of · lities'2 associated with the karyotype 46,XY,rec(16), which results in partial trisomy for 16pl3-1-÷pter and partial monosomyfor 16q22-+qter. Ascontrols, DNA was obtained from

Journal ofMedical Genetics 1987, 24, 39-46

High resolution gene mapping of thehuman oa globin locusR D NICHOLLS*, J A JONASSONt, J O'D McGEEt, S PATIL§,V V IONASESCU§, D J WEATHERALL*, AND D R HIGGS*From *the MRC Molecular Haematology Unit, Nuffield Department of Clinical Medicine, and tthe NuffieldDepartment ofPathology, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU; tthe DepartmentofMedical Genetics, Churchill Hospital, Headington, Oxford; and §the Department of Pediatrics, Division ofMedical Genetics, The University ofIowa Hospitals and Clinics, Iowa City, Iowa 52242, USA.

SUMMARY A combination of polymorphic DNA markers, cytogenetic analysis, and in situhybridisation has been used for the high resolution assignment of the human a globin gene clusteron chromosome 16. Multiallelic DNA probes from within the a globin cluster were used todetermine the number of copies of this locus in three cell lines containing trisomies of the short armof chromosome 16 and one with a familial inversion, inv(16). The breakpoints in theserearrangements flank the a globin locus and locate a shortest region of overlap to 16pl 3-1. Ameiotic crossover was also localised to this band. In situ hybridisation of biotinylated DNA probesto normal and inverted chromosomes 16 [inv(16)(p13 1;q22)] showed hybridisation sites atopposite ends of the chromosomes, consistent with this regional localisation.

The precise regional localisation of highly poly-morphic DNA markers is of fundamental impor-tance in establishing a human genetic linkage map.'With accurately localised markers, the combined useof molecular studies and high resolution karyotypicanalysis will provide a greater understanding of thenormal structure of the human genome and theprocess of recombination, ultimately establishing therelationship between physical and genetic distancewithin chromosomes. Furthennore, this combinedapproach will provide insight into the chromosomalrearrangements and aneuploidies that underly manygenetic diseases.Accurate localisation of the human a globin

complex would enable it to be used as a useful modelfor this combined cytogenetic and molecular stra-tegy. It is a multiallelic locus2 localised to chromo-some 163 within the broad limits of pl2--qter.4 5 Asyet it is the only cloned segment ofDNA on the shortarm of chromosome 16 but 200 kb of this region hasnow been characterised (R D Nicholls and D RHiggs, in preparation). It is an important diseaselocus; mutations in this region ((a thalassaemias) are amajor cause of morbidity and mortality in manypopulations.67 Moreover, it is genetically linked to

Rcceived for publication 16 June 1986.Rcvised version accepted for publication 14 July 1986.

other important, poorly localised polymorphic genes8and disease loci, such as that for adult polycystickidney disease (APCK.D),9 and is associated with anuncommon form of mental retardation.'0

In order to obtain a more precise intrachromo-somal localisation of the ax globin locus we havedetermined the number of copies of this region in celllines obtained from four subjects with well definedcytogenetic abnormalities of chromosome 16, usingpolymorphic DNA markers and in situ hybridisationstudies. The combined molecular and karyotypicdata localise the human aE globin gene locus to16pl3-1.

Materials and methodsCELL LINES AND OTHER HUMAN DNASAMPLESCell lines from two subjects with partial trisomy 16pand one parent of each carrying a balanced trans-location involving 16p were obtained from theHuman Genetic Cell Repository, Institute for MedicalResearch, Camden, NJ [GM2324, balanced, 46,XX,t(16;22Xpl3;ql2) and GM2325, unbalanced, 47,XX,+der(22) (that is, partial trisomy for both 16pl3-aopter and 22ql2-*pter); GM6227, balanced,46,XX,t(l;16)(q44;pl3- 11) and GM6226, unbalanced,46,XY,- 1, +der(l) (that is, partial trisomy for

