The Glial Glutamate Transporter Complementarv DNA in Patiehts with Amyotrophic Lateral Sclerosis Thomas Meyer, MD,* Astrid Speer, MD, PhD,t Bientje Meyer, MD,’ Wolfgang Sitte, MD,* Gerald Kiither, MD,$ and Albert C. Ludolph, MD, PhD*

Here, we report a mutation screening by single-stranded conformational analysis of the astroglial human brain glutamate transporter (HBGT) I1 complementary DNA in patients with amyotrophic lateral sclerosis. The confor- mational analysis data indicate a lack of sequence varia- tions in the HBGT I1 coding region in 6 patients with amyotrophic lateral sclerosis and the same number of nonneurological control subjects. In both groups, three variants of the HBGT I1 5’ untranslated region were iso- lated. We have no evidence that the reported complemen- tary DNA variants are disease specific.

Meyer T , Speer A, Meyer B, Sitte W , Kiither G, Ludolph AC. The glial glutamate transporter

complementary DNA in patients with amyotrophic lateral sclerosis. Ann Neurol 1996;40:456-459

Glutamatergic neurotransmission is followed by an im- mediate removal of glutamate from the synaptic cleft. High-affinity glutamate carriers clear the excitatory neurotransmitter glutamate from the synapse [I]. The glutamate reuptake is known to be critical, as an ele- vated synaptic glutamate level results in postsynaptic neuronal death through excitotoxic mechanisms [2]. Molecular cloning studies led to the isolation of a neuron-specific [ 3 ] , an astroglial [4] , and a third trans- porter complementary DNA (cDNA) in both glia and neurons [5]. Most recently, another aspartate/gluta- mate transporter, which is expressed predominantly in the cerebellum, was cloned [6]. The subtypes of gluta- mate transporters have been imniunohistochemically localized to distinct regions of the central nervous sys- tem [71.

Pharmacological inhibition of glutamate transporters reportedly results in a model of chronic anterior horn

From the Departments of *Neurology and ?Internal Medicine- Molecular Biology, Humboldt University (ICharitit), Berlin, and the $Department of Neurology, Medizinische Hochschule, Hannover, Germany.

Received Feb 7, 1996, and in revised forrn Mar 28. Accepted for publication Apr 1, 1996.

Address correspondence to Dr Ludolph, Department of Neurology, Humboldt University (Chariti.), Schumannstr. 20/21, 101 17 Ber- lin, Germany.

cell degeneration [S] . Deficient glutamate transport has been observed in synaptosomes from motor cortex and spinal cord in patients with amyotrophic lateral sclero- sis (ALS) [9]. More recently, Rothstein and coworkers [lo] showed a substantial loss of immunoreactive glial transporter protein in ALS brain tissue. The selective reduction of the astroglial glutamate transporter may be responsible for the glutamate transport defect re- ported in ALS patients [ 101.

Here we report a mutation screening of the astroglial glutamate transporter cDNA isolated from ALS and control brain tissue. We screened the human brain glu- tamate transporter (HBGT) I1 coding region for possi- ble sequence abnormalities. From normal and diseased tissue, 5’-end variants of the HBGT I1 cDNA were isolated. The presence of the cDNA variants in the ALS tissue was studied.

Materials and Methods Material Brain and spinal cord tissues from 5 patients with sporadic ALS and 1 patient with familial ALS were obtained by au- topsy between 5 and 16 hours (mean, 9.2 hours) post mor- tem. The patients were between 52 and 68 years old (mean, 61.0 years). Brain and spinal cord tissues from 6 nonneuro- logical individuals served as control samples. The mean age of the control patients was 57.0 years (range, 50-75 years). The mean postmortem delay in this group was 20.2 hours (range, 12-26 hours). The cause of death of these patients was myocardial infarction or neoplasma without central ner- vous system (CNS) involvement. The control specimens were kindly provided by the Medical Research Council Alz- heimer’s Disease Brain Bank, London. Motor cortex speci- mens from 3 patients with sporadic ALS and the patient with familial ALS, spinal cord tissues from 2 patients with sporadic ALS, and control brain and spinal cord tissues were shock frozen and stored at -80°C until use. All ALS patients studied here met the international clinical criteria (“El Esco- rial”) for the diagnosis of ALS [ 1 11.

