Neuroscience Letters 180 (1994) 91-95
Changes in c-fos mRNA in the neonatal rat brain following hypoxic ischemia Ulrika
~d6n a, Elsa Bona b, Henrik Hagberg b'c, Bertil
B. F r e d h o l m a'*
~Department of Physiology and Pharmacology, Section of Molecular Neuropharmacology, Karolinska Institute, Sol 71 77 Stockholm, Sweden bDepartment of Anatomy and Cell Biology, CDepartmentof Obstetrics and Gynecology, University of G6teborg, G6teborg, Sweden Received 23 February 1994; Revised version received22 August 1994; Accepted 29 August 1994
We used quantitative in situ hybridization to study changes in the expression of c-fos following hypoxic-ischemia (H-I) in the neonatal rat brain. 7-day-old rat pups were subjected to a unilateral ligation of the common carotid artery followed by a 2 h 15 min hypoxic period (7.7% 02 in N2). This resulted in the expected ipsilateral infarction of cortex, lateral hippocampus, lateral-superior aspects of the striatum and the white matter of the corpus callosum. Brain damage was not seen in the contralateral hemisphere subjected only to hypoxia, c-fos mRNA levels increased in the contralateral hemisphere immediately after the hypoxia and had returned towards normal levels 2 h thereafter. In the ipsilateral hemisphere, the expression of c-fos was delayed but very marked at 2 h. Animals subjected only to hypoxia showed little or no increase in c-fos mRNA. Thus the earliest recorded increase in c-fos after hypoxic ischemia, which occurred on the non-ischemic, contralateral side, may represent a generalized response to a more localized insult. Key words." Infarction; Immediate early gene; In situ hybridization; Neonatal rat
Brain damage induced by neonatal hypoxic-ischemia (H-I) can have long-term neurological sequelae in term and pre-term infants [10,25,27]. The mechanisms that lead to damage are largely unknown but neurotransmitters [11,15,21], calcium [20,23], carbohydrate metabolism , and oxygen free radicals  appear to be involved in the pathogenetic process. Following pioneering studies  measurements of Fos protein or c-fos m R N A have been used to map neurons that are activated by a variety of experimental treatments. The c-fos changes appear to be related to seizures , burst activity  or spreading depression and trauma  and the changes are often attenuated by EAA antagonists [6,9]. Expression of the c-fos gene may thus reflect activation of neuronal pathways that are also implicated in neuronal cell damage after hypoxia and/or ischemia. Although the bulk of the data obtained relate to adult animals, there is some information also from immature animals. The c-fos gene is activated after neo-
*Corresponding author. Fax: (46) (8) 33 16 53. 0304-3940/94l$7.00© 1994 Elsevier ScienceIreland Ltd. All rights reserved S S D I 0304-3940(94)00655-5
natal H - I [3,8]. In the present study we therefore compared the regional distribution of m R N A encoding c-fos in a model of hypoxic ischemia (H-I) in 7-day-old rats . The early changes revealed an unexpected pattern of c-fos induction in that it started on the non-ischemic side of the brain. Twenty-five 7-day-old rat pups of either sex (inbred Wistar F strain) from 3 litters were used. The rat pups were housed at 24°C in humidified air and fed by their dams until the start of the experiments. All rat pups were subjected to H - I [2,17] as follows: anesthesia was induced with halothane (2%) and maintained with 0.6% halothane in O//nitrous oxide (30/70). The left common carotid artery was cut between ligatures of prolene suture (9-0). Animals were allowed to recover from anesthesia for at least 2 h. H - I was induced by exposure to 7.7 + 0.01% oxygen in nitrogen for 2 h 15 min in a humidified chamber at 36°C. Rats were sacrificed 0 h (n = 4), 1 h (n = 5) or 2 h (n = 5) after H - I for measurement of c-fos m R N A and adenosine receptors (see below). Control rats (n = 5) were ligated but not subjected to hypoxia.
