Rearrangement of Variable Region T Cell Receptor y Genesin Acute Lymphoblastic LeukemiaVy Gene Usage Differs in Mature and Immature T Cells

Junichi Hara,* Stephen H. Benedict,* Keiko Yumura,* Kyungsae Ha-Kawa,l and Erwin W. Gelfand**Division of Basic Sciences and the Sackler Foundation Laboratory, Department ofPediatrics, National Jewish Center for Immunologyand Respiratory Medicine, Denver, Colorado 80206; tDivision of Immunology/Rheumatology, Research Institute, The HospitalforSick Children, Toronto, Ontario, Canada M5GIX8; and the tDepartment of Pediatrics, Osaka University Hospital, Osaka, Japan 553

Abstract

Using probes recognizing variable regions (Vet) and joiningregions (Jy) of the T cell receptor (TCR) -y gene, we haveanalyzed the usage of V-y genes in 24 patients with T cell acutelymphoblastic leukemia (ALL) and 36 patients with B-precur-sor ALL. In CD3- T-ALL derived from immature T cells, V-ygenes more proximal to Jy were frequently rearranged; Vy8,V'y9, Vy10, and Vy1 1 were used in 19 of 24 rearrangements.In contrast, CD3YT-ALL derived from a more mature stage ofT cell ontogeny, showed a high frequency of rearrangementsinvolving V-y genes distal to Jy; V'y2, Vy3, V'y4, and Vy5 wereused in 17 of 25 rearrangements. In B-precursor ALL, no nota-ble bias of Vy gene usage was observed. This probably reflectsthe possibility that TCRgenes may not rearrange according toa T cell hierarchy when under control of a B cell gene program.Furthermore, deletions of those Vy genes located 3' to rear-ranged Vy genes were observed in all patients analyzed. Thissupports the theory that loop deletion is a major mechanismfor TCR--y gene rearrangement.

Introduction

In addition to the T cell receptor (TCR)'a and : genes, a thirdgene, y, has been shown to undergo somatic rearrangement inT cells (1, 2). Similar to other rearranging genes, the TCR-ygene is composed of variable (V), joining (J), and constant (C)regions (2, 3). The number of Vy genes is much smaller thanobserved with other rearranging genes; to date 14 Vy genes,including 6 pseudogenes have been identified upstream to Jyand &y (4). These Vy genes are divided into four families(V'y, V-yII, VyIII, and VyIV) (4) and are shared by two sets ofJy and Cy genes that show a strong homology to each other (2,3, 5).

Dr. Hara's present address is Department of Pediatrics, Osaka Univer-sity Hospital, Osaka, Japan 553. Address reprint requests to Erwin W.Gelfand, Department of Pediatrics, National Jewish Center for Immu-nology and Respiratory Medicine, 1400 Jackson Street, Denver, CO80206.

Received for publication 29 August 1988 and in revised form 18November?1988.

1. Abbreviations used in this paper: ALL, acute lymphoblastic leuke-mia; C, constant region; J, joining region; TCR, T cell receptor; V,variable region.

The protein product of the TCR-y gene has been shown toform a T cell receptor in association with the CD3 moleculeson the cell surface and may be involved in some types ofnatural killer function (6-9). Although the specific roles of thisTCR and the T cell subset expressing this receptor are stillobscure, in general the TCR-y gene along with the other puta-tive TCRgenes probably encode a diverse set of receptor pro-teins involved in antigen recognition (8, 9).

In previous studies, we and others demonstrated a highincidence of TCR-y gene rearrangement not only in T cellacute lymphoblastic leukemia (T-ALL) but also in B-precursorALL (10-12). In this study we have used probes recognizingVy and Jy, and have analyzed the usage of Vy genes in 24T-ALL and 36 B-precursor ALL. Vy genes more proximal toPy were frequently used in T-ALL derived from immaturecells. In contrast, T-ALL derived from more mature T lineagecells demonstrated a high frequency of rearrangements in-volving Vy genes more distal to Jy and the most upstream(distal) functional Vy gene showed the highest frequency ofrearrangement. In B-precursor ALL, no particular clusteringwas observed. This observation may suggest that TCR-y generearrangement in B-precursor ALL is adventitious and doesnot follow the hierarchy observed in T lineage cells. Further-more, deletions of Vy genes located downstream to rearrangedV~y genes were observed in all patients analyzed, indicatingloop deletion (13, 14) as the main mechanism for TCR-y generearrangement.

Methods

Cell samples. Mononuclear cells were obtained from bone marrow byFicoll-Hypaque gradient centrifugation at the time of diagnosis andbefore initiation of treatment. The samples evaluated contained > 90%malignant cells. Reactivity of malignant cells with a panel of MAbwasassessed by indirect immunofluorescence (15). The diagnosis ofT-ALL was based on the expression of the T cell-associated antigensCD1, CD2, CD3, CD4, CD5, CD7, and CD8. T-ALL were divided intotwo subgroups according to CD3expression (Table I). The diagnosis ofB-precursor ALL was based on the expression of the B-cell-associatedantigens CD19 and/or CD20 and the lack of surface Ig and T cell andmyeloid-associated antigens. The B-precursor ALL studied here fellinto two subgroups (16) (Table II): group III, Ia', CD19', and CDI0+;and group IV, Ia', CDl9', CD10, and CD20'.