Methods The guanidinium thiocyanate method as described [ 121 was used for total RNA isolation from 1 gm of each of the tissue specimens. Based on the sequence information of the glial glutamate transporter HBGT I1 cDNA [13], we designed primers for reverse transcription of HBGT I1 messenger RNA (mRNA) and subsequent polymerase chain reaction (PCR) amplifications. Reverse transcription of the coding re- gion and the 5’ untranslated sequence was carried out using murine Moloney Murine Leukemia Virus Reverse Tran- scriptase (Gibco-BRL) followed by 30 cycles of PCR ampli- fication ( Taq DNA polymerase, Boehringer Mannheim). The reactions were denatured at 94°C for 1 minute followed by 30 cycles of denaturing at 94°C for 1 minute, annealing at 60°C for 1 minute, and extension for 2 minutes at 72°C. The first amplification products were used as template for a second PCR amplification. For single-stranded conforma- tional analysis (SSCA) mutation screening, we generated nine fragments with a maximum number of 350 bp (primer se-

456 Copyright 0 1996 by the American Neurological Association

-173 -129 +1

HBGT IIC b\\\\\\\l ORF

HBGT IIB

HBGT IIA

-125 -96 -12 +1

ORF

- 40 -11 +1

ORF+5AA

HBGT I1 -159 -130

Fig 1. 5' Heterogeneit~ o f the glial glutamate transporter complementary DNA human brain glutamate transporter (HBGT) II. Schematic representation of the 5' noncoding sequences of the transporter variants in relation to the previously described HBGT 11 sequence [13]. Diagrammatically shown is a diversity in size and sequence of the 5' untranslated region (UTR) and the 5' part o f the coding region in HBGT IIA. UTR mot$ that are homologous to the HBGT 11 sequence are displayed a( gray; the position o f the open reading fiame (0 RF), as dark gray rectangles. Novel sequences are shaded. The 5' junking primer sequence is not shown. The numbers indicate nucleotide (nt) positions of the corresponding regions o f homology in HBGT 11 and the HBGT 11 variants. I n HBGT IIA, nt -40 to -12 correspond to the absolute 5' end o f the HBGT 11 UTR whereas the sequence - I I to - I is homologous to the same region in the HBGT 11 UTR. I n HBGT IIB and HBGT IIC the 5' UTRs are composed o f novel and previously described HBGT 11 sequences

quences available on request to authors). The amplification of the glial transporter 5' end was achieved by primers posi- tioned at the absolute 5' end of the untranslated region (UTR, nucleotide [nt] -178 to -161) and the 5' part of the coding domain (nt 379-400). The PCR products again were used as template for a second and seminested amplifi- cation (nested primer: nt 190-213). For reverse transcription we used the primer PI; for first amplification, P1 and P2. Seminested amplification was performed with P2 and P3 (Pl : 5' GGATAGCCAAGACCAGAATGAC; P2: 5' CAGC TGCCGACTCCGCTG; and P3: 5' GAAGGCTATTAAC ATAACCACATC). The PCR products were run on a 5% nondenaturating acrylamide gel at 20°C, 10°C, and 4"C, followed by silver staining.

The sequence of the cDNA fragments was determined as direct sequencing of purified PCR products (Sequenase 2.0, USB) using the dideoxy chain termination method [14]. For sequence confirmation, both strands of the DNA were se- quenced. To study the presence of the HBGT IIA to IIC variants in ALS and control tissue, we designed variant- specific primers (for HBGT IIA: 5' GACGTGGGAGG ATGGGTGC; HBGT IIB: 5' GGTAGAATTCAGCGCT GTACC; and HBGT IIC: 5' AGGAGAGGTTAGATGTC- AGCAG).

Results From both patient and control tissues the cDNA of the transporter HBGT I1 was generated and amplified.

