J. Cell Set. 53, S45-254 (198a) 245Printed in Great Britain © Company of Biologists Limited 1982

REGIONAL MAPPING OF THE GENE CODING

FOR ENOLASE-2 ON HUMAN CHROMOSOME 12

MARTHA LIAO LAW AND FA-TEN KAOEleanor Roosevelt Institute for Cancer Research and The Department of Biochemistry,Biophysics and Genetics, University of Colorado Health Sciences Center, Denver, Col.80262, U.S.A.

SUMMARY

Enolase-2 (ENO2), previously termed 14-3-2 protein, is an isozyme of enolase that isenriched in neuronal tissue. The gene coding for ENO2 was previously assigned to humanchromosome 12. The present study presents data for a regional mapping of gene ENO2 usingcell hybrids containing various deletions of human chromosome 12. These deletions wereproduced by treatment with chromosome-breaking agents. Cytogenetic analysis has allowedassignment of ENO2 to the short arm of chromosome 12, in the region of pter-pi2O5. Thisassignment is consistent with the segregation pattern of the 93 hybrid clones analysed. Thesegregation pattern has also established the linear order of 6 genes on chromosome 12:pter - TPI- GAPD - LDHB - ENO2 - centromere - SHMT - PEPB - qter. Estimation ofthe relative distances between the 6 genes on chromosome 12 has been made by a statisticalmapping analysis of the segregation data of the hybrid clones. A set of deletion hybrids con-taining various combinations of these 6 markers has been established for a rapid regionalmapping of genes in one of these regions on chromosome 12.

INTRODUCTION

Previously we reported a regional assignment of 5 genes on human chromosome 12(Law & Kao, 1979). These genes code for the following enzymes: triose phosphateisomerase-i (TPIi), glyceraldehyde-3-phosphate dehydrogenase (GAPD), lactatedehydrogenase B (LDHB), serine hydroxymethyltransferase (SHMT), and peptidaseB (PEPB). The linear order and the map positions of these genes have been determinedby both cytogenetic and statistical mapping analyses. In this paper, we present datafor a-regional assignment of another gene on human chromosome 12, coding forenolase-2 (ENO2; EC 4.2.1.11), an enzyme previously termed neuronal-specific14-3-2 protein. The chromosomal assignment of the ENO2 gene was made byGrzeschik (1975) and Herbschleb-Voogt et al. (1978). An abstract reporting prelimi-nary studies of this regional assignment has been presented (Law & Kao, 1980).

MATERIALS AND METHODS

Cells and cell culture

The various hybrid cells used in this study were derived from fusions between human cellsand the glycine-requiring mutant gly'A of the Chinese hamster ovary (CHO-K1) cells. Thechromosomal deletions induced in human chromosome 12 were achieved by treatment of thecells with either X-rays or 5-bromodeoxyuridine +near-visible light, as described in detail

246 M. L. Law and F.-T. Kao

previously (Law & Kao, 1978, 1979). The hybrid clone 12A containing theglyh genome and asingle human chromosome, 12, was grown in glycine-free medium F12D supplemented with10% of the macromolecular fraction of foetal calf serum. The CHO-K1 cells and the gly'Amutant were grown in complete F12 medium supplemented with 8% foetal calf serum.

The following cells were used to assay for enolase-2 activities: IMR-32, a human neuro-blastoma cell line obtained from the American Type Culture Collection; P2, a mouseneuroblastoma cell line obtained from K. Spuhler; FS10, a diploid human fibroblasticculture established from a foreskin biopsy. These cells were grown in complete Fi2 mediumsupplemented with 8 % foetal calf serurn. In addition, the enolase-2 activity has been assayed invarious tissue cells obtained from human biopsy, human brain grey and white matter, andChinese hamster brain tissue biopsy. These tissue cells were used directly for enzyme assayswithout in vitro culture.