39

on April 24, 2020 by guest. P

rotected by copyright.http://jm

g.bmj.com

/J M

ed Genet: first published as 10.1136/jm

g.24.1.39 on 1 January 1987. Dow

nloaded from

http://jmg.bmj.com/

40 R D Nicholls, J A Jonasson, J O'D McGee, S Patil, V V Ionasescu, D J Weatherall, andD R Higgs

l6pl3-11-->pter and partial monosomy for 1q44-+qter)]. The latter two have a/P globin chain synthesisratios of 1:1 and 3:2, respectively, in peripheral bloodreticulocytes." GM2325 is a fibroblast line from an11 day old child with multiple congenital anoma-lies." No further data were reported for the otherthree, lymphoblast, cell lines." A fibroblast cell line(SJ') was isolated from an asymptomatic carrier ofa familial pericentric inversion of chromosome16: 46,XX,inv(l6)(pl3 l;q22).'2 DNA was also obtain-ed from the peripheral blood of her son (SJ) who hasa severe myopathy and other congenital abnorma-lities'2 associated with the karyotype 46,XY,rec(16),which results in partial trisomy for 16pl3-1-÷pterand partial monosomy for 16q22-+qter. As controls,DNA was obtained from the peripheral blood ofnormal British subjects.

DNA STUDIESSouthern blot hybridisation studies were carried outessentially as described'3 except that Hybond-Nmembranes (Amersham Int) were used so thatprehybridisation, hybridisation, and washing ofmembranes were performed with the addition of SDSto 05"',. Final wash stringency was 68'C in 0015mol/l NaCl/0 0015 mol/l Na- citrate/055% SDS. Theprobes used were a 4 kb Hinfl fragment of the 3'-HVR (hypervariable region) subclone pSEAI'4 and a1 8 kb SstI *C probe2 (fig la). pDH8, a clonecontaining a 1.5 kb PstI a globin genomic fragment,'6was used for in situ hybridisation.

IN SITU HYBRIDISATIONHuman metaphase cells were prepared'7 from celllines of the inv(16) heterozygote (SJ') and three othersubjects. pDH8 was labelled with biotin by nicktranslation as described'8 except that the meanbiotinylated probe size was 150 bp. The details of thein situ hybridisation protocol and detection systemare described elsewhere.'9

ResultsREFINEMENT OF a GLOBIN ASSIGNMENT TO

pl 13 lI-pterThe human a globin genes were previously localisedto 16pl2---pter by in situ hybridisation4 and in-directly by finding a raised oc/5 globin chain synthesisratio in a person trisomic for this distal segment.5The availability of further subjects with rearrange-ments of 16p and the identification of two highlyinformative, multiallelic genetic markers within thea globin gene cluster (IZHVR and 3'-HVR2 1420)provided the means to refine this localisationsignificantly.The cell line GM2325 has an unbalanced karyo-

type derived by a 3:1 type of meiotic disjunction2' inthe mother (GM2324), whose karyotype shows abalanced translocation. The breakpoints l6pl3 and22ql2 in this rearrangement have been reanalysedbecause of an improvement in cytogenetic bandingtechniques and refined to 16pl 3-11 and 22q 1 1-21 (MM Aronson, 1985, personal communication). In theindex case, one rearranged chromosome (the der(22)carrying distal 16p) has been transmitted togetherwith the normal 16 homologue from the mother andtherefore this cell line contains three copies of16p 13 11 --+pter (two maternal and one paternal). The3'-HVR probe (fig la) which usually distinguishesallelic a globin gene clusters shows that GM2325 hasinherited both maternal alleles plus a third (presum-ably paternal) 3'-HVR allele (fig lb).Therefore thetrisomic segment in this cell line includes the a globingene locus which is thus localised to 16pl 3-11- pter.A second pair of balanced and unbalanced chromo-

somes 16 with a breakpoint at 16pl3 11 is representedby GM6226 and GM6227.2' The cytogenetic findingssuggest that GM6226 also contains three copies of16pl3 11-÷pter and that this trisomy results fromadjacent I type segregation. That is, one of the tworearranged chromosomes (the der(l)) is transmittedwith the normal homologue of the other chromo-some (the normal 16).21The mother (GM6227) has two 3'-HVR alleles (fig

