclei of the hypothalamus, AVP immunoreactive fibers and perikarya have been visualized in several extra-hypothalamic CNS sites. Immunoreactive perikarya have been found in the suprachiasmatic nucleus (SCN), the bed nucleus of the stria terminalis (BNST), the medial amygdala (MA), the locus coeruleus (LC), and accessory magnocellular groups. Immunoreactive fibers from these cell bodies, particularly from the BNST, project to a variety of extra-hypothalamic regions, including the septum (2,3). Vasopressinergic neurons in these regions have been implicated in a variety of functions that are disturbed in aging. An age-related decrease in the AVP content of the dorsal septum, organ vasculosum of the lamina terminalis, and LC, has been demonstrated by radioimmunoassay (7). Similarly, immunocytochemicai studies of senescent Brown-Norway rats have revealed a decrease in AVP immunoreactive fiber density in the MA, substantia nigra (SN), ventral hippocampus, LC, central grey, and nucleus ambiguus (8). Through the technique of in situ hybridization histochemistry, we have previously reported the detection of AVP messenger RNA (mRNA) in cells within
Testosterone
Aging
the BNST (18) and MA (23). In the current study, we use in situ hybridization histochemistry and quantitative autoradiography to examine age-related changes in the biosynthetic capacity of AVP neurons in the BNST. METHOD Three-month-old (n = 6), 14-month-old (n = 6), and 24-monthold (n = 6) virgin male Fisher 344 rats were obtained from the Harlan Sprague-Dawley National Institute of Aging breeding facility. The animals were housed for 21 days in filter-bonnetted cages and allowed ad lib access to commercial food and water. The rats were sacrificed by decapitation; their brains were removed and immediately frozen on dry ice. Twenty-micron thick serial brain sections were cut on a cryostat and thaw-mounted onto RNAse-free slides. The slides were stored at - 7 0 ° C . Trunk blood was collected for the measurement of plasma concentration of testosterone (T) by radioimmunoassay. Reagents for the assay were supplied by the WHO Matched Reagent Program. Details of this assay have been described previously (14). The minimum sensitivity of the assay was less than 10 pg/tube (0.1 ng/ml). The intraassay coefficient of variation was 8.4%. In situ hybridization was performed according to the protocol that we have previously described (17). Slides from all animals
~Requests for reprints should be addressed to Dorcas J. Dobie, M.D., RP-10, Department of Psychiatry, University of Washington, Seattle, WA 98195.
419
42O
DOBIE ET AL 100'
80 .J .M IM
ffl .1
80'
u
60 ¸
o r~ t~ .J ,J ~J
60"
,( =J
W D.
.... ... / / / / / / / / 1/11//// / / / / / / / / / / / / / / / / / / / / / i / / i l l / / i l l /111/111 / i / i / / / i
_z < 0c
40"
40" ////////
U.
0
20"
I
/ / / / / / / / /
~
z
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
/ / / / / / / / /
0
/ / / / / / / / /
tu m
3
mos.
14 mos.
//////// / / / / / / / / /i/111// / i / / / / / / / / / / / / / / / / / / / i / / / / / / / / / i
~/////Z
0
20'
/ / / / / / / / IJJJJ/J/ / / / J / / / / / / / / / / / /
24 mos.
3 mos.
///j///.,I ,.//~-.,.
/ / A
~'/////A t / / / / / / ~ e.~.//.i/1~ ~.//j/llll ~.//////A
/.¢//...A /////'///1
14 mos.
2 4 mos.