Cellogel assay for enolase-z

The crude cell extracts were prepared from the cultured cells and cell hybrids using theprocedures previously described (Law & Kao, 1978). Similar methods were used for cells fromtissue biopsies. Horizontal Cellogel electrophoresis (Meera Khan, 1971) was carried out atroom temperature using coi M-phosphate buffer (pH 6-2), for detecting enolase activity. Thestaining solution of Chen & Giblett (1976) was used, which contained c i M-Tris-HCl buffer(pH 7-8), 8 mM-MgSO4, o-i M-KCI, I mM-2-phosphoglycerate, 2 mM-ADP, 1 mM-NADH,10 units per ml of pyruvate kinase, 30 units per ml of lactate dehydrogenase. After staining, thedark enolase bands can be seen against a fluorescent background, by long-wave ultravioletillumination. Since ENO2 is generally less active than enolase-i (ENOi), it is necessary tostain longer for detecting ENO2 activity. In order to avoid over-staining of ENOi in some gels,the reaction was carried out first for 30 min; the gel was then cut and ENOi bands wereremoved. The gel was stained for another 90 min to reveal ENO2 bands.

Cytogenetic analysis

The trypsin-banding techniques described previously for karyotypic analysis were used (Kao,Jones & Puck, 1976).

RESULTS

Enolase-z activities in various tissues and cultured cells

Fig. i A shows the enolase patterns of the various neuronal cells. Enolase is a dimerconsisting of a and /? subunits. In Cellogel assays, 3 bands are formed: the cathodalband aa, the anodal band /?/?, and the middle band aft. The human homodimer a ^ istermed human enolase-1; the human homodimer Pxft\ is termed human enolase-2;the middle band is a heterodimer ct1fiv Chen & Giblett (1976) have shown that ENO2is abundant in human brain tissue cells, but this isozyme is also present in lowerconcentrations in other tissues and in cultured skin fibroblasts.

In Fig. 1 A, lane 1 shows a mouse neuroblastoma cell line P2, which contains 2bands, presumably a homodimer oioux type (lower band) and a heterodimer ay? (upperband). Lanes 2, 4 and 5 show tissue cells obtained from, respectively, human brain,human grey matter, and human white matter. Each of these cell types has 3 bandscorresponding to (from the bottom): a ^ , cLxfSx and fixP\- Lane 3 shows a humanneuroblastoma cell line IMR-32. The homodimer fix^ is weak, but shows a definiteband after 2 h of staining.

Regional mapping of human enolase-z 247

Enolase-z assay in hybrid cells

In Fig. IB, lane 1 shows the enolase assay in the hybrid clone 12A, which containsa single human chromosome 12. The cathodal band in lane 1 is the homodimer a^tz

in which the a2 subunit is coded for by a CHO gene. The band migrated to the sameposition as the cathodal band of CHO-Kl (lane 2). After 2 h of staining, a second bandappeared above the a^xt band in 12A (lane 1), but not in CHO-Kl (lane 2). This bardis interpreted to be a heterodimer composed of subunits at^1 in which the fix subunit

Fig. 1. Cellogel electrophoretic patterns stained for enolases. Lanes in A: I , mouseneuroblastoma cell line P2; 2, human brain tissue cells; 3, human neuroblastoma cellline IMR-32; 4, human brain grey matter; 5, human brain white matter. Lanes inB: 1, 12A; 2, CHO-Kl; 3, Chinese hamster brain tissue cells; 4, human brain tissuecells; 5, cultured human fibroblasts FS10. The human enolase subunits are designatedas at and filt and the Chinese hamster enolase subunits are designated as otj and j3t.The band a-ifii in B, lane 1, was very faint after staining for 45 min, but became darkerafter longer staining. The gels in A and B were stained for 2 h. The subunit composi-tion of each enolase band is indicated by arrows.

is coded for by a gene on human chromosome 12. The band a ^ migrated slightlyfaster than the Chinese hamster heterodimer ai^l (second band from top in lane 3),and considerably faster than the human heterodimer a2y32 (lanes 4, 5). The presenceof the band a ^ in the 93 hybrid clones assayed has been used as indication of thepresence of the ^ subunit of the ENO2 marker. In each individual assay for ENO2activity in the hybrid clones, cell extracts from CHO-Kl and 12A were alwaysincluded as controls and, in every case, the band a2/?! was absent in CHO-KI andpresent in 12A.

Grzeschik (1975) showed the formation of interspecies heteropolymers betweenhuman and mouse enolase subunits. We demonstrated here that similar heteropoly-mers can also be formed between human and Chinese hamster subunits.