~ ,IU&l,/ J/ i \,t) 1 9 7 4 9 / 9 .
Fig. I. H 1 brain damage was seen in the cerebral cortex and striaimn of anterior sections (a) and cerebral cortex, strialum and hippocampus of posterior sections (b).
Two pups from each litter were allowed to survive for 14 days for histological evaluation of brain damage. They were perfusion-fixed with 4% formol saline in a 0.1 M phosphate buffer through the ascending aorta. The animal was stored at 4°C for 24 h before the brain was removed. The brains were cut into 2 mm thick coronal slices starting 2 3 mm from the frontal pole of the intact hemisphere. The specimens were embedded in paraffin and sectioned. The sections were stained with hematoxylin-eosine. For in situ hybridization, coronal sections were cut with a Leitz cryostat at the following levels +8 mm, +5.9 mm, +4.7 ram, +3.5 mm, +2.6 mm, +2.0 mm, +1.2 mm, and +0.8 mm from bregma  (anterior to posterior). 14 /am thick sections were thaw-mounted on polyL-lysine (50/ag/ml)-coated slides for in situ hybridization. The in situ hybridization for c-/bs, using a 48-mer c-~)s probe complementary to rat amino acids 137--152 of Fos was performed as described earlier . After hybridization for 16 h at 42°C the sections were apposed to Hyperfilm fl-max from Amersham for 1 4 weeks. The results from the quantitative in situ hybridization were analyzed by multivariate ANOVA using the procedures in the SYSTAT program. Where appropriate, the
F-values are given. Statistical hypotheses were considered significant if P < 0.05. All brains subjected to histopathological evaluation exhibited severe cerebral damage in the hemisphere ipsilateral to ligation (Fig. 1), in agreement with previous data using this method [2,17]. There was a low expression of c:fbs in untreated control animals (Fig. 2), but immediately after the hypoxic period there was a small increase in the hemisphere contralateral to the ligation (Fig. 2). One hour after the hypoxic period, there was a small increase on the ipsilateral side and a marked increase on the contralateral side. After 2 h the level of cqbs on the contralateral, non-inlitrcted side returned towards control whereas the increase in the ipsilateral hemisphere was very marked (Figs. 2 and 3), but the level was only some 70% of the peak level on the contralateral side (P < 0.001 ). The effects were similar throughout cortex from front to back (Fig, 2) and in hippocampus. It is clear that those areas of the cerebral cortex that would eventually show damage were those that showed the lowest levels of c-fi)s in the ipsilateral H I hemisphere after 1 h (compare Fig. l b and Fig. 3). There were similar changes also in other brain regions, which were quantitated in hippocampus (CA1, CA3 essentially pyramidal cell layer and dentate gyrus, essentially granule cell layer) and striatum (see legend to Fig. 3). In agreement with previous reports , the cqbs expression in ligated animals, not subjected to hypoxia, did not differ from that seen in completely untreated animals (data not shown). Importantly, when sham-ligated animals were exposed to hypoxia (2 h 15 lnin) there was a very small increase of c-lbs from 0.073 +_ 0.035 to 0.103 + 0.027 arbitrary OD units I h alter the hypoxia. This small 39% increase is thus almost 10-fold lower than that found on the contralateral side in the ligated animals at the same time point. Hypoxia alone had no effect at all (not shown). Previous studies in the adult have demonstrated a rapid increase in c-los mRNA, Fos-immunoreactivity or AP-1 transcriptional activity on the