Molecular Brain Research, 19 (1993) 9-21 9 © 1993 Elsevier Science Publishers B.V. All rights reserved 0169-328x/93/$06.00

BRESM 70609

GABAA/benzodiazepine receptor y2 subunit gene expression in developing normal and mutant mouse cerebellum

Vera Luntz-Leybman a, Adrienne Frostholm a, Lawrence Fernando b, Angel De Bias b and Andrej Rotter a

aDepartment of Pharmacology and the Neuroscience Program, The Ohio State University, Columbus, OH 43210 (USA) and b Division of Molecular Biology and Biochemistry, School of Biological Sciences, University of Missouri - Kansas City, Kansas City, M O 64110 (USA)

(Accepted 12 January 1993)

Key words: y-Aminobutyric acid; Y2 Subunit mRNA; In situ hybridization; Purkinje cell degeneration; Lurcher; Reeler; Ontogeny; CerebeUar cortex

Recent studies have identified several subunits (or,/3, 3' and 8) of the y-aminobutyric acid A/benzodiazepine receptor; each consists of several variants. The ~'2 subunit appears to mediate the interaction of the a and /3 subunits making the receptor capable of modulation by benzodiazepines. In the present studies, the expression of mRNA encoding the Y2 subunit was examined in the cerebellum during development and in adult Purkinje cell degeneration, lurcher and reeler mutant mice. In the normal adult cerebellum, in situ hybridization with [35S]cRNA probes revealed a strong signal over the Purkinje cell layer and deep cerebellar nuclei, and a weaker signal over basket, stellate and granule cells. Labeling over Purkinje cells was detectable at birth, gradually becoming stronger and more punctate during postnatal weeks 1 and 2, as Purkinje cells formed a monolayer between the molecular and granule cell layers. Adult levels of grain density were reached by P20. The external germinal layer, which contained proliferating granule cells, was unlabeled throughout development; however, weak labeling was detected over the internal granular layer at the end of postnatal week 1, as granule cells began their migration across the molecular layer. During the second postnatal week, punctate labeling became visible over the molecular layer in a distribution indicative of basket and stellate cells. In adult Purkinje cell degeneration and lurcher mutants, in which Purkinje cells have degenerated, no punctate labeling characteristic of mature Purkinje cells was detected. In adult and developing reeler mutants, where all classes of cells are malpositioned throughout the cerebellum, the punctate hybridization signal was present and clearly associated with Purkinje cells in all cortical regions. Our results suggest that developing Purkinje cells express the Y2 gene at a time prior to receiving GABAergic inhibitory input, and that the continued expression in the adult is not affected by the absence of afferents.

INTRODUCTION

T h e y - a m i n o b u t y r i c a c i d A / b e n z o d i a z e p i n e

( G A B A A / B Z ) r e c e p t o r is a mu l t imer i c G A B A - g a t e d

ch lo r ide channe l c o m p o s e d o f severa l d i f fe ren t sub-

units , inc lud ing a , /3, y and 6. Each subuni t exists in

severa l va r i an t forms (for reviews, see refs. 5, 29, 34,

49), some o f which a re de r ived f rom a l t e rna t ive R N A spl icing 2'25'5s. I t has b e e n d e m o n s t r a t e d tha t a subuni t s

con ta in the b e n z o d i a z e p i n e b ind ing si te 36'37'5°, /3 sub-

uni ts con ta in the y - a m i n o b u t y r i c acid ( G A B A ) b ind ing si te 7'36'37, and tha t r e sponses to G A B A in vivo can be

m o d u l a t e d by the b e n z o d i a z e p i n e s ( rev iewed in ref.

35). However , r e c o m b i n a n t G A B A g r ecep to r s ex-

p r e s sed in X e n o p u s oocytes , c o m p o s e d en t i re ly of a

and /3 subuni ts , show only a weak modi f i ca t ion of

G A B A responses by b e n z o d i a z e p i n e s 3°'47. T h e inclu-

s ion of Y2 o r 73 subuni ts , in combina t i on with a a n d / 3

subuni ts , confers b e n z o d i a z e p i n e sensi t ivi ty on r ecom-

b inan t G A B A A - r e c e p t o r s 24'39, ind ica t ing tha t the y

subuni t is r equ i r ed for the c o m p l e t e b e n z o d i a z e p i n e

response .

T h e y subuni t was first iden t i f i ed when a b ra in

c D N A l ibrary was s c r eened with a p r o b e based on

sequences c o m m o n to the ot and /3 subuni t s 39'62. Cur-

rent ly, t h r ee var ian ts of the y subuni t a re known 2°. In

add i t ion , two forms o f the 'Y2 subuni t , "Y2s (short form)

and "Y2L ( long form), which a re g e n e r a t e d by a l t e rna- t ive R N A splicing, have been iden t i f i ed 25'56'58. T h e TEL

sequence dif fers f rom the Y2s by the inser t ion of an

Correspondence: A. Rotter, Department of Pharmacology and the Neuroscience Program, The Ohio State University, 333 West 10th Ave., Columbus, OH 43210, USA. Fax: (1) (614) 292-9805.

I ' , !

e i g h t a m i n o ac id s e q u e n c e in t h e i n t r a c e l l u l a r l o o p

j o i n i n g t r a n s m e m b r a n e s e g m e n t s M 3 a n d M 4 eS. T h e

a d d i t i o n a l 8 a m i n o ac id p e p t i d e in Yzt. c o d e s fo r a

c o n s e n s u s t a r g e t s e q u e n c e fo r p r o t e i n k i n a s e C 25 a n d

c o n f e r s s ens i t i v i t y to a l c o h o l s~. R N A s e p r o t e c t i o n as say

a n d N o r t h e r n b l o t a n a l y s i s h a v e s h o w n t h a t t h e two

f o r m s a r e d i f f e r e n t i a l l y e x p r e s s e d in t h e b r a i n d u r i n g

d e v e l o p m e n t 57.

T h e l o c a l i z a t i o n o f t h e v a r i o u s G A B A A / B Z s u b u n i t

m R N A s h a s b e e n e x t e n s i v e l y s t u d i e d 3°'39'4s'6°'t't. R e c e n t

i m m u n o c y t o c h e m i c a l 3 a n d in s i tu h y b r i d i z a t i o n ls'~j

s t u d i e s in t h e r a t h a v e s h o w n t h a t t h e Y2 s u b u n i t

p e p t i d e a n d its m R N A a r e l o c a l i z e d in b r a i n r e g i o n s

c o n t a i n i n g ~1 a n d /32//33 s u b u n i t s o f G A B A A / B Z

r e c e p t o r . In t h e p r e s e n t s t u d i e s , we h a v e u s e d in s i tu

h y b r i d i z a t i o n o f [ 3 S S ] c R N A p r o b e s to l oca l i ze t h e Y2

s u b u n i t m R N A in t h e d e v e l o p i n g m o u s e c e r e b e l l u m ,

a n d in t h e c e r e b e l l u m o f a d u l t P u r k i n j e cel l d e g e n e r a -

t i o n ( p c d / p c d ) , l u r c h e r ( L c / + ) , a n d r e e l e r ( r l / r l )

m u t a n t m i c e , to d e t e r m i n e t h e r e l a t i o n s h i p b e t w e e n

t h e t e m p o r a l a n d s p a t i a l e x p r e s s i o n o f t h e T2 s u b u n i t

m R N A a n d t h e f o r m a t i o n o f c e r e b e l l a r s y n a p t i c c i r -

cu i t ry .