248 M. L. Law and F.-T. Kao

The ENO2 activity and the activities of the 4 other isozymes were assayed in 93hybrid clones and the results are presented in Table 1. Of these 93 clones, 44 wereisolated by method B (Law & Kao, 1978), in which human cells were treated withchromosome-breaking agents followed by fusion with the gly~A mutant. SHMT+hybrids were selected in F12D for the retention of human chromosome 12, eitherintact or partial. The other 49 hybrids were isolated either by method A (Law & Kao,1978), or by a modified method A (Law & Kao, 1979), in which the 12A cells weretreated with chromosome-breaking agents and the survivors were isolated.

Table 1. Isozyme analysis of 93 hybrid clones each possessing at least one human markeron chromosome 12

Human isozyme marker— Number of

TPI GAPD LDHB ENO2 SHMT PEPB hybrids

Method B + + + + + + 18— + + + + + 4+ + + + + - 3- - + + + + 4— + — + + + 1+ + — + + — 2— — — + + + i— — + + + — 1— — — + + — 1— — — — + + 1- - - - + - 8

Method A + + + + + + 44— — — — + + 2

The 93 hybrid clones can be grouped into various phenotypic classes and arrangedin order according to the map position of the genes on the chromosome. From thesegregation patterns shown in Table 1, it is logical to place ENO2 between LDHB andSHMT. Moreover, ENO2 was assayed and shown to be positive in the 2 previouslydescribed independent clones MAi and MA2, which had large deletions in the longarm of chromosome 12 with the breakpoint at qi2 (Law & Kao, 1979). Thus, based onthe karyotype and the phenotype (TPI+GAPD+LDHB+EN^+SHMT-PEPB-) ofthese 2 clones, ENO2 can be placed in the region pter-qi2, or between LDHB and qi2.

Cytogenetic analysis

The hybrid clone A9 was derived from 12A after treatment with bromodeoxyuridine+ near-visible light (Law & Kao, 1979). The phenotype of A9 is TPI-GAPD"-LDHB-ENO2-SHMT+PEPB+. This clone has a terminal deletion in the short armwith the breakpoint at pi205 (Fig. 2). Thus, like TPI, GAPD and LDHB, ENO2 canalso be assigned to the region pter-pi2O5. This regional assignment is consistentwith the location of ENO2 being pter-qi2 as derived from clones MAi and MA2.

Regional mapping of human enolase-2 249

Statistical mapping analysis

Regional mapping by statistical analysis of the segregation data (Table 1) wasperformed following closely the procedures described in our previous paper (Law &Kao, 1979). We used the methods of Goss & Harris (1975, 1977a, b) to derive thelinear order of the 6 genes on chromosome 12 and to estimate the relative distances ofthese genes in relation to the selected locus SHMT.

B

D

IFig. 2. Trypsin-banded metaphase chromosomes in the 12A cell (A) containing a singlehuman chromosome 12 (Hui2, arrow), and the A9 cell (B) containing a partial humanchromosome 12 with a terminal deletion pter-pi2os (arrow). In the lower section arepresented the enlarged human chromosome 12 (c) and the partial chromosome 12 (D).

The segregation data of the 44 deletion hybrids (Table 1, method B) show that TPIis lost in 21 hybrids (48 ±7%); GAPD is lost in 16 hybrids (36 ±7%); LDHB is lostin 14 hybrids (32 ±7%); ENO2 is lost in 9 hybrids (20 ±6%); and PEPB is lost in 15hybrids (34 ± 7%). The numbers in brackets refer to the frequency ± standard error.

These segregation frequencies reflect the relative distances between SHMT andany one of these markers. Thus the following order of relative distances can be arran-ged, beginning with the smallest distance: ENO2, LDHB, PEPB, GAPD, TPI. Thisorder of relative distances is consistent with our previous results (Law & Kao, 1979),and also adds the ENO2 marker, which is closest to the locus of selection.

Our previous data based on statistical calculation and cytogenetic analysis havealso shown that TPI, GAPD and LDHB are on the short arm of chromosome 12,

9 CEL53

250 M. L. LawandF.-T. Kao

while SHMT and PEPB are on the long arm. It is evident from the segregation patternsthat, when ENO2 is lost, TPI, GAPD and LDHB are also lost. However, the reverseis not true; that is, when TPI, GAPD or LDHB is lost, there is on average a 50%chance that ENO2 is still retained in the hybrid. Therefore, we conclude that ENO2is closer to SHMT than the other markers. PEPB appears to segregate independentlyof ENO2, indicating that they are on opposite sides of the selected locus SHMT.Indeed, our cytogenetic data support this conclusion.