Southern blot analysis. High molecular weight DNAwas extractedby standard procedures from the mononuclear cells and the DNAsamples were digested with restriction endonucleases. The digestedDNAwas subjected to electrophoresis in agarose gels and transferredto nylon membranes. DNAsamples bound to membranes were thenhybridized to probes labeled with 32P by the random primer method(17). Membranes were hybridized sequentially after removal of hybrid-ized probes. Complete removal was confirmed by a test exposure be-fore the next hybridization.

Usage of Vy Genes in Lymphoblastic Leukemia 1277

J. Clin. Invest.© The American Society for Clinical Investigation, Inc.0021-9738/89/04/1277/07 $2.00Volume 83, April 1989, 1277-1283

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.

Wl YVyA Vyll Vylll VyIVI"1l Vy2 Vv3 Vy4 Vy5 'Vy5 YVy6 YVy7 Vy8 vy V0lo Villo I__ 0 r...I , , , ,1 to , , ,, II la

E E E E E E E E EE E E E

A

G 53 10 20do1

Jyl.l Jyl.2 Jy1.3..., I

B E HEI E H E H E

HJy

CylIJy2.1 Jy2.3 Cy2

I I * I' 0YIIII 1 11 liiH BHEESHEEE HEHE HE H B E

HJy

Hkb

G (.40)-'-I-V7 *(-3 5) -

*$_' i 1'PV5 +TPV6.. E~~(.4 - AM_

f?3\V2+V3 (70)0_

zL Add ~~~V8(3.8). _- V8*V'5 TV7 (3.6)- -

3V4(.1)- -

5.V4 (2.4)1

Figure 1. (Top) Diagram of Vy region. E denotes Eco RI restrictionsites. (Bottom) Restriction mapof Jy and C'y region and the Jyprobe used in this study. B, E, and H denote BamHI, Eco RI, andHind III restriction sites, respectively. H* indicates a polymorphicHind III restriction site.

-V3i (16)

5 V2 (1.8)-i-V2- (11.7)

VyIBam HI

Gene probes. A restriction map of the TCRy region is presented inFig. I and was derived from data presented in references 3, 4, and 18.The Jy probe used in this study was an 0.7-kb Eco RI-Hind III genomicfragment provided by Dr. T. H. Rabbitts (2). The JPy probe hybridizedto Jy 1.3 and Jy2.3 but not to Jy 1l.1, Jy 1.2, and Jy2. I (Fig. 1) ( 18). TheV-y probes used in this study, provided by Dr. T. W. Mak, were as

follows: VyI (360-bp Eco RI-Ava II fragment of cDNAHGPO3[19]),V'yII (290-bp Eco RI-Acc I fragment of cDNA HGPO2(19); VyIII(90-bp Eco RI-Apa I fragment of cDNAHGPO6[19]). The V'yI andVyII probes contained complete sequences of the VyI and V'yII genes,

respectively. The V'yIII probe contained only that portion 3' to the EcoRI site located in the VyIII gene. The VyI probe hybridized to all V'YIfamily members (OVy 1, Vy2, Vy3, Vy4, Vy5, IV-y5, IVy6, 'Vy7,and Vy8) and detected eight bands in Eco RI digests of germline DNA(4). Designations for each band are shown in Fig. 6. The VyI probedetected a single germline band after Barn HI digestion indicating thatall VyI genes lie within a single BamHI fragment. After hybridizationwith the V'yII probe, Bam HI, Eco RI, and Hind III digests yieldedsingle germline bands. Of these, Eco RI and Hind III bands containedonly V-y9. The VyIII probe detected single germline bands after BamHI, Eco RI, and Hind III digestion. Hind III bands contained onlyV'y 1O, but BamHI and Eco RI bands contained both V-y9 and Vy 10,

and VylIO and V-y I1, respectively. The order of genomic Vy genes isshown in Fig. 1 (4, 20, 21). Rearrangements to Vy 11 may representrearrangements of 'V'yB (4) since they localize to the same Eco RIfragment. Based on the sizes of the rearranged fragments, no rearrange-ments of PV-yB were found.

Results

C

G 39 40

V9* (4.0)- %.

G (2.4)- OM

VYlIHind III

D

G 1 10

E- VI * (5. 7)

G (3.8)-

VyIIIHind III

Figure 2. Representative rearrangement patterns of the VY genes.

The patient numbers are noted above each lane. Lane G representsthe germline DNAcontrol. The germline position and size of the in-dividual bands recognized are indicated by bars at the side of eachpanel. (A and B) BamHI and Eco RI digests, respectively, probedwith the VyI probe. Cis a Hind III digest probed with the VyIIprobe. D shows a Hind III digest probed with the VYII1 probe. Spe-cific rearranged bands are indicated by *.