I b), one linked to the a complex on the normalchromosome 16 and the other to the a globin geneson either the der( l) or der( 16) depending on whetherthe locus is distal or proximal to the breakpoint,respectively. From the results for GM2325 (seeabove) and the globin chain synthesis data (seeMaterials and methods) it was predicted thatGM6226, with the 16p trisomy, should have three 3'-HVR alleles (two maternal and one paternal).Nevertheless, only one maternal and one paternal 3'-HVR allele could be identified (fig lb and legend).However, subsequent gene dosage experiments usinga 4C gene probe, which detects the multiallelicIZHVR, showed that there are two identical copiesof the maternally inherited oE globin gene cluster, asdescribed below. Thus, GM6226 has a total of threecopies of the a globin gene cluster, again consistentwith the localisation to 16p13-l 11-pter.LOCALISATION OF A MEIOTIC CROSSOVERNEAR TO THE a GLOBIN CLUSTERThe presence of three copies of the a globin complexin GM6226 is most easily demonstrated using BglIIand the *C probe (fig 1 a), which detects an invariant12.6 kb band, 3' of *C, an inter-c fragment ofvariable size, 10 5 to 1I 5 kb, and an invariant 18 kbband 5' of 4 which is not seen on this autoradiograph(fig lc). In subjects with two inter-C bands differing in



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/

High resolution gene mapping of the human a globin locus

size, these represent paternal and maternal alleles.The ratio of upper:middle bands thus shows a 2:2 or2:1:1 gene dosage effect in normal subjects (forexample GM2324). In contrast, for GM2325 this is3:1:2, consistent with the previous results demon-strating three copies of the a globin locus in this cellline.

Subjects GM6226/GM6227 display an uncommon

i ~ ~~~~~~------ -; ---

- -

4. 5o

4. 3ej

2. 0

BglIH polymorphism (±2) 3' of the 4C gene (fig la)which results in a 46 kb band (+) rather than theusual 12 6 kb band (-) and thus distinguishesmaternal (+) and paternal (-) 3' *C bands. Thisallows quantification of bands in GM6226 as 1:1:2:2and in GM6227 as 1:1:1:1. These findings indicatethat GM6226 is monosomic for paternal bands anddisomic for maternal bands (fig 1c). This estimation

FIG I Molecular localisation of the human a globin locus by Southern blot analysis of c ell lines carrying 16p rearrangements.(a) Restriction map of the a globin genev. Closed boxes represent genes and hvpervariable region (HVR) sequences aredenoted by the symbols IZHVR (inter-C) and 3'-HVR. The two probes used are shown underneath. Polymorphic enzyme sitesare indicated by +. B= BamHI, Bg= BglII, E= EcoRI, Pv= PvuII, S= Sstl. (b) Distribution of3'-HVR alleles using PvuII.Faint background bands resultfrom lower stringency detection ofa minisatellite family using the 3' HVR probe.'5 This probedetects long tandem arrays ofa 17 bp repeat and thus the hybridisation intensity is dependent on the repeat copy number.This makes the probe unsuitablefor the quantification ofgene dosage effects. Each allele detected by PvuII has about 700 bpnon-HVR'sequence. N=normal subject, SJ'=inv(16) heterozygote, SJ=rec dup (16p) (see fig 3), 4=GM2324.5= GM2325, 6= GM6226, 7= GM6227, A= size markers (kb). (c) Detection of'aneuploidy at the IZHVR and 3'-HVRmultiallelic loci. These were identified with the 4C and 3'-HVR probes, respectively. See textfor details ofgene dosage effects.Comparison of3'-HVR BglJI digests with EcoRI and PvuII (fig Ib) identifies a BglII site polymorphism near the 3'-HVR(fig la) in both of these families, also seen in other unrelated persons.

41

i

.



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/

42 R D Nicholls, J A Jonasson, J O'D McGee, S Patil, V V Ionasescu, D J Weatherall, and D R Higgs

of gene dosage from band quantification is alsoconsistent with the intensity of the BanHI C specificfragments (fig ic).These data show that although it is only possible

to demonstrate one of the two maternal 3'-HVR (figlb) and IZHVR (fig lc) alleles in GM6226, there aretwo identical copies of the region of the maternalchromosome containing the a globin cluster. Themost likely mechanism to explain this is a geneticcrossover between adjacent, non-sister chromatids inthe region between the a globin locus and thetranslocation breakpoint, preceding adjacent 1 typesegregation at maternal meiosis (fig 2).

