GFAPβ mRNA Expression in the Normal Rat Brain and after Neuronal Injury

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
 5
 
  GFAPβ mRNA is an alternative transcript of the glial fibrillary acidic protein (GFAP) gene, whose transcriptional start site is located 169 nucleotides upstream to the classical GFAPα mRNA. By an RT-PCR method with primers on separate exons, we were
Share
Transcript
  Neurochemical Research, Vol. 24, No. 5, 1999, pp.  709-714 GFAPB mRNA  Expression  in the  Normal  Rat  Brain and  after Neuronal Injury D.  F. Condorelli, 1,2  V. G. Nicoletti, 1  P. Dell Albani, 1  V. Barresi, 1  A. Caruso, 1 S. G.  Conticello, 1  N.  Belluardo, 3  and A. M.  Giuffrida  Stella 1  Accepted  November  10,  1998) GFAPB  mRNA is an alternative transcript of the glial  fibrillary  acidic protein (GFAP) gene,whose transcriptional start site is located 169 nucleotides upstream to the classical GFAPA mRNA.  By an RT-PCR method with primers on separate exons, we were able to confirm thepresence of GFAP transcripts with a longer 5' untraslated region in all the examined areas of rat  brain and in primary cultures of astroglial cells. Northern blot analysis, using an oligoprobespecific for the 5' region of GFAPB, revealed a single hybridization band of 2.9 kb in all the brain  regions examined and in primary cultures of astroglial cells. The availability of the quan- titative  Northern blot assay allowed  further  studies on the regulation of GFAPB expression in  vivo. Since it is well-known that neuronal brain  injury  is one of the most powerful inducers of  GFAP, we examined the expression of GFAPA and B after a neurotoxic lesion in the rat hip- pocampus.  Results  obtained  show  a  parallel increase  in both  GFAP  transcripts  with  an  identi- cal  time-course, suggesting that regulatory regions  of the  gene influence  in  similar  way the  rateof transcription at the two different start  sites  (A and B) or that a similar post-transcriptionalmechanism  is  involved  in  regulating both mRNA isoforms. KEY  WORDS:  GFAP; gene; alternative splicing; intermediate filament. INTRODUCTION Glial fibrillary acidic protein (GFAP) is the major subunit  of  intermediate filaments  of  astroglial  cells  and it  has  been extensively used  as a  specific marker  for  this cellular  type in the  central  nervous  system (CNS)  (1,2). All  evidence indicates that  the  gene coding  for GFAP is present in a single copy in the genome ofmouse and man. However, different GFAP gene tran-scripts have been described, generated by usage of al-ternative transcriptional start site or differential splic- 1  Institute  of  Biochemistry, Faculty  of  Medicine, University  of  Cata- nia,  Italy. 2  Address reprint requests to: Prof. Daniele F. Condorelli, Istituto di Chimica  Biologica, Facolta  di  Medicina, Universita  di  Catania, Viale A.  Doria, 6 95125 Catania, Italy. Telephone +39 095  336990  FAX +39 095  333206.  Email: CONDORDA@ICTUNIV.UNICT.IT. 3  Institute  of  Human  Physiology,  University  of  Palermo,  Italy. 709 ing  (3,4).  The  transcriptional start site  of the  first  iden- tified  and  most abundant GFAP mRNA (GFAPA)  is located 13-14 bases upstream  the  translation start site.The transcript is 2.9 kb long and is encoded by 9 exonsspread over a 10 kb genomic region. Feinstein et al. (3) reported the  existence  of a second alternative  form  of GFAP mRNA (GFAPB), whose transcriptional startsite is located 169 nucleotides upstream of the GFAPAone; quantitative reverse transcription-polymerase chain  reaction (RT-PCR) showed that GFAP  B  mRNA is the  predominant GFAP mRNA form  in the  periph-eral nervous system, but represents only the 5-10% oftotal GFAP mRNA in the CNS (5).Nevertheless, Brenner (6) pointed out that most ofthe experiments characterizing GFAPB mRNA have de-pended upon RT-PCR, a technique highly susceptibileto  trace  contamination by genomic DNA. In order to provide  further evidence  for the  presence  of  this tran- 0364-3190/99/0500-0709 16.00/0  C  1999 Plenum Publishing Corporation  7 Condorelli et al script  and for its  size,  in the  present work, Northern blotanalysis was performed with a specific  oligoprobe  forthe 5' untraslated region of the  GFAPB  mRNA.Previous studies (5) on the regulation of  GFAPB expression were performed only in models in vitro,whose physiological meaning remains questionable.We  considered  interesting to extend the studies on reg- ulation  of  GFAPB expression  to in  vivo  experimental conditions. Since  it is  well established that brain  injury is the main inducer of  GFAP  mRNA expression (7,8,9),the regulation of both  GFAP  mRNA isoforms (a and (3) was  examined  in an  experimental model  of  neuro-toxic  injury. EXPERIMENTAL PROCEDURE Primary Rat Cell Cultures.  