Analysis of TCR-,y gene rearrangement. Fig. 2 demonstrates

how the various Vy rearrangements are recognized. In Fig. 2

A, members of the Vy family are recognized using BamHI

and the VyI probe. Rearrangements of the V2, V3, and IV7

genes are recognized using this enzyme. Fig. 2 B shows thatrearrangements of the V4, V5, and V8 genes are recognizedusing Eco RI and the VyI probe. Fig. 2 Cdemonstrates detec-tion of rearrangement of Vy9, the only member described forthe VyII family using Hind III and the VyII probe. Finally,Fig. 2 D uses Hind III and the VyIII probe to recognize rear-

rangement of Vy 1O, the only member as yet described for theVyIII family.

Fig. 3 demonstrates recognition of rearranged Jy genes

using the Jy probe and the three restriction enzymes. As de-scribed by Forster et al. (4), those Vy genes rearranged to Jy 1.3or Jy2.3 in a particular cell were assigned based on sizes of

rearranged Jy bands after separate digestion with three en-

zymes (Bam HI, Eco RI, and Hind III) and using a Jy probe.WhenVy genes were rearranged to Jy genes other than Jy 1.3or Jy2.3, the usage of Jy and Vy genes was estimated byhybridization patterns obtained using specific Vy probes. Foreach representative patient, rearranged Jy bands are indicatedby a line and the particular Vy gene that has rearranged to thatJy is given next to that line. Results presented in Figs. 2 and 3,in conjunction with numerous other experiments, confirm thepublished results (4) that Vy and Jy rearrangements are com-

paratively limited in number and can be assigned on the basisof size. The data also indicate that of six pseudo Vy genes,

'Vy7 seems most capable of being rearranged in T cell or

B-precursor acute lymphoblastic leukemia. Results describedin this paper used these procedures to investigate Vy and Jy

gene rearrangement in acute lymphoblastic leukemia. In pre-

1278 J. Hara, S. H. Benedict, K. Yumura, K. Ha-Kawa, and E. W. Gelfand

B

G 58

- -V5*

VylEcoR I

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.

B

G 1 34 10 20

V8 ,

11 (2n)- -v1o.F

0Ik~r-ul--- vioV9 ..V3 Jl,J2

J2(13)- 4- (2.1). ..NV11 -V2

G 15 17 13 G 55 26

-vi1

V3NaoO" N*-V

V8 . NVJ2 (3.2) - "O

moo T~~~V7V9

Ji (1.5)-:--

V2

BamHi Hind III EcoR I

Figure 3. Representative36 rearrangement patterns of

J' genes. Patient num-bers are noted above eachlane. Lane Gshows thegermline control, and the

V3 germline position and es-timated size are indicatedat the left of each panel.Specific rearranged bandsfor each patient are indi-cated by a small bar lead-ing to the particular VTgene which the size of theband suggests has rear-ranged in that patient asnoted by Forster et al.(4). (A, B, and C) showDNAcut with BamHI,Hind III, and Eco RI, re-spectively. All wereprobed with the J-y probe.

vious reports (10, 11) we demonstrated rearrangements ofTCRy in 24 of 24 T-ALL and 36 of 57 B-precursor ALL. Inthis study, we have analyzed Vy and Jy usage in these samepatients. Results of these studies are summarized in Table I forT-ALL and Table II for B-precursor ALL. Data from the indi-vidual blots are shown only to make specific points.

All T-ALL and 16 B-precursor ALL had biallelic rear-rangements and 20 B-precursor ALL had single allelic rear-rangements of the TCR-,y gene. In 22 T-ALL, germline bandscorresponding to 'y 1 and y2 loci were not detected after BamHI, Eco RI, or Hind III digestion indicating rearrangements toJy2.3 on both alleles (Table I). In contrast, the majority ofB-precursor ALL (14 of 16 patients) with biallelic rearrange-ments retained y2 in germline configuration. Both y 1 and y2germline bands were deleted in only 2 B-precursor ALL (TableII). Therefore, it seems that most rearrangements in B-precur-sor ALL occurred in the y 1 locus.

Usage of Vy genes in CD3- T-ALL. The usage of individ-ual Vy genes in T-ALL is summarized in Table I and Fig. 4. 12patients were examined and these contained a total of 24 rear-rangements. In CD3- T-ALL, 19 of 24 rearrangements of theVy genes were located proximal to Jy. The usage of Vy8 wasmost frequent; 9 of 24 rearrangements showed involvement ofV~y8. Vy9, belonging to the VyII family, was also frequentlyused; Vy9 was rearranged on five alleles. Rearrangements ofVy 10 (VyIII) and Vy 11 (VyIV) were observed on two andthree alleles, respectively. In contrast, involvement of V'yIgenes located in an upstream region were less frequent; V'y2and Vy4 rearranged on one allele each, and Vy3 rearrangedon three alleles. Involvement of IV-y 1, IV-y5, 'Vy6, IVy7,and IVyA was not detected.