MOLECULAR ANALYSIS OF AN INVERSION (16)

The above data indicate that the a locus lies betweenl6pl3.11 and pter. In order to obtain a moreaccurate distal limit, we analysed DNA from asubject with a familial inversion of chromosome 16,inv(l6)(pl3 1;q22), and her son with a recombi-nant chromosome 16 resulting in partial trisomy16pl 3 -*pter and partial monosomy 1 6q22-+qter,

p

der(l) (

p q

16 j j der (16)

associated with mental retardation and multiplecongenital abnormalities.'2 The recombinant chromo-some can only arise from a crossover within thepaired, inverted segment at maternal meiosis (fig 3).This will result in duplicated a globin genes if the alocus is distal to the p13-1 breakpoint, but not if it isproximal, and will thus result in the inheritance oftwo or one maternal 3'-HVR alleles, respectively.The mother (SJ') is heterozygous at the 3'-HVR andthe son (SJ) inherits only one maternal 3'-HVR (figI b). Gene dosage estimations at other regions of the acomplex were also consistent with the inheritance ofa single maternal a globin locus (data not shown) andtherefore this locus lies proximal to the 16pl3-1breakpoint. The combined data (summarised in fig 4)thus show that the a globin genes lie between p13 1 1and p13-2 and assign the shortest region of overlap(SRO) to 16pl3 1.

E s-E meioticE

O, pairinga .

or-06a.-W

FIG 2 Diagram ofchromosomes involved in the reciprocaltranslocation t(1;16) (q44,p13 11) at pachytene. Duringmeiosis, a putative crossover (X) between the a globincomplex and the translocation breakpoint, followed byalternate I type segregation, produces unbalanced gametesthat include the two chromosomes der(l) and 16 with thesame a globin allele (a'). Alternative crossover events or

modes ofsegregation *would generate molecular or

karyotypic differencesfrom those found.

FIG 3 Schematic illustration of the meiotic behaviour ofcells heterozygous for the inv( 16): meiotic pairing with onecrossing over within the inversion segment and both resultingrecombinants. SJ is the rec(16) used in this study. Theinversion loop model shown is that commonly accepted.21Alternative models have been suggested since there was noevidence of inversion loops on meiotic analysis ofpachytenecellsfrom a single inversion heterozygote,22 although theoutcome ofa crossover is the same. Distal 16q is shaded anddistal 16p is cross hatched.

I



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/


%O Lr%

C.D C.M )L

13.3

13213.1313.1213.11

p 12.312212.1

11.2

11.111.1

112

12.112213

q 21

22.1

22222.323.123223.324.124224.3

i Igl)I ; SRO

inv( 16)

IN SITU HYBRIDISATIONIn situ hybridisation studies of an a globin probe tometaphase cells (fig 5) were also consistent with thisregional localisation. Chromosome 16 was speci-fically labelled in several normal karyotypes, asexpected, although we also observed several weaklylabelled regions which may represent dispersedpseudogenes (fig 5a). Two further subjects wereanalysed for hybridisation sites on chromosome 16only. From a total of 16 sites on the karyotypicallynormal chromosomes of one subject, 10 (63%) werelocalised at pl2-*pl3.1 (fig 5b). From fig 4 it can bepredicted that the a globin genes will be relocated inthe inv(16) hererozygote (SJ'), and this was con-firmed (fig 5c) by showing that the a globin locus(proximal p13) now abuts the distal q22 segment.

Discussion

II

16

%O L(N

%O J

-~ ~ ~ ~ ~ ~~-

CD CD Ln

3'-HVR 2 3 2a loci 3 3 2

FIG 4 Summary ofmolecular data localising a globin to theshortest region ofoverlap (SRO), 16p13 1. Top: the extentof the 16p duplication in the three partial trisomies studiedare indicated by hatched boxes and the inversion by arrows.

The inherent three band uncertainties in the assignment oftranslocation breakpoints23 are indicated by bars and alsoapply to the gene localisation. Bottom: the table summarisesthe number of3'-HVR size alkeles detected as infig lb andthe total number ofa gene complexes deduc-edfrom data infig I (and unpublished data). Data in the first three columnsallows placement ofa globin to the SRO as di4gssed in thetext.