Table 2. Estimation of target size {i.e. distance between 2 loci) using co-transfer frequency(F) of unselected markers with SHMT

Unselected marker

No. of clones exhibiting unselectedmarkers and SHMT

F-log F- l o g F '

Total number of clones = 44

TPI

230-520-28I'OO

GAPD

280-640-19o-68

LDHB

3°0 6 80-170 6 1

ENO2

350 8 0

o-io0-36

PEPB

290 6 60-180-64

TPIGAPDLDHBEN02

SHMT

PEPB

•TPI-GAPD•LDHB•ENO2

•SHMT

•PEPB

s6

CO

6

L9

0

CO0-

6

-

Fig. 3. Diagrams showing cytogenetic map (left) and statistical gene map (right) ofhuman chromosome 12.

Finally, the relative distance between 2 loci can be estimated using a simple targettheory (Goss & Harris, 1975, 1977a, b), which to a first approximation equates therelative distance between 2 markers to — log of the frequency of hybrids possessingboth markers. Table 2 presents the calculations based on these analyses. The last rowrepresents the relative distance between SHMT and each of the syntenic markers,normalized to a distance of i-oo between TPI and SHMT. Corrections for interstitialdeletions and other complex chromosomal rearrangements as described by Goss andHarris were not attempted here because of the small sample size of our deletionhybrids. Thus, the relative distance presented here is only an estimate.

Regional mapping of human enolase-2 251

Comparison of cytogenetic and statistical maps of human chromosome 12

Fig. 3 presents the gene maps of human chromosome 12 constructed by bothcytogenetic and statistical analyses. In general, the order of the 6 markers on chromo-some 12 is consistent in the 2 maps. A significant difference in the 2 maps exists in therelative positions of the 4 genes on the distal part of the short arm. The cytogeneticmap assigns the 4 genes to the region pter-pi2O5, but the statistical map shows thatthe 4 genes occupy a larger section, with ENO2 extended to the band pi 1.

DISCUSSION

Enolase, or 2-phospho-D-glycerate hydrolase, reversibly converts 2-phosphogly-cerate to phosphoenol pyruvate in the glycolytic pathway. Three enolase isozymeshave been identified: a non-neuronal enolase (ENOi), a neuronal enolase (ENO2),and a hybrid form of ENOi and ENO2 (Schmechel et al. 1978). ENOi is present inall tissues and consists of 2 identical subunits ct^x.lt each of about 39000 molecularweight. ENO2 is enriched in neuronal tissue but is also present in smaller quantities inother tissues. This enzyme consists of 2 identical subunits filfi1 each of about 43 500molecular weight. The third form of enolase is a heterodimer consisting of cc^isubunits.

The locus for human a subunit ( a j has been mapped to chromosome 1 both byfamily study (Giblett, Chen, Anderson & Lewis, 1974) and by cell hybrid analysis(Meera Khan, Deppert, Hagemeijer & Westveld, 1974). The locus for human fi sub-unit (/fj) has been mapped to chromosome 12 using cell hybrids (Grzeschik, 1975;Herbschleb-Voogt et al. 1978). Here we present data to assign regionally the fix locusfor the enolase-2 phenotype to the short arm of chromosome 12, in the distal regionpter-pi2O5. This regional assignment is consistent with the segregation data of 6syntenic genes on human chromosome 12 (Table 1). The cytogenetic and segregationdata also place ENOz between LDHB and the breakpoint pi205. Previously, TPIand GAPD were regionally assigned to pi3 by Serville et al. (1978), and LDHB toI2-I-I2-2 by Rethore et al. (1975). In their nomenclature, the band pi2 is dividedinto 3 equal parts designated 12-1, pi2-2 and 12-3, with 12-1 being contiguous withpi 1. Thus, the region pi2-i-pi2-2 is roughly equivalent to the region 1201-1207 inthe system recommended by the Paris Conference Supplement (1976). Based on thisregional mapping of LDHB and our assignment of ENOz between LDHB and pi205,we can further localize both LDHB and ENO2 to P1207-P1205, with the order of the 2genes being LDHB distal and ENO2 proximal to the centromere.