Usage of V-y genes in CD3+ T-ALL. The Vy gene usage inCD3+T-ALL is summarized in Table I and Fig. 4. 12 patientswere examined and these contained a total of 25 rearrange-ments. In contrast to the findings in CD3- T-ALL, VyI genesdistal to Jy were rearranged in the majority of CD3+ T-ALL.Amonga total of 25 rearrangements, Vy2 was most frequentlyused; eight alleles showed rearrangements of Vy2. Vy3 and

V~y5 were also frequently used; 3 and 5 alleles had rearrange-ments to V'y3 and Vy5, respectively. Vy4 and TVy7 rear-ranged on a single allele, respectively, and Vy8 showed rear-

Table L Surface Antigen Expression and Vy Gene Usage in 24Patients with T Cell Acute Lymphoblastic Leukemia

Surface markers*

Patient CD3 CD4 CD8 CD2 CD7 Vy gene rearranged to Jy

1 - - - - + V-y8,VyI12 - - - + + Vy8, Vy93 - - - + ND Vy8, Vy94 - - + + ND Vy8, Vy95 - - + + ND Vy3, Vy86 - + - + ND Vy8,Vyll7 - + + + ND Vy3, Vy88 - + 30 20 ND Vy8, Vy99 - + + + ND Vy2,Vy4

10 - + 42 + ND Vy3,Vy0O11 - + + + ND V710,V'11I*12 - ND ND + + Vy8, Vy913 + + - - + Vy2, Vy1114 + - 20 + ND Vy4, 4IVy7, Vy9*15 + 40 40 + ND Vy3, Vy816 + 19 - + ND V72, Vy817 + 10 + + ND V75,Vy518 + 29 + + ND Vy3,Vy519 + + + + ND Vy2, Vy820 + - - + + Vy2,VylO21 + 27 - - ND Vy3,Vy822 + 42 45 + ND Vy2,Vy523 + - - + ND V72,Vy524 + + - + ND Vy2,Vy2

* A minus sign denotes <10% positive cells and a plus sign >50%positive cells; numbers are specific percentages of positive cells.t Rearrangements to Jy 1. 1, Jy 1.2, or Jy2. 1.

Usage of Vy Genes in Lymphoblastic Leukemia 1279

A C

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.

Table I. Phenotype and V'y Gene Usage in 36 Patients with B-Precursor Acute Lymphoblastic Leukemia

Germlinet GermlinetVy gene rearranged

Patient Phenotype* y 1 y2 V7 gene rearranged to J'y Patient Phenotype y 1 'y2 to JY

25 III + + Vy4 43 IV + + Vy326 III + + VY9 44 IV + + VY927 III + + IV'y7 45 III - + Vy5, IVy728 III + + V'y4 46 III - + V-y4, Vy929 III + + V-y2 47 III - + Vy9, Vy9130 III + + Vy4 48 III - - Vy2, Vy231 III + + V'5 49 III - + Vy4, Vy932 III + + Vy5 50 III - + Vy2, V-yI§1133 III + + Vy9 51 III - + IVy7, Vy934 III + + V-y9 52 III - + Vy2, Vy335 III + + Vy9 53 III - + IVy7, Vyl l36 III + + Vy3 54 III - + Vy2, IVy737 III + + Vy9 55 III - + Vy2, IV'y738 IV + + Vy9 56 III - + VYIlI, VyIN39 IV + + Vy8 57 III - + Vy2, Vy540 IV + + Vy4 58 III - - V-y5, Vy841 IV + + Vy9 59 III - + V'y3, Vy4, Vy942 IV + + Vy8 60 IV - + Vy2, IV-y7

* III: Ia+, CDl9+, and CD10+, IV: Ia', CD19+, CD10+, and CD20+. $ Results after Bamr HI digestion and hybridization with the Jy probe. +,retention of germline bands; -, deletion of germline bands. § Rearrangements to Jy 1.1, Jy 1.2, or J-y2. 1. 1" Rearrangements to one of the VYIfamily.

rangements on four alleles. V-y9 (VyII), Vy 10 (VyIII), andV'y 11 (VyIV) were used on only one allele each.

Usage of Vy genes in B-precursor ALL. The usage of indi-vidual Vy genes in B-precursor ALL is shown in Table II andFig. 5. 57 patients with B-precursor ALL were examined ofwhich 36 showed rearrangement of the TCRy gene. In these36 patients, a total of 53 rearranged genes were observed. Vy2and V'y9 were rearranged on a total of 9 and 14 alleles, respec-tively. Involvement of Vy3, Vy4, Vy5, and Vy8 were ob-served on four, seven, five, and three alleles, respectively. Re-arrangements of IVy7 were also frequently observed; sevenalleles showed involvement of IVy7. The usage of Vy 1(VyIV) was observed on two alleles. Rearrangements of Vy 10(V~yIII) and pseudogenes other than IVy7 were not observed.

z

0zccco

wcr

a:rw

z

10-

8-

6-

4-

0-~~.J- LL0 1 iVy1 Vy2 V13 Vy4 Vyw5 TVy TVi 'Vy7 Vy IVyA Vy9Vy10Vyi 1VyGENE

Figure 4. Usage of Vy genes in 24 patients with T-ALL. Columns in-dicate the particular Vy gene rearranged in 12 CD3- T-ALL (i) and12 CD33 T-ALL patients (m).