side of the lesion after unilateral focal ischemia [1,24]. We therefore expected the expression of c-lbs to be particularly pronounced in the part of the brain ipsilateral to the ligation of the carotid artery. However, the expression of c-los m R N A was first increased in the contralateral hemisphere where a clearcut increase was observed already at the end of the hypoxic period. In this hemisphere a further increase was observed 1 h after the insult, but at 2 h after the insult the c-Jos levels approached control. Expression of c-/bs appeared later in the hemisphere ipsilateral to the ligation of the carotid artery (denoted H- I) with a significant rise 2 h after hypoxic exposure. In situ hybridization  and Northern blot techniques  revealed a similar ipsilateral activation of the c-/bs gene 2 h after H I in 7-day-old rat pups. In these studies, however, the early contralateral rise was not examined,
U. ,4d~n et al,/Neuroscience Letters 180 (1994) 91 95
well-known decrease in protein synthesis following hypoxia or ischemia . This could explain why no decrease in c-fos m R N A was observed in the present study. The early c-fos increase even in the H - I hemisphere does not exactly correspond to the area of eventual brain damage. On the contrary, c-Jos expression was absent in regions with severe infarction in adult animals [1,24]. This is the case also in the present series of experiments where c-fos was most heavily expressed in non-injured tissue (Figs. 2 and 3). The situation may be different if c-fos changes are followed for longer periods than done
1 hr post
2 hr post
•'+t- m,,.,., m,,._, - i m~,u-,,.,
+,41= ++ 1 1111 Control
t hr post
2 hr post
Fig. 2. Expression of c-fos m R N A in cortex at the levels shown. The location of the areas measured are shown in Fig 3b. Note that the increase o f c-los m R N A was similar throughout cortex at all measured levels. The results from the statistical evaluation are given below as F-ratios and p-values - using P < 0.05 (*), 0.01 (**) and 0.001 (***) as standard levels. The interaction term demonstrates that the timedependent changes are significantly different between the two sides. Section no.
Level ( m m anterior to Bregma)
1. 2. 3. 4. 5. 6. 7. 8.
8.0 5.9 4.7 3.5 2.6 2.0 1.2 0.8
17.2"** 24.3*** 14.7'** 13.6"** 12.3"** 11.7"** 10.8"** 12.4"**
2.8 5.4* 5.4* 4.0* 4.1" 3.9 5.0* 5.8*
14.7"** 31.7"** 14.2"** 10.6"** 12.6"** 10.5"** 7.2*** 13.0"**
possibly because they used a longer period of hypoxia (3 h). Interestingly, in older rats (21-days) a marked increase in Fos protein was observed in the contralateral hemisphere soon after hypoxia , whereas a decrease in Fos protein was often observed in the H-I hemisphere . The decrease in Fos protein might depend on the
(c) Fig. 3. Expression ofc-fos m R N A at section 6 (as described in Fig. 2). In situ hybridization was performed with the labelled oligonucleotide specific for c-los m R N A . (a) Untreated animals. (b) One hour after hypoxia a general c-fos activation occurred, which was more marked in the contralateral (right) hemisphere. (c) Two hours after H 1, the c-Jos activation on the contralateral side (see Fig. lb) was less pronounced whereas the ipsilateral side was activated. The areas measured are indicated in panel b. The time-dependent changes, as well as the difference in these changes between sides, were highly significant (P << 0.001 ) in both the CA1 and the CA3 regions as well as the dentate gyrus of the hippocampus (all measured in the section shown) and in the striatum (measured in section no. 4).
