Quantitative differences in the circadian rhythm of locomotor activity and vasopressin and vasoactive intestinal peptide gene expression in the suprachiasmatic nucleus of tau mutant compared to wildtype hamsters

BRAIN RESEARCH ELSEVIER

Brain Research 736 (1996) 251-259

Research report

Quantitative differences in the circadian rhythm of locomotor activity and vasopressin and vasoactive intestinal peptide gene expression in the suprachiasmatic nucleus of t a u mutant compared to wildtype hamsters Kathryn Scarbrough *, Fred W. Turek Center for Circadian Biology and Medicine, Department of Neurobiology and Physiology, Northwestern University, Evanston, 1L 60208, USA

Accepted 11 June 1996

Abstract The activity profiles of homozygous tau mutant hamsters bred in our colony exhibit several differences when compared to wildtype golden hamsters. In addition, tau mutant hamsters respond to saturating white light pulses presented between circadian time (CT) 11 and CT 16 with extremely large phase shifts (type 0 resetting) after prolonged time in constant darkness. We measured five parameters of the activity rhythm early during exposure to constant darkness (DD) (cycles 5-9), and after 44-48 cycles in DD, and we confirmed the tau mutants' unusual phase shifting response to light. Next we determined whether neurotransmitter peptide mRNA levels in the SCN differed between wildtype and tau mutant hamsters exhibiting these divergent activity patterns and responses to light. After 49 circadian cycles in DD, tau mutant hamsters responded to a 1 h light pulse at CT 15 with phase shifts averaging 10.19 ± 0.35 h. Among wildtype hamsters the mean phase shill was 1.22 ± 0.34 h and the largest phase shift observed was 3.67 h. Total wheel revolutions/circadian cycle were significantly lower in tau mutants (4022 _+ 1103) vs. wildtypes (7528 _+ 458) and there was a significant decrease in wheel-running activity after prolonged exposure to DD, particularly among the wildtype hamsters (tau = 3045 ± 972, wildtype =4362_+ 388 rev/circadian cycle). When analyzed by 5 min segments throughout the circadian cycle, the highest intensity wheel-running activity did not differ between groups and there was no significant effect of length of time in DD on this measure ( t a u = 38.5 + 6.3 and 38.4 ± 4.7 r e v / m i n , wildtype = 46.8 ± 1.7 and 41.4 + 2.7 r e v / m i n early or late in DD, respectively). The precision of activity onset differed greatly between groups with tau mutants exhibiting a much higher daily deviation from mean z (1.00 ± 0.24 h) than wildtypes (0.14 _+ .02 h). Activity onset became significantly less precise with increased time in DD: tau = 1.66 ± 0.21 h, wildtype = 0.45 ± 0.14 h after 44-48 circadian cycles. The length of the active period, a, was significantly shorter in tau mutants than in wildtypes (7.2 ± 0.2 h vs. 8.0 ± 0.2 h) but a was a similar percentage of ~- in the two groups ( t a u mutant = 36%, wildtype = 33%). After 48 circadian cycles in DD, a measured 7.2 _+ 0.5 h in tau mutants and 8.9 ± 0.6 h in wildtypes, thus there was no significant effect of time in DD on this parameter. Activity records of tau mutant animals appear more fragmented to the eye and we quantitated this with a computer-aided analysis of the number of bouts of wheel-running per active period. Wildtype hamsters exhibited 2.8 + 0.2 bouts of wheel-running activity early in DD and 3.1 _ 0.2 bouts per circadian cycle later in DD. The activity records of tau mutant hamsters were significantly more fragmented but this group actually showed some consolidation of bouts per circadian cycle after prolonged time in DD (4.7 ± 0.3 vs. 3.9 ± 0.3 bouts per cycle). Wildtype and tau mutant hamsters were killed after 66-71 cycles in DD at either CT 4 or CT 16 and in situ hybridization was performed for vasopressin (AVP) and vasoactive intestinal peptide (VIP). Levels of AVP and VIP mRNA were significantly lower in tau mutant than wildtype hamsters at CT 16. We conclude that the tau mutation causes these differences in gene expression and we speculate that differences in the peptidergic output of the clock may have some relevance for the differences in the quantitative aspects of the activity rhythm and the response to light pulses exhibited by these animals. Keywords: Circadian rhythm; Tau mutant hamster; Suprachiasmatic nucleus; Vasopressin; Vasoactive intestinal peptide; Locomotor behavior

1. Introduction

* Corresponding author. Fax: (1) kscarbro @casbah.acns.nwu.edu

(847) 467-4065;

E-mail:

T h e t a u mutation in g o l d e n hamsters segregates as a single a u t o s o m a l g e n e and causes a dramatic shortening o f the free-running period o f circadian rhythms [24]. In the h e t e r o z y g o u s state the period is 22 h and this is further

0006-8993/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved. Pll S 0 0 0 6 - 8 9 9 3 ( 9 6 ) 0 0 7 0 9 - 3

