Characterization of the T Cell Receptor Repertoire of Neonatal T Cells by RT-PCR and Single Strand Conformation Polymorphism Analysis

Home , Junctional diversity

Bone Marrow Transplantation (2000) 26, 83–89  2000 Macmillan Publishers Ltd All rights reserved 0268–3369/00 $15.00 www.nature.com/bmt Characterization of the T cell receptor repertoire of neonatal T cells by RT-PCR and single strand conformation polymorphism analysis

E Alfani1, AR Migliaccio2, M Sanchez1, AM Passarelli3 and G Migliaccio1

1Laboratory of Cell Biology and 2Clinical Biochemistry, Istituto Superiore di Sanita`, Rome; and 3Ospedale Civile Tivoli, Tivoli, Italy

Summary: by other molecular mechanisms which include imprecise joining of V(D)J recombination,6 N-region diversification We have analyzed by reverse transcriptase-polymerase (random addition of nucleotides by terminal deoxy-ribonu- chain reaction (RT-PCR) the individual non-germ line cleotidyl transferase)7,8 and insertions of palindromic configurations of the T cell receptor (TCR) V␤ chains nucleotides9 at specific points of the VD, DJ and VJ junc- expressed by T cells from eight individual cord blood tions. As a result of allele exclusion,10 each T cell clone specimens. cDNA from each cord blood was amplified expresses one and only one specifically rearranged TCR using a common primer coupled with a primer specific V␤ chain. Therefore, the number of TCR V␤ chain for each of 22 variable elements of the V␤ chain family rearrangements expressed by a given population correlates and the amplified fragments were separated under high with its T cell clonality. resolution conditions. With cDNA from adult blood (as Cord blood (CB) has been proven to be a good source a control), all of the TCR chains were amplified as a of stem/progenitor cells for related and unrelated trans- smear consistent with the extensive polyclonality of plants.11,12 Successful engraftment with low graft-versus- adult T cells. In contrast, a heterogeneous pattern of host disease (GVHD) has been observed even in the case amplification was observed with cDNAs from cord of unrelated poorly matched transplants.13,14 The reduced blood: only 26.7 ± 21.9% of the 22 V␤ chains analyzed GVHD severity observed in allogeneic CB transplantation were amplified as a smear. The majority of them were is explained by the low immunologic reactivity of the neo- amplified as a discrete number of bands (up to 10) (in natal T lymphocytes.15–17 The reasons for such immaturity 68.2 ± 18.7% of samples) and some of them as a single is not yet fully understood but it is possible that it is fragment (4.0 ± 7.8%). Only one of the eight samples mediated by cooperation between different mechanisms. analyzed expressed the majority (72.7%) of its V␤ For example, some studies have reported that CB contains chains as a smear, consistent with an adult-like TCR reduced numbers of CD45RO+ memory cells,18 while repertoire. In conclusion, cord blood expressed, on aver- others have suggested an intrinsic immaturity of CB T cells, age, a less complex TCR repertoire than adult blood. independent of the CD45 isoform they express.19–21 Other Bone Marrow Transplantation (2000) 26, 83–89. authors have observed that neonatal T lymphocytes do pro- Keywords: cord blood; T cell receptor; single strand liferate in response to several antigens, suggesting that anti- conformation polymorphism; cord blood transplantation; gen primed T cells are present in neonatal blood,22–25 and graft-versus-host disease that it is a reduced functional capacity of accessory cells, such as APC, which underlies the ‘immaturity’ of the neonatal immune system.26 During human ontogeny, complete functional TCR V␤ T lymphocytes are cells produced in the thymus that, once chain transcripts begin to be expressed by the thymocytes activated, are responsible for the humoral and cellular in the thymus at 15 weeks of gestation,27–29 and almost all immune responses. The process of T cell activation is the different TCR V␤ genes are expressed by the 17th mediated by the T cell receptor (TCR), a membrane-bound week.6 Since the molecular mechanisms which determine ␣ ␤ heterodimer composed of an and a chain, that recog- the non-germ line V␤ gene diversity are not fully active nizes antigens once coupled with the major histocompat- until the 32nd week,29 it is debated whether the T cells ibility complex expressed by antigen presenting cells circulating in CB are as polyclonal as those circulating in ␤ (APC). During the process of T cell differentiation, the adult blood. Schelonka et al30 have observed, in 35 separate chains undergo rearrangements through somatic recombi- CB samples analyzed by spectra-typing, a highly polyclonal nation among multiple coding segments, namely the vari- TCR repertoire before the 28th week of gestation and oligo- 1,2 able (V), diversity (D), joining (J) and constant (C) genes. clonal T cell expansion at 29th–33rd weeks. This oligoclon- ␤ At present, more than 