Astroglial cultures were prepared from  the  cerebral cortex, striatum, mesencephalon,  and  brainstem  ofnewborn  rats  (10).  The  meninges  were removed and brain  areas  dis-sected and passed through a sterile nylon  sieve  (pore size, 82 um) into  nutrient medium.  The  basal nutrient medium consisted  of  Dul- becco  modified  Eagle's  medium (DMEM), containing 10% heat- inactivated  foetal calf serum,  2 mM  glutamine, penicillin  (50  U/ml) and  streptomycin (0.05 mg/ml). Cells were seeded into plasticFalcon Petri dishes at a plating density of 0.5 x 10 5  cells/cm 2 . The cultures  were incubated at 37° C in a humidified 5%  CO 2 /95%  airatmosphere.  The  culture medium  was  changed after  6  days  and  thentwice a week. Cultured  astroglial cells were identified  by  immunocytochemi- cal  detection  of the  glial  fibrillary  acidic protein (GFAP),  an as- troglial specific marker. Cells, grown  on  35-mm diameter dishes, were  fixed  by 10 min  incubation  at  -20°  C in 5%  acetic  acid  inethanol or by  incubation  for 20 min in 4%  paraformaldehyde  in phosphate  buffered  saline,  pH  7.4, followed  by  permeabilization with 0.2 %  Triton X-100.  Two  different  preparations  of  rabbit polyclonal antibodies  against GFAP (one commercial, DAKO,  and the  otherprepared  in the  laboratory) gave identical results; both  an  immuno-fluorescence technique with fluorescein-labeled antibodies  or an im- munoperoxidase labeling technique (Peroxidase Vectastain Elite ABC  Kit, Vector Laboratories, Burlingam,  CA,  USA) were used withidentical results. RNA  Isolation, RT-PCR.  Total  RNA was  extracted  from  primaryastroglial cultures and brain areas, as described by Chomczynski andSacchi (11). In order to study GFAP A- and B-transcripts RT-PCR experiments  were performed using  a  cDNA obtained with MuLVReverse Transcriptase (RNA PCR Kit, Perkin Elmer) and randomhexamers.  The  cDNA  was  amplified with AmpliTaq  DNA  poly-merase. Specific oligonucleotide primers were designed: a forwardprimer  localized  on the GFAP exon I corresponding to bases +2 to+21 of the GFAP gene (El: GAAGCAGGGCAAGATGGAGC); an- other  forward primer corresponding to  bases  –19 to +1 (E1B: TGA-CATCCCAGGAGCCAGCA) and a reverse primer complementary to  bases  +1323  to  +1342  in  exon  II  (RP2:  AGGTTGGTTTCA- TCTTGGAG).  The primer localization on  different  exons  permits to differentiate  cDNA  or  genomic  DNA  amplification products, pre- venting  misinterpretation  of the  results.  The  amplification program was as  follows:  a  denaturation step (94°  C) of 2  min, followed  by35  cycles (94°  C for 45 sec and 68° C for 1  min)  and a  final  extension of  7 min at 68° C.  Amplification products were examined  by  elec-trophoresis  in  1.8% Agarose  IX TBE  gels  and  ethidium  bromide staining. Electrophoresis  and  Hybridization.  Total  RNA was  electro-phoresed through 1.1% agarose-2.2M formaldehyde gels, stained  withethidium  bromide, photographed under  UV  light,  and  then  blottedonto nitrocellulose (Hybond-C extra; Amersham). An  oligonucleotide corresponding to bases -53 to -10 of ratGFAP gene  was  designed  to  detect  rat  GFAPB mRNA  and an  oligonu-cleotide corresponding to bases -48 to -4 of  human  GFAP for  human GFAPB mRNA. Oligonucleotides were 3' end labeled  with terminal transferase  using [A- 32 P] dATP.  To  detect total GFAP  mRNA,  a  full length  cDNA probe (12)  was  linearized with HindIII. Probes were 32 P-labelled  by the  random-primed DNA-labelling method (13). Pre- hybridization and  hybridization were performed  as  described  in  Con-dorelli  et al.  (7). Blots were exposed  to  X-ray  film  at  -70°  C  using  in- tensifying  screens. Intrahippocampal  Injection  of  Ibotenic Acid.  Male albino Wis- tar  rats  (200-220  g body weight) were used. The animals were anes- thetized  with ether  and  placed  in a  stereotaxic apparatus. Ibotenicacid (0.05  M  final  concentration)  was  dissolved  in  phosphate buffered  saline (PBS)  and the pH  adjusted  to 7.2  with  a  small  volume of  NaOH.  