252

K. Scarbrough, F. W. Turek / Brain Research 736 (1996) 251 259

shortened to 20 h in homozygous mutants. Studies of the changes in circadian locomotor behavior of these mutants have shown the phase shifting effects of both photic and non-photic stimuli differ between tau mutant and wildtype hamsters. The amplitude of the non-photic phase response curve (PRC) is increased in homozygous tau mutants and the peak of the advance region occurs at earlier phases of the circadian cycle [16]. The PRC to light appears to be more labile in tau mutant hamsters than in wildtypes, and large phase shifts of nearly 12 circadian hours can be induced in tau mutants by a standard light pulse if the animals have been maintained for prolonged periods in constant darkness (DD) or on certain T cycles [28]. In the present study, other characteristics of the circadian rhythm of locomotor activity between wildtype and tau mutant hamsters were examined by quantitating total activity per circadian cycle, the intensity of wheel-running activity, the precision of activity onset, the length of the active period each cycle and the consolidation activity during the active period. The biological clock driving most 24 h rhythms in mammals is located in the suprachiasmatic nucleus (SCN) of the hypothalamus [12]. Studies in which fetal SCN from either tau mutant or wildtype hamsters is transplanted into lesioned hosts of the opposite genotype demonstrate unequivocally that this brain area drives system-wide circadian rhythms with the period characteristic of the transplanted fetal tissue [23]. Thus, the tau mutation acts at the level of the SCN to modify circadian behavior. Anatomical a n d / o r neurochemical differences in the SCN of wildtype and tau mutant animals that may account for the differences in circadian phenotype have not been identified. The SCN appears morphologically normal in tau mutant hamsters and several neurotransmitter peptides found in wildtype SCN are also found in tau mutant SCN [23]. While no qualitative differences within the SCN of wildtype and tau mutant hamsters have been identified, few attempts have been made to look for possible quantitative differences. We focused our attention on two prominent suprachiasmatic neurotransmitter peptides with different regional distributions within the nucleus to investigate possible quantitative differences in neuropeptide gene expression in these two groups of hamsters. Vasopressin (AVP) and vasoactive intestinal peptide (VIP) are two peptides that are synthesized by many SCN neurons and are found in synaptic specializations throughout the nucleus [4,35,36]. AVP and its cognate mRNA exhibit a robust circadian rhythm in the wildtype SCN [3,8,27,32,33,37]. Suprachiasmatic VIP mRNA exhibits rhythmicity when animals are kept in a light:dark cycle [1,39] but this rhythmicity does not persist when animals are transferred to constant conditions [11,31]. Suprachiasmatic VIP is depressed by light [29] so levels remain constitutively low during exposure to constant light or constitutively high during exposure to constant darkness (DD). The AVP rhythm may be phase shifted by light

because within the SCN, expression of this peptide and its mRNA maintain a specific phase relationship with the light:dark cycle [3,32]. Continuous light has no monotonic effect on levels of this peptide, instead, the AVP rhythm freeruns in continuous light or constant darkness [3,32]. In the studies reported here, we employed a paradigm where the behavioral response to saturating white light pulses differed between wildtype and tau mutant hamsters. We monitored locomotor activity in groups of wildtype and tau mutant hamsters and measured suprachiasmatic AVP and VIP mRNA by in situ hybridization at two phases, one during the subjective day and the other during the subjective night.

2. Materials and methods 2.1. A n i m a l s a n d r e c o r d i n g o f l o c o m o t o r actit~it3,

Male tau mutant hamsters (n = 10) were the first generation offspring from homozygous mutant adults obtained from the colony at the University of Virginia (Charlottesville, VA). Wildtype male golden hamsters ( M e s o c r i c e t u s a u r a t u s , n = 10) were purchased from Charles River (Lak:LVG/SYR). 9 - 1 2 week old wildtype and tau mutant animals were individually housed in cages with running wheels and maintained on a 14:10 light:dark cycle (wildtype) or on 11.7:8.3 light:dark cycle ( t a u mutants) in a temperature-controlled room. Food and water were available ad libitum. Wheel-running activity was continuously recorded on computer with the Chronobiology Kit (Stanford Software Systems, Stanford, CA). After stable entrainment was evident in all animals they were transferred to constant darkness (DD). Animal maintenance in DD was aided by use of an infrared viewer (Find-R-Scope, FJW Optical Industries). After 49 circadian cycles in DD, each animal received a saturating fluorescent white light pulse (200-300 lux) 1 circadian hour in length at circadian time 15 (CT 15, about 3 h after the onset of activity for wildtype hamsters and approximately 2.5 h after activity onset for tau mutant hamsters). Animals were returned to DD after the light pulse. 2.2. Tissue c o l l e c t i o n a n d in situ h y b r i d i z a t i o n

Hamsters were killed 16-21 circadian cycles after the light pulse. With the aid of an infrared viewer, each animal was removed from its cage, anesthetized briefly with CO 2, and killed in total darkness. The optic chiasm was crushed before the lights were turned on and then the brain was removed quickly from the skull and frozen on dry ice. Brains were stored at - 8 0 ° C until 10/xm sections through the SCN were prepared on a cryostat. Brain sections were stored with desiccant at - 8 0 ° C until assayed for AVP or VIP mRNA. The hybridization procedure has been previously described in detail [22]. Briefly, sections were fixed