td'Cll ~/ ~11.:,\cino.~(lctl~c Lcllcr.~ ]h?i
in the present and most previous studies. Thus, it has been shown that a secondary, prolonged c:lbs expression could be a halhnark of terminal differentiation and precede programmed cell death . The rise in c-lbs expression on the contralateral side is not simply due to hypoxia, nor is it due simply to the ligation of the x,essels to the other side. Thus, the stimulus for the contralateral increase in Cribs mRNA appears to be the same as that required to induce damage on the ipsilateral side. The question is how the effects on the side of hypoxic ischemia spread to the other side. The increase of c-lb,s mRNA or Fos protein in neurons in the contralateral hemisphere might depend on seizure activity, since both the N M D A receptor antagonist MK-80t and carbamazepine produced a parallel decrease in the contralateral rise of Fos protein and in seizures . However, m contrast to the 21-day-old animals used by these authors, there were no overt seizures in 7-day-old rats subjected to H 1, but EEG recordings demonstrate burst-activity . Burst activity can also lead to reduction of crlbs . Alternatively, wide-spread activation of crib,s mRNA may be a consequence of spreading depression [6,28] seen in the penumbra after focal ischemia in adults . However, it is uncertain whether spreading depression occurs m neonatal rats . It was also notable that the rise m c:/i~,s, expression occurred sooner on the contralateral than on the ipsilateral side. We do not know the reason l\)r this. If we assume that the damaging insult on the ischemic side of the brain induces a generalized stimulus that causcs an increase in cs'b.s' expression, then the ischemia must be assumed to delay the expression. In principle the reason for this could be either that the decrease in blood flow reduces the magnitude of the stimulus on the ipsilateral side or conversely that the transcription of c-/us m R N A is reduced when blood flow is low. Whatever the mechanism the important aspect of the present work is that a restricted insult causes a generalized effect of such a magnitude that immediate early genes are induced even in areas of the brain that apparently suffer no adverse consequences over the long-term. This work was supported by the Swedish Medical Research Council (9455, 2553), Seen Jerring foundation, The 1987 Foundation for Strokeresearch, Loo and Hans Ostermans foundation, ,~ke Wiberg l\~undation, /~hlen foundation, Magnus Bergwall foundation, Konung Gastat V's 80firs lk)undation, Frimurare Barnhus foundation and First May Flower Annual Campaign.  An, G.. Lin. T.N.. Liu, J.S.. Xuc, ,I.J., He, Y.Y. and Hsu, C.Y., Expression of c:l~xs and crjm~ famil} genes aftcr focal cerebral ischemia, Ann. Ncurol., 33 (1993) 457 464.  Andind. P.. Thordstein. M.. Kjellmer. I., Nordberg, C., Thiringer, K., Wennberg. E. and Hagberg, H., Evaluation of brain damage in ~ 11~l[ model of neonatal hspoxic ischemia. J. Neurosei. Methods. 35 (19901 253 2(o.
I9~)4, 91 (J.'~
Regional expression o1 c:/;~ and heat shock protcin-70 mRNA folio\ring hypoxia ischemia m imnwturc rat brain. J. ("erc'b. Blood Ih~\~ Mctab.. 12 (1992)987 99~ ( h e n . R.. Sib, crslem, l.S.. ,.\ldrigc, J.W. and Johnston. M.V, t.~ olulion ol'aCtlte cortical EE(J changes in experhnental perinatal hM~oxic ischcmic cncephalopalh>. Neurology, 36 Suppl. I (1986} ~'~ ,";7. I)ouglas. R.M., l)ragunov~, Ms and Robertson, lhA.. High frequency discharge of dentate granule cells, but not long-lcrm potentiation, induces c-l\~s protein. Mol. Brain Res., 4 (1988) 259 262. Dragunow, M., Goulding. M.. Faull. R.L.M., Ralph, R., Mee. E. and Frith, R., Induction of c:/i,s mRNA and protein in neurons and glia after traumatic brain injury: pharmacological characterization, Exp. Neurol., 107 (1990) 236 248. Dwyer, B.E., Nishimura, R.N,, Powell, C.L. and Mailheau, S.L.. Focal protein synthesis inhibition in a model of neonatal hypoxie ischemic brain injury, Exp. Ncuroh. 95 (1987) 277 28tL Gubits. ll.M.. Burke, R.E., Casey-Mclntosh, G., Bandele, A. and Munell. F., Immediatc earl~ gent induction atier nconatal hvpoxiu ischemia, Mol. Bruin Res., 18 il993)228 23,";. (jl.lnn. A..I., Draguil,ow, M., Faull, R.L. aLl.d Gluckman, P.I)., F,Ifeets of hypoxia-ischemia and seizures on neuronal and glial-likc c-M~ protein levels m lhc infant rat. Brain Res.. 531 119901 105 I1(,.