SSCA of nine overlapping cDNA fragments that cover the complete coding region of the transporter cDNA resulted in an identical migration pattern. The SSCA data indicate a lack of sequence variations such as mu- tations and polymorphisms in the HBGT I1 coding region of the patients and control subjects studied here.

Gel electrophoresis and direct sequencing revealed a diversity in size and sequence of the 5' amplification products in normal and ALS tissues. The amplification product (P2-3) generated at the 5' end of the trans- porter cDNA was expected to cover 213 nt of the cod- ing region and 178 nt of the 5' UTR with a total size of 391 nt. None of the PCR products in 6 control subjects and the same number of diseased individuals showed the predicted size. In contrast, we amplified a cDNA fragment of 271 nt from both patients and con- trols. Two other PCR products with 356 and 404 nt were found in motor cortex tissue of single control individuals. W e propose to name the transcript variants HBGT IIA, H B G T IIB, and HBGT IIC.

In H B G T IIA the 5' UTR consists of 40 n t (if sub- tracted the primer sequence) that show complete ho- mology to two regions of the HBGT I1 5' UTR. It is lacking a 119-nt-long fragment of the previously de- scribed UTR (Fig 1). Apparently, H B G T IIA encodes for a protein that possesses five additional N-terminal

Brief Communication: Meyer et al: Glial Glutamate Transporter cDNA in ALS 457

HBGT IIA M G A N N M

5' acctgggacccccagacgtgggagg atg ggt gcc aac aat atg

HBGT IIB M G W V C L P T G * .

5' acctgggaccctccagacgtgggagg atg ggg TGG GTG TGC CTG CCT ACT GGT TGA CGGTCAGGTGGGCTGATG~CTCCCTGCTACAGTGGTAG~~CAGCGCTGTACCTagtgcc~caa~tg

HBGT IIC M G G V P A C E L G G "

5' TGACCCAACGTGGAGG ATG GGT GGT GTG CCT GCC TGT GAG 'ITT Ggg ggg tga cctagtgtcttccag tagggcaatc tctggacatctttatctccccagtacctctccagatcctg~gcctggggccagggaggagagg~gatgtcagcagtggaaggacagtgcc~caa~tg

Fig 2. Nucleotide (nt) sequence proJle ~lf the 5' untranslated region (UTR) variants human brain glutamate transporter (HBGT) IIA, HBGT IIB, and HBGT IIC (5' $anking primer sequence not shown). Lower case is used for HBGT II homologous sequences. A TG codons shown in bold indicate upstream translation start codons. In HBGT IIA, an additional translation start codon is positioned 15 nt upstream fionz the authentic ATG. The top strand shows a one-letter display ofthe deduced nmino acid sequence o f the additional N-terminal amino acid motif and the small upstream open reading frdmes.

Distribution o f the 5' End of Human Brain Glutamate Transporter (HBGT) II Variants in ALS Patients and Controls'