Thus, no notable selectivity of Vy gene usage was observed inB-precursor ALL. However, when these patients were subdi-vided into two subgroups; 20 patients with single allelic rear-rangements and 16 patients with biallelic rearrangements,most Vy2 and 'Vy7 rearrangements were observed in pa-tients with biallelic rearrangements.

Deletion pattern of Vy genes. Eco RI digests probed withV7yI afford the opportunity to observe all of the VyI genes onthe same blot. This was demonstrated above (Fig. 2 B). Inaddition to rearrangements, the complexity of the pattern fa-cilitates detection of deletions that occur during rearrange-ment. To analyze the deletion pattern of V7 genes, we exam-ined the Eco RI digests that had been hybridized with the VyIand VyII probes in 12 patients with T-ALL. BamHI and Hind

zw

a:

z

w

w

zU

10 -

8-

6-

4-

0- -0VVyl Vy2 Vy3 Vy4 Vi'PVy6PVy7 Vy8 t'VyA Vy9VyVyii1

VyGENE

Figure 5. Usage of Vy genes in 36 patients with B-precursor ALL.Figure indicates which Vy gene rearranged in 20 patients with singleallelic rearrangements (X) and 16 patients with bi-allelic rearrange-ments (n).

1280 J. Hara, S. H. Benedict, K Yumura, K. Ha-Kawa, and E. W. Gelfand

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.

1 3 9 1 9 1 7 20_- 1

_ _ IJ,-...%~~~~~~~4wl~~~~~~~~~ ar<Zit.

V2*,V4* V2*I

v2*

Figure 6. Representative deletion patterns of VyI genes using EcoRI. The patient numbers are noted above each lane. Lane Gand G'show the germline controls cut with Eco RI. The specific Vy geneand the size (in kilobases) corresponding to each germline band are

indicated by bars between blots. *Rearranged Vy genes.

III digests hybridized with the V-yII and VyIII probes were alsoexamined. As illustrated in Fig. 6 and summarized in TableIII, most patients showed deletion of Vy genes interveningbetween the rearranged Vy genes and the 3'-most terminus.These findings support the hypothesis that loop deletion (13,14) is the primary mechanism for Vy gene rearrangement. Asdescribed earlier (Tables I and II) Pt. 24 was found to be rear-

ranged to Vy2 on both alleles, and Fig. 6 shows that Pt. 24 hasdeleted all the V region genes intervening between V72 andthe constant region. Pt. 13 was found to be rearranged to V-y2and Vy 11, and therefore an apparent germline pattern is ob-

served with this patient. This is solely due to the allele thatrearranged to Vy 1 1. Pt. 9 was shown to rearrange to Vy2 andVy4, and Fig. 6 shows that all genes located 5' to Vy4 are ingermline whereas those genes located between V-y4 and the

constant region have been deleted. A similar situation oc-curred with Pt. 17. Pt. 1 with rearrangements to V-8 andV~y 11 showed retention of all VyI, VyII, and VyIII genes, andno germline V'yIV (Vy 1 1) genes identified by the VyIII probeand Eco RI digestion.

Interesting hybridization patterns of the VyI probe wereobserved in three patients (Pts. 10, 19, and 20) with rearrange-ments involving Vy3 and Vy10, Vy2 and Vy8, and Vy2 andV710, respectively. In these patients, 7.0-kb bands corre-sponding to Vy2 and Vy3 and 2.4-kb bands corresponding to5'Vy4 were lost with the retention of 9.4-kb bands corre-sponding to IVy5 and IV'y6. Because the intensities of the9.4-kb bands increased in spite of single allelic deletions of'Vy5 and IVy6 caused by rearrangements to Vy2 or V'y3, itis possible that an Eco RI restriction site between Vy3 andVy4 may be polymorphic and this polymorphic Eco RI frag-ment containing Vy2, Vy3, and 5Vy4 may be overlaid on thenormal 9.4-kb band. This possibility is supported by the find-ings of increased intensities of the 9.4-kb bands compared withthe 7.0-kb bands in germline DNA(Fig. 6, lane G'), and by thelack of aberrant rearranged bands when samples cut with HindIII or BamHI were probed.