M A T E R I A L S A N D M E T H O D S

Riboprobe preparation

Cloning of human Y2 S subunit cDNA The cDNAs from postmortem adult human cerebral cortex were

obtained from poly(A) + RNA with reverse transcriptase and a com- bination of oligo-dT and random priming, as described previously. A 3 kb 3% (short form) subunit clone was isolated 23.

3': riboprobe preparation from human cDNA The sequence encoding the intracellular loop of the "/2 (short

form) subunit, which included 254 bases and is located between the M3 and M4 transmembrane segments, was synthesized by PCR with specific forward (CTC GAA "ITC CAT TAT TTT GTC AGC AAC) and reverse (CTC GAA TTC GGA GTC CAT TTT GGC AAT) oligonucleotide primers. Each contained 18 bases of the specific sequence (1295-1313 and 1537-1555 reverse complementary), plus a 9-mer nucleotide sequence added to the 5' end of each. The latter included an EcoRI restriction site. Using the whole length human Y2 cDNA as template, DNA coding for the intracetlular region of the receptor was amplified by PCR. The amplified DNA was digested with EcoR1 (United States Biochemical) and ligated (T4 DNA ligation kit, US Biochemical) into an EcoRI-treated vector pSP65 (Promega). E. coli HB101 (Bethesda Research Laboratories) were transformed with the ligated DNA. Bacterial colonies containing the insert DNA, in the sense and antisense orientation to the SP6 polymerase promoter, were selected. The orientation of the insert in the pSP65 vector was determined by using combinations of primers from the y2 insert and the pSP65 vector.

72 riboprobe preparation from mouse cDNA Probes were generated by polymerase chain reaction (PCR), as

follows: Poly(A) + RNA was isolated from mouse brain according to method of Badley et all . First strand cDNA and PCR reactions were conducted according to the Perkin Elmer Cetus GeneAmp RNA PCR protocol (Perkin Elmer Cetus, Norwalk), briefly described as

follows: I ,~g of poly(A)" RNA was reverse transcribed at 42 ( t~,~ 30 min in a reaction mixture containing 5 mM MgC[:> 50 mM K('l, ll) mM Tris (pH 8.3), I mM dNTPs. I U/~tl of RNase inhibitor. 2.5 U//al cloned Moloney Murine Leukemia Virus Reverse transcrip- tase and 2.5 ~M random hexamer, in a final vohmac of 21) #1. The mixture was heat inactivated at 99o( ̀ for 5 rain and prepared for PCR.

PCR reaction Two to 8 /~1 of the reverse transcription reaction mixture was

amplified for 30 cycles, with a denaturation step at 94°C for I min (first cycle for 2 rain), an annealing step at 60°C for 1 rain, and an extension step at 72°C for 1 min (last cycle for 5 rain) in a total volume of 100 ~tl containing 2 mM MgC12, 50 mM KCI, 10 mM Tris (pit 8.3), 200 mM dNTPs, 25 U AmpliTaq DNA polymerase and 0.2 /.tM of upstream and downstream primer. Primers, used for amplifi- cation of 3'2 probe, spanned a region of the M3-M4 fragment, the most variable portion of the GABA A receptor subunits. At the 5' end of each primer, a HindIIl or BamHI restriction site, plus three additional nucleotides, were added. The sequence of the primer, based on the published mouse sequence 25 was as follows: 5'-ATA AAG CTT TCA GCA ACC GGA AGC CAA GCA-3' (upstream y2 primer) and 5'-ATA GGA TCC ACT GGC ACA GTC CTT GCC ATC-3' (downstream Y2 primer). An amplified fragment of 196 b was obtained from the PCR reaction. The PCR product was digested with HindIIl and BamHI restriction enzymes and ligated 'in gel' in HindlII-BamHl treated, dephosphorylated pBluescript II SKI+] phagemid vector (Stratagene, La Jolla) as previously described ~-~ Recombinant plasmid was purified, alkali denatured, and sequenced using Taq polymerase (Perkin Elmer Cetus) as described by the manufacturer. The fragment was then subcloned into pBluescript SK plasmid 22.

lit citro transcription

The mouse y2-containing plasmid was linearized with Hindlll and BamHI (Boehringer, Indianapolis) for the subsequent produc- tion of antisense and sense RNA probes, respectively. Both sense and antisense human 3'2 plasmids were linearized with Xbal for production of corresponding 254 bp cRNA probes after transcription with SP6 polymerase. The linearized templates were purified over a Nensorb 20 column (NEN, Wilmington) and transcribed at 37°C for 60 min in a reaction mixture containing 40 mM Tris, 8 mM MgCI> 50 mM NaC1, 2 mM spermidine, 10 mM dithiothreitol (DTT), 0.5 units RNase-Block II, 500 ~M ATP, 500 ~M CTP, 500 /zM GTP (Stratagene, La Jolla), 500 ng of linearized template, 25/zM [35S]UTP (Amersham, S.A. > 800 Ci/mmol); the reaction was initiated by the addition of 25 units of T3 RNA polymerase (for antisense mouse cRNA), or 25 units of T7 RNA polymerase (for sense mouse cRNA) or 25 units of SP6 RNA polymerase (Stratagene, La Jolla) (for both sense and antisense human cRNA), pH 8.0, to a final volume of 10 /xl. After digestion of template with 5 units of DNase (Stratagene, La Jolla) for 30 min at 37°C, riboprobes were purified over a Nuctrap push columns (Stratagene, La Jolla) and DTT was added to a final concentration 100 mM. Probes were stored at - 20°C. Specific activ- ity of the riboprobes was 1-1.5 X 10 9 dpm//xg.

Northern blots

The number and size of the mRNA species recognized by the Yz probe was determined by Northern blot hybridization, according to previously described methods u'64.

In situ hybridization

Normal C57BL/6 and Purkinje cell degeneration, lurcher and reeler mutant mice were bred from stocks purchased from Jackson Laboratories (Bar Harbor, Maine). Animals were decapitated after inhalation anesthesia (Metofane, Pitman-Moore Inc., NJ); brains

were rapidly removed, frozen on dry ice and stored at -70°C. Coronal or sagittal sections, 20/~m thick, were thaw-mounted onto 3 x subbed (300 bloom gelatin & chrome alum) slides and stored at