Patients Controls

1 2 3 4" 5' 6' 7 8 9 10 11 12 -

H B G T I I A + + + O + + + + + + + + H B G T I I B + 0 0 + + 0 + 0 + 0 + + H B G T I I C + 0 + 0 + + + 0 + 0 + +

~~~ ~

'Distribution of the 5'-end HBGT I1 variants was determined by the presence of polymerase chain reaction amplification products using primers specific for glial glutamate transporter 5' variants HBGT IIA 10 IIC. Brain tissue samples from 6 ALS patients (num- bered 1-6) and 6 nonneurological controls (7-12) were investi- p ted : , Familial form of ALS.

'Spinal cord tissue.

+ = amplifiable; 0 = not amplifiable; A13 = amyotrophic lareral sclerosis.

amino acid residues (MGANN) if compared to the protein deduced from the HBGT I1 cDNA (Fig 2) [13]. A second and a third variant, HBGT IIB and HBGT IIC, are composed of novel and previously de- scribed HBGT I1 UTR sequences (see Fig 1).

HBGT IIA was isolated in tissues from 5 of the 6 patients and each of the control subjects, whereas HBGT IIB and HBGT IIC were found in a single control individual. The amplification with variant- specific primers revealed a broalder presence of the HBGT IIB and HBGT IIC transcripts in both patients and control subjects: HBGT IIB was present in half of the patients and in 4 of the 6 control subjects. In 4 patients and the same number of control subjects, HBGT IIC could be amplified. Amplification demon- strated HBGT IIA in 5 of the 6 patients and in all 6 controls (Table). Therefore, none of the transcript vari-

ants HBGT IIA through IIC is disease specific. The amplification from 7 control spinal cord specimens showed the presence of HBGT IIA in all individuals but a lower frequency of HBGT IIB (in 4 specimens) and HBGT IIC (in 2 specimens).

Discussion Functional abnormalities of high-affinity glutamate transport had been reported in patients with sporadic ALS [9]. Furthermore, a dramatic decrease of the astro- glial transporter protein has been detected in CNS re- gions typically involved in this disease [lo]. Our data suggest that sequence alterations of the HBGT I1 cod- ing region are unlikely to be the primary cause of the deficient glutamate transport reported in ALS.

Previous reports [13, 15, 161 and our own observa- tions indicate that the 5' UTR region and the 5' end of the protein-coding domain of the glial glutamate transporter cDNAs are variable. The abundance of these variants in human brain remains to be demon- strated. Since data on the genomic structural basis of the transporter mRNA are not complete yet, we have to consider that the variant HBGT IIB may actually arise from an unspliced mRNA precursor. However, this possibility does not affect our conclusions with re- gard to ALS. Alternative 5' UTRs are known to be involved in posttranscriptional expression regulation such as inhibition of translation, alteration of mRNA stability, ot binding of regulatory cytosolic factors [ 17, 181. The functional consequences of the 5' divergence of the astroglial glutamate transporter have yet to be demonstrated.

Studies of the HBGT I1 variants in patients and control subjects revealed that the 5' heterogeneity of the glial transporter cDNA does not seem to be disease

458 Annals of Neurology Vol 410 No 3 September 1996

specific. Our findings contribute to the hypothesis that a still unknown cellular dysfunction may lead to the selective vulnerability of the astroglial glutamate trans- port [IO]. In future studies, patients with a biochemi- cally or immunohistochemically defined transport de- fect should be screened. Furthermore, quantitative investigations of the transporter mRNA expression will allow us to distinguish between a transcriptional and posttranscriptional cause of the reported loss of trans- porter expression in ALS. Further work is required to identify a functional role of each of the transporter variants and their eventual differential expression in normal and diseased tissue.

This work was supported by a research grant for Drs Ludolph, Speer, and Meyer from the Deutsche Forschungsgemeinschaft (grant Lu 336/5-1) and the Universitatsklinikum Charit6 at Hum- boldt University, Berlin.

We cordially thank Dr N. Cairns at the Medical Research Council’s Alzheimer’s Disease Brain Bank for supporting the project by pro- viding normal postmortem brain and spinal cord tissue. We are grateful to Dr G. Kuther at the Medizinische Hochschule Hannover and Dr A. Weindl at the Technical University, Munich, for provid- ing ALS brain and spinal cord tissue.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