Discussion

Acute lymphoblastic leukemia results from the clonal expan-sion of cells arrested at different stages of lymphoid ontogeny(22, 23). These leukemic cells provide models for analyzing themolecular events that occur during lymphocyte development(10, 11, 15, 24-26). In this study, we have analyzed the usageof Vy genes in patients with T-ALL and B-precursor ALL.Based on the expression of the CD3 antigen, the patients withT-ALL were divided into two subgroups: CD3- T-ALL cellspresumably representing T lineage cells arrested early in on-togeny; CD3+ T-ALL cells presumably derived from a moremature stage of T cell ontogeny. In CD3- T-ALL, rearrange-ments frequently involved Vy8 and Vy9, which are locatedproximal to Jy, and rearrangements of Vy genes located distal

Table III. Patterns of Vy Gene Deletions in 12 Patients with T Cell Acute Lymphoblastic Leukemia

V-yI V-I* V-yIII,

Patient 'V'y I 5V-y2 Vy2, Vy3 5V-y4 V'y4 I'V-y5, *V-y6 V-y5, *V-y7 V-y8 V-y9 Vy 10 Usage of V-y genes

1 + + + + + + + + + + Vy8,Vy18 + + + + + + + + R/D - Vy8, Vy99 + + + + - - - - - - Vy2,V-y4

10 + + - - + + + + + R/- Vy3,V-ylO13 + + + + + + + + + + Vy2, VyI 115 + + + + + + + - - - Vy3, V-y817 + + + + + - - - - - Vy5, Vy519 + + _ _ + + + - - - V'y2, V-y820 + + - - + + + + + R/- Vy2,VylIO22 + + + + + - - - - - Vy2, Vy523 + + + + + - - - - - Vy2,Vy524 + + - - - - - - - - V'y2,Vy2

+, germline, -, deleted, and R, rearranged. * Results after Eco RI digestion. Results after Eco RI and Hind III digestion. § Results afterHind III digestion.

Usage of V"y Genes in Lymphoblastic Leukemia 1281

G G 2

_#Arm& _ _Ti1V5 + V6(9V4)(.3' V2 + V3(7.(0)TV-11 (5.8) L

as V8 (3.8) -l_ _ W - ~V5,'i-'V7 (3.6)L

3' V4 (3- 1)L

5 4V4 (2.4)

5' V2 (t .8)n

- V2'

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.

to Jy were less frequently observed. In contrast to the findingsin CD3- T-ALL, leukemic cells from CD3+T-ALL frequentlyshowed rearrangement of Vy2, which is the most upstreamfunctional Vy gene detected so far and in these cells, those V'ygenes located distal to Jy (Vy2, Vy3, Vy4, and Vy5) wererearranged on the majority of alleles. In contrast to a recentpublication (27), we were unable to detect a bias for or againstusage of specific Jy loci, even when cells displayed a decidedbias for Vy usage. This selectivity of Vy gene usage is similarto that observed in Ig heavy chain gene rearrangements; VHgenes proximal to JH are preferentially used in murine imma-ture B lineage cells and during further differentiation, expres-sion of VHgenes proximal to JH is replaced by expression of VHgenes located more 5' (distal) to JH (28, 29).

Others have observed that the proportion of rearrange-ments involving Jy genes distal to Cy was greater in immatureT cells than mature cells (19). Wehave demonstrated similarfindings for the TCR-a gene; immature T lineage cells prefer-entially used Ja genes located proximal to Va genes and thiswas in sharp contrast to results using mature T lineage cells(1 1, 30). From work presented here, it appears that V'y genesproximal to Jy may initially rearrange to Jy genes locateddistal to the C'y region in the more immature T cells andduring further differentiation the cells may successively rear-range using more 5' Vy genes and 3' Jy genes with concomi-tant deletion of the preexisting VJy complex. It is also possiblethat 5' Vy genes may undergo secondary rearrangements re-sulting in Vy gene replacement without Jy gene replacementin an analogous manner to that of the Ig heavy chain gene (31,32). Because our conclusions are drawn from examination ofleukemic cells, the possibility exists that the CD3- leukemic Tcell does not represent the normal precursor for the CD3' Tcell but are rather dead-end cells. Using the Vy probes, dele-tions of Vy genes located 3' to recombined Vy genes wereobserved in all 12 patients analyzed. These findings suggestthat loop deletion is the major mechanism for TCR--y generearrangement and for TCR-a and (3 genes (13, 14).

Leukemic cells from B-precursor ALL showed no notablepreference of Vy gene usage. When patients were subdividedbased on single or biallelic rearrangements, most rearrange-ments to V72 were observed in cells with rearrangements onboth alleles and are presumably more differentiated with re-spect to TCR-'y gene rearrangement (1 1). Leukemic cells withrearrangements on single alleles rarely demonstrated rear-rangement of Vy2 and rearrangements of Vy9 were the mostfrequent. However, unlike the case with T-ALL, definition ofthe stage of development by V region selection in B-precursorALL does not coincide with definition of developmental stageusing cell surface markers. It seems probable, therefore, thatrearrangement of the TCR-y gene in B-precursor ALL doesnot follow the maturational hierarchy outlined above forTCR-y gene rearrangement in T-ALL.