The human a globin complex contains two highlypolymorphic regions which, together with severaldimorphic RFLPs, increase the heterozygosity at thislocus to almost 10 in many populations.2 Althoughprevious studies have emphasised the importance ofsuch multiallelic markers in establishing a humangenetic linkage map,'2 these loci are also of greatvalue in establishing a physical map of the genome.Their presence or absence can be readily shown inwell characterised chromosomal rearrangements andhence their regional localisation can be accuratelydetermined. Using this approach, we have localisedthe human a globin locus to band 16pl3 1. Con-sistent with this is a recent preliminary study thatplaced it distal to the fragile site at 16pl213.28 Thislocalisation increases the scope of further studiesusing a similar approach; for example, it will allowthe identification of subcytogenetic rearrangementsin this region'0 and provide the means to examine theorigin of chromosome 16 trisomies, so frequentlyfound in early spontaneous abortions.29 Usingsimilar multiallelic loci this general strategy will be ofvalue in the identification and characterisation ofother common chromosomal abnormalities.21 30 31

It has previously been shown that the a globinlocus is linked to the loci for phosphoglycolatephosphatase (PGP) and APCKD and is associatedwith an unusual form of mental retardation.8-'0 Theregion of chromosome 16 shown to contain the aglobin genes (16pl 3-1) lies about mid-way along thegenetic span of 16p (approximately 50 cM in themale32 33), so that many other genes on the short armof chromosome 16 will be linked within 20 cM eitherside of this. The identification of translocation andinversion breakpoints closely flanking the a locuswill enable determination of the chromosome order

43



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/


205 (a)15-

10'-O

p q p q p q p q a p q1 2 3 4 5 6

c 20* 15C7. 10-°~ o-.1 I*1 . .1I. . . ag I, a . .. .I. .

--l p q Il pTI a p Q7 8 9 10

p p a p ai11 1? 13

25 -

20 -

15 -10 -

O-.0 I I1I I 1. ,^ I I A Li IL

P q p q p p p q P q P pP q p q p p q14 15 16 17 18 19 20 21 22 X Y

C h r o m o s o m e s

(b) (c)13.313.213.1

12 P

0

21

22

23

24

q

UEENl

normal 16 inv(16)(p13.1q22)

FIG 5 In situ hybridi.sation ofbiotinylateda globin DNA probes localises the gene cluster to 16p12-pl311. (a) Distributionofchromosome associated hybridisation sitesfrom an analysi.s ofmetaphase chromosomes: 337 sites were scoredfrom 71 and36 cells, respectively, from two normal haploid karyotypes. The major cluster ofsites is on chromosome 16 andfrom a total of46 sites on chromosome 16, 25 (55%) were localised at 16p12-p13 1. Hybridisation above background is observed atI lq22-23 and 22qter (in both subjects) and possibly at 16pter. This most likely represents sequences related to a globin,possibly a globin pseudogenes, that wc'ere previously undetected. Low stringency Southern hybridisation also detects specificfragments that are not on chromosome 16 (R D Nicholls, unpublished data). These related sequences may be analogous to themouse a globintJ genes a3 anda4 on mouse chromosomes I5 and 17, respectively.2425 Indeed, several genes on theformerappear to be part ofa syntenic linkage group cc'ith genes on human chromosome 22qter26 and a human a globin relatedsequence has recently been found on chromosome 22.27 (b) Regional localisation ofax globin on chromosome 16 from a third

subject. The bar at the top represents a single hybridisation site. The dot at the heterochromatic region (16qh) representsstaining of this region due to the Giemsa banding. This may be misinterpreted as hybridisation sites. It was not observed in theexperiments in (a). (c) Relocation ofa globin labelled sites on the inverted chromosome inv( 16) (p13 1;q22). The arrows

identify the breakpoints. Of26 hybridisation sites on the inv( 16), 17 (65%) were located at p12-p 13 1. Analyses were

performed on 55 chromosome spreadsfrom the inv(16) heterozygote SJ'.

, t p13prox q22

.. prox p13dist q22

.-q



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/


and orientation of these linked genes with respect tothe centromere. This in turn will provide strategiesfor establishing a physical linkage of neighbouringgenes by long range chromosome 'walking' tech-niques.3435 Unfortunately, none of the chromosomalrearrangements described here have breakpointswithin or close to (<25 kb) the a globin locus, as

determined by genomic mapping with a, C, and 3'-

HVR probes (fig I and unpublished data).The combined use of high resolution karyotype