K. Scarbrough, F.W. Turek / Brain Research 736 (1996) 251 259

in 4% paraformaldehyde, rinsed and treated with 0.25% acetic anhydride. The tissues were dehydrated through graded alcohols and delipidated with chloroform. Oligodeoxynucleotide probes were a mixture of 4 sequences, each 48-56 bases in length corresponding to conserved sequences within the AVP and VIP genes [10]. The probe mixtures were 3'-tailed with [35S]dATP (New England Nuclear, Boston, MA) and terminal deoxynucleotidyl transferase (Promega, Madison, WI). Unincorporated [35S]dATP was separated from the probe by chromatography on a NENsorb column (Dupont). Probe was diluted to 3 p m o l / m l / s e q u e n c e (3.6 x 10 6 c p m / p m o l for AVP and 8.6 × 10 6 c p m / p m o l for VIP) in standard hybridization buffer containing 50% formamide, applied to the brain sections, coverslipped with Parafilm and incubated overnight at 33°C (AVP) or 27°C (VIP). Coverslips were floated off in 1 X standard sodium citrate (SSC), slides were washed 4 times in 1 X SSC at 50°C, and 2 times for at least 1 h in 1 X SSC at room temperature. The slides were then dehydrated and air dried. Slides were dipped in NTB2 photographic emulsion, exposed at 4°C for approximately 3 weeks, developed, counterstained with 0.01% toluidine O blue and coverslipped. For each assay, sections through the SCN of each animal were anatomically matched with sections appearing in the atlas by Paxinos and Watson [17]. Other AVP cell groups labeled on the slides (supraoptic, accessory supraoptic, nucleus circularis, paraventricular) allowed fine resolution anatomical matching for this assay. The hybridization signal on a dark field image of each half of the SCN was outlined and the average gray level (0-256 arbitrary units) was measured using Image-1, version 4.14c (Universal Imaging Corp., Westchester, PA). A background reading (the average gray level of the same area over the adjacent anterior hypothalamus) was subtracted from the signal present in each half SCN. 2.3. Data analysis: activity rhythm

Two tau mutant hamsters did not receive a light pulse due to irregularities in the activity record. To estimate the phase shifts of the activity rhythm for those animals that did receive a light pulse, regression curves were eye-fitted to the onsets of activity during the last 7 - 1 0 cycles before the light pulse and projected to the day of the light pulse. Similarly, eye-fit lines through activity onsets during 10 cycles after the pulse were retrojected to the day of the light pulse and the phase shift in hours was determined as the difference between these lines. Data were converted to circadian hours for comparison of wildtypes and taus. Because many of the phase shifts exhibited by tau mutant hamsters were so large that they were difficult to categorize as either a delay or an advance, the group means are presented as an absolute value in circadian hours. The total wheel revolutions per circadian cycle and the intensity of wheel-running activity (revolutions per minute

253

within a given 5 rain bin of data) were obtained directly from Chronobiology Kit software. Data were analyzed at 2 or 3 time periods during DD by 2-way ANOVA with repeated measures. Determination of daily onset and offset of wheel-running activity was required for the calculation of the length of the active period (o~), and also to determine precision of activity onset. To determine onset and offset of locomotor activity, data for each cycle were divided into 5 rain bins of time. Onset of wheel-running activity was defined to have occurred when at least one of the following criteria was met: (A) the first 5 rain bin in a given cycle in which the hamster attained 20% of his maximal wheel-running intensity for that cycle, or (B) the time when 10% maximal intensity was followed by that level of activity in 3 out of the next 6 bins, or (C) when there was 15-30 rain of continuous activity and at least one bin was 10% of maximal intensity for that circadian cycle. Offset for a given cycle was defined as the last time 5% maximal intensity was detected. Application of these criteria defined onsets and offsets for each circadian cycle which were in good agreement with visual inspections of the actograms. Alpha ( a , the time in hours between activity onset and offset) was determined for cycles 5 - 9 in DD and again during cycles 44-48 in DD for each animal. The effect of group and time in DD was analyzed by 2-way ANOVA with repeated measures. We defined precision of activity onset as the daily deviation from each animal's mean free-running period. Specifically, we took the mean of the periods elapsed between activity onsets and calculated the standard deviation of this value. Onsets for wheel-running behavior for cycles 5 - 9 in the beginning of DD and cycles 44-48 late in DD were used to compare the precision of activity onset between wildtype and tau mutant hamsters. The data were analyzed by 2-way ANOVA with repeated measures. Division of activity into bouts was performed with the method of Slater and Lester [30] as described previously by our laboratory [20]. Briefly, we determined that an interval of rest for 15 rain or longer separated intra-bout intervals from inter-bout intervals using a log survivor function of all rest intervals over the course of several cycles from several different animals. To eliminate low amplitude noise artifacts in the records an additional requirement of wheel-running intensity of at least 4 revolutions/rain was introduced into the bout detection algorithm. These criteria was applied to several cycles from each animal and the number of bouts of activity per circadian cycle was compared both early and late in DD by 2-way ANOVA with repeated measures. 2.4. Data analysis: in situ hybridization

For technical reasons, tissues hybridized with the AVP probe had to be run in two separate assays. Sections from one animal were present in both assays and the signal from this animal was set equal to 100%. Data from every other

•aA!lao.l'qns se,~ dnoag luelntu nm oql u! slj!qs ~e,lOp "~;Aaaue^pv, jo lUatUu~!ssg "(I(1 u! Salaaa tll?!pl~aJ!36'¢ JoUe g l ID m (xnl 00£ 00g) aslnd lq~![ altqt~ q [ e UaA!~ letu!ue q,3ea &q pal!q!qxa slj!qs aseqd aql jo apnl!ugelN :tuollo~J "(. &q po~lJetu) anoq ue.!peaJ!a 1 Joj lq~'!l luaasaaonlj al!q~ Xq uo!le[ntu!lS "(1(I u! ala~3 q16~ oql uo g [ iDle, poluasoad aslnd lqg!l e aaLle pue aJojaq Jolsmeq lue,mtu m)l e jo Kl!a[lae Jolotuoao I ~u!uuna-aaaj jo (~l!ae,ia Joj z olnpom pallold alqnop) tueagolae aAIlmuasaada'd :zlpp!lN "(. &q pa>lJmu) anoq ue,!peaa!,~ I Joj lq~!l luaasaaOnLl al!q~ ~q uo!le,Intu!l S '(IQ u! aIaKa q1617aql uo gl I D l e paluasaad aslnd lq:~!l e, JOlje pue aao~aq aalstugq ad3hpl!A~ e, jo ~lIa!lae Jolotuoao I gu!uuna-aoJj jo (~l!aela Joj pallold aiqnop) umagolae oA!leluasoJdakI :do& 'I "g!d mmna~ nm

ad,qpl!.~x

_cI[[-

~-

0 c

~-; _=i.