Acknowledaaments

Weare indebted to Dr. T. W. Mak and A. Goldstein for their help incarrying out the studies, to the members of the Division of Hematol-ogy/Oncology for providing patient specimens, and to Dr. G. Mills forhelpful discussions. Dr. T. W. Mak and T. H. Rabbitts kindly providedthe DNAprobes.

This work was supported in part by the Medical Research Council(to E. W. Gelfand), by grant IM-5 10 from the American Cancer Soci-

ety (to S. H. Benedict) and by grant GM-40767091 from the NationalInstitutes of Health (to S. H. Benedict). Dr. Gelfand is a Scholar of theRaymond and Beverly Sackler Foundation. Dr. Hara was a recipient ofthe Terry Fox Fellowship Award.

References

1. Saito, H., D. M. Kranz, Y. Takagaki, A. C. Hayday, H. N. Eisen,and S. Tonegawa. 1984. A third rearranged and expressed gene in aclone of cytotoxic T lymphocytes. Nature (Lond.). 312:36-40.

2. Lefranc, M. P., and T. H. Rabbitts. 1985. Two tandemly orga-nized human genes encoding the T-cell y constant-region sequencesshow multiple rearrangement in different T-cell types. Nature (Lond.).316:464-466.

3. Lefranc, M. P., A. Forster, R. Baer, M. A. Stinson, and T. H.Rabbitts. 1986. Diversity and rearrangement of the human T cell rear-ranging y genes: nine germline variable genes belonging to two sub-groups. Cell. 45:237-246.

4. Forster, A., S. Huck, N. Ghanem, M. P. Lefranc, and T. H.Rabbitts. 1987. New subgroups in the human T cell rearranging Vygene locus. EMBO(Eur. Mol. Biol. Organ.) J. 6:1945-1950.

5. Quertermous, T., C. Murre, D. Dialynas, A. D. Duby, J. L.Strominger, T. A. Waldmann, and J. G. Seidman. 1986. HumanT-cell'y chain genes: organization, diversity and rearrangement. Science(Wash. DC). 231:252-255.

6. Brenner, M. B., J. McLean, D. P. Dialynas, J. A. Strominger,F. L. Smith, J. G. Owen, J. G. Seidman, S. Ip, F. Rosen, and M. S.Krangel. 1986. Identification of a putative second T-cell receptor. Na-ture (Lond.). 322:145-149.

7. Bank, I., R. A. De Pinho, M. B. Brenner, J. Cassimeris, F. W. Alt,and L. Chess. 1986. A functional T3 molecule associated with a novelheterodimer on the surface of immature human thymocytes. Nature(Lond.). 322:179-181.

8. Borst, J., R. J. van de Griend, J. W. van Oostveen, S. L. Ang,C. J. Melief, J. G. Seidman, and R. L. H. Bolhuis. 1987. A T-cellreceptor y/CD3 complex found on cloned functional lymphocytes.Nature (Lond.). 325:683-688.

9. Brenner, M. B., J. McLean, H. Scheft, J. Riberdy, S. L. Ang, J. G.Seidman, P. Devlin, and M. S. Krangel. 1987. Two forms of the T-cellreceptor y protein found on peripheral blood cytotoxic T lymphocytes.Nature (Lond.). 325:689-694.

10. Tawa, A., S. H. Benedict, J. Hara, N. Hozumi, and E. W.Gelfand. 1987. Rearrangement of the T-cell receptor y-chain gene inchildhood acute lymphoblastic leukemia. Blood. 70:1933-1939.

11. Hara, J., S. H. Benedict, T. W. Mak, and E. W. Gelfand. 1897.T-cell receptor a-chain gene rearrangements in B-precursor leukemiaare in contrast to the findings in T-cell acute lymphoblastic leukemia.A comparative study of T-cell receptor gene rearrangement in child-hood leukemia. J. Clin. Invest. 80:1770-1777.

12. Chen, Z., D. Le Paslier, J. Dausset, L. Degos, G. Flandrin, D.Cohen, and F. Sigaux. 1987. Human T cell y genes are frequentlyrearranged in B-lineage acute lymphoblastic leukemias but not inchronic B cell proliferations. J. Exp. Med. 165:1000-1015.

13. Fujimoto, S., and H. Yamagishi. 1987. Isolation of an excisionproduct of T-cell receptor a-chain gene rearrangements. Nature(Lond.). 327:242-243.

14. Okazaki, K., D. D. Davis, and H. Sakano. 1987. T cell receptor,B gene sequences in the circular DNAof thymocyte nuclei. Directevidence for intramolecular DNAdeletion in V-D-J joining. Cell.49:477-485.

15. Ha, K., N. Hozumi, and E. W. Gelfand. 1985. Lineage specificclassification of leukemia. Results of the analysis of sixty cases ofchildhood leukemia. Br. J. Haematol. 61:237-249.