For  intrahippocampal injections,  the  coordinates determi- nated  according to the atlas of Pellegrino et al. (14) were: AP: -3.8 mm  (from  bregma); vertical axis inclined  of 24° ; L:  1.9mm; V:7,5.5, 4 mm below dura. At each vertical point 0.5 ul of solution was injected.  Some animals were used  for  microscopic  examination  of the lesion (cresyl-violet staining) performed with standard histologicaltechniques.  As  previously reported (15), these experimental condi- tions  for unilateral intrahippocampal  ibotenic  acid  injection  produce a rapid decrease (-50% compared to control at 4 days post  injection) of the  activity  of the  enzyme glutamate decarboxylase,  a  neuronalmarker.  In the  same experimental conditions, glutamine synthetase activity, a  marker enzyme  of  astroglial cells, increases sharply  9 and 15  days  after  the  lesion. RESULTS RT-PCR Assay for GFAP mRNA.  As a  first  step, we  analyzed, by a RT-PCR assay, the presence of GFAPB  transcripts in the rat central nervous systemand in cultured glial cells. On the basis of the rat GFAPgene sequence, we designed three primers (Fig. la): areverse primer complementary to 20 nucleotides ofexon II (RP2); one forward primer (El) correspondingto  bases  +2 to +21 (exon I) and another forward primer(E1B)  corresponding  to  bases  -19 to +1 (+1  indicates the  GFAPA  transcription start  site).  In RT-PCR exper-iments primers E1 and RP2 detect both forms of GFAPmRNA (A and B), while the E1B/RP2 primer pair am- plifies  only longer alternative transcripts  of the  B-type. As  shown in  Fig.1b  the primer pair E1/RP2 and E1B/RP2  amplified PCR product of the expected  sizes(489 and 509 respectively) using cDNA  from  primary astroglial  cultures and 35 cycles of amplification. How- ever,  after 25 PCR cycles, the E1B/RP2 amplifica-tion product was not detectable by ethidium bromide  GFAPB  Expression  in the Rat  Brain 711 Fig. 1.  RT-PCR assay for GFAP  alternative  transcripts.  A schematic drawings of the GFAP 5' region indicating the positions of the primers isreported  in  (a).  Filled  boxes represent exons. Empty boxes indicate 5'-untranslated region  of  GFAPA mRNA. Single lines represent  intron  and 5' flanking sequences. The bent arrow indicates the transcriptional start site (+1). The translational start codon (ATG), the CAAT boxes, theTATA  box and the  -alternative starting site  are  also indicated.  The  results obtained  with  primers E1/RP2, that  simultaneously  amplify  bothGFAPA  and B  mRNAs,  are  indicated with  A  (All)  in  (b,c).  The  results obtained with primers E1B/RP2, that selectively  amplify  GFAPB  mRNA, are  indicated with  L  (Long)  in  b,c).  Total  RNA was  extracted  from  primary astroglial cultures  (b) and  different  brain regions (c). Cycle number (25 or 35) is  reported  on the top of the  corresponding lane  in  (b).  35  cycles were performed  in  assays reported  in  (c).  CX:  cerebral cortex;  ST: striatum;  CB:  cerebellum;  HP:  hypothalamus;  HC:  hippocampus;  OB:  olfactory  bulb;  BS:  brainstem.  M  (markers):  100 bp  ladder. staining, thus suggesting that the GFAPB mRNA level is  much  lower than that  of  GFAPA mRNA.  In  order  to detect GFAP transcripts subsequent assays were per- formed  at 35 cycles, a saturating condition that prevents quantitation.As  shown in Fig. 1c, RT-PCR experiments withprimer pairs  E1/RP2  and E1B/RP2 confirmed the  pres- ence of alternative transcripts of the a- and B-subtypes in  all examined brain areas (cerebral cortex, stria- tum,  cerebellum, hypothalamus, hippocampus, olfactory bulbs and  brainstem). Northern Analysis of  GFAPB  mRNA.  To  establish mRNA  size  of the  B-subtypes  of  GFAP transcripts  weperformed  Northern blot experiments using a specificantisense oligonucleotide probe complementary to bases -53 to –10 of the  GFAP  gene.  