[101 ttagbcrg, B. and Hagberg, G.. The changing panorama of inlimtilc h.',droccphalus and cerebral palsy over forty years. A Swedish survey, Brain Dee., II (1989) 368 373. [I I] Hagbcrg, H., Hypoxic-ischemic damage in the neonatal brain: excitalor} amino acids, De\. Pharmacoh Ther., 18 (1992) 139
144.  Johansson, B., Lindstr6m, K. and Fredholm, B.B., Differences m Ihe regional and cellular localization of c-lbs m R N A induced by amphetamine, cocaine and caffeine in the rat, Neuroscience. 59 (1994) 837 849.  Kjelhner, 1., Mechanisms of perinatal brain damage, Ann. Med., 23 (1991) 675 679.  Mares. U.,Kriz, N.. Bruzck. G. andBurcs, J . . A n o x i c c h a n g e s o f extracellular potassium concentration m the cerebral cortex of young rats, Exp. Neurobiol., 53 (1976) 12 20.  McDonald, J.W. and Johnston, M.V., Physiological and pathophysiological roles of excitaloD amino acids during central ncrxous system development, Brain Res. Rev., 15 (1990)41 70. [161 Nedcrgaard. M. and Aslrup, J., Infarct rim: effect of hyperglycclnia on current potential and ['aC]2-deoxyglucose phosphorylalion. J. Ccreb. Blood Flow Metab.. 6 (1986)607 615. [I 7] Rice..I.E., Vannucci. R.C. alld Briefly. J.B., lnfluencc of immaturity on hypoxic ischemic brain damage in the rat, Ann. NeullOh, tl 11981) 131 141. [181 Sager, S.M., Sharp, F.R. and Curran, T.. Expression of cdk~,,, protein in brain: metabolic mapping at the cellular le;ei. Science. 240 11988) 1328 1331. I19] Sherwood. N.M. and Timiras. P.S., A stereotaxic atlas of the dc~eloping ral brain, l. lnixersity of Calili)rnia Press, Berkeley, t 970 [201 Siesj6, B.K., Mechanisms o1" ischemic brain damage, Crit. ('arc Med., 16 (1988) 954 963.  Sil~erstein, F. and Johnston, M.V., Effects of" hypoxia ischemia on monoamine metabolism in the immature brain. Anna. NenllOl.. 15 11984) 342 347.  Smeyne, R.J., Vendrell, M., Hayward, M., Baker, N.J., Miao. G.G., Schilling, K., Robertson, L.M., Curran, T. and Morgan, J.l.. Continuous c-fos expression precedes programmed cell death m \leo, Nature, 363(1993) 166 169.  Stem, D.T. and Vannucci, R.C., Calcium accumulation during the evolution of hypoxic dschemic brain damage in the immature rat. I. Cereb. Blood Flow Metab.. 8 (1988) 834 842,
U. AdOn et al./Neuroscience Letters 180 (1994) 91-95
 Uemura, Y., Kowall, N.W. and Moskowitz, M.A., Focal ischemia in rats causes time-dependent expression of c-fos protein immunoreactivity in widespread regions of ipsilateral cortex, Brain Res., 552 (1991) 99-105.  Vannucci, R.C., Acute perinatal brain injury: hypoxia-ischemia. In W.R. Cohen, D.B. Acker and E.A. Friesman (Eds.), Management of Labor, Aspen Publishers, Rockville, 1989, pp. 186244.
 vannucci, R.C. and Yager, J.Y., Glucose, lactic acid, and perinatal hypoxi~ischemic brain damage, Pediatr. Neurol., 8 (1992) 3 12.  Volpe, J.J., Neurology of the Newborn, W.B. Saunders, Philadelphia, 1987, pp. 1-716.  Welsh, F.A., Moyer, D.J. and Harris, V.A., Regional expression of heat shock protein-70 mRNA and c-fos mRNA following focal ischemia in rat brain, J. Cereb. Blood Flow Metab., 12 (1992) 204~212.