Nervous System-Specific Proteins

15. Wilkinson, K. I). (1087) Anti-Cancer Drug Des. 2, 2 1 1-220 16. Jentsch, S.. Seufert. W. & IIauser, H.-1’. (1991) Hiochim. Hiophys. Acta 1089, 127-1 30 17. Finley, L)., Ozkaynak, E. & Varshavsky, A. (1987)

18.

10.

20. 21.

22.

23.

24.

Cell 48. 1035-1040 Knight, E., Jr.. Fahey, I)., Cordova, H., Hillman. M.. Kutney. K.. Keich, N. & Hlomstrom, L). ( 1 988) J. Hiol. Chem. 263,4520-4522 Wilkinson, K. I)., Cox, M. J., Mayer, A. N. & Frey, T. ( 1080) Hiochernistry 25, 0044-hh40 Mayer, A. N. & Wilkinson, K. I). (1080) Hiochemistry 28, 1 Oh- 172 Wilkinson. K. I).. I,ee. K., I Ieshpande, S.. Iluerksenllughes. I’., Hoss, J. M. & I’ohl. J. (1080) Science 246, 670-073 Miller, H. I., Henzel, W. J., Kidgeway, J. IL, Kuang, W., Chisholm. V. & I,iu. C. (1080) Hiotechnology 7. 008-704 Wilson, 1’. 0. G., Barber, 1’. C., Hamid, Q. A., Power, 13. F., Dhillon, A. I’., Rode, J., Day, I. N. M., Thompson. K. J. & l’olak9 J. M. (1088) Hr. J. Exp. I’athol. 69, 01-104 I)ay, I. N. M., Hinks, I,. J. & Thompson, K. J. (1090) fhchem. J. 268, 521-524

25. Cole, G. M. & Timiras, 1’. S. (1987) Neurosci. I x t t . 79,207-2 12 26. Perry, G.. Friedman, K..Shaw. G. & Chau, V. (1 087) Proc. Natl. Acad. Sci. USA. 84, 3033-3036 27. Mayer, J., Arnold, J., Laszlo, I,., 1,andon. M. & I,owe, J. (1991) Biochim. Hiophys. Acta 1089, 141-157 28. Cox, M. J., Haas, A. I,. & Wilkinson, K. I). (1086) Arch. Riochem. Biophys. 250, 400-400 29. Hadari. T., Warms, J. V., Rose, I. A. & Hershko, A. ( 1992)J. Riol. Chem. 267,7 19-727 30. Lowe, J., McDermott. H., Imclon, M., Mayer, K.J. & Wilkinson, K. D. (1990) J. I’athol. 161. 153-100 31. Kwak, S., Masaki, T., Ishiura, S. & Sugita. H. (1001) Neurosci. Lett. 128, 21-24 32. Heggie, P.. Hurdon. T.. Lowe, J.. Idandon,M., Ixnnox, G.,Jefferson. I). & Mayer, K. J. (1 980) Neurosci. 1,ett. 102,343-348 33. Nowak, T. S.Jr., Hond. U. & Schlesinger, M. J. (1000) J. Neurochem. 54,451-458 34. tlonore, H., Kasmussen, H. H., Vandekerckhove, J. & Celis, J. E (1991) FEHS Lett. 280, 235-240 35. Hizzi, A,, Schaetzle, H., I’atton, A., Gambetti. 1’. & Autilio Gambetti, 1,. (1001) Brain Kes. 548, 202-200 Received 16 April 1992

Enolases and PGP9.5 as tissue-specific markers I. N. M. Day University Clinical Biochemistry, Level D, South Laboratory and Pathology Block, Southampton General Hospital, Tremona Road, Southampton SO9 4XY, U.K. T h r e e isoenzymes of enolase occur in man and in other mammals. Evidence has accumulated from biochemical, genetic, somatic cell genetic and molecular genetic studies that there are at least three different genes (see [ 1 I). Various nomenclatures have been used previously and these have been clarified recently [ 1 I. 1,ocus E N 0 1 expresses a subunit, locus E N 0 2 expresses y subunit and locus E N 0 3 expresses /3 subunit: enolase subunits form homo- o r heterodirners. aa-enolase also being known as non-neuronal enolase (NNE), PP-enolase as muscle-specific enolase (MSE) and y y-enolase as neuron-specific enolase (NSE). Many neurons express predominantly y subunit, whereas some neurons and various neuroendocrine cells express a mixture of y and a subunits, and skeletal and cardiac muscle express predominantly /3 subunit [ 2, 31: in ziivo all possible diniers except P y have been

Abbreviations used: MSE. muscle-specific enolase; NNE, non-neuronal enolase; NSE, neuron-specific enolase; I U l , ubiquitin C-terminal hydrolase.

observed. T h e level of expression of /3 subunit seems to correlate with fibre type, being higher in glycolytic (type 2) fibres [ 3 ] . T h e expression of a unique isoenzyme of enolase in neurons first became clear when the brain-specific protein 14-3-2 identified by onedimensional PAGE was shown to possess enolase activity [4]. More recently, the application of twodimensional electrophoresis to the comparison of different human tissues in a search for novel cellular markers resulted in the identification of a protein (designated PGP9.5) specific to the cytosol of almost all neuronal types [S] at a concentration similar to, o r higher than, that of NSE (1 -5% w/w of total soluble protein). Protein and c D N A sequencing [6], followed by database entry of a n homologous protein [7], resulted in the identification of PGP9.5 as the neuron-specific member amongst (apparently) three similar isoenzymes acting as ubiquitin C-terminal hydrolases. Further consideration of PGP9.5 expression as a model for neuronal gene activation is given in another paper in this issue [8]. Although there may be 50000