- 70"C. Hybridization was conducted as follows: Slides were brought to

room temperature and dried. Sections were fixed for 30 rain in 4% paraformaldehyde in 1 × PBS (pH 7.4), and washed (2 × 5 min) in I×PBS. Sections were then acetylated (0.25% acetic anhydride diluted in 0.1 M triethanolamine/0.9% NaCI, pH 8.0) for 10 rain at room temperature (RT), and dehydrated (1 min each) through 70%, 80%, 90%, and 100% EtOH. Each slide was covered with 100 ~1 of hybridization buffer (50% formamide, 4 × SSC, 500 mg/ml salmon sperm DNA, 250/~g/ml tRNA, 1 ×Denhardt 's , 100 mM DTT, and 10% dextran, containing 2.0× 107 dpm/ml riboprobe), coverslipped with parafilm and incubated for 20 h at 50"C. The parafilm was then removed in 1 × SSC; sections were washed in 2 × SSC at RT for 10 min, followed by incubation in RNAse A solution (20/zg/ml RNAse A in 10 mM Tris-HCI, 500 mM NaCI, 1 mM EDTA, pH 8.0) for 30 rain at 37"C. RNase treated sections were washed at increasing stringency, as follows: 2 × SSC (10 rain), 1 × SSC (10 min), 0.5 × SSC (10 min), and 0.5xSSC (60 rain) at 70"C. After final washes in 0.5 × SSC (10 min) at RT, sections were dehydrated (1 min each) through 70%, 80%, 90% and 100% EtOH. All washing solutions, except RNAse A solution, contained 10 mM 2-mercaptoethanol to prevent non-specific binding of riboprobe. The non-specific hy- bridization signal was determined by exposing adjacent sections to sense RNA probes.

Autoradiography

Slide-mounted sections were dipped in Ilford K5-D photographic emulsion at 42"C, dried overnight, exposed for 5 days at 4°C. Slides were developed in Kodak D-19 developer (diluted 1 : 1 with distilled water) for 3.5 rain at 20"C, fixed for 3.5 min in Kodak Rapid-Fix, and washed in distilled water for 30 rain. Emulsion covered sections were counterstained with Cresyt fast violet and coverslipped. Since the Cresyl fast violet stain partly obscures the grain density over regions in which cells are densely compacted, such as the cerebellar granule cell layer, additional autoradiograms were generated as follows: Acid-washed coverslips (25 × 77 ram, No. 0, Corning Glass Works, NY), previously coated with a uniform layer of Kodak NTB-2 photo- graphic emulsion, were apposed to the slide-mounted sections and clamped together with hinder clips under minimum sodium safelight illumination. The assemblies were placed in lightproof boxes contain-

11

ing desiccant and exposed for 5 days at 4°C. Coverslips containing autoradiographs were developed in Kodak Dektol developer (diluted 1:1 with distilled water) for 2 min at 17°C, fixed and washed as above, and mounted onto microscope slides with Depex (Bio/medi- cal Specialties, Santa Monica, CA). Autoradiographic grain density was measured with a Nikon 'Image-One' optical density program. Sections and autoradiograms were photographed (Kodak Panatomic- X film) under bright-field or dark-field illumination using a Nikon SMZ-10 stereomicroscope (low magnification), or a Nikon Optiphot compound microscope (high magnification).

R E S U L T S

Probe design and specificity The two probes used in these experiments were

designed to recognize the region of mRNA which codes for a portion of the intracellular loop between transmembrane spanning domains M3 and M4 of the 3'2 subunit. This intracellular loop exhibits the least sequence identity among the various GABAA/BZ re- ceptor subunit cDNAs, and is, thus, unique to the T2 subunit mRNA. This region is conserved among sev- eral species. As may be seen in Fig. 1, the correspond- ing regions from human, bovine, rat, mouse and chicken show considerable sequence identity. Although some differences between the nucleotide sequence of mouse and that of other species were noted, they do not result in change in the predicted amino acid sequence of the receptor protein. Northern blots revealed bands of approximately 4.3 and 2.5 kb, using either mouse 3"2 L or human 3'2 S as probes (data not shown). No cross- hybridization with other subunit mRNAs was observed. In situ hybridization with both the long and short forms of the 3'2 probe revealed identical distributions in adult, developing and mutant cerebellum, since each

=L CA.TT~rrTGr crrcrl"tX~A TrCaE b.CACTA T~ ACCA'I'I'C-~T c.CATTA =ut I'l JC~;Ti ~ A C A ~ T r G A C d. CATTA T - - i ~ l I I r ~ - ! I ~ A C C A T T G A T e,

H Y F V S N R K P S K D K D K K K K N P P P R M F S F K A P T I D

I R P R S A T I O M N N A T H L Q E R D E E Y G Y E C L D G K D C A S

I I II ICI~I,~i4~I II l~IA~.'.'.'.'.'.'.'i~ . . . . - T . ~ . . . A T C C G T A ~ ' I ' G G A C T D C I II I IL~l~l.~Ir~l II It,.~5~i~ ~ , ~ , ~ T A ~ ~ ~

" T ~ ' T ~ T . a ~ A T A ' I ~ ' I ' T ~ ' r ~ F F C C F E D C R T G A W R H G R I H I R I A K M D S

Fig. 1. Comparison of sequences of 72-cDNAs from mouse, rat, bovine, human and chicken brain corresponding to a region coding for the intracellular loop between transmembrane domains M3 and M4 (amino acids 318-404). a: mouse long form25; b: rat short form3°; c: bovine long form5S; d: human long form39'58; e: chicken long form 17. Although 100% homology was found in amino acid sequence, nucleotides were not identical across species (underlined mouse nucleic acids indicate difference with at least one other species). Mouse DNA compared with rat,

bovine, human and chicken DNAs were, respectively, 96.14%, 92.98%, 94.04%, and 88.42% homologous.

12

p robe hybr id izes to bo th forms of the ye m R N A . Sense

p robes r evea led a very low signal, uni formly d i s t r ibu ted

over each sect ion (da ta not shown).

Adult distribution The Y2 subuni t m R N A had a wide d i s t r ibu t ion

t h r oughou t the brain. The highest concen t r a t i ons were

obse rved in the o l fac tory bulb, h i p p o c a m p u s and cere-

be l lum (Fig. 2A). In the adu l t c e r ebe l l um (Fig. 2 B - D ) ,

Purkin je cells, deep ce rebe l l a r neurons , baske t and

s te l la te cells were the most p rominen t ly l abe led neu-

rons. The g ranu le cell layer was also diffusely l abe led

at a lower level.

Del,eloping norrnal animals

Although , at bir th, the overal l 3/2 signal was lower in

all b ra in regions than in adul t animals , the ce rebe l lum

was among those most densely l abe led (Fig. 3A). At

pos tna ta l day 1 (P1), the "/2 subuni t hybr id iza t ion sig-

nal was visible as a diffuse band over Purkinje cells

which, at this deve lopmen ta l stage, were packed sev-

eral cells deep , be tween the ex te rna l germinal and the

in terna l g ranu la r layers (Fig. 3B,C). The label ing den-

sity was low relat ive to adul t levels and lacked the

punc t a t e d i s t r ibu t ion assoc ia ted with adul t Purkinje

cells. A dis t inct hybr id iza t ion signal was also p resen t

over the d e e p ce rebe l l a r nuclei at b i r th (Fig. 3B), which

I

~J W

Fig. 2. Localization of the GABA A receptor Y2 subunit mRNA in the adult mouse brain and cerebellum. A: dark-field photomicrograph of a sagittal section through the adult mouse brain. B: a coronal section showing the laminar structure of the adult cerebellum. C: bright-field photomicrograph of an emulsion-dipped section, counterstained with Cresyl fast violet, showing autoradiographic grains over cells within the adult molecular layer. D: a bright-field autoradiogram showing the punctate labeling over the Purkinje cell layer. Large arrow indicates the deep cerebellar nuclei; g, granule cell layer: m, molecular layer; P, Purkinje cell layer: w, white matter. Bars = I mm (A); 1 mm (B); 20 ~m (C);

50 ~xm (D).