9c-

9 6

.~.

_CI QQ u! soIOXa 6t7/ 3Olau2 ~1 Z D re asjnd ~qf~!l~noq

,,

~"

.

W

.";~.,,~.

,

'

,";.-"d

OE .~1:

.,,#..,.:

0~

"'"

00'0~

00'0~

O0'OE

3> -.<

0

00"0~

~NV£ '

.

W.,,

O~

,~

' W

"~lr '11" "

If'"

'111~1~'

'

"1[11~ I~"

ll" '

"~

!

'~r

,'r- ,

~<;~

0~

.,~,~

U3 3> -<

g

',iml~ t,

E~

VE

E~

~E 9±b4

lqg!l oql aJO]toq s a p 3 ~ 3 0 £ - g E lStl oql uo s t u t J g o p o ! J a d zX fiq p a s s a s s e s t q •0'0 T- I"e~; p o S t J a A t d n o J 8 adglPI!A~ aql p u t q £0"0 T- I'0Z; j o po.uad 8u!uunJ-aa.~j a~tJOAt Ut p t q ~pnls s!ql u! p a s n sJOlStutq l U t l n t u m)g j o dnoa~ oqSL

n m p u e odXlPI!~ Joj elep p~z!letUJouun oql j o s u o s u e d t u o 3 ul p~lt[O!a SeA~ OaUmJ~a jO £1!OU~OtUOq JO u o [ l d t u n s -sg o q i "ls~l /'/ £oul[qA,~-uutlAI ~tll q l ! ~ otu!l u t ! p t a J ! 3 q3e~ 113 p ~ J e d t u o 3 oao~, sletu[ue l u e l n t u n o l p u e ad£1pI!~

£q pol!q!qxo VN~ItU d A V j o SlOAOI "VNHtU JO SDAO I OA!lel~J Se passoJd×o oJe £oql 'pazIIeUIJOU oJaA~ el~p oql ~ 3 u ! s "ietuiue Duo g q pol!q!qxa 3~l!SUalU! ll3U~'IS ~ql ol poz! -letUaou OSle aaaa¢ £ e s s t uo!ltz!p!JqKq nl!s u! d l A aql tuoJj t l e p ' u o ! l t l u a s a J d e l t p u! &aUOlS!SUO3 aO~l 'IttUIUt p J t p u t l s oql ttloJj ltUg!S aql j o lUaaJod t s t p a s s a J d x a aJOA~ ImUiUt

sllnsaH "£

"lSal f / 3~oul!qAk-UUtlA[ a ! J l a t u t J t d u o u oql q l ! ~ p a z X l t U t OSlt aJaA~ sdnoJ~ osaql os 'VN~ltU d l A

6fig- l ffC (966l) 9 U ttaJr)as'a8 u!r)'~,Sl/ ~ta'mi "/!4"d 'qg,noaqJooS "31

~g'C

255

K. Scarbrough, F. W. Turek / Brain Research 736 ('1996)251-259

-V ~=

8

I

iiiiiiiiiiiiilt i!ii!iJi!i!t

©

6

+1 ~=

• Wildtype [] Tau

• Wildtype [] Tau

= -= "a

1.0

.~

0.5

...l...

iiiiiiil

4

x.-!

iiiiiiiiiiiiiil

2

"N

0.0

0 5-9

44 - 48

5-9

Cycles in DD

4 4 - 48

Cycles in DD

Fig. 2. Total wheel revolutions per circadian cycle early in DD (cycles 5 - 9 ) and later in DD (cycles 44-48) for tau mutant and wildtype hamsters. 2-way ANOVA with repeated measures indicates a significant effect of group (tau mutant vs. wildtype, P = 0.003) and a significant effect of the length of time in DD ( P = 0.01).

Fig. 4. The precision of daily activity onset, quantitated as the daily deviation from mean 7, early in DD (cycles 5 - 9 ) and later in DD (cycles 44-48) for wildtype and tau mutant hamsters. Tau mutant hamsters were much less precise than wildtypes ( P = 0.0001) and the effect of prolonged time in DD was also significant ( P = 0.03).

pulse. As shown previously [28], the response to a saturating light pulse after 49 cycles in DD was dramatically different in wildtype vs. tau mutant hamsters (Fig. l). Wildtype hamsters uniformly showed phase delays and the group average was 1.22 + 0.34 h. T a u mutant hamsters uniformly exhibited extremely large phase shifts which were difficult to categorize as either advances or delays. The absolute value of these shifts averaged 10.19 _+ 0.35 circadian hours. Thus, regardless of whether they were categorized as advancing or delaying phase shifts, the onset of locomotor activity moved almost halfway through the circadian cycle in response to a single light pulse presented at CT 15 in the tau mutant hamster.

Total wheel-running activity per circadian cycle was significantly lower ( P = 0.003) among tau mutant hamsters compared to wildtypes (Fig. 2). There was also a significant ( P = 0.01) effect of length of time in DD on this parameter. Inspection of the graph demonstrates that the effect of prolonged time in DD is particularly obvious among wildtype hamsters where the mean number of' wheel revolutions per cycle dropped 40% between the two timepoints studied. The differences in total wheel-running activity between wildtypes and t a u s is similar even if the difference in the length of the circadian cycles is taken into

60 e~3

•

Wildtype

,.