analysis and extremely polymorphic loci provides a

way of analysing the recombination events thatunderly the chromosomal rearrangements that occurin germ cells21 36 and some malignant somaticcells.3742 In one of the cases reported here, forexample, we identified a meiotic crossover in therelatively small cytogenetic distance between the aglobin locus and a translocation breakpoint at16pl 3-1 1. Although we consider a meiotic crossoverto be the most likely explanation of the data, wecannot exclude an interchromosomal gene con-version event of the distal portion of 16p including atleast the ac locus. At the molecular level little isknown about the sequences involved in largechromosomal rearrangements. Recently, molecularcloning has enabled characterisation of sequencesinvolved in homologous and illegitimate recom-bination, associated with gene deletions of betweenabout 4 kb to greater than 65 kb, that cause athalassaemia43-45 (R D Nicholls and D R Higgs, inpreparation). These may provide a useful back-ground with which to compare the local sequenceevents occurring at translocation, inversion, andchiasma breakpoints in this relatively small region ofthe genome.The accurate identification and characterisation of

genes that are closely associated with translocationbreakpoints may be important for understanding thefunctional effect of changing the position andneighbouring sequences of a gene. Pericentric inver-sions of chromosome 16 have been observed inphenotypically normal subjects, with breakpoints atplIql2-13,4647 pl3ql2-13 (D Bianchi and S A Latt,1986, personal communication), and pl3- Iq22'2 (thispaper). Indeed, 16p breakpoints in the latter two areproximal (R D Nicholls, unpublished data) and distalto the oa locus, respectively. An inv( 16) (p1 3- 1q22) hasalso been implicated in the evolution of theleukaemic cell clone in many cases of acute non-

lymphocytic leukaemia (ANLL) with abnormal bonemarrow eosinophils.48-52 The inversion breakpointin these cases splits the metallothionein (Mt) genecluster.53 It has been suggested53 that an enhancerelement in the Mt-2 gene promoter54 activates a

proto-oncogene at 16p 13. Comparison of the familial

and leukaemic inv(16) will clearly be of practical (seeabove) and biological interest. Furthermore, iso-lation of this putative proto-oncogene at i6p13 maybe of importance in understanding the molecularbasis of diseases closely linked to the a globin genelocus at l6pl3 1.

We thank Dr C Potter for culture of the cell lines,H Teal for DNA extractions, Dr M Aronson forcommunication of unpublished cytogenetic data, andL Roberts for typing the manuscript. We also thankDr J B Clegg for support and encouragement andDrs K E Davies, S T Reeders, W G Wood, and J BClegg for comments on the manuscript. RDN wasinitially supported by a Research Scholarship fromthe Royal Commission for the Exhibition of 1851and for the latter part of this project was supportedby the Rockefeller Foundation.

References

Botstein D, White RL, Skolnick M, Davis RW. Construction ofa genetic linkage map in man using restriction fragment lengthpolymorphisms. Am J Hum Geniet 1980;32:314-3 1.

2 Higgs DR, Wainscoat JS, Flint J, et al. Analysis of the human a-globin gene cluster reveals a highly informative genetic locus.Proc Nail Acad Sci USA 1986;83:5165-9.

3 Deisseroth A, Nienhuis A, Turner P, et al. Localization of thehuman a-globin structural gene to chromosome 16 in somaticcell hybrids by molecular hybridization assay. Cell 1977;12:205-18.

4 Barton P, Malcolm S, Murphy G, Ferguson-Smith MA.Localization of the human x-globin gene cluster to the short armof chromosome 16 (16p12-16pter) by hybridization in situ. JMol Biol 1982;156:269-78.Wainscoat JS, Kanavakis E, Weatherall DJ, et al. Regionallocalization of the human a-globin genes. Lancet 198 l;ii:301-2.

6 Weatherall DJ, Clegg JB. The thalassaemiiia syndromes. 3rd ed.Oxford: Blackwell, 1981.

7 Higgs DR, Weatherall DJ. Alpha-thalassemia. Curr TopHematol 1983;4:37-97.

8 Reeders ST, Breuning MH, Corney G, et al. Two geneticmarkers closely linked to adult polycystic kidney disease onchromosome 16. Br MedJ 1986;292:851-3.

9 Reeders ST, Breuning MH, Davies KE, et al. A highlypolymorphic DNA marker linked to adult polycystic kidneydisease on chromosome 16. Nature, 1985;317:542-4.

10 Weatherall DJ, Higgs DR, Bunch C, et al. Haemoglobin Hdisease and mental retardation. N Engl J Med 1981;305:607-12.NIGMS Human Genetic Mutant Cell Repository. Catalog ofcell lines. Camden, NJ: National Institutes of Health, 1985:187,350.