65 different V genes, a quarter of ality would indicate a reduction in the complexity of the which are represented by pseudogenes, have been ident- circulating T cell repertoire around birth. In contrast with 3–5 ified. The V(D)J junctional diversity is further increased these observations, Garderet et al,31 by analyzing the lengths of the complementarity determining region 3 (CDR Correspondence: Dr G Migliaccio, Laboratory of Cell Biology, Istituto 3) by immunoscopy (a computer-based program developed Superiore di Sanita`, Viale Regina Elena 299, 00161 Rome, Italy to analyze the numbers of fluorescent RT-PCR products Received 19 November 1999; accepted 7 March 2000 obtained with a given pair of primers and separated on a Reduced T cell receptor repertoire in cord blood E Alfani et al 84 sequencing gel) in nine separate CB, have reported that CB RNA extraction and cDNA synthesis T cells express a mature polyclonal TCR repertoire like adult T cells. The discrepancy between these two sets of Nucleated cells were counted and washed twice in HBSS. data could be due to technical limitations. In fact, both After centrifugation, the cell pellet was lysed with guani- spectra-typing and immunoscopy do not discriminate dine isothiocyanate and mRNA extracted with oligo(dT) between V␤ chain rearrangements identical in length but spun columns (Quick Prep Micro mRNA Purification kit; different in sequence. Amersham Pharmacia Biotech) as described by the manu- To clarify whether CB T cells express a reduced com- facturer. RNA was retro-transcribed with Superscript II plexity of the TCR repertoire, the non-germ line configur- reverse transcriptase and oligo(dT) primers (Gibco-BRL, ations of 22 (out of the 25 described) V␤ chain families Grand Island, NY, USA). expressed by eight CB specimens (and four adult PB as control) were compared by the very sensitive method Polymerase chain reaction analysis of the expression of recently described,32,33 which combines reverse tran- the non-germ line configurations of the TCR V␤ chains scriptase-polymerase chain reaction (RT-PCR) amplification of the V␤ chain transcripts with the separation of Individual non-germ line configurations of the TCR V␤ the amplified products by single strand conformation poly- elements were amplified with a common primer coupled morphism (SSCP) analysis. This technique, although not as with a primer specific for each of the 22 variable elements sensitive as heteroduplex analysis,34 allows discrimination of the V␤ chain as described.35,36 cDNA obtained from pur- of V␤ products differing by a single nucleotide substi- ified mRNA extracted from an equivalent number of cells tution.33 The results obtained indicate that CB expresses, (Ӎ1 × 106) was used in each reaction. Hypoxanthine phos- on average, a reduced T cell clonality, although one out of phoribosyl transferase (HPRT) was also amplified as a con- the eight specimens analyzed had an adult-like repertoire. trol for the quality and quantity of cDNA. cDNA was We conclude that the immaturity of the neonatal T cell sys- amplified with Pfu DNA polymerase (0.5 U/reaction, Strat- tem includes, at least in some CB specimens, a reduced agene USA, La Jolla, CA, USA) for 35 cycles in a PTC- number of T cell clones. 100 Thermocycler (MJ Research, Watertown, MA, USA). The amplified products were denatured and separated on a 5% acrylamide 10% glycerol non-denaturing gel.33 The gel was blotted on to Hybond N+ membrane (Amersham Phar- Materials and methods macia Biotech). The specificity of the amplification was confirmed by probing with end-labelled oligonucleotides (␥-32P-ATP, specific activity = 3000 Ci/mMol, Du Pont, Specimen collection and cell purification Firenze, Italy and T4 polynucleotide kinase end-labelling kit, Gibco BRL) specific for the constant portion of the Umbilical cord blood was obtained at the time of delivery V␤ elements.37 After hybridization, the membranes were from uncomplicated pregnancies with previous written washed twice under stringent conditions and exposed to X- informed consent from the mother. Peripheral blood (PB) ray films (Eastman Kodak Company, Rochester, NY, was obtained from healthy adult donors through the collab- USA). All the procedures were conducted according to oration of the Italian Red Cross. The human subject proto- standard protocols.38 col was approved with the premise that no personal data would be linked with the samples analyzed. Mononuclear cells (MC) were isolated from a total of 11 cord blood (CB) Statistical analysis and four PB specimens (all fresh) by density-cut separation Analysis of variance (ANOVA) was performed with the ␳ Ͻ ( 1.077) (Ficoll-Paque; Amersham Pharmacia Biotech, Origin 4.10 software for Windows (Microcal Software, Uppsala, Sweden). CBMC and PBMC were either pro- Northampton, MA, USA). cessed directly or cryopreserved with dimethyl sulfoxide (DMSO 10% v/v, in Hanks’s balanced salt solution, HBSS, Sigma, St Louis, MO, USA).35 Results