16. Nadler, L. M., S. J. Korsmeyer, K. C. Anderson, A. W. Boyd, B.Slaughenhoupt, E. Park, J. Jensen, F. Coral, R. J. Mayer, S. E. Sallan,J. Ritz, and S. F. Schlossman. 1984. B cell origin of non-T cell acutelymphoblastic leukemia. A model for discrete stages of neoplastic andnormal pre-B cell differentiation. J. Clin. Invest. 74:332-340.

1282 J. Hara, S. H. Benedict, K Yumura, K Ha-Kawa, and E. WGelfand

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.

17. Feinberg, A. P., and B. Vogelstein. 1983. A technique for ra-diolabeling DNArestriction endonuclease fragments to high specificityactivity. Anal. Biochem. 132:6-13.

18. Quertermous, T., W. M. Strauss, J. J. M. van Dougen, and J. G.Seidman. 1987. Human T cell -y chain joining regions and T celldevelopment. J. Immunol. 138:2687-2690.

19. Yoshikai, Y., B. Toyonaga, Y. Koga, N. Kimura, H. Griesser,and T. W. Mak. 1987. Repertoire of the human T cell gammagenes.High frequency of non-functional transcripts in thymus and mature Tcells. Eur. J. Immunol. 17:119-126.

20. Strauss, W. M., T. Quertermous, and J. G. Seidman. 1987.Measuring the human T-cell receptor y chain locus. Science (Wash.DC). 237:1217-1219.

21. Huck, S., P. Dariavach, and M. P. LeFranc. 1988. Variableregion genes in the human T-cell rearranging gamma(TRG) locus. V-Jjunction and homology with the mouse genes. EMBO(Eur. Mol. Biol.Organ.) J. 7:719-726.

22. Greaves, M. F., G. Janossy, J. Peto, and H. Kay. 1981. Immu-nologically defined subclasses of acute lymphoblastic leukemia inchildren: their relationship to presentation features and prognosis. Br.J. Haematol. 48:179-197.

23. Greaves, M. F., and G. Janossy. 1978. Patterns of gene expres-sion and the cellular origins of human leukemias. Biochim. Biophys.Acta. 516:193-230.

24. Korsmeyer, S. J., A. Arnold, A. Bakhshi, J. V. Ravetch, U.Siebenlist, P. A. Hieter, S. 0. Sharrow, T. W. Lebien, J. H. Kersey,D. G. Poplack, P. Leder, and T. A. Waldmann. 1983. Immunoglobulingene rearrangement and cell surface antigen expression in acute lym-phocytic leukemias of T cell and B cell precursor origins. J. Clin.Invest. 71:301-313.

25. Tawa, A., N. Hozumi, M. Minden, T. W. Mak, and E. W.Gelfand. 1985. Rearrangement of the T-cell receptor $-chain gene innon-T-cell non-B-cell acute lymphoblastic leukemia of childhood. N.Engl. J. Med. 313:1033-1037.

26. Hara, J., K. Kawa-Ha, K. Yumura, S. Ishihara, S. Doi, H.Yabuuchi, S. Konishi, and A. Nishikawa. 1987. Heterogeneity of acuteundifferentiated leukemia at the immunoglobulin and T-cell receptorgenes level. Jpn. J. Cancer. Res. 78:170-175.

27. Boehm, T. L. J., A. Werle, A. Ganser, B. Kornhuber, and D.Drahovsky. 1987. T cell receptor sy chain variable gene rearrangementsin acute lymphoblastic leukemia of T and B lineage. Eur. J. Immunol.17: 1593-1597.

28. Yancopoulos, G. D., and W. Alt. 1986. Regulation of the as-sembly and expression of variable-region genes. Annu. Rev. Immunol.4:339-368.

29. Sugiyama, H., T. Maeda, Y. Tani, S. Miyake, Y. Oka, T.Komori, H. Ogawa, T. Soma, Y. Minami, N. Sakato, and S. Kishi-moto. 1987. Selective use of the VHQ52family in functional VHto DJHrearrangements in a B-precursor cell line. J. Exp. Med. 166:607-612.

30. Hara, J., S. H. Benedict, E. Champagne, T. W. Mak, M. Min-den, and E. W. Gelfand. 1988. Comparison of T cell receptor a, ,B, and-y gene rearrangement and expression in T cell acute lymphoblasticleukemia. J. Clin. Invest. 81:989-996.

31. Reth, M., P. Gehrmann, E. Petrac, and P. Wiese. 1986. A novelVH to VHDJHjoining mechanism in heavy-chain-negative (null) pre-Bcells results in heavy-chain production. Nature (Lond.). 322:840-842.

32. Kleinfield, R., R. R. Hardy, D. Tarlinton, J. Dangl, L. A.Herzenberg, and M. Weigert. 1986. Recombination between an ex-pressed immunoglobulin heavy-chain gene and a germline variablegene segment in a Lyl+ B-cell lymphoma. Nature (Lond.). 322:843-846.

Usage of V-y Genes in Lymphoblastic Leukemia 1283

Downloaded November 14, 2014 from The Journal of Clinical Investigation. doi:10.1172/JCI114012.