12 lonasescu VV, Patil S, Hart M, Rhead W. Abnormality ofchromosome 16 associated with myopathy. Aoi J Med Genet1 985;37:60A.

13 Old JM: Higgs DR. Gene analysis. The thalassaemias. MethodvHaemaiol 1983;6:74-102.

14 Nicholls RD, Hill AVS, Clegg JB, Higgs DR. Direct cloning ofspecific genomic DNA sequences in plasmid libraries followingfragment enrichment. Vucleic Acids Res 1985;13:7569-78.

ES Jarman AP, Nicholls RD, Weatherall DJ, Clegg JB, Higgs DR.Molecular characterisation of a hypervariable region down-stream of the human a-globin gene cluster. EMBO J 1986;5:1857-63.

16 Higgs DR, Goodbourn SEY, Lamb J, et al. a-thalassaemia

45



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/


caused by a polyadenylation signal mutation. Nature 1983;306:398-400.

'7 Burns J, Chan VTW, Jonasson JA, et al. Sensitive system forvisualising biotinylated DNA probes hybridised in situ: rapid sexdetermination of intact cells. J C/in Pathol 1985;38:1085-92.

18 Chan VTW, Fleming KA, McGee JO'D. Detection of sub-picogram quantities of specific DNA sequences on blot hybridi-zation with biotinylated probes. Nucleic Acids Res 1985;13:8083-91.

9 Ferguson DJP, Burns J, Harrison D, Jonasson JA, McGeeJO'D. Chromosomal localization of genes by scanning electronmicroscopy using in situ hybridisation with biotinylated probes:Y chromosome repetitive sequences. Histochem J 1986;18:266-70.

20 Goodbourn SEY, Higgs DR, Clegg JB, Weatherall DJ.Molecular basis of length polymorphism in the human C-globingene complex. Proc Nail Acadl Sci USA 1983;80:5022-6.

21 de Grouchy J, Turleau C. Clinical aila.s of human chroniosonies.New York: Wiley, 1984:436-47.

22 Winsor EJT, Palmer CG, Ellis PM, Hunter JLP, Ferguson-Smith MA. Meiotic analysis of a pericentric inversion, inv(7)(p22q32), in the father of a child with a duplication-deletion ofchromosome 7. C)togenet Cell Genet 1978;20:169-84.

23 Savage JRK. Assignment of aberration breakpoints in bandedchromosomes. Nature 1977;270:513-4.

24 Leder A, Swan D, Ruddle F, D'Eustachio P, Leder P.Dispersion of a-like globin genes of the mouse to three differentchromosomes. Nature 1981*293:196-200.

25 Popp RA, Lalley PA, Whitney JB, Anderson WF. Mouse a-globin genes and a-globin-like pseudogenes are not syntenic.Proc Natl Acad Sci USA 1981;78:6362-6.

26 Buckle VJ, Edwards JH, Evans EP, et al. Chromosome maps ofman and mouse II. C/in Genet 1984;26:1-1 1.

2 t Marks J, Shaw JP, Shen CKJ. Sequence organization andgenomic complexity of primate 01 globin gene, a novel a-globin-like gene. Nature 1985;321:785-8.

28 Sutherland G. Abstract.s of the Human Genetics Society ofAustralia 1986.

29 Hassold TJ, Jacobs PA. Trisomy in man. Ann Rev Genet1984;18:69-97.

30 Antonarakis SE, Kittur SD, Metaxotou C, Watkins PC, PatelAS. Analysis of DNA haplotypes suggests a genetic pre-disposition to trisomy 21 associated with DNA sequences onchromosome 21. Proc Natl Acad Sdi USA 1985;82:3360-4.

31 Hassold TJ, Kumlin E, Leppert M, Takaesu N. Determinationof the parental origin of sex chromosome monosomy usingrestriction fragment length polymorphisms. Amt J Hum Genet1985;37:965-72.

32 Keats B. Genetic mapping: chromosomes 6-22. Am7 J Humi1Genet 1982;34:730-42.

33 Robson EB. The human gene map. In: Human genetics. Part A.The unfolintg genome. New York: Liss, 1982:85-101

34 Schwartz DC, Cantor CR. Separation of yeast chromosome-sized DNAs by pulsed field gradient gel electrophoresis. Cell1 984;37:67-75.