V␤ gene rearrangements expressed by T cells from cord Cell immune-phenotyping blood and adult pheripheral blood Cells were washed twice in HBBS and incubated (3– The analysis by RT-PCR and SSCP of the TCR V␤ chain 5 × 105 cells per tube) at 4°C for 30 min with monoclonal repertoire expressed by adult PB and CB specimens is antibodies against CD4, CD7, CD8, CD45, CD45RA and presented in Figure 1 and in Table 1. In this analysis, the CD45RO (all from Sigma), CD2, TCR␥␦ and TCR␣␤ presence of a smear-like pattern for a single amplified V␤ (PharMingen, San Diego, CA, USA) labelled with appro- chain transcript indicates the presence of many individual priate fluorochromes. Aliquots of the cells were also incu- T cell clones expressing that particular V␤ chain, while the bated with appropriate isotype-matched antibodies as con- presence of a single (or few distinct) band(s) indicates that trols for non-specific binding. Two- and three-color the T cell clones that express that particular V␤ segment cytofluorimetric analyses were performed with the Epics are mono- or oligoclonal. The TCR repertoire was defined Elite ESP Cell Sorter (Coulter, Miami, FL, USA). as mono-, oligo- or polyclonal when a single band, 2–10

Bone Marrow Transplantation Reduced T cell receptor repertoire in cord blood E Alfani et al 85

Figure 1 (a) TCR V␤ gene rearrangements expressed by T cells from a representative adult PB. The numbers on the top of each lane indicate the V␤ family analyzed. With adult blood, all the V␤ chain transcripts were amplified as a smear. Similar results were observed in another three separate adult PB. (b) TCR V␤ chain transcripts expressed by three different CB samples (each panel a separated CB unit). The number on the top of each lane indicates the V␤ family analyzed. In the CB unit shown in the bottom panel (H96 of Table 1), at least 16 (out of the 22 V␤ chain transcripts analyzed) were amplified as a smear. Therefore, this unit expressed a polyclonal TCR repertoire similar to that expressed by adult PB and shown in Figure 1a. In the other two CB units (top and middle panel), the majority of V␤ chains were amplified as a single or discrete number of fragments consistent with the presence of mono- or oligoclonal T cell populations. All the cDNA used for amplification had been normalized for the expresssion of HPRT (not shown).