13

was associated with large cell bodies in each of the

three subnuclei (lateral, in termedia te and medial) on

e i ther side o f the midline. The external germinal layer

was unlabeled at bir th and th roughou t deve lopment

(Fig. 3C). By P6, Purkinje cells were becoming aligned

into a monolayer (Fig. 4A); au toradiographic grain

Fig. 3. Localization of the GABA A receptor 3'2 subunit mRNA in the P1 mouse brain and cerebellum. A: dark-field photomicrograph of a sagittal section through the mouse brain at birth. B: a sagittal section through the newborn cerebellum, showing a band of high grain density below the external germinal layer. C: bright-field pho- tomicrograph of an emulsion-dipped section, counterstained with Cresyl fa~t violet, showing autoradiographic grains over Purkinje cells stacked between the external germinal and internal granular layers. Small arrows indicate the cerebellar cortex; large arrow indicates the deep cerebellar nuclei; e, external germinal layer; i, internal granular layer; P, Purkinje cell bodies. Bars= 1 mm (A); 500 /~m (B);

25 ~m (C).

Fig. 4. Localization of the GABA A receptor 3'2 $ubunit mRNA in the P6 mouse cerebellum. A: dark-field photomicrograph of a sagit- tal section through the cerebellum. B: a single cerebeilar folium, showing a band of high grain density over the developing Purkinje cell layer. C: a bright-field photomicrograph of an emulsion-clipped section, counterstained with Cresyl fast violet, showing autoradio- graphic grains over Purkinje cells, taken at high magnification. White arrow indicates the deep cerebellar nuclei; black arrow, labeled Purkinje cells; e, external germinal layer; i, internal granular layer; m, molecular layer; P, Purkinje cell layer; w, white matter. Bars = 500

/~m (A); 100/tin (13); 25/zm (C).

density became more puneta te and concent ra ted over

the cell bodies (Fig. 4B,C). Label ing was absent over

prol iferat ing granule cells in the external germinal layer th roughou t development . W e a k labeling became de-

tectable in the internal granular layer (Fig. 4C) at approximately P5-7 , as granule cells began their mi-

grat ion into this region.

14

During postnatal weeks 2 and 3, Purkinje cell label-

ing increased considerably (Fig. 5A-F) . The grain den- sity over the granule cell layer also increased progres- sively as more cells migrated from the external germi-

nal layer, across the molecular layer and into the

internal granular layer (Fig. 5A,B). Grain density was low over the molecular layer until the second postnatal

week (P9-13), when clusters of moderate density, be- came visible in regions corresponding to basket and stellate cells (Fig. 5C,D). Adult levels of grain density (Fig. 2) were reached in all cerebellar regions between

P15 and P20 (Fig. 5E,F).

Purkinje cell degeneration attd Lurcher mutant.~ Adult Purkinje cell degeneration (pcd/pcd) and

Lurcher (Lc /+ ) mutant mice are both characterized by the absence of Purkinje cells. The overall appear- ance of the adult pcd/pcd cerebellum (Fig. 6B), ~d-

though somewhat smaller, was very similar to that of

normal animals (Fig. 6A). Foliation and laminar ar-

rangement of the cerebellar cortex was essentially nor- mal, although the pcd/pcd molecular layer was dis- tinctly narrower than that of control animals (Figl 6B). At high magnification, the absence of Purkinje cells in

the pcd/pcd was clearly evident, and in situ hybridiza-

Fig. 5. The autoradiographic distribution of Y2 subunit mRNA in a saginal section through the developing cerebellum, and in a single folium. A,B: P9; C,D: P13; E,F: P20. White arrow, deep cerebellar nuclei; e, external germinal layer; g, granule cell layer: m, molecular layer: P, Purkinje

cell layer: w. white matter. Bars = 1 mm (A): 50(I p.m (C,E): 200 ~m (B,D,F)

tion with the T2 probes revealed that the punctate signal, dearly present in the unaffected littermate (Fig. 7A), was completely absent from the interface of the molecular and granule cell layers (Fig. 71]). Calibrated

. _ , , 3 : ~' ~ - : .

15

optical density values were measured in autoradio- graphs processed by the coverslip method (10 sections from each animal, mean + S.D.), where there was no counterstain to obscure the autoradiographic grains: Increased grain density was observed over the molecu- lar layer in the P140 pcd/pcd (63.7 + 8.2 vs 45.9 ± 8.5) and over the deep cerebellar nuclei of both P40 (83.6 ± 13.3 vs 70.3 + 7.4) and P140 (84.5 ± 14.9 vs 68.6 + 12.4) mice. In each of these regions, the hybridization signal was clustered, suggesting association with the cell bodies of basket and stellate cells, and with deep nuclear neurons, respectively. Similar results were ob- tained in the adult (P108) Lc /+ cerebellum (Figs. 6C and 8A). As in the pcd/pcd mouse, the distinct 3'2 hybridization signal at the interface of the molecular and granule cell layers, seen in control animals, was missing in the heterozygous lurcher cerebellum (Fig. 8B). Labeling was again higher than that of controls over the Lc /+ molecular layer (45.3 + 5.1 vs 25.5 ± 6.0) and deep cerebellar nuclei (66.0 ± 11.9 vs 40.6._+ 8.4). In each of the mutants, the granule cell layer was labeled at, or slightly below, the respective control level.

Reeler (rl / rl) mutants Reeler is an autosomal recessive mutation in which

the migration of Purkinje cells is arrested early in development. At the pial surface, a thin band of cortex contains the molecular, Purkinje and granule cell lay- ers; the organization of this region resembles, to some extent, the normal cerebellar cortex, although the cor- tex of the reeler is largely unfoliated and extensively atrophied (Fig. 6D). Grain density over the superficial cortex of the reeler was similar to that of controls, inasmuch as low grain density was present over the molecular and granule cell layers, and much higher, punctate labeling was associated with Purldnje cells (Fig. 10). Below the narrow cortical region, central

Fig. 6. The autoradiographic distribution of ~'2 subunit mRNA in the normal, Purkinje cell degeneration, lurcher and reeler mutant cere- bellum. A: a dark-field autoradiograph of control mouse cerebellum showing a punctate band of autoradiographic grains over Purkinje cells; Cresyl fast violet counterstain partially obscures grain density over the granule cell layer. B: cerebellum of a P140 pcd/pcd mouse; note absence of Puridnje cells at the interface of the molecular and granule cell layers. C: cerebellum of a P108 Lc/+ mouse; as with pcd/pcd animals, Purkinje cells are absent throughout the coi, tex. D: cerebellum of a P20 rl/rl mouse; punctate clusters of autoradio- graphic grains are present over Purkinje cell bodies throughout the malformed cortex, g, granule cell layer; m, molecular layer; P, Purkinje cell layer. Arrows indicate deep cerebeilar nuclei; I, II and III indicate the three subcortical Purkinie cell masses of the reeler

cerebellum. Bars = 500 ~m (A,B,D); 250 t~m (C).