As  shown  in  Fig.  2, a  sin-gle hybridization band of 2.9 kb was clearly detected in all  brain  areas  examined (cerebral cortex, striatum,  cere- bellum, hippocampus, brainstem and olfactory bulbs) and  in primary astroglial cultures derived from various brain  regions  (cerebral cortex,  striatum, brainstem andmesencephalon). For a comparison the same sampleswere hybridized with a cDNA GFAP probe  which  de-tected all A/B transcripts. As expected, a single intense2.9 kb band was observed (Fig. 2). Therefore, the dif- ferent  mRNA size between the a- and B-isoforms is notenough to allow a resolution of the two transcripts in thesame Northern analysis (Fig. 2a). Since it has been pre-viously reported the B transcript represents only the5-10% of total GFAP mRNA level (5), results obtained using  the cDNA  probe  are mainly indicative of the level of  the  abundant  GFAPA  mRNA. Interestingly,  the  rankorder of GFAPB expression among the  different  cere- bral areas  brainstem >  olfactory  bulb > hippocampus >cerebellum  >  cerebral cortex  >  striatum)  is identical tothat observed for GFAP mRNA, thus suggesting a clearcorrelation between expression of GFAP A- and B-iso- forms  (Fig.2b). The same correlation was also present in  primary astroglial cultures derived  from  various cere-bral  areas  (Fig. 2c).Northern  blot  experiments were  also  performed with  RNA  extracted  from  human cerebral cortex  and  712 Condorelli et al Fig. 2.  Northern blot analysis of GFAP mRNA.  Using a specific  oligoprobe  for GFAPB  mRNA  a single band of 2.9 kb was observed in rat hippocampus (a) in  different  brain areas  of  adult  rat  brain  (b) and in  primary astroglial cultures prepared  from different  brain areas (c).  The same blots were stripped  and  reprobed with  a  full-length  GFAP cDNA (GFAP mRNA).  The  entire length  of the  autoradiogram  is  reported  in(a) to  show  the  identical position  of  GFAPB  and  total GFAP  mRNA  bands.  The  ethidium bromide stained ribosomal RNAs (28S  and  18S)  are also shown. CX: cerebral  cortex;  ST: striatum; CB: cerebellum; HC: hippocampus; BS; brainstem; OB: olfactory bulb; MES: mesencephalon. glioblastoma.  An oligoprobe, corresponding to  bases -48 to –4 of the  human  GFAP  gene (16), revealed  a band  of 2.9 kb, confirming the existence of GFAP mRNA  in the human  species  (data not shown). Up-Regulation  of  GFAPB  mRNA  after  Neuronal Injury.  The  availability  of the  quantitative Northernblot assay allowed  further  studies  on the  regulation  of GFAPB expression. Since GFAP gene expression  is strongly upregulated  by in  vivo neuronal  injury,  in the present study we investigated the effect of an experi-mental brain lesion on the expression of the GFAPBtranscripts. Ibotenic  acid,  a powerful toxin which de-stroys neuronal cell bodies but spares glial cells (17), was  injected unilaterally  in the  hippo-campus  of  adultrats and total RNA was  extracted  from lesioned and in-tact hippocampi  at  different days after injection.  As shown in Fig. 3 GFAPB mRNA increased in the le- sioned side at 3  days, reached  the  maximum  at 9  days and  persisted elevated until  15  days after injection.The quantitative and temporal pattern of induction isidentical to that observed for GFAPA transcripts. DISCUSSION The  first  evidence  for the  existence  of the  alter- native  transcript called GFAPB  was  found  by  Feinsteinet al. (3) in peripheral glial cells. In a subsequent work, the  same authors  (5)  reported that  low  levels  of  thistranscript were also detectable in the central nervoussystem by the sensitive RT-PCR assay. However, it was  argued that  the  extreme sensitivity  of  this assay was  also one of its limits, due to the possible amplifi-cation  of  minute amount  of  contaminating genomic DNA  (6). In the present study, the primers for the RT-PCR assay were chosen on separate exons, in order toeasily distinguish  the  amplification  of  cDNA  from  that of  contaminating genomic DNA. With this method, wewere able to confirm the results of Galea et al. (5), showing  the  presence  of  GFAP transcripts  with  a longer 5' untraslated region in all the examined re-gions of rat brain and in primary cultures of astroglial cells.  This observation is also supported by the resultsobtained  by  Northern blot analysis using  an  oligoprobe  GFAPB Expression  in the Rat  Brain 713 Fig. 3.  Northern blot analysis of  GFAPB  mRNA  after  neurotoxicbrain lesion.  Total  RNA was  extracted  from  lesioned hippocampus(L) and contralateral unlesioned hippocampus (C) at different days (3, 9, 15)  after  the  unilateral injection  of  ibotenic acid. Similarresults were obtained  in two  different  experiments. specific  for the 5' region of GFAPB. This probe reveals a  single band  of  hybridization  of 2.9 kb in all the  brainregions examined and in primary cultures of astroglialcells. The approximate size of this message is identi- cal to  that  of the  classical  and  abundant form  of  GFAP mRNA  (GFAPA). Indeed, the  resolution power  of the electrophoretic separation used for the Northern blot analysis  is not enough to resolve two transcripts dif- fering  only for  0.1-0.2kb.  