Bone Marrow Transplantation Reduced T cell receptor repertoire in cord blood E Alfani et al 86 Table 1 Summary of the expression of 22 elements of the TCR V␤ found in the expression of most of the T lymphocyte anti- chain family in eight individual CB and four individual PB gens analyzed. CBMC and PBMC contained similar numbers of CD4 and CD8 T lymphocytes as well as of T cells TCR V␤ segments expressing the TCR ␣/␤ and/or CD2 and CD7. The slightly higher number of T cells expressing CD4 and TCR ␣/␤ in Not Monoclonal Oligoclonal Polyclonal CB remained at the limit of statistical significance. The expressed only significant difference in T cell phenotype between CB No. (%) No. (%) No. (%) No (%) and PB was found in the frequency of cells expressing the CD45RA and CD45RO isoforms. In fact, CB contained a statistically significant higher number of CB unit + − H79 2 (9.1) 5 (22.7) 15 (68.2) 0 (0) CD45RA CD45RO T cells (85.8 ± 8.3%) compared with H111 0 (0) 1 (4.5) 19 (86.4) 2 (9.1) PB (59.0 ± 7.1%) (Table 2, P Ͻ 0.0001), while the number H88 0 (0) 0 (0) 18 (81.8) 4 (18.2) of T cells expressing the CD45RA−CD45RO+ and the H87 0 (0) 0 (0) 17 (77.3) 5 (22.7) + + H25 0 (0) 0 (0) 17 (77.3) 5 (22.7) CD45RA CD45RO phenotype was higher in PB compared H84 0 (0) 0 (0) 15 (68.2) 7 (31.8) with CB (13.3 ± 10.8% vs 2.1 ± 1.0%, P Ͻ 0.005 and H121 0 (0) 1 (4.5) 13 (59.1) 8 (36.4) 27 ± 15.7% vs 11.4 ± 8.6%, P Ͻ 0.05, respectively) H96 0 (0) 0 (0) 6 (27.3) 16 (72.7) (Table 2). mean 0.3 (1.1) 0.9 (4.0) 15.0 (68.2)a 5.9 (26.7)a Std 0.7 (3.2) 1.7 (7.8) 4.1 (18.7) 4.8 (21.9)