16

Fig. 7. The autoradiographic distribution ol T_, subunit m R N A m the Purkinje cell degeneration mutant . A: a cerebellar folinm of a normai control mouse showing a band of autoradiographic grains at the interfacc of the molecular and granule cell layers, photographed under dark power, at high magification. B: the pod ~pod mutant ccrebellar cortex; note the absence of labeled Purkinje cells (arrowheads). g, granule cell

layer: m, molecular layer: P. Purkinje cell layer. Bars = 50/zm.

masses largely comprised of Purkinje cells (correspond- ing to the medial, intermediate and lateral cell groups described by Goffinet et al.lS), were present on either side of the midline. The majority of ectopic Purkinje cells were largely concentrated within the cerebellar

masses, or embedded within the granule cell layer; very few migrated to their normal position at the interface of the granule cell and molecular layers. A strong 72 hybridization signal was observed over Purkinje cells in each of these regions (Fig. 6D). At the ventral surface, the deep cerebellar nuclei were of lower grain density, than those of behaviorally normal littermates. The ar- chitectonic defect in the reeler cerebellum was obvious at birth (Fig. 9A). Although the hybridization signal at P1 was low relative to that of older mice, reeler Purk- inje cells expressed the Y2 subunit mRNA throughout postnatal development (Figs. 9 and 10), irrespective of their location within the cerebellum, The hybridization signal was punctate and clearly associated with Purk- inje cells in each region; grain density increased in a manner comparable to that of normal controls.

D I S C U S S I O N

Fig. 8. The autoradiographic distribution of "/2 subunit m R N A in the heterozygous lurcher mutan t cerebellar cortex. A: bright-field, Cresyl fast violet-stained section showing a single folium of the Lc/+ cortex. B: corresponding dark-field autoradiograph of the same sec- tion; note the absence of labeled Purkinje cells at the interface of the molecular and granule cell layers (arrowheads). g, granule cell layer;

m, molecular layer; w, white matter. Bars = 50 ~zm.

A variety of experimental approaches have revealed

that a l, /32/3 and T2 subunits of the native GABA A receptor complex are frequently associated. Im- munoaffinity purification studies, using an antibody to the a subunit, have shown that the GABA A receptor "/2 subunit is co-precipitated with different populations of a subunits 9. Both benzodiazepine and GABA bind- ing sites are present within the precipitated receptor complex when immunoprecipitated with an antibody to the /32/3 or "~2 subunitt°'23'36, indicating the presence of both a and /3 subunits, respectively. Furthermore, im- munocytochemical studies have identified sets of neu- rons in which the a t,/32/3 and Y2 subunit peptides are all expressed s't2'44. The close association of a t and 3'2 subunits is also suggested by recent chromosomal local- ization studies which have indicated that members of the GABA A receptor gene family colocalize in small

17

Fig. 9. The autoradiographic distribution of Y2 subunit mRNA in the developing reeler mutant cerebellum. A: P1; B: P5; C: Pll; D: P13. Arrowheads indicate deep cerebellar nuclei. Bars = 1/zm.

gene clusters which are widely distributed throughout the genome: The Y2 and a 1 subunits map closely together on human chromosome 5 59 , suggesting that their developmental expression may also be closely linked.

Within the cerebellum, the association of the Y2 mRNA hybridization signal with Purkinje cells is evi- dent from the results obtained in Purkinje cell degen- eration and Lurcher animals. The pcd/pcd mouse is an autosomal recessive mutation in which ataxia be- comes observable at 3-4 weeks of age. This results from the degeneration of all Purkinje cells between P15-18 and P45, after an apparently normal early development. During this time, Golgi, granule, basket and stellate cells appear unaffected 26,33. Severe granule cell loss occurs later (between 6 and 12 months), prob-

ably as a consequence of Purkinje cell degeneration 16. Our in situ hybridization studies on these mutants show that the absence of Purkinje cells is accompanied by a loss of punctate labeling at the boundary of the granule cell and molecular layers by P40, thus provid- ing direct evidence that the Purkinje cells of normal animals express the Y2 subunit gene. A similar result was observed in the heterozygous Lurcher mouse. Lurcher is an autosomal semi-dominant mutation in which virtually all Purkinje cells, and the majority of granule cells, degenerate by P266. Homozygous ani- mals (Lc/Lc) die during the perinatal period, while heterozygous littermates (Lc/+ ), which are also af- fected by the mutation, survive. The loss of granule cells is much greater, and occurs much earlier, than in pcd/pcd mice, which results in a cerebellar cortex

~J

Fig. 10. The distribution of 3'2 subunit mRNA within the cerebellar cortex ol' the reeler mutant. A: a dark-field photomicrograph show- ing labeled ectopic Purkinje cells within the granule cell layer of a P20 mouse. B: labeled Purkinje cells within the cerebellar masses, g, granule cell layer; m. molecular layer; P, Purkinje cells; w, white matter; llI, lateral Purkinje cell mass. Bars = 51) p.m (A): 25 #m (B).

which is considerably smaller than that of normal lit- termate controls 6'3s'-~3. Some labeling is retained in the

granular layer of Lc/+ cerebellum, which suggests

that, although many granule cells die, the remaining ones are normal with respect to their "/2 subunit m R N A expression.

Although Purkinje cells, which contain both "/2 and a~ 63 subunit mRNAs, are completely absent in pcd/pcd and Lc/+ cerebellar cortex, paradoxically, only a small decrease in [3H]flunitrazepam binding site density has been observed in the molecular layer of each mutant 45. This was surprising, since Purkinje cell dendrites, which are thought to possess benzodiazepine binding sites 2~'52, are missing from the pcd/pcd and Lc/+ molecular layer. It has been suggested that other cell types, in addition to Purkinje cells, contribute to the high den- sity of [3H]flunitrazepam binding sites in the molecular layer ~4"55. Previous in situ hybridization experiments

have indicated that basket and stellate cells of the mouse cerebellum express G A B A A / B Z receptor a~ subunit m R N A 63. In the present experiments, "/~ sub-

unit m R N A is shown to be present over cell bodies in the molecular layer of pcd/pcd and Lc/ + mice, at a

higher level than in control animals. The reasons for the increased grain density are unclear. It is possible that any loss of benzodiazepine receptors present on

degenerating Purkinje cell dendrites is balanced by an

increase in G A B A A receptors in the afferent basket and stellate cells. Alternatively, the increase may be due to a generalized upregulation in response to dis- connection from afferent granule cells; these cells are known to be decreased in number in older mutant

animals 6'~6. Similarly, in the deep cerebellar nuclei of

adult Purkinje cell degeration and lureher animals, where GABA-containing Purkinje cell terminals have

degenerated, the expression of Y2 subunit m R N A is again increased. No such increases were observed in the deep cerebellar nuclei of reeler animals, where Purkinje cell axons remain intact. Although this change in grain density may reflect a form of denervation

supersensitivity, similar to that observed in other re- gions of the CNS 4, the overall cross-sectional area of the molecular layer and deep cerebellar nuclei are slightly reduced in these mutants45'55; such shrinkage

could contribute to an apparent increase in grain den- sity.