50

[]

Tau

'

40

-~

10

• Wildtype [] Tau

e'~

<

30 2

l0 5 - 9

4

25

44

-

48

48 Cycles in DD

Cycles in DD Fig. 3. The highest intensity wheel-running activity (rpm in a 5 rain bin) at 3 points during the free run. We found no significant effect of group and no significant effect of length of time in DD on this parameter.

Fig. 5. The length of the active period per circadian cycle (~x), early in DD (cycles 5 - 9 ) and later in DD (cycles 44-48) for wildtype and tau mutant hamsters. The effect of group was significant ( P = 0.02), and there was no significant effect of time in DD.

256

K. Scarbrough, F. W. Turek / Brain Research 736 (1996) 251-259

• Wildtype [] Tau

5

< Z

•

Wildtype

E ;>

~8 ~5

400

300-

2oo-

3 100-

az L"

2 CT 4

0 5-9

44 - 48 Cycles in DD

Fig. 6. The number of bouts of activity per circadian cycle early in DD (cycles 5-9) and later in DD (cycles 44-48) for tau mutant and wildtype hamsters. Mutant hamsters exhibited a greater number of bouts of wheelrunning activity per circadian cycle than wildtype hamsters (P = 0.0001). There was also a significant effect of time in DD on this measure ( P ~ 0.001), however a significant interaction term ( P - 0.001) suggests that the effect of time in DD depends upon group.

account; t a u mutant hamsters ran 4 0 - 5 0 % f e w e r wheel revolutions than w i l d t y p e hamsters early in D D and they p e r f o r m e d 2 0 - 3 0 % f e w e r w h e e l revolutions than wildtypes after p r o l o n g e d t i m e in DD. The highest intensity w h e e l - r u n n i n g activity in r e v o l u t i o n s / r a i n in a 5 rain bin did not differ b e t w e e n groups and there was no significant effect of length of t i m e in D D on this aspect of l o c o m o t o r b e h a v i o r (Fig. 3). T a u mutant hamsters exhibited m u c h l o w e r precision o f daily activity onset than wildtype h a m sters (Fig. 4, P = 0.0001). W e applied objective criteria (see Section 2) to d e t e r m i n e the time o f activity onset on a cycle by cycle basis and calculated the daily deviation f r o m mean r for cycles 5 - 9 in D D and again for cycles 4 4 - 4 8 . Early in D D the onset of activity in wildtype hamsters varied by less than 10 rain per cycle. In contrast, the daily deviation f r o m m e a n ~- exhibited by t a u mutant hamsters was nearly 1 h at this point. Both groups b e c a m e less precise with p r o l o n g e d time in DD, P = 0.03. Fig. 5 shows that the length of the active period, c~, was significantly shorter in t a u mutants than in wildtypes, P = 0.02. H o w e v e r , cr was a similar percentage of ~- in the two groups ( t a u mutant = 36%, wildtype = 33%). W e found no significant effect of time in D D on this parameter. The f r a g m e n t e d appearance o f the t a u mutant hamster actograms as c o m p a r e d to the w i l d t y p e was quantitated in our analysis of the n u m b e r of bouts of w h e e l - r u n n i n g activity per circadian cycle. Fig. 6 shows that wildtype hamsters exhibit about 1 / 3 f e w e r bouts of w h e e l - r u n n i n g activity per cycle than t a u mutants (effect o f group, P = 0.001). W i l d t y p e activity patterns are m o r e consolidated into longer bouts of w h e e l - r u n n i n g with little rest time.

CT 16

Fig. 7. Vasopressin gene expression in the SCN of wildtype and tau mutant hamsters. Brains were collected in DD, 16-21 circadian cycles after a saturating white light pulse. Tau mutant hamsters exhibit significantly less AVP mRNA as compared to wildtype at both CT 4 (P = 0.05) and CT 16 (P 0.008).

There was also a significant effect of time in D D on this measure ( P = 0 . 0 0 1 ) and a significant interaction term ( P = 0.001) which indicates that the effect of D D depends upon group. Indeed, whereas the m e a n n u m b e r of bouts of activity per cycle changes very little a m o n g wildtype animals, there appears to be a trend for the t a u mutant hamsters to consolidate their activity into f e w e r bouts after living in D D for a p r o l o n g e d period of time. Both groups o f hamsters exhibited a circadian rhythm in A V P m R N A , as has been reported previously in the rat [3] with high levels of m R N A detectable at C T 4 and low levels at C T 16 (Fig. 7). In general, levels of S C N A V P m R N A were noticeably l o w e r than any of the other A V P cell groups (i.e. supraoptic, accessory magnocellular) present on the same brain section (data not shown). At C T 4 w i l d t y p e animals tended to exhibit higher A V P levels than t a u mutant animals ( P = 0.05), but at C T 16 the difference

< Z

•

Wildtype

[] Tau

200

150-

lO0~O

"-d

50-

0 CT 4

CT 16

Fig. 8. V1P gene expression in the SCN of wildtype and tau mutant hamsters. Brains were collected in DD, 16-21 circadian cycles after a saturating white light pulse. Levels of VIP mRNA were not significantly different between wildtype and tau mutant hamsters at CT 4 but tau mutants exhibited significantly lower levels of VIP mRNA than wildtype hamsters at CT 16 (P = 0.009)

K. Scarbrough, F. W. Turek / Brain Research 736 (1996) 251-259

between groups was highly significant ( P = 0.008) with wildtype hamsters having approximately double the abundance of mRNA compared to t a u mutants. Based on observations made previously in the rat [31] wildtype and t a u mutant hamsters were not expected to exhibit a free-running circadian rhythm in VIP mRNA. Levels of VIP mRNA were not significantly different between wildtype and t a u mutant hamsters killed at CT 4. In contrast, VIP mRNA at CT 16 was significantly lower among t a u mutant hamsters ( P = 0.009) than wildtypes (Fig. 8). These data suggest the possibility of a circadian rhythm in VIP mRNA in t a u mutant hamsters, however this experiment was designed to compare wildtypes and t a u s and not specifically to study that question.