AGE
FIG. 1. Number of cells in the bed nucleus of the stria terminalis (BNST) showing labelling for AVP mRNA following in situ hybridization (*p<0.05, **p<0.01, compared to 24-month animals). were run in a single assay and were distributed evenly throughout the assay. Sections were postfixed in 4% paraformaldehyde, rinsed in phosphate and triethanolamine buffers, treated with acetic anhydride, and delipidated and dehydrated through alcohols and chloroform. A 35S-labelled oligonucleotide probe was applied to slides in 45 Ixl aliquots at a concentration of 2.5 pmol/ml. This probe concentration was determined by performing a saturation curve. The slides were incubated overnight at 37°C, subjected to a series of washes in 1 x standard saline citrate (SSC), dehydrated, and coated with Kodak NTB-2 photographic emulsion. They were exposed for three weeks, developed, and stained with cresyl violet. The probe used was a 48-base oligonucleotide complementary to the mRNA coding for the last 16 amino acids of the glycopeptide portion of the vasopressin precursor. This portion of the vasopressin molecule does not overlap in sequence with oxytocin, hence allowing specific recognition of the vasopressin precursor molecule. We have previously verified the specificity of the signal obtained with this probe by RNAse pretreatment, by displacement with excess unlabelled probe, and by comparison with other methods of anatomical localization, including immunocytochemistry (18). Additionally, Young et al. detected homogenous bands of appropriate size on northern analysis of RNA extracted from vasopressinergic brain regions using this identical oligonucleotide sequence (24). The probe was end-labelled with 35S-dATP using terminal deoxyribonuceotidyl transferase, purified on a NENSORB column, and heat denatured. It was then diluted in a buffer containing 50% formamide, 10% dextran sulfate, 0.3 M NaC1, 10 mM Tris (ph 8.0), 1 mM EDTA (ethylene diaminetetraacetate), 1 x Denhardt's solution, 0.5 mg/ml yeast tRNA, and 10 mM dithiothreitol (DTT). The specific activity was estimated to be 6320 Ci/mmol (8.7 x 10s dpm/~g). Processed sections through the BNST of 3-month-old (n = 4), 14-month-old ( n = 4 ) , and 24-month-old animals ( n = 5 ) were anatomically matched at atlas levels 6360--6790 of the stereotaxic atlas of K6nig and Klippel (12). The sections were coded
FIG. 2, Number of labelled grains per cell in the BNST following in situ hybridization for AVP mRNA. (*p<0.01, 24-month animals compared to 3- or 14-month animals).
and read blindly using dark-field optics. The total number of labelled cells present in the BNST of one matched hemisection from each slide was recorded. A cell was considered to be labelled if its signal to background ratio was greater than 14:1. The number of grains per cell was measured using an automated image analysis system that has been developed by Dr. Donald Clifton, Department of Obstetrics and Gynecology, University of Washington (20). This system consists of an IBM-AT computer equipped with a video imaging board. Images were obtained by a Dage model 65 camera attached to an Olympus photomicroscope The images were acquired in dark-field using a 40 x objective. Background and signal grain density were calculated, as were the number of grains per cell. No more than ten labelled cells were grain-counted in each hemisection. An exhaustive discussion of the details of the methodology has been recently published (5).
Statistical Analysis Variables are expressed as m e a n - S E M . The testosterone levels, the number of labelled cells, and the grains per cell were compared among the three age groups using one-way analysis of variance. When ANOVA revealed significance, Duncan's multiple range test was used to assess the magnitude of the significance of the differences between groups. RESULTS There were significantly fewer labelled cells in the BNST of the 24-month-old animals (35.4_+ 11.2) than there were in the BNST of the 3-month-old group (77.5 _+6.5; p<0.01) or the 14month-old group (67.8___4.8; p<0.05) (Fig. 1). Similarly, the 24-month-old animals demonstrated significantly fewer grains per cell (36.65-+3.3) than did the 3-month-old animals (56.25 -+ 3.4; p<0.01) or the 14-month-old animals (53.03 _ 2.6; p<0.01) (Fig. 2). Plasma T levels were markedly lower in 24-
FACING PAGE FIG. 3. Dark-field photomicrographs of hybridized sections through the BNST at atlas level 6670. Twenty-four-month-old rats (Ax 37.2) show fewer and less intensely labelled cells than do 3-month-old rats (B x 37.2).
AGE-RELATED DECLINE OF VASOPRESSIN mRNA
421
422
DOBIE ET AL.