Purkinje cells are generated early in the mouse cerebellum, between embryonic days 11 and 13 32'54.

They originate in the ventricular germinal zone and

migrate radially to form a multicetlular region beneath the spreading external germinal layer. At birth, the

multiply stacked Purkinje cells are not yet tully differ- entiated, lacking the elaborate dendritic trees charac- teristic of fully mature cells 4243'54. The Y2 signal fol-

lowed a similar distribution: the diffuse band of Yz labeling detected at birth was present throughout the multicellular region. During the first postnatal week, dendritic arborization increases and Purkinje cell bod- ies become organized into a monolayer at the interface of the internal granular and molecular layers 42'43. Simi-

larly, between P5 and 7, autoradiographic grains no longer spanned the entire molecular layer, but were

clustered over Purkinje cell bodies. The intensity of the signal increased up to P15-20 when adult levels were reached.

The observed developmental pat tern of `/2 subunit m R N A expression in the murine cerebellum is similar to that of other G A B A A / B Z receptor subunit mark- ers. The closest correspondence is with a163 and /3~ ~'4

subunit mRNAs, which are also detectable in the de- veloping Purkinje cell layer at birth. At the end of the first postnatal week, the al and/32hybridization signals are punctate and unambigaously associated with Purk- inje cell bodies. Both the a~ and 72 mRNAs are

19

presumably translated into their respective subunits and incorporated into GABAA/BZ receptors, since [3H]flunitrazepam binding sites, which are indicative of functional benzodiazepine receptors, are also present in the same region on P163. Morphological 21 and elec- trophysiological studies 51 have traced the formation of synaptic contacts during cerebellar development. Dur- ing the first postnatal week, Purkinje cell bodies are contacted by climbing fibers and granule cell parallel fibers 21. Since both of these inputs are excitatory in nature, they are unlikely candidates for initiating the appearance of GABAA/BZ receptors in the Purkinje cell. Inhibitory GABAergic inputs to the Purkinje cell arise from basket and stellate cells which migrate from the external germinal layer into the molecular layer, beginning at the end of the first postnatal week. Both cell types form synaptic contacts with their target, the Purkinje cell, at approximately P8-10 21'27. It is un- likely, therefore, that the initial expression of Y2 mRNA at, or before, birth is initiated by inhibitory afferent contact. Taken together, these results indicate that Purkinje cells express the genes for the al,/32 and Y2 subunits and form a GABAA/BZ receptor capable of binding [3H]flunitrazepam at an early stage of their maturation and prior to the formation of synaptic connections.

The absence of Y2 subunit mRNA from the external germinal layer indicates that both proliferating neuro- blasts and postmitotic cells, which give rise to granule, basket and stellate cells, are not, at this point, commit- ted to the expression of this subunit of the GABAA/BZ receptor. As cell division ceases, each granule cell begins to extend its axon; the cell body descends through the molecular layer in close association with a Bergrnann glial cell 4° trailing the axon which bifurcates and remains within the molecular layer, forming the parallel fibers 43. The great majority of granule cells move through the forming Purkinje cell layer to their definitive location in the internal granular layer after P7. When the granule cell reaches its final destination, its axon forms an excitatory synaptic contact with the target Purkinje cell 2~. Granule cell labeling increased progressively as more cells migrated from the external germinal layer, across the molecular layer into the internal granular layer. Although proliferating granule ceils in external germinal layer have been shown to express /33 subunit mRNA 64, 3'2 subunit mRNA is expressed later, in conjunction with t he o~163 and fiE 64 subunit hybridization signals.

Since the Purkinje cell hybridization signal becomes clearer and more intense at the end of postnatal week 1, at a time when basket and stellate cells begin to form inhibitory synapses with Purkinje cells, there is

the a priori possibility that contact with appropriate afferents may tend to stabilize ~/2 mRNA expression at some point during neuronal maturation. Our experi- ments in the reeler mutant mouse do not support this, however. In the reeler cerebellar cortex, the migration of Purkinje cells is arrested early in development and the majority of cells remain clustered in three masses (central, intermediate and lateral) on either side of the midline. A small number of cells migrate further, but become embedded in the sparse granule cell layer; a few cells are able to reach their normal position at the border of the molecular and granule cell layers t8'19'31'41.

Purkinje cells remaining in the subcortical masses, un- like the cells which attain normal positions in the reeler cortex, are devoid of inhibitory inputs from basket and stellate cells 31. However, reeler Purkinje cells expressed 3/2 mRNA regardless of their position within the cerebellar cortex. As in control animals, the signal was present at birth and increased in intensity from the end of the second week into adulthood. Similar results were obtained in previous studies on the expression of a 1 subunit mRNA 13, and with the fiE subunit 46, in developing reeler mutants. Thus, our data on the reeler mutant are in agreement with our find- ings in developing normal animals, in that they suggest that inhibitory afferents do not appear to have a major role in the acquisition and maintenance of GABA A subunit mRNA expression.

The present studies demonstrate that the GABA/ BZ 3/2 subunit gene is expressed in mouse cerebellum during the early stages of postnatal development, prior to the full maturation of Purkinje cells and before the adult pattern of synaptic connections is formed. Fur- thermore, the expression of "Y2 mRNA persists in Purkinje cells which are devoid of the normal comple- ment of afferent synapses. The idential pattern of expression of Otl, f12 and 3'2 in the reeler mutant strongly supports their colocalization within the Purk- inje cell, and indicates that the entire complex may be coordinately regulated.

Acknowledgments. We would like to thank Roxanne Prachthauser for technical assistance. This work was supported by grants from the National Institute of Alcohol Abuse and Alcoholism and the Na- tional Institute of Neurological Disorders and Stroke. V.L.-L was supported by USPHS Training Grant NS07291.

REFERENCES

1 Badley, J.E., Bishop, G.A., St. John, T. and Frelinger, J.A., A simple, rapid method for purification of poly A + RNA, Biotech- niques, 6 (1988) 114-116.

2 Bateson, A.N., Lasham, A. and Darlison M.G., ~-Aminobutyric acid A receptor heterogeneity is increased by alternative splicing of a novel y-subunit transcript, J. Neurochem., 56 (1991) 1437- 1440.

3 Benke, D., Mertens, S., Trzeciak, A., Gillessen, D. and Mohler,

20

H., GABA A receptors display association of y2-subunit with a~- and /3z/~-subunits, J. Biol. Chem., 266 (19911 4478-4483.

4 Biggio, G., Corda, M.G., Concas, A. and Gessa, G.L.. Dcnerva- tion supersensitivity for benzodiazepinc receptors in the rat sub- stantia nigra, Brain Res., 220 (1981) 344-349.

5 Burt, D.R. and Kamatchi G.L., GABA A receptor subtypes: from pharmacology to molecular biology, FASEB .l.. 5 (19911 2916- 2923.

6 Caddy, K.W.T. and Biscoe, T.J., Structural and quantitative stud- ies on the normal C3H and Lurcher mutant mouse, Phil. Trans. R. Soc. Lond. Ser. B, 287 (1979) 167-201.