By primer extension exper- iments  Feinstein et al. (3) suggested that GFAPB is 169  nucleotides  longer than GFAPA. However,  it has pointed out that the size of primer-extension products can  be underestimated due to secondary structures thatcause  a  premature stop  of  transcription.  Our  Northernblot results suggest that the real size of GFAPB mRNAcannot  be  very different from that determined  by Fe- instein  et al.  (3).  Of  course, these data  do not  exclude that  another small exon, upstream exon I, could takepart to the structure of GFAPB. However, Northernblot analysis with probes recognizing the 5' flankingregion  of the GFAP gene  (18)  indicates that no  exon  is present  in the  1.8-kb  5'  region upstream exon  I  (data not  shown).The presence of this longer alternative transcript isnot restricted to rodents. Using a specific oligoprobe for the  region  48/ 4  nucleotides  5' of the  classical  tran- scriptional  start site,  we can  detect  a  clear band  ofhybridization in  samples derived  from  human cerebralcortex and  from  tumors of glial srcin. During the preparation of  this manuscript Riol  et al.  (19) reported the  presence  of  GFAPB mRNA  in the  human brain  by RT-PCR. Although we  have very  few  indications  on the  pos-sible physiological meaning of GFAPB, it has been sug-gested that regulation of its expression can be different from  that  of  GFAPA.  Galea  et al. (5)  reported  thatthe A- and B-isoforms can be differentially regulated in  vitro: GFAPB mRNA levels increase (cultured astro-cytes) or remain unaffected (brain slices), while GFAPA mRNA  slightly (astrocytes) or dramatically (brain slices)decreases  after  interferon-G treatment. Therefore,  in both experimental conditions the GFAPB/GFAPA ratio is  increased  by  interferon-G.  Since  it is  well-known thatneuronal brain  injury  is one of the  most powerful  in- ducers of GFAP expression in the brain (9), we exam-ined the expression of GFAPA and B  after  a neurotoxiclesion induced by ibotenic acid in the rat hippocampus. Results obtained  show  a  parallel increase  in  both GFAPtranscripts with an identical time-course, suggestingthat regulatory regions  of the  gene influence  in  similar way the  rate  of  transcription  at the two  different  startsites (A and B) or that a similar post-transcriptionalmechanism  is  involved  in  regulating both mRNA iso- forms.  It is interesting that the two zones of the 5'  flank- ing region of the GFAP gene involved in the ciliary neu-rotrophic factor (CNTF)-induced GFAP expression areupstream to both transcriptional start sites and CNTFmay  represent  one of the signals that  trigger  the  GFAP upregulation  and reactive astrogliosis (20,21).The different secondary structure of the 5' untrans-lated sequences might influence translation and Fein-stein et al. (3) suggested that the distinct  5'-leader length could influence  the  choice  of two  possible  startcodons present in GFAP mRNA. GFAPB mRNA trans-lation could initiate at the downstream AUG, producing a  protein identical  to the  full-length form  of the  protein, but  lacking  the  first  17  amino acids. Evidence  for the existence  of a  shorter GFAP  protein  have been providedby Feinstein et al. (3) and Riol et al. (19).In conclusion in the present work we provide fur-ther evidence for the existence of longer GFAP tran-scripts of the B-type and show a parallel upregulation of  GFAPA  and B  after brain injury. Further studies  are necessary  to  confirm  the  influence  of the  different mRNA  leader sequences on translation and the role ofthe putative shorter GFAP protein. ACKNOWLEDGMENT This work  was  partially supported  by a  grant  from  the  Italian  Consiglio Nazionale delle Ricerche REFERENCES 1.  Eng,  L. F.,  Vanderheagen,  J. J.,  Bignami,  A., and  Gerstl,  B.1971. An  acidic protein isolated  from fibrous  astrocytes.  Brain Res 28:351–354. 2.  Bignami, A., Eng, L. F., Dahl, D., and Uyeda, C.T . 1972. Local- ization of the  glial  fibrillary  acidic protein  in  astrocytes  by im-munofluorescence.  Brain Res 43:429–435.
Related Search
Similar documents
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x