In the IV International Workshop on Human Gene Mapping, Bruns & Reginareported a possible second TPI locus (TPI2) on human chromosome 12 coding for aheat-labile subunit of the TPI isozyme. In our assays for this heat-labile TPI2isozyme, we found no segregation between TPIi and TPI2 in the 93 hybrids analysed,as we reported previously for a smaller number of these hybrids (Law & Kao, 1980).In the present paper, we used the term TPI to represent both of these isozymes; the

9-2

252 M. L. Law and F.-T. Kao

precise nature and the identity of TPI2 require further biochemical and geneticstudies.

The set of hybrids carrying various marker deletions can divide chromosome 12 intothe following 8 regions: pter -1- - TPI -2- - GAPD -3- - LDHB -4- - ENO2 -8- -centromere -6- - SHMT -7- - PEPB -8- - qter. Thus a set of a minimum of 5 clonesexhibiting unique combinations of these markers can be selected from Table 1 andused for rapid regional mapping of other genes assigned to chromosome 12. However,the assignment to regions 5 and 6 will also require karyotypic analysis. A similar set ofdeletion hybrids has been established for the X chromosome by Becker et al. (1979).

We previously used the statistical mapping analysis of Goss & Harris (1975, 1977 a,b)to estimate distances for the 5 genes on human chromosome 12 (Law & Kao, 1979).The statistical map in general agrees with the cytogenetic map. However, Servilleet al. (1978) assigned TPI and GAPD to pi3 by cytogenetic analysis, while ourstatistical map places GAPD in pi2. Moreover, the statistical map position for ENO2is more proximal to the centromere than its position in the cytogenetic map (Fig. 3).By comparing the 2 maps for the 4 genes on the short arm, it appears that the genes onthe statistical map are extended to a greater distance. The significance of this disparityrequires further study.

It is worthwhile to point out that in the statistical mapping of the human chromo-some 1, Goss & Harris (19776) found a similar non-coincidence, one interpretationof which was that radiation-induced rearrangements occurred preferentially inGiemsa-stained light material.

It should also be pointed out that the cytogenetic map is constructed on the basisof the location of genes in the highly condensed metaphase chromosomes, whereas thestatistical map is based on the location of genes in the extended state of the interphasechromosome in which radiation-induced breaks occur. Thus, while the statisticalmap measures the distance between genes in the extended DNA sequences, the cyto-genetic map measures the distance between genes in the condensed metaphasechromosomes. Since meiotic crossing-over takes place during close-pairing ofhomologous chromosomes in an extended state, the statistical map should resemblemore closely the genetic map based on recombination events observed in higherorganisms.

The regional assignment of ENO2 to the short arm of chromosome 12 is particularlyinteresting. It is the fourth enzyme of the glycolytic pathway for which the gene hasbeen assigned not only to the same chromosome, but also to the same arm. Since these4 genes are all separated by some distances as shown in the map, they are clearly notcontiguous in the DNA sequences. However, the assignment of 4 genes related in acommon pathway to a specific region of the chromosome may have some significancein evolution and possibly in gene regulation. The possible relationship between thecoordinate regulation of functionally related genes and the physical linkage of thesegenes on a segment of the chromosome certainly requires further investigation.Mapping of these 4 genes in other species may provide additional insight into thesignificance of this linkage relationship.

Regional mapping of human enolase-2 253

This investigation is a contribution from the Eleanor Roosevelt Institute for Cancer Researchand the Florence R. Sabin Laboratories for Development Medicine (Contribution no. 350),and the Department of Biochemistry, Biophysics and Genetics, University of Colorado HealthSciences Center, Denver, Colorado. This work was supported by grants from the AmericanCancer Society (CD105) and the National Institutes of Health (GM26631, HD02080). Wethank Drs T . T. Puck and D. Patterson for critical reading of the manuscript. This paper isno. 34 in the series entitled 'Genetics of Somatic Mammalian Cells'. The preceding paper isby Meisler, Wanner, Kao & Jones (1981).

REFERENCES

BECKER, M. A., YEN, R. C. K., ITKIN, P., Goss, S. J., SEEGMILLER, J. E. & BAKAY, B. (1979).

Regional localization of the gene for human phosphoribosylpyrophosphate synthetase on theX chromosome. Science, N. Y. 203, 1016-1019.

CHEN, S. H. & GIBLETT, E. R. (1976). Enolase: Human tissue distribution and evidence forthree different loci. Arm. hum. Genet. 39, 277-280. ^

GIBLETT, E. R., CHEN, S. H., ANDERSON, J. E. & LEWIS, M. (1974). A family study suggestinggenetic linkage of phosphopyruvate hydratase (enolase) to the Rh blood group system.Cytogenet. Cell Genet. 13, 91-92.