7 Casalotti, S.O., Stephenson, F.A. and Barnard, E.A., Separate subunits for agonist and benzodiazepine binding in the 3, aminobutyric acid A receptor oligomer, ,L Biol. Chem., 261 (1986) 15013-15016.

8 De Bias, A.L., Vitorica, J. and Friedrich, P., Localization of GABA A receptor in the rat brain with a monoclonal antibody to the 57,000 M r peptide of GABA A receptor/benzodiazepine receptor/Cl channel complex, Z Neurosci., 8 (1988) 602-614.

9 Duggan, M.J., Pollard, S. and Stephenson, F.A., Immunoaffinity purification of GABA A receptor a-subunit iso-oligomers, J. Biol. Chem., 266 (1991) 24778-24784.

10 Duggan, M.J., Pollard, S. and Stephenson, F.A., Quantitative immunoprecipitation studies with anti--/-aminobutyric acid A re- ceptor 3'2 1-15 Cys antibodies, J. Neurochem., 58 (1992) 72-77.

11 Fourney, R.M., Miyakoshi, J., Day III, R.S. and Paterson, M.C., Northern blotting: efficient RNA staining and transfer. Focus, 10 (1988) 5-7.

12 Fritschy, J.M., Benke, D., Mertens, S., Oertel, W.H., Bachi, T. and Mohler, H., Five subtypes of type -/-aminobutyric acid recep- tors identified in neurons by double and triple immunofluores- cence staining with subunit-specific antibodies. Proc. NatL Acad. Sci. USA, 89 (1992) 6726-6730.

13 Frostholm, A., Zdilar, D., Chang, A. and Rotter, A., Stability of GABA A/benzodiazepine receptor a t subunit mRNA expression in reeler mouse cerebellar Purkinje cells during postnatal devel- opment, Det,. Brain Res., 64 (1991) 121-128.

14 Fry, J.P., Rickets, C. and Biscoe, T.J., On the location of gamma-aminobutyrate and benzodiazepine receptors in the cere- bellum of normal C3H and Lurcher mutant mouse, Neuroscience 14 (1985) 1901-1101.

15 Gambarana, C., Beattie, C.E., Rodriguez, Z.R. and Siegel, R.E., Region-specific expression of messenger RNAs encoding GABA A receptor subunits in the developing rat brain, Neuroscience, 45 (1991) 423-432.

16 Ghetti, B., Truex, L, Sawyer, B., Strada, S. and Schmidt, M., Exaggerated cyclic AMP accumulation and glial cell reaction in the cerebellum during Purkinje cell degeneration in pcd mutant mice, J. Neurosci. Res., 6 (1981) 789-801.

17 Glencorse, T.A., Bateson, A.N. and Darlison, M.G., Sequence of the chicken GABA A receptor y2-subunit cDNA, Nucl. Acids Res., 18 (19911 23.

18 Goffinet, A.M., So, K.-F., Yamamoto, M., Edwards, M. and Caviness, V.S., Architectonic and hodological organization of the cerebellum in reeler mutant mice, De~,. Brain Res., 16 (1984) 263-276.

19 Hamburgh, M., Analysis of the postnatal developmental effects of "reeler" a neurological mutation in mice. A study in developmen- tal genetics, Det,. Biol., 8 (19631 165-185.

20 Herb, A., Wisden, W., Luddens, H., Puia, G., Vicini, S. and Seeburg, P.H., A third y subunit of the GABA A receptor family, Proe. Natl. Acad. Sci. USA, 89 (1992) 1433-1437.

21 Herndon, R.M., Seil, F.J. and Seidman, C., Synaptogenesis in mouse cerebellum: A comparative in vivo and tissue culture study, Neuroscience, 6 (1981) 2587-2598.

22 Kalvakolanu, D.V.R. and Livingston III, W.H., Rapid and inex- pensive protocol for generating > 95% recombinants in sub- cloning experiments, BioTechniques, 10 (1991) 176-177.

23 Khan, Z.U., Fernando, L.P., Escriba, P., Busquets, X., Mallet, J., Miralles, C.P., Filla, M. and De Bias, A.L., Antibodies to the human Y2 subunit of the GABA A receptor, J. Neurochem., in press.

24 Knoflach, F., Rhyner, T., Villa, M., Kellenbergcr. S., Drcschc~, U., Malherbc, P., Sigel, E. and Mohler, tt., File y3 subunit of the GABAA-receptor confers sensitivity to benzodiazepinc rcccplor ligands, FEBS Lett.. 293 (19911 191 194.

25 Kofuji, P., Wang, J.B., Moss, S.J., [tuganir, R.L. and Burr, D.R., Generation of two forms of the y-aminobutyric acid A receptor ye-subunit in mice by alternative splicing, ,I Neurochent., 56 (199t) 713-715.

26 Landis S.C. and Mullen, R.J., The development and degeneration of Purkinje cells in the pcd mutant mice, Z Comp. NeuroL, 177 (19781 125-144.

27 Larramendi, L.M.H., Analysis of synaptogenesis in the cerebel- lum of the mouse, In R. Llinas (Ed.), Neurobiologv of Cerebel&r Et'olution and DeL:elopment, Am. Med. Assoc. Educ. Res. Fdn., Chicago, 1969, pp. 803-843.

28 Lippa, A.S., Sano, M.C., Coupet, J., Klepner, C.A. and Beer, B., Evidence that benzodiazepine receptors reside on cerebel!ar Purkinje cells: studies with 'nervous' mutant mice, Life Sci., 23 (1978) 2213-2218.

29 Luddens, H. and Wisden, W., Function and pharmacology of multiple GABA A receptor subunits, TIPS, 12 (1991) 49-51,

30 Malherbe, P., Sigel, E., Baur, R., Persohn, E., Ricbards, J.G. and Mohler, H., Functional characteristics and sites of gene expres- sion of the ~l, 3t, Y2 - - isoform of the rat GABA A receptor, J. Neurosci., 10 (19901 2330-2337.

31 Mariani, J., Crepel, F., Mikoshiba, K., Changeux J.-P. and Sotelo, C., Anatomical, physiological and biochemical studies of the cerebellum from reeler mutant mouse, Phil. Trans. R. Soe. Lond. B, 281 (19771 1-28.

32 Miale, 1. and Sidman, R.L., An autoradiographic analysis of histogenesis in the mouse cerebellum, Exp. NeuroL, 4 (19611 277-296.

33 Mullen, R.J., Eicher, E.M. and Sidman, R.L., Purkinje cell de- generation: a new neurological mutation in the mouse, Proc. Natl. Acad. Sci. USA, 73 (19761 208-212.

34 Olsen, R.W. and Tobin, A.J., Molecular biology of GABA a receptors, FASEB J., 4 (199(I) 1469-1480.

35 Olsen, R.W. and Venter, C.J. (Eds.), Benzodiuzepine/GABA Receptors and Chloride Channels; Structural and Functional Prop- erties, Alan R. Liss, New York, 1986.

36 Park, D. and De Bias, A.L., Peptide heterogeneity of GABAA/benzodiazepine receptors in bovine cerebral cortex and cerebellum, Brain Res., 550 (1991) 279-286,

37 Park, D., Vitorica, J., Tous, G. and De Blas, A.L., Purification of the ~,-aminobutyric acidA/benzodiazepine complex by im- munoaffinity chromatography, J. Neurochem., 56 (19911 1%2- 1971.

38 Phillips, R.J.S., "Lurcher'. A new gene in linkage group XI of the house mouse, J. Genet., 57 (1960) 35-42.

39 Pritchett, D.B., Sontheimer, H., Shivers, B.D., Ymer, S., Ketten- mann, H., Schofield, P.R. and Seeburg, P.H., Importance of novel GABA A receptor subunit for benzodiazepine pharmacol- ogy, Nature, 338 (1989) 582-585.

40 Rakic, P. and Sidman, R.L., Sequence of developmental abnor- malities leading to granule cell deficit in cerebellar cortex of weaver mutant mice, ,L Comp. NeuroL, 152 (1973) 103-132.

41 Rakic, P., Synaptic specificity in the cerebellar cortex: study of anomalous circuits induced by single gene mutations in mice. Cold Spring Harbor Symp. Quant. Biol., XL (1975) 333-346.

42 Ramon y Cajal, S., Histologie du Systeme Nert,eux, Vol. II, Mal- oire, Paris, 1911, pp. 80-119.

43 Ramon y Cajal, S., Studies on lmJertebrate Neurogenests, Trans- lated by L. Guth, Charles C. Thomas, Springfield, IL, 1960.

44 Richards, J.G., Schoch, P., Haring, P., Takacs, B. and Mohler, H., Resolving GABA A/benzodiazepine receptors: cellular and sub- cellular localization in the CNS with monoclonal antibodies, J. Neurosci., 7 (1987) 1866-1886.

45 Rotter, A. and Frostholm, A., Cerebellar benzodiazepine recep- tors: cellular localization and consequences of neurological muta- tions in mice, Brain Res., 444 (1988) 133-146.

46 Rotter, A., Zdilar, D., Luntz-Leybman, V. and Frostholm, A., Expression of Purkinje cell GABA A receptor subunit mRNAs

precedes, and is independant of, synapse formation, J. Cell Biochem., Suppl. 16E (1992) 234.

47 Schofield, P.R., Darlison, M.G., Fujita, N., Burt, D.R., Stephen- son, F.A., Rodriguez, H., Rhee, L.M., Ramachandran, J., Reale, V., Glencorse, T.A., Seeburg, P.H. and Barnard, E.A., Sequence and functional expression of the GABA A receptor shows a lig- and-gated receptor super-family, Nature, 328 (1987) 221-227.

48 Shivers, B.D., Killisch, I., Sprengel, R., Sontheimer, H., Kohler, M., Schofield, P.R. and Seeburg, P.H., Two novel GABA g recep- tor subunits exist in distinct neuronal subpopulations, Neuron, 3 (1989) 327-337.

49 Sieghart, W., Molecular basis of pharmacological heterogeneity of GABA A receptors, Cell. Sig., 4 (1992) 231-237.

50 Sigel, E., Stephenson, F.A., Mamalaki, C. and Barnard, E.A,, A y-aminobutyric acid/benzodiazepine receptor complex of bovine cerebral cortex, Z Biol. Chem., 258 (1983) 6965-6971.

51 Shimono, T., Nosaka, S. and Sasaki, K., Electrophysiological study of the postnatal development of neuronal mechanisms in the rat cerebellar cortex, Brain Res., 108 (1976) 279-294.

52 Skolnick, P., Syapin, P.J., Paugh, B.A. and Paul, S.M., Reduction in benzodiazepine receptors associated with Purkinje cell degen- eration in 'nervous' mutant mice, Nature, 277 (1979) 397-398.

53 Swisher, D.A. and Wilson, D.V., Cerebellar histogenesis in the Lurcher (Lc) mutant mouse, J. Comp. Neurol., 173 (1977) 205- 217.

54 Uzman, L.L., The histogenesis of the mouse cerebellum as stud- ied by its tritiated thymidine uptake, J. Comp. Neurol., 114 (1960) 137-160.

55 Vaccarino, F.M., Ghetti, B. and Nurnberger, Sr., J.I., Residual benzodiazepine (BZ) binding in the cortex of pcd mutant cere- bella and qualitative BZ binding in the deep cerebellar nuclei of control and mutanf mice: An autoradiographic study, Brain Res., 343 (1985) 70-78.

56 Wafford, K.A., Burnett, D.M., Leidenheimer, N.J., Burt, D.R., Wang, J.B., Kofuji, P., Dunwiddie, T.V., Harris, R.A. and Sikela, J.M., Ethanol sensitivity of the GABA A receptor expressed in Xenopus oocytes requires 8 amino acids contained in y2 L sub- unit, Neuron, 7 (1991) 27-33.

21

57 Wang, J.B. and Burt, D.R., Differential expression of two forms of GABA A receptor YE-SUbunit in mice, Brain Res. Bull., 27 (1991) 731-735.

58 Whiting, P, McKernan, R.M. and lversen, L.L., Another mecha- nism for creating diversity in y-aminobutyrate type A receptors: RNA splicing directs expression of two forms of y: subunit, one of which contains a protein kinase C phosphorylation site, Proc. Natl..4cad. Sci. USA, 87 (1990) 9966-9970.

59 Wilcox, A.S., Warrington, J.A., Gardiner, K., Berger, R., Whit- ing, P., Altherr, M.R., Wasmuth, J.J., Patterson, D. and Sikela, J.M., Human chromosomal localization of genes encoding the Yl and Y2 subunits of the y-aminobutyric acid receptor indicates that members of this gene family are often clustered in the genome, Proc. Natl. Acad. Sci. USA, 89 (1992) 5857-5861.

60 Wilson-Shaw, D., Robinson, M., Gambarana, C., Siegel, R.E. and Sikela, J.M., A novel y subunit of the GABA A receptor identi- fied using the polymerase chain reaction, FEBS Lett., 284 (1991) 211-215.

61 Wisden, W., Laurie, D.J., Monyer, H. and Seeburg, P.H., The distribution of 13 GABA A receptor subunit mRNAs in the ~'at brain. I. Telencephalon, diencephalon, mesencephalon, J. Neu- rosci., 12 (1992) 1040-1062.

62 Ymer, S., Draguhn, A., Wisden, W., Werner, P., Keinanen, K., Schofield, P., Sprengel, R., Pritchett, D. and Seeburg, P.H., Structural and functional characterization of the Yl subunit of the GABA A/benzodiazepine receptor, EMBO J., 9 (1990) 3261- 3267.

63 Zdilar, D., Rotter, A. and Frostholm, A., Expression of GABAA/benzodiazepine receptor a 1 subunit mRNAs and [3H]flunitrazepam binding sites during postnatal development of mouse cerebellum, Dee. Brain ICes., 61 (1991) 63-71.

64 Zdilar, D. Luntz-Leybman, V., Frostholm, A. and Rotter, A., Differential expression of GABA A / benzodiazepine receptor ill, /t2 and /33 subunit mRNAs in the developing mouse cerebellum, J. Comp. Neurol., in press.