References 1. Nicholls D, Atnvell D. The release and uptake of excitatory

amino acids. Trends Pharmacol Sci 1990;11:462-468 Choi DW, Maulucci-Gedde M, Kriegstein AR. Glutamate neurotoxicity in cortical cell culture. J Neurosci 1987;7:357- 368 Kanai Y, Hediger M. Primary structure and functional charac- terisation of a high-affinity glutamate transporter. Nature

Pines G, Danboldt N, Bjoras M, et al. Cloning and expression of a rat brain L-glutamate transporter. Nature 1992;360:464- 467 Storck T , Schulte S, Hofmann T, Stoffel W. Structure, expres- sion and functional analysis of a Na+-dependent glutamate/ aspartate transporter from rat brain. Proc Natl Acad Sci USA

Fairman W, Vandenberg R, Arriza J, et al. An excitatory amino-acid transporter with properties of a ligand-gated chlo- ride channel. Nature 1995;375:599-604

1992;360:467-471

1992;89:10955-10959

13. Shashidharan P, Wittenberg I, Plairakis A. Molecular cloning of human brain glutamate/aspartate transporrer 11. Biochim Bi- ophys Acta 1994;1191:393-396

14. Sanger F, Niklen S, Coulsen A. DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 1977;74:

15. Arriza J, Fairman W, Wadiche J, et al. Functional comparisons of three glutamate transporter subtypes cloned from human motor cortex. J Neurosci 1994;14:5559-5569

16. Manfras B, Rudert W, Trucco M, et al. Cloning and character- ization of a glutamate transporter cDNA from human brain and pancreas. Biochim Biophys Acta 1994;195:185-188

17. Pierrat B, Lacroute F, Losson R. The 5‘-untranslated region of the PRPl regulatory gene dictates rapid mRNA decay in yeast. Gene 1993;131:43-51

18. Romeo D, Park K, Roberts A, et al. An element of the trans- forming growth factor beta 1 5’-untranslated region represses translation and specifically binds a cytosolic factor. Mol Endo- crinol 1993;7:759-766

5463-5467

MELAS Associated with a Mutation in the Valine Transfer RNA Gene of Mitochondria1 DNA R. W. Taylor, PhD,* P. F. Chinnery, MRCP,* F. Haldane,* A. A. M. Morris, MRCP,* L. A. Bindoff, MRCP,* 1. Wilson, FRCP,? and D. M. Turnbull, FRCP*

~ _ _ _ _ _ _ ~ ~~

We describe a patient with the mitochondrial myopathy, encephalopathy, lactic acidosis, and strokelike episodes (MELAS) phenotype in whom initial investigations in skeletal muscle failed to show any histochemical or bio- chemical defect. Subsequent analyis of the mitochondrial genome identified a new heteroplasmic mutation in the valine transfer RNA gene, the first described in this re- gion.

Taylor RW, Chinnery PF, Haldane F, Morris AAM, Bindoff LA, Wilson 1, Turnbull DM.

MELAS associated with a mutation in the valine Rothstein JD, Martin L, Levey AI, et al. Localization of neu- ronal and glial glutamate transporters. Neuron 1994;13:713- 725

transfer RNA gene of mitochondrial DNA. Ann Neurol 1996;40:459-462

Rorhstein JD, Jin L, Dykes-Hoberg M, Kuncl R. Chronic inhi- bition of glutamate uptake produces a model of slow toxicity. Proc Natl Acad Sci USA 1993;90:6591-6595 Rothstein ID, Martin LJ, Kuncl RW. Decreased glutamate transport by the brain and spinal cord in ALS. N Engl J Med

Rothstein JD, Van Kammen M, Levey AI, et al. Selective loss of glial glutamate transporter GLT-1 in amyotrophic lateral sclerosis. Ann Neurol 1995;38:73-84 Tandan R. Clinical features and differential diagnosis of classi- cal motor neuron disease. In: Williams AC, ed. Motor neuron disease. London: Chapman and Hall Medical, 1974:l-27 Chomczynski P, Sacchi A. Single step method of RNA isola- tion by acid guanidinium thiocyanate-phenol-chloroform ex- traction. Anal Biochem 1987;162: 156-159

1992;326: 1464-1468

Over 80% of cases of mitochondrial myopathy, en- cephalopathy, lactic acidosis, and strokelike episodes (MELAS) are associated with an A-to-G substitution at position 3243 in the tRNAhU(UUR) gene of the mito-

From the *Department of Neurology, University of Newcastle upon Tyne, Newcastle upon Tyne, and ?Great Ormond Street Hospital, London, United Kingdom.

Received Nov 8, 1995, and in revised form Mar 25, 1996. Accepted for publication Apr 1, 1996. Address correspondence to Prof Turnbull, Department of Neurol- ogy, The Medical School, Framlington Place, Newcastle upon Tyne, NE2 4HH, United Kngdom.

Copyright 0 1996 by the American Neurological Association 459