4. Discussion

The major findings reported here are (1) a number of parameters of the activity rhythm differ between t a u mutant and wildtype hamsters and (2) the differences in the pattern of locomotor activity and the response to light pulses between the two groups of hamsters correlate with differences in the level of suprachiasmatic mRNA encoding AVP and VIP. Wildtype hamsters have been shown previously to perform fewer wheel revolutions per day, to exhibit greater lability in activity onset and to have more fragmented actograms under conditions of short photoperiod when compared to animals maintained on long days [9]. Activity records from the wildtype group in the present study showed similar changes with prolonged time in DD. However, relative to wildtype hamsters, t a u mutants generally exhibited decreased total wheel-running activity, greater lability in activity onset, and more fragmented activity records, even early in DD. Activity onset of both wildtype and t a u mutants became less precise with prolonged time in DD but activity onset within the t a u mutant group was vastly more labile at both time points studied. Among the wildtype hamsters the records of individual animals did appear somewhat more fragmented after 48 cycles in DD but the group average number of bouts per cycle showed only a small tendency to increase over the time interval studied. These data are not at variance with observations made previously [9] because in the study cited above, the animals were held in short photoperiod for at least twice as long as the animals were in DD in the current study. The t a u mutant hamsters records, in contrast, were much more fragmented at the start and the activity period actually became somewhat more consolidated after a prolonged time in DD. The significance of this observation is unknown but two other observations regarding the regulation of stability of activity onset and fragmentation of the activity rhythm may be relevant. First, decreased activity, lability in activity onset and

257

fragmentation of the active period have been correlated with the decreasing levels of testosterone characteristic of male hamsters undergoing testicular regression in response to short photoperiods. When the testes of hamsters maintained on inhibitory photoperiods begin to spontaneously recrudesce there is a concomitant consolidation of bouts of activity and an increased stability in activity onset [9]. Testicular recrudescence usually takes approximately 180 cycles to occur in wildtype animals, a much longer time than the duration of this study. Regulation of the precision of activity onset and fragmentation of the active period diverged in the t a u mutant hamsters in this study because mutants showed some consolidation of activity bouts after 44-48 cycles in DD but the precision of activity onset continued to decline with prolonged time in DD. Therefore, the present observations are not fully consistent with fast regression and recrudescence but they do raise the question of whether the testes of wildtype and t a u mutant hamsters used in this study regressed at similar rates. We did not measure testosterone levels in the two groups of hamsters but we determined paired testes weights at the end of the study. All mutant hamsters were fully regressed (the group mean was 0.23 _+ 0.02 g) after approximately 70 cycles in DD. Interestingly, six out of 10 wildtype hamsters had paired testes weights greater than 1 g and the group average was 1.97 + 0.44 g. Apparently, the light pulses which interrupted otherwise constant darkness had different effects on the reproductive system of homozygous t a u vs. wildtype hamsters. This was surprising since both groups responded to the light pulse with robust phase shifts and since the rate of regression in DD has been reported to be the same in heterozygous t a u mutants and wildtype hamsters [14]. In any case, the consolidation of bouts of activity in the mutant group late in DD cannot be explained by testicular recrudescence. The second observation relevant to stability of activity onset and fragmentation of the activity rhythm involves the role of the monoamine afferent input to the SCN. We have shown previously that monoamine depletion via a single injection of reserpine causes increased lability in activity onset and increased fragmentation of the locomotor activity rhythm [18]. A more selective serotonin neurotoxin, p-chloroamphetamine, does not appear to have the same effects on these parameters of the activity rhythm (see activity records presented in [19]), suggesting that norepinephrine, epinephrine a n d / o r dopamine may influence the stability of activity onset and the consolidation of the active period. Noradrenergic afferents terminate in a capsule surrounding the SCN and probably make synaptic contact with the dendritic tree of S CN neurons[ 15]. Whether or not the t a u mutation alters catecholaminergic neurotransmission is not known. The decrease in total activity, lack of precision in activity onset, fragmentation of the active period and the propensity for homozygous t a u mutant hamsters to generate large phase shifts is reminiscent of activity patterns

258

K. Searbrough, F. W. Turek / Brain Research 736 (1996) 251-259

exhibited by old-aged wildtype hamsters [20,26,38]. Extending the analogy further, it is interesting to note that previous studies in the rat have demonstrated an age-related decrease in suprachiasmatic AVP and VIP immunoreactivity [5,25]. We speculate that the decreased levels of AVP and VIP mRNA found in this study are somehow related to the different characteristics of the overt activity rhythm exhibited by wildtype and t a u mutant hamsters. Shimomura and Menaker have suggested that the coupling between individual oscillators of the circadian clock may be weakened by the t a u mutation [28]. A change in coupling could affect the free-running period as well as change the phase shifting response [6,21]. AVP may be a good candidate for a role in coupling individual suprachiasmatic oscillators. AVP-containing neurons make synapses within the nucleus [7,34] and cross the midline to connect the two nuclei [35]. Unlike suprachiasmatic VIP or GRP (gastrin releasing peptide)[29], AVP does not respond to light with either tonic suppression or tonic stimulation but instead its rhythm freeruns in constant light [32]. Previous studies aimed at elucidating the role of AVP in the circadian system by employing peptide infusions or receptor antagonists [2,13] have demonstrated that AVP may have some small effect on wakefulness and arousal. Since the t a u mutant hamster provides an animal model where the circadian response to light pulses can be manipulated in a predictable manner, correlating changes in the abundance of AVP and other suprachiasmatic peptide mRNAs with these circadian responses may help tease out the role of these peptides in the regulation of circadian outputs. On the other hand, VIP does appear to be light responsive [29], these neurons make synapses intrinsic to the nucleus [7] as well as project out of the SCN [36] and may function as a neurotransmitter which broadcasts the circadian clock's response to light to other parts of the SCN (i.e. the dorsomedial SCN containing AVP neurons) as well as to other brain regions controlling metabolic and behavioral rhythms. It seems reasonable to think that the low levels of VIP mRNA found at CT 16 may somehow be related to the t a u mutants' aberrant locomotor response to light at CT 15. The issue of a direct response to light was not addressed in the experiments described here so whether VIP mRNA is photoinhibited in wildtype and t a u mutant hamsters in a similar manner as that seen in rats [29], remains to be determined. In conclusion, the present results demonstrate differences between wildtype and t a u mutant hamsters in the pattern of locomotor activity and in biochemistry at the level of the SCN. Differences in gene expression of two major suprachiasmatic neurotransmitter peptides may, by an as yet unknown mechanism, contribute to the differences in overt circadian rhythmicity exhibited by t a u mutant when compared to wildtype hamsters.

Acknowledgements The authors wish to thank Michael Menaker for generously providing homozygous t a u mutant hamsters and Terrance Pyles and Yi-Rong Ge for technical help with this project. This work was supported by AG-10870 to F.W.T.

References [1] Albers, H.E., Stopa, E.G., Zoeller, R., Kauer, J., King, J., Fink, J., Mobtaker, H. and Wolfe, H., Day,night variation in prepro vasoactive intestinal peptide/peptide histidine isoleucine mRNA within the rat suprachiasmatic nucleus, Mol. Brain Res., 7 (1990) 85-89. [2] Arnauld, E., Bibene, V., Meynard, J., Rodriguez, F. and Vincent, J., Effects of chronic icy infusion of vasopressin on sleep-waking cycle of rats, Am. J. Physiol.. 256 (1989) R674-R684. [3] Cagampang, F., Yang, J., Nakayama, Y., Fukuhara, C. and Inouye, S., Circadian variation of arginine-vasopressin messenger RNA in the rat suprachiasmatic nucleus, Mol. Brain Res., 24 (1994) 179-184. [4] Card, J., Brecha, N., Karten, H. and Moore, R., ]mmunocytochemical localization of vasoactive intestinal polypeptide-containing cells and processes in the suprachiasmatic nucleus of the rat: light and electron microscopic analysis, J. Neurosci., 1 (1981) 1289-1303. [5] Chee, C., Roozendaal, B., Swaab, D., Goudsmit, E. and Mirmiran+ M., Vasoactive Intestinal Polypeptide Neuron Changes In The Senile Rat Suprachiasmatic Nucleus, Neurobiol. Aging, 9 (1988) 307-312. [6] Daan, S. and Berde, C., Two coupled oscillators: simulations of the circadian pacemaker in mammalian activity rhythms, ,L Theor. Biol.. 70 (1978) 297-313. [7] Daikoku, S., Hisano, S. and Kagotani, Y.. Neuronal associations in the rat suprachiasmatic nucleus demonstrated by immunoelectron microscopy, J. Comp. Neurol., 325 (1992) 559 571. [8] Earnest, D. and Sladek, C., Circadian rhythms of vasopressin release from individual rat suprachiasmatic explants in vitro, Brain Res., 382 (1986) 129-133. [9] Ellis, G. and Turek, F., Changes in locomotor activity associated with the photoperiodic response of the testes in male golden hamsters, J. Comp. Physiol., 132 (1979) 277-284. [10] Genetics Computer Group, Program Manual j~)r the GCG Package, Genetics Computer Group, Madison. W1, 1991. [11] lnouye, S., Shinohara, K., Tominaga, K., Takeuchi, J., Nagasaki, H., lsobe, Y., Fukuhara, C., Otori, Y., Yang, J., Cagampang, F., Yamazaki. S. and Tokumasu, A., Circadian rhythms in peptides and their precursor messenger RNAs in the suprachiasmatic nucleus. In: H. Nakagawa, Y. Oomura and K. Nagai (Eds.), New Functional Aspects ~]" the Supruchiasmatic Nucleus r~]"the 14vpothalamus, John Libbey and Company, New York, 1993, pp+ 219-233. [12] Klein, D., Moore, R. and Reppert, S., Suprachiasmatic Nucleus: The Mind's Clock, Oxford University Press, New York, 1991. [13] Kruisbrink, J., Mirmiran, M., Van der Woude, T. and Boer, G., Effects of enhanced cerebrospinal fluid levels of vasopressin, vasopressin antagonist or vasoactive intestinal polypeptide on circadian sleep-wake rhythm in the rat, Brain Res., 419 (1987) 76 86. [14] Menaker, M. and Refinetti, R., The tau mutation in golden hamsters. In: M.W. Young (Ed.), Molecular Genetics ~[" Biological Rhythms, Marcel Dekker, New York, 1993, pp. 255 269. [15] Moore, R., Organization and function of a central nervous system circadian oscillator: the suprachiasmatic hypothalamic nucleus, Fed. Proc., 42 (1983) 2783-2789. [16] Mrosovsky, N+. Salmon. P.A., Menaker, M. and Ralph, M.R.,

K. Scarbrough, F.W. Turek / Brain Research 736 (1996) 251-259

[17] [18]

[19]

[20]

[21]

[22]

[23]

[24] [25]

[26] [27] [28] [29]

Nonphotic phase shifting in hamster clock mutants, J. BioL Rhythms, 7 (1992) 41-49. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego, CA, 1986. Penev, P., Turek, F. and Zee, P., Monoamine depletion alters entrainment and the response to light of the circadian activity rhythm in hamsters, Brain Res., 612 (1993) 156-164. Penev, P., Turek, F. and Zee, P., A serotonin neurotoxin attenuates the phase-shifting effects of triazolam on the circadian clock in hamsters, Brain Res., 669 (1995) 207-216. Penev, P., Zee, P. and Turek, F.W., Quantitative analysis of age-related fragmentation in the circadian pattern of hamster locomotor activity, Soc. Res. Biol. Rhythms, (Abstract) (1996). Pittendrigh, C. and Daan, S., A functional analysis of circadian pacemakers in nocturnal rodents. V. Pacemaker structure: a clock for all seasons, Z Comp. Physiol., 106 (1976) 333-355. Porkka-Heiskanen, T., Urban, J., Turek, F. and Levine, J., Gene expression in a subpopulation of luteinizing hormone-releasing hormone (LHRH) neurons prior to the preovulatory gonadotropin surge, J. Neurosci., 14 (1994) 5548-5558. Ralph, M., Foster, R., Davis, F. and Menaker, M., Transplanted suprachiasmatic nucleus determines circadian period, Science, 247 (1990) 975-978. Ralph, M.R. and Menaker, M. A mutation of the circadian system in golden hamsters, Science, 241 (1988) 1225-1227, Roozendaal, B., van Gool, W., Swaab, D., Hoogendijk, J. and Mirmiran, M., Changes in vasopressin cells of the rat suprachiasmatic nucleus with aging, Brain Res., 409 (1987) 259-264. Rosenberg, R., Zee, P. and Turek, F., Phase response curves to light in young and old hamsters, Am. J, Physiol., 261 (1991) R491 -R495. Schwartz, W., Coleman, R. and Reppert, S., A daily vasopressin rhythm in rat cerebrospinal fluid, Brain Res., 263 (1983) 105-112. Shimomura, K. and Menaker, M., Light-induced phase shifts in tau mutant hamsters, J. Biol. Rhythms, 9 (1994) 97-110. Shinohara, K., Tominaga, K., Isobe, Y. and Inouye, S., Photic regulation of peptides located in the ventrolateral subdivision of the suprachiasmatic nucleus of the rat: Daily variations of vasoactive intestinal polypeptide, gastrin-releasing peptide, and neuropeptide Y, J. Neurosci., 13 (1993) 793-800.

259

[30] Slater, P. and Lester, N., Minimising errors in splitting behaviour into bouts, BehaL,iour, 79 (1982) 153-162. [31] Takeuchi, J., Nagasaki, H., Shinohara, K. and lnouye, S., A circadian rhythm of somatostatin messenger RNA levels, but not of vasoactive intestinal polypeptide/peptide histidine isoleucine messenger RNA levels in rat suprachiasmatic nucleus, Mol. Cell. Neurosci., 3 (1992) 29-35. [32] Tominaga, K., Shinohara, K., Otori, Y., Fukuhara, C. and Inouye, S., Circadian rhythms of vasopressin content in the suprachiasmatic nucleus of the rat, NeuroReport, 3 (1992) 809-812. [33] Uhl, G. and Reppert, S., Suprachiasmatic nucleus vasopressin messenger RNA: Circadian variation in normal and Brattleboro rats, Science, 232 (1986) 390-393. [34] van den Pol, A. and Gorcs, T., Synaptic relationships between neurons containing vasopressin, gastrin-releasing peptide, vasoactive intestinal polypeptide, and glutamate decarboxylase immunoreactivity in the suprachiasmatic nucleus: dual ultrastructural immunocytochemistry with god-substituted silver peroxidase, J. Comp. Neurol., 252 (1986) 507-521. [35] van den Pol, A. and Tsujimoto, K., Neurotransmitters of the hypothalamic suprachiasmatic nucleus: immunocytochemical analysis of 25 neuronal antigens, Neuroscience, 15 (1985) 1049-1086. [36] Watts, A. and Swanson, L., Efferent projections of the suprachiasmatic nucleus: II. Studies using retrograde transport of fluorescent dyes and simultaneous peptide immunohistochemistry in the rat, J. Comp. Neurol., 258 (1987) 230-252. [37] Yamase, K., Takahashi, S., Nomura, K., Haruta, K. and Kawashima, S., Circadian changes in arginine vasopressin level in the suprachiasmatic nuclei in the rat, Neurosci. Lett., 130 (1991) 255-258. [38] Zee, P., Rosenberg, R. and Turek, F., Effects of aging on entrainment and rate of resynchronization of circadian locomotor activity, Am. J. Physiol., 263 (1992) R1099-R1103. [39] Zoeller, R., Broyles, B., Earley, J., Er, A. and Albers, H., Cellular levels of messenger ribonucleic acids encoding vasoactive intestinal peptide and gastrin-releasing peptide in neurons of the suprachiasmatic nucleus exhibit distinct 24-hour rhythms, J. Neuroendocrinol., 4 (1991) 119-124.