APB unit Discussion PB220999 0 (0) 0 (0) 0 (0) 100 (100) PB240399 0 (0) 0 (0) 0 (0) 100 (100) RT-PCR amplification followed by SSCP separation of the PB101198 0 (0) 0 (0) 0 (0) 100 (100) PB041198 0 (0) 0 (0) 0 (0) 100 (100) amplified transcripts is one of the most effective methods mean 0 (0) 0 (0) 0 (0) 100 (100) available to date to evaluate the diversity of TCR rearrangements expressed by a T lymphocyte population.35,37,39,40 The TCR repertoire is defined as mono-, oligo- and polyclonal when a This technique was used in the present study to compare single band, 2–10 bands or a smear was detected after SSCP of the ampli- the TCR repertoire expressed by eight CB units and by four fied products. The number of V␤ chain families that expressed a specific adult PB specimens. The results obtained indicate that CB clonal profile is indicated by No., while the number in parenthesis indicates their percentage of the total number of V␤ chains analyzed. expresses, on average, a T cell clonal diversity less complex −5 aSignificantly different (P Ͻ 5 × 10−5) by one-way ANOVA analysis from than that expressed by adult PB (Ͻ5 × 10 ). However, the the values obtained with PB. complexity of the TCR repertoire expressed by the different CB units was heterogeneous: one unit expressed a T cell clonality similar to that expressed by adult PB; another did bands or a smear was detected, respectively, after PCR not express two V␤ chains and expressed all the others amplification. As expected, all the V␤ chains were ampli- with a mono-/oligoclonal profile, whilst the other six units fied as a smear from PB samples (Figure 1a). In contrast, expressed a TCR complexity with a random clonal distri- the analysis of the CB samples indicated the presence of bution intermediate between these two (Figure 1 and reduced TCR polyclonality. However, the number of TCR Table 1). These observations are consistent with the oligo- rearrangements expressed by individual CB was variable clonal T cell expansion in CB units from full-term deliver- as shown by the three representative samples presented in ies reported by Schelonka et al.30 Both our data and the Figure 1b. One CB out of the eight samples analyzed did data from Schelonka et al30 are, however, in contrast to the not express two V␤ chains and expressed all of the others paper by Garderet et al31 that describes CB T cells express- as mono- or oligoclonal TCR repertoires (H79, Table 1). ing a polyclonal TCR repertoire similar to that of adult PB. Another case (H96 in Table 1 and Figure 2, bottom panel) As mentioned earlier, the discrepancy between the two sets expressed, on average, a TCR rearrangement complexity of data could be due to technical reasons. However, one of very similar to that of adult PB. All of the other CB our CB samples displayed a full adult clonality similar to (Figure 1b, top and middle panel and Table 1) expressed an the pattern presented by Garderet et al31 and both intermediate TCR complexity. A smear-like amplification Schelonka et al30 and ourselves (two fetal blood, results not of TCR V␤ chains was observed on average only in shown) have observed a polyclonal adult-like T cell profile 26.7 ± 21.9% of the cases (Table 1) and in the majority when preterm (before the 28th week) CB was analyzed. (68.2 ± 18.7%) of V␤ chains analyzed were amplified as a It is, therefore, possible that the complexity of the T cell discrete number of bands while in some of them clonality expressed by a CB unit decreases with its (4.0 ± 7.8%) they were even amplified as a single band. gestational age. How and why the number of T cell clones circulating in Flow cytometric analysis of the expression of T cell CB decreases at the end of gestation have not been eluci- specific antigens by CB and PB mononuclear cells dated as yet. The fact that the total number of circulating T cells remains constant during the ontogenesis, suggests The expression of T cell-specific antigens (CD2, CD4, that the T cell clonality is decreased by selective expansion CD7, CD8, TCR␥␦, TCR␣␤, CD45, CD45RA and of some of the circulating T cell clones and not by their CD45RO) in CBMC and PBMC is compared in Figure 2a, negative selection. Although a temporal correlation has b and in Table 2. No statistically significant difference was been found between an oligoclonal T cell profile and a peak

Bone Marrow Transplantation Reduced T cell receptor repertoire in cord blood E Alfani et al

PBMC CBMC 87 ab1023 1023 Side scatter Side scatter R1 R1 0 0 0 1023 0 1023 Forward scatter Forward scatter

R1 gated cells R1 gated cells

104 104 104 104 34.5 0.8 5.6 78.7 12.2 0.2 2.8 73.8 103 103 103 103 26.6 38.1 10.2 5.2 25 62.6 23.3 0.1 102 102 102 102 CD7 CD8 CD7 CD8 101 101 101 101

100 100 100 101 102 103 104 100 100 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 CD4 CD2 CD4 CD2

104 104 104 104 55.8 19.6 0.3 0.5 84.1 1.8 0.1 0.2 103 103 0.2 24.2 32.3 66.9 103 13.4 0.7 103 32.2 67.5

d / d

g / 102 g 102 102 102 TCR TCR CD45RA

101 CD45RA 101 101 101

100 100 100 100 101 102 103 104 100 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 CD45RO TCR a/b TCR a/b CD45RO

Figure 2 Two-color cytofluorimetric analysis of CD4/CD8, CD2/CD7, CD45RA/CD45RO and TCR␣/␤/TCR␥/␦ expression in CB (a) and PB (b) mononuclear cells. The light scatter with the lymphocyte gate (R1) used for the analysis is shown in the top panels. The numbers inside each dot plot indicate the percentages of the lymphocyte-gated cells present in the corresponding quadrants. One representative unit out of 11 separate CB and four separate adult PB analyzed are presented. For further details see Table 2. of circulating maternal IgG in the blood of the fetus at 29– Table 2 T cell subsets in the mononuclear fraction from CB and PB 33 weeks of gestation41 it is not clear as yet if these two determined by three-color fluorimetric analysisa phenomena are causally linked as a consequence of a maternal ‘mechanism’ to induce a ‘tolerant immune CBMC PBMC P system’ in the fetus just before birth.30 Although recipients of CB grafts experience, on average, CD4+ 59.5 ± 13 46.0 ± 6.9 Ͼ0.05 CD8+ 21.2 ± 6.8 26.7 ± 6.3 Ͼ0.1 a lower incidence of GVHD, this is not always the case. + + CD4 /8 0.5 ± 0.6 0.3 ± 0.3 Ͼ0.1 Sometimes patients transplanted with similarly mismatched TCR␣/␤+ 83.1 ± 12.4 69.2 ± 0.8 Ͼ0.1 CB grafts experience very different degrees of GVHD. As TCR␥/␦+ BD BD an example, in the case of two three antigen-mismatch CD7+2+ 81.2 ± 12.9 74.5 ± 6.1 Ͼ0.1 + − transplants recently performed with CB units from the New CD7 2 3.2 ± 3.0 3.3 ± 1.9 Ͼ0.5 CD45RA+/RO− 85.8 ± 8.3 59.0 ± 7.1 Ͻ0.0001 York Cord Blood Program, one experienced acute GVDH CD45RA−/RO+ 2.1 ± 1.0 13.3 ± 10.8 Ͻ0.005 and died within 1 month, while the other remained CD45RA+/RO+ 11.4 ± 8.6 27.0 ± 15.7 Ͻ0.05 completely GVHD free (P Rubinstein, personal communication). The reason for such a dramatic difference aThe results are expressed as the mean (±s.d.) of 11 individual CBM and is not known and many studies are under way to define a four PBM (24–40 years old healthy donors), respectively. The P values predictive index for the development of GVHD in cases of were obtained by one-way ANOVA analysis. Because of the relatively high frequency of erythroblasts in CBMC, all the data presented were CB transplantation. It is widely accepted that activation of calculated from the cells in the lymphocyte gate which were also positive mature donor T lymphocytes, after exposure to the antigens for CD45 staining. *Significant P values are indicated in bold. of the recipient, is the main event involved in the onset of BD = below detectable levels.

Bone Marrow Transplantation Reduced T cell receptor repertoire in cord blood E Alfani et al 88 acute GVHD.42,43 For this reason, most of the studies on gene use in peripheral blood and multiple synovial membranes the immaturity of the CB immune system are focused on during rheumatoid arthritis. Hum Immunol 1995; 42: 331–339. the evaluation of the relative numbers of activated (defined 6 Siu G, Clark S, Yoshikai Y et al. The human T-cell antigen as CD45RO+) with respect to naive or unprimed (defined receptor is encoded by variable, diversity and joining gene as CD45RA+) T cells present in these specimens. We con- segments that rearrange to generate a complete V gene. Cell 44 1984; 37: 393–401. firm the previously reported differences between CB and 7 Gilfillan S, Dierrich A, Lemeur M et al. Mice lacking TdT: adult PB in the number of T cells expressing CD45RA or mature animals with an immature lymphocyte repertoire. CD45RO. All of the CB analyzed contained more + − − + Science 1993; 261: 1175–1178. CD45RA /CD45RO and less CD45RA /CD45RO cells 8 Komori T, Okada A, Stewart V et al. Lack of N regions in than adult PB (Figure 2 and Table 2), irrespective of their antigen receptor variable region genes of TdT-deficient lym- clonal TCR profile. A less characterized T cell population phocytes. Science 1993; 261: 1171–1175. is represented by the CD45RA+/CD45RO+ cells, thought 9 Meier J, Lewis S. P nucleotides in V(D)J recombination: a by some authors to represent a CD45RA−/CD45RO+ pre- fine-structure analysis. Mol Cell Biol 1993; 13: 1078–1092. cursor.45 It is interesting that CB also contained lower num- 10 Malissen M, Trucy J, Jouven-Marche E et al. Regulation of ␣ ␤ bers of CD45RA+/CD45RO+ T cells than adult blood TCR and genes allelic exclusion during T-cell develop- (P Ͻ 0.05, Figure 2a, b and Table 2). However, since all of ment. Immunol Today 1992; 13: 315–322. 11 Cairo MS, Wagner JE. Placental and/or umbilical cord blood: 11 CB samples analyzed contained a similar lower number an alternative source of hematopoietic stem cell for transplan- of activated T cells, it is unlikely that this fact alone can tation. Blood 1997; 90: 4665–4678. explain the lower incidence of GVHD experienced by some 12 Rubinstein P, Rosenfield RE, Adamson JW et al. Stored pla- of the recipients of CB grafts. It is possible that the varia- cental blood for unrelated bone marrow reconstitution. Blood bility in the number of T cell clones present in CB 1993; 81: 1679–1690. described here and by Schelonka et al30 plays a role in the 13 Gluckman E, Rocha V, Boyer-Chammard A et al. Outcome development of GVHD at least as a predictive index of the of cord blood transplantation from related and unrelated establishment of a ‘tolerant immune status’ in the fetus. donors. New Engl J Med 1997; 337: 373–381. However, a correlation between reduced complexity of the 14 Rubinstein P, Carrier C, Scaradavou A et al. Outcomes among T cell repertoire and development of GVHD after CB trans- 562 recipients of placental-blood transplants from unrelated plantation remains to be proven. In the absence of such a donors. New Engl J Med 1998; 339: 1565–1577. + + proven correlation, it is not appropriate to recommend the 15 Hassan J, Reen DJ. Cord blood CD4 CD45RA T cells achi- characterization of the T cell clonality of CB unit stored in eved a lower magnitude of activation when compared with their adult counterparts. Immunology 1997; 90: 397–401. banks by RT-PCR and SSCP. Moreover, this technique is 16 Harris DT, Schumacher MJ, Locascio J et al. Phenotypic and very cumbersome. Further studies are necessary to prove functional immaturity of human umbilical cord blood T lym- whether such a correlation exists and to clarify the mech- phocytes. Proc Natl Acad Sci USA 1992; 89: 10006–10010. anism which reduces T cell clonality in the blood at birth. 17 Risdon G, Gaddy J, Horie M et al. Alloantigen priming induces a state of unresponsiveness in human umbilical cord blood T cells. Proc Natl Acad Sci USA 1995; 92: 2413–2417. Acknowledgements 18 Lewis DB, Wilson CB. Developmental immunology and role of host defences in neonatal susceptibility to infection. In: Remington J, Klein J (eds). Infectious Diseases of the Fetus This study was supported by institutional funds from Istituto and Newborn Infant. Saunders Inc: Toronto, 1995, p 20. ` Superiore di Sanita, Rome, Italy, and by contract No. BIO4- 19 Takahashi N, Imanishi K, Nishida H et al. Evidence for CT96–0646 of the European Community. Dr E Mannella, director immunologic immaturity of cord blood T cells. Cord blood T of the Italian Red Cross in Rome, and Dr H Valenzise, Depart- cells are susceptible to tolerance induction to in vitro stimu- ment of Obstetrician and Gynecology, University of Tor Vergata, lation with a superantigen. J Immunol 1995; 155: 5213–5219. Rome, are gratefully acknowledged for providing adult peripheral 20 Clement LT, Yamashita N, Martin AM. The functionally dis- blood and fetