306 nature neuroscience • volume 3 no 4 • april 2000

news and views

characteristic phasing is just what wouldbe expected if the horizontal sections hadexposed the M and E oscillators. This newapproach, and its attendant change in per-spective, holds the promise of understand-ing the basis of seasonal rhythms.

But have Jagoda et al. really identified theM and E oscillators? Several arguments sup-port this possibility. First, when slices weretaken from hamsters exposed to longer daysthan nights (a light: dark cycle of 14:10 ver-sus 8:16), the onset of the morning peak wasadvanced and its duration lengthened rela-tive to the evening peak; the phases of thetwo peaks are not mutually locked. Second,the two peaks respond in the expected wayto a pulse of glutamate, the neurotransmit-ter that normally conveys light signals fromthe eye to the SCN. Thus, glutamate givenafter dusk delayed the evening peak but notthe morning peak, whereas glutamate givenbefore dawn advanced the morning but notthe evening peak. Nevertheless, althoughthese findings are consistent with indepen-dent behavioral output from dual oscilla-tors, they are not sufficient to prove thecase—and indeed the authors are careful toavoid committing themselves as to whethertheir two peaks actually represent the M andE oscillators. For proof, one would like tosee distinct outputs from the two oscillators,to confirm that they can display differentperiod lengths, and to see them run acrosseach other assuming in passing all possiblephase angles. Ideally, one would also like tobe able to manipulate the two peaks indi-vidually through genetic or pharmacologicalmeans and confirm that they really are inde-pendent. To do all this in an in vitro prepa-ration whose viability is limited will bechallenging, to say the least.

There are excellent precedents in the ani-mal literature for dual separable oscillators.Neuronally coupled clocks have beendescribed in the optic lobes of the cockroach(Leucophaea maderae)8, and in the eyes ofmarine molluscs such as Aplysia or Bulla (forexample, see refs 9 and 10), and data—bothphysiological and pharmacological—con-sistent with dual oscillators in the SCN havebeen around for years (for example, refs. 1and 5). If dual oscillators are present in theSCN but haven’t been seen, then somethingeither lost or gained in the horizontal sec-tion is exposing them. Both are possibilities,because distinct subregions exist within theSCN that have distinct neuropeptide andgene expression profiles (for example, see ref.11). Moreover, connections within the SCNare complex, so the consequences of cuttingthe tissue at different angles are hard to pre-dict. For instance, the first clock - specificimpact of light in the SCN appears to be the

tors in the isolated SCN, but it does raiseinteresting and important questions, atleast some of which are now made moreexperimentally tractable. The conceptualleap in this work could drive a branch ofcircadian experiments over much of thenext decade, and one can only agree withthe authors’ understated conclusion: “Wepredict that the horizontal slice prepara-tion will prove to be a useful experimentalsystem”. Yup, it probably will.

1. Klein, D. C., Moore, R. Y. & Reppert, S. M.Suprachiasmatic Nucleus: The Mind’s Clock(Oxford, New York, 1991).

2. Welsh, D. K., Logothetis, D. E., Meister, M. &Reppert, S. M. Neuron 14, 697–706 (1995).

3. Herzog, E., Takahashi, J. & Block, G. Nat.Neurosci. 1, 708–713 (1998).

4. Earnest, D. J., Liang, F. Q., Ratcliff, M. &Cassone, V. M. Science 283, 693–695 (1999).

5. Gillette, M. U. et al. Ciba Found. Symp. 183,134–144 (1995).

6. Pittendrigh, C. S. & Daan, S. J. Comp.Physiol. A106, 333–355 (1976).

7. Jagota, A., de la Iglesia, H. O. & Schwartz, W. J.Nat. Neurosci. 3, 372–376 (2000).

8. Page, T. Science 216, 73–75 (1982).

9. Block, G. D., Geusz, M., Khalsa, S. & Michel, S.Sem. Neurosci. 7, 37–42 (1995).

10. Eskin, A. Fed. Proc. 38, 2573–2579 (1979).

11. Moore, R. Y. Prog. Brain Res. 111, 103–119 (1996).

12. Shigeyoshi, Y. et al. Cell 91, 1043–1053 (1997).

13. Lydic, T., Albers, H. E., Teppers, B. & Moore-Ede, M. C. J. Comp. Neurol. 20, 225–237 (1982).

induction of Per1 in an area of the ventro-lateral SCN that receives inputs from the reti-na12. From here, the effect of light on theclock gradually moves to the dorsolateralSCN, whose cells do not receive a directinput from the retina. It is plausible that neu-ronal connections between the paired SCNlie dorsally, in the part that is removed in thehorizontal plane of sectioning but that isretained in the coronal section (Fig. 1). Thenagain, the typical coronal section removesaspects of the caudal and rostral extent of theSCN, where yet another oscillator could lie,and the connections that might tie the mostrostral parts of the paired SCN are not fullyunderstood. Three-dimensional recon-structions have suggested that hamster SCNsare actually fused posteriorally13. Even if theyaren’t actually joined in other species, thatthey are in hamsters suggests that there arelikely to be tight connections in all species,and that these connections could be lost incoronal sections.

Unravelling all these connections with-in the circuitry of the SCN should be greatfun for all concerned, but what of thebroader significance? I believe that Jagotaet al.7 is a major step forward, in that it suc-ceeds in bridging the behaviorally-basedmodel of the E and M oscillators with cel-lular/tissue level phenomena that are like-ly to point to its anatomical andphysiological underpinnings. It doesn’tprove the existence of the E and M oscilla-

Which way, honey?

Evidence indicates that men performbetter than women at spatialnavigation tasks, such as finding theirway in an unfamiliar environment. Onpage 404, Riepe and colleagues nowestablish a neural correlate for thisgender difference, by showing thatmen and women show differentpatterns of brain activation duringnavigation. The authors asked subjectsto navigate their way out of a virtualreality maze displayed on a computerwhile their brain activity was monitored by fMRI. As anticipated from previous work,they found that men are faster than women at finding their way out of the maze.However, although some brain areas were equally activated in men and women, therewere several differences. Men showed activation of the left hippocampal region (whichwas previously shown to be involved in spatial tasks) whereas women showedactivation of the parietal and right prefrontal cortices. The significance of thisdifferential activation is less clear, but the results raise the possibility that men andwomen may be using different cognitive strategies to solve the navigation task. Now,could this have anything to do with the notorious male reluctance to ask for directions?

Kalyani Narasimhan

© 2000 Nature America Inc. • http://neurosci.nature.com©

200

0 N

atu

re A

mer

ica

Inc.

• h

ttp

://n

euro

sci.n

atu

re.c

om