month-old animals (0.156---0.045 ng/ml) than in either 3-monthold (1.103---0.243 ng/ml; p<0.01) or 14-month-old animals (0.905__+0.276 ng/ml; p<0.05). Figure 3 presents high power dark-field photomicrographs of representative hybridized sections through the BNST of 3-month-old and 24-month-old animals; 24-month-old animals show fewer and less intensely labelled cells. There were no statistically significant differences between 3-month and 14-month-old animals in any of the parameters measured. DISCUSSION The results of this experiment indicate that aged Fischer 344 rats show not only a reduced number of AVP-synthesizing cells in the BNST, but also a diminished capacity for synthesizing AVP within the remaining active cells. These neurons and their projections to the septum have been implicated in a variety of CNS functions that are disturbed in aging, including memory, social behaviors, conditioned taste aversion, antipyresis, and regulation of blood pressure. The ability of centrally administered vasopressin to prevent the extinction of conditioned avoidance responses, prevent or reverse amnesia, and enhance performance in positively rewarded behavior has been repeatedly demonstrated in increasingly sophisticated studies in rodents (13,22). It is plausible to consider the possibility that the vasopressin deficiency in extra-hypothalamic brain regions contributes to disturbances in a variety of CNS functions in aged animals. Aged animals provide an interesting model for studying the effects of low levels of androgens on AVP synthesis (10,19). Our laboratory has previously shown that AVP synthesis in the BNST of young rats is a sexually dimorphic, T-dependent phenomenon (16). The finding of a decrease in AVP synthetic activity in the BNST of inherently androgen-deficient aged animals is consistent with gonadal steroid regulation of these neurons. Whether neurons in the BNST of aged animals are still capable of responding to T remains to be explored. Studies are underway to further evaluate the mechanisms by which T modulates AVP biosynthetic activity in senescence. That the interaction between vasopressin and gonadal hormones may be relevant to an understanding of CNS changes in aging is suggested by behavioral studies. For example, aged rats show impairment on conditioned taste aversion (CTA) paradigms; this impairment can be reversed by vasopressin (6). The underlying neurobiology of this observation has not yet been elucidated. If vasopressin is important in normal memory processes, it is possible that the age-related impairment in performance on the CTA paradigm may be due to a decreased synthesis of AVP in cell groups such as the BNST that project to the septum. Moreover, as CTA has been shown by our group and others to be an androgen-dependent phenomenon (1,4), the age-
related impairment on this conditioned avoidance paradigm could be attributed to the influence of low levels of circulating T on AVP synthesizing cells in the BNST of aged animals. Our observations are consistent with previously cited reports of decreased vasopressinergic immunoreactive fiber density in extra-hypothalamic brain regions in senescent rats (8) and with a report that this decrease in fiber density can be reversed by administration of T (9). However, the findings obtained by immunocytochemistry are difficult to interpret. The decrease in immunoreactivity of a neuron could be explained by decreased peptide synthesis, increased peptide degradation, or increased peptide release. Moreover, changes in the activity of neuronal systems may be detected more immediately by measuring mRNA than by measuring peptide immunoreactivity. We have recently shown that, after castration, AVP gene expression in the BNST as measured by in situ hybridization declines much more rapidly than does AVP immunoreactivity (15). Thus it appears that in situ hybridization is a more direct method than immunocytochemistry to assess the synthetic activity of the neurons in this extra-hypothalamic vasopressin system. Although in situ hybridization remains a semiquantitative technique, the number of grains per cell has been demonstrated to be proportional to the mRNA content of a cell by both Northern analysis (11,21) and solution hybridization (5). The exact mathematical nature of the relationship is unclear. A standardization process that would allow the number of silver grains over a cell to be related to the absolute number of mRNA copies has not been developed. Nonetheless, it is possible to quantify relative changes in gene expression using this technique, and so to estimate differences in vasopressin biosynthetic activity between groups in the same assay. We have used in situ hybridization to show that AVP synthesis in the BNST is diminished in aged Fischer 344 male rats. This is consistent with reports of a decrease in vasopressinergic fiber density in areas that project from the BNST in senescent Brown-Norway rats (8) and extends the observation to a strain of rat which has been more consistently used for studies of aging. In situ hybridization provides an advantage over immunocytochemical reports in that it allows the observation of a decrease in vasopressin content to be more convincingly explained as being due to a decrease in AVP synthesis. That these extra-hypothalamic neurons have been shown to be steroid sensitive in young animals suggests that this system may be a useful model in which to investigate the regulation of AVP gene expression in aging. ACKNOWLEDGEMENTS We would like to thank Ms. Cheryl Refsdal, Ms. Kimberly Donnell, and Mr. Carl Sikkema for their invaluable technical assistance. This work was supported by the Department of Veterans' Affairs, by NIH grants AG8419 and N.S. 20311, and by NARSAD.
REFERENCES 1. Brot, M. D.; Berstein, I. L.; Dorsa, D. M. Chronic central administration of vasopressin antagonist hastens extinction of conditioned taste aversion in rats. Soc. Neurosci. Abstr. 15:1069; 1989. 2. Buijs, R. M.; Swaab, D. F.; Dogterom, J.; Van Leeuwen, F. W. Intra and extra hypothalamic vasopressin and oxytocin pathways in the rat. Cell Tissue Res. 186:423-433; 1978. 3. Caffe, R.; Van Leeuwen, F. W. Vasopressin-immunoreactive cells in the dorsomedial hypothalamic region, medial amygdaloid nucleus, and locus coeruleus of the rat. Cell Tissue Res. 233:23-33; 1983. 4. Chambers, K. C. Hormonal influences on sexual dimorphism in rate of extinction of a conditioned taste aversion in rats. J. Comp. Physiol. Psychol. 90(9):851-856; 1976.
5. Chowen, J. A.; Steiner, R. A.; Clifton, D. K. Semiquantitative analysis of cellular somatostatin mRNA levels by in situ hybridization histochemistry. Methods Neurosci.; in press. 6. Cooper, R. L.; McNamara, M. C.; Thompson, W. U. Vasopressin and conditioned flavor aversion in aged rats. Neurobiol. Aging 1:53-57; 1980. 7. Dorsa, D. M.; Bottemiller, L. A. Age-related changes of vasopressin content in microdissected areas of rat brain. Brain Res. 242: 151-156; 1982. 8. Fliers, E.; deVries, G. J.; Swaab, D. F. Changes with aging in the vasopressin and oxytocin innervation of the rat brain. Brain Res. 348:1-8; 1985. 9. Goudsmit, E.; Fliers, E.; Swaab, D. F. Testosterone supplementa-
AGE-RELATED DECLINE OF VASOPRESSIN mRNA
10. 11.
12.
13. 14.
15.
16.
17.
tion restores vasopressin innervation in the senescent rat brain. Brain Res. 473:306-313; 1988. Kaler, L. W.; Neaves, W. B. The androgen status of the aging male rat. Endocrinology 108:712-719; 1981. Kelsey, J. E.; Watson, S. J.; Burke, S.; Akil, H.; Roberts, J. L. Characterization of proopiomelanocortin mRNA detected by in situ hybridization. J. Neurosci. 6(1):38; 1986. KSnig, J. F. R.; Klippel, R. A. The rat brain: A stereotaxic atlas of the forebrain and lower parts of the brain stem. New York: Krieger; 1963. Koob, G. F.; Bloom, F. E. Behavioral effects of neuropeptides: endorphins and vasopressin. Annu. Rev. Physiol. 44:571-582; 1982. Matsumoto, A. M.; Paulsen, C. A.; Hopper, B. R.; Rebar, R. W.; Bremner, W. J. Human chorionic gonadotropin and testicular function: stimulation of testosterone, testosterone precursors, and sperm production despite high estradiol levels. J. Clin. Endocrinol. Metab. 56:730; 1983. Miller, M. A.; Dorsa, D. M.; DeVries. G. J. Decline of vasopressin mRNA and immunoreactivity in the BNST following castration: a time course study. Soc. Neurosci. Abstr. 16:661; 1990. Miller, M. A.; Urban, J. H.; Dorsa, D. M. Steroid dependency of vasopressin neurons in the bed nucleus of the stria terminalis by in situ hybridization. Endocrinology 125:2335-2340; 1989. Miller, M. A.; Urban, J. H.; Dorsa, D. M. Quantification of mRNA in discrete cell groups of brain by in situ hybridization histochemis-
423
try. Methods Neurosci. 1:164-181; 1989. 18. Miller, M. A.; Zoeller, R. T.; Dorsa, D. M. Detection of vasopressin messenger RNA in cells within the bed nucleus of the stria terminalis by in situ hybridization histochemistry. Neurosci. Lett. 94:264-268; 1988. 19. Ravid, R.; Fliers, E.; Swaab, D. F.; Zurcher, C. Changes in vasopressin and testosterone in the senescent Brown-Norway (BN/BiRij) rat. Gerontology 33:87-98; 1987. 20. Rogers, K. V.; Vician, L.; Steiner, R. A.; Clifton, D. A. The effect of hypophysectomy and growth hormone administration on preprosomatostain mRNA in the periventricular nucleus of the rat hypothalamus. Endocrinology 122:586-591; 1988. 21. Sherman, T. G.; Akil, H.; Watson, S. J. Vasopressin mRNA expression: a northern and in situ hybridization analysis. In: Schrier, R., ed. Vasopressin. New York: Raven Press; 1985:475--483. 22. Strupp, B. J.; Levitsky, D. A. A mnenomic role for vasopressin: The evidence for and against. Neurosci. Biobehav. Rev. 9:399-411; 1985. 23. Urban, J. H.; Miller, M. A.; Drake, C. A.; Dorsa, D. M. Detection of vasopressin mRNA in cells of medial amygdala but not locus coeruleus by in situ hybridization. J. Chem. Neuroanat. 3:27783; 1990. 24. Young, W. S., III; Bonner, T. I.; Brann, M. R. Mesencephalic dopamine neurons regulate the expression of neuropeptide m/RNAs in the rat forebrain. Proc. Natl. Acad. Sci. USA 83:9827-9831; 1986.