35 Collins FS, Weissman SM. Directional cloning of DNAfragments at a large distance from an initial probe: acircularization method. Proc Natl Acad Sci USA 1984; 81:6812-6.

36 Kazazian HH, Antonarakis SE, Wong C, et al. Ring chromo-some 21: characterization of DNA sequences at sites of breakageand reunion. Ann NY Acad Sci 1985;450:33-42.

37 Cavenee WK, Dryja TP, Phillips RA, et al. Expression ofrecessive alleles by chromosomal mechanisms in retinoblastoma.Nature 1983;305:779-84.

3 Murphree AL, Benedict WF. Retinoblastoma: clues to humanoncogenesis. Science 1984;223:1028-33.

3 Koufos A, Hansen MF, Copeland NG, et al. Loss ofheterozygosity in three embryonal tumours suggests a commonpathogenetic mechanism. Nature 1985;316:330-4.

40 Dracopoli NC, Houghton AN, Old U. Loss of polymorphicrestriction fragments in malignant melanoma: implications fortumor heterogeneity. Proc Natl Acad Sci USA 1985;82:1470-4.

41 Hansen MF, Koufos A, Gallie BL, et al. Osteosarcoma andretinoblastoma: a shared chromosomal mechanism revealingrecessive predisposition. Proc Natil Acad Sdi USA 985;82:62 16-20.

42 Raizis AM, Becroft DM, Shaw RL, Reeve AE. A meioticrecombination in Wilms tumor occurs between the parathyroidhormone locus and I lpl3. Hum Genet 1985;70:344-6.

43 Michelson AM, Orkin SH. Boundaries of gene conversionwithin the duplicated human a-globin genes: concerted evolutionby segmental recombination. J Biol Chem 1983;258:15245-54.

44 Higgs DR, Hill AVS, Bowden DK, Weatherall DJ, Clegg JB.Independent recombination events between the duplicatedhuman a-globin genes: implications for their concerted evo-lution. Nucleic Acids Re.s 1984; 12:6965-77.

45 Nicholls RD, Higgs DR, Clegg JB, Weatherall DJ. a'-thalassemia due to recombination between the al-globin geneand an Alul repeat. Blood 1985;65:1434-8.

46 Fonatsch C. New chromosome polymorphism: inv(l 6) (p1 lq 12or 13). Cytogenet Cell Genet 1977;18:106-7.

47 Miller K. Pericentric inversion 16 in man-a second case. CliiGenet 1986;29:181-2.

48 LeBeau MM, Larson RA, Bitter MA, et al. Association of aninversion of chromosome 16 with abnormal marrow eosinophilsin acute myelomonocytic leukemia. N Entgl J Med 1983;309:630-6.

-9 Arthur DC, Bloomfield CD. Letter to the editor. Blood1 983;62:93 1.

50 Tantravahi R, Schwenn M, Henkle C, et al. A pericentricinversion of chromosome 16 is associated with dysplasticmarrow eosinophils in acute myelomonocytic leukemia. Blood1983;63:800-2.

5' Hogge DE, Misawa S, Parsa NZ, Pollak A, Testa, JR.Abnormalities of chromosome 16 in association with acutemyelomonocytic leukemia and dysplastic bone marrow eosino-phils. J Cliii Oncol 1984;2:550-7.

52 Berger R, Bernheim A, Daniel MT, ct al., Cytogenetic studieson acute myelomonocytic leukaemia (M4) with eosinophilia.Leukemia Res 1985;9:279-88.

53 LeBeau MM, Diaz MO, Karin M, Rowley JD. Metallothioneingene cluster is split by chromosome 16 rearrangements inmyelomonocytic leukaemia. Nature 1985;313:709-1 1.

54 Haslinger A, Karin M. Upstream promoter element of thehuman metallothionein-IIA gene can act like an enhancerelement. Proc Natl Acad Sci USA 1985;82:8572-6.

Correspondence and requests for reprints to Dr R DNicholls, MRC Molecular Haematology Unit, Room7501, John Radcliffe Hospital, Headington, OxfordOX3 9DU.



g.bmj.com

/J M



nloaded from

http://jmg.bmj.com/

Documents

of · lities'2 associated with the karyotype 46,XY,rec(16), which results in partial trisomy for 16pl3-1-÷pter and partial monosomyfor 16q22-+qter. Ascontrols, DNA was obtained from