I992

637

Biochemical Society Transactions

638

human genes [9], there can be few gene products expressed at levels comparable with NSE or PGP9.5 and marking most neurons. The existence of isoproteins and isogenes can have advantages at one or more levels (e.g. enzymological, regulation, control of gene expression) and many different examples have been identified (for an overview, see [ 101). The advent of molecular cloning has enabled more detailed investigations of structural differences between isoproteins and of the evolution and differential regulation of members of a gene family. Such investigations should enable a better understanding of NSE and PGP9.5 at the protein and gene levels, in the context of both the evolution and regulation of their gene families, in relative terms in their context within the neuron, and also in their clinical application as cell-specific markers. I Iere, some recent work in this laboratory is reviewed. We have previously described protein sequence and cI)NA analysis for human NSE and PGP9.S [h, I l l , and also isolation and characterization of human genomic clones representing both PGP9.5 [I21 and MSE [13, 141. The NSE locus is now designated ubiquitin C-terminal hydrolase I,1 (UCH1,l) [ 1.51. Characterization of the MSE locus enabled a direct prediction of both protein and mRNA sequences [ 131. Our analysis of human NNE D N A clones (L. J. Hinks & I. N. M. Day, unpublished work) is in good agreement with that of Giallongo et al. [ l h ] . Human NSE cDNA sequence from other laboratories is in close agreement with our analysis. In addition, MSE cDNA analysis [17] is consistent with our analysis at the genomic level. The sequencing of loci E N 0 1 [18] and E N 0 2 [ 191 appears to complete the set for active enolase loci in man. Parallel sets of investigations have taken place for rat (refs in [ZO]) and, less completely, for several other species. This collection of data and cloned DNA fragments has opened the way for many possible investigations. Two avenues are considered here.

Structural and immunochemical aspects of the protein, NSE The solubility in plasma, patterns of cellular specificity and high level of expression of major isoenzymes lends them, in the field of clinical biochemistry, to development as diagnostic markers. Indeed, this was one of the original reasons for the discovery of PGP9.5 [S], which has proved particularly valuable in the immunohistochemical detection of very fine nerve processes in tissue sections 1211. However, as well as their

Volume 20

capacity for marking neurons, NSE and PGP9.5 have both been found to mark particular tumour cells [22, 231. Whether this represents de nono activation of E N 0 2 or UCH1,I genes in some tumour cells, or whether the tumour cells have amplified from cells initially expressing these genes to some extent, is not known. The major potential clinical application of NSE is in the monitoring of small cell lung cancer [24 1, which frequently possesses other neuroendocrine characteristics [25]. In this instance, at least, it is possible that the tumours arise from neuroendocrine cells lining the normal lung [25]. Small cell lung cancer accounts for 24% of all lung tumours, which themselves account for 26.8% of all cancer deaths in the U.K. [26]. At present there are no established biochemical markers either for screening, diagnosis, monitoring or prognosis. In addition little is understood of the genetic regulation of small cell tumours. Conventional polyclonal immunoassays have been used to monitor tumourderived NSE in the serum of patients undergoing therapy [241. More recently, several monoclonal antibodies to NSE have been described [27-20]. W e have reconstructed a full-length NSE cDNA from partial-length cDNA clones, first to adopt a more directly structural approach to further the immunochemical analysis of NSE, by antigen expression, and secondly to enable in vitro expression and analysis of functional NSE. In the long term the latter approach will be of particular interest with respect to the neuronal environment, for example in relation to the known chloride resistance of NSE [ 301 and axoplasmic movement of NSE [ 3 11 versus possible functional specializations of the other isoproteins (e.g. [ 321). The reconstructed cDNA, verified by restriction mapping, sequencing and expression as a full-length fusion protein (G. W. Quinn & I. N. M. Day, unpublished work), was expressed using several prokaryotic vectors including plJC19, plJEX and derivatives [33). Selected restriction fragment subclones and clones generated by sonication-shotgun cloning were also analysed. Standard colony blotting and Western blotting of SDS-PAGE of colony extracts using several rabbit and sheep polyclonal antisera to human NSE were employed to analyse antigenic and epitope regions in native and recombinant NSE. The results of these studies are shown in Fig. 1. The recent determination at 2.25A of the crystal structure of Saccharomyces cermisiae enolase-1 [34] and the direct alignment possible between human enolases and this sequence [14] make it possible to try to interpret our immunochemical data and human sequence comparisons in

Nervous System-Specific Proteins

Fig. 1

Identification of major epitope regions of human NSE by cDNA expression The positions of peptides (expressed from cDNA fragments cloned by sonication shotgun into the prokaryotic expression vector pUEX) that react positively with pooled rabbit or sheep antisera raised t o native human NSE are shown a t the top in relation to their nucleotide position The positions of likely a helices (solid) and p strands (hatched) in NSE are also displayed in relation to their cognate cDNA these positions are deduced on the basis of direct protein sequence alignment (see Fig 2) between human NSE and yeast enolase- I , the crystal structure of the latter being known at 2 25 A (see text)

..... --

*

i

C% *I.".

.*I

100

-4-

...

Enolases and PGP9.5 as tissue-specific markers.

Nervous System-Specific Proteins 15. Wilkinson, K. I). (1087) Anti-Cancer Drug Des. 2, 2 1 1-220 16. Jentsch, S.. Seufert. W. & IIauser, H.-1’. (1991...
613KB Sizes 0 Downloads 0 Views