Goss, S. & HARRIS, H. (1975). New method for mapping genes in human chromosomes.Nature, Lond. 255, 680-684.

Goss, S. & HARRIS, H. (1977 a). Gene transfer by means of cell fusion. I. Statistical mapping ofthe human X-chromosome by analysis of radiation-induced gene segregation.,7. Cell Set. 25,17-37-

Goss, S. & HARRIS, H. (19776). Gene transfer by means of cell fusion. II. The mapping of 8 locion human chromosome 1 by statistical analysis of gene assortment in somatic cell hybrids.J.CellSci. 25, 39-58.

GRZESCHIK, K. H. (1975). Assignment of human genes: /?-glucuronidase to chromosome 7,adenylate kinase-i to 9, a second enzyme with enolase activity to 12, and mitochondrial IDHto 15. Cytogenet. Cell Genet. 16, 142-148.

HERBSCHLEB-VOOGT, E., MONTEBA-VAN HEUVEL, M., WIJNEN, L. M. M., WESTERVELD, A.,

PEARSON, P. L. & MEERA KHAN, P. (1978). Chromosomal assignment and regional localizationof CS, ENO,, GAPDH, LDHB, PEPB, and TPI in man-rodent cell hybrids. Cytogenet.Cell Genet. 22, 482-486.

KAO, F. T., JONES, C. & PUCK, T . T. (1976). Genetics of somatic mammalian cells: Genetic,immunologic, and biochemical analysis with Chinese hamster cell hybrids containing selectedhuman chromosomes. Proc. natn. Acad. Sci. U.S.A. 73, 193-197.

LAW, M. L. & KAO, F. T . (1978). Induced segregation of human syntenic genes by 5-bromo-deoxyurdine + near visible light. Somat. Cell Genet. 4, 465-476.

LAW, M. L. & KAO, F. T. (1979). Regional assignment of human genes TPI t , GAPDH,LDHB, SHMT, and PEPB on chromosome 12. Cytogenet. Cell Genet. 24, 102-114.

LAW, M. L. & KAO, F. T. (1980). Regional assignment of human chromosome 12 of sevengenes TPI1 ( TPI . , GAPDH, LDHB, ENO,, SHMT and PEPB. Cytogenet. Cell Genet. 25,179-180.

MEERA KHAN, P. (1971). Enzyme electrophoresis on cellulose acetate gel: Zymogram patternsin man-mouse and man-Chinese hamster somatic cell hybrids. Archs Biochem. Biophys.I4S. 47°-483-

MEERA KHAN, P., DOOPERT, B. A., HACMEIJER, A. & WESTERVELD, A. (1974). The human loci forphosphopyruvate hydratase and guanylate kinase are syntenic with the PGD-PGM! linkagegroup in man-Chinese hamster somatic cell hybrids. Cytogenet. Cell Genet. 13, 130-131.

MEISLER, M. H., WANNER, L., KAO, F. T . & JONES, C. (1981). Localization of the uroporphy-rinogen I synthase locus to human chromosome region nq i3 -q te r and interconversion ofenzyme isomers. Cytogenet. Cell Genet. (In Press.)

RETHORE, M. O., KAPLAN, J. C , JUNIEN, C , CRUVEILLER, J., DUTRILLAUX, B., AURIAS, A.,

CARPENTIER, S., LAFOURCADE, J. & LEJEUNE, J. (1975). Augmentation de l'activite de laLDH-B chez un garcon trisomique I2p par malsegrcgation d'une translocation maternellet ( i2 : i4) (q i2 :pn) . Annls Genet. 18, 81-87.

254 M. L. Law and F.-T. Kao

SCHMECHEL, D., MARANCOS, P. J., Zis, A. P., BRIGHTMAN, M. & GOODMAN, F. K. (1978).Brain enolases as specific markers of neuronal and glial cells. Science, N.Y. 199, 313-315.

SERVILLE, F., JUNIEN, C, KAPLAN, J. C, GACHET, M., GADOUX, J. & BROUSTET, A. (1978).Gene dosage effect for human triosephosphate isomerase and glyceraldehyde-3-phosphatedehydrogenase in partial trisomy 12P13 and trisomy i8p. Httm. Genet. 45, 63-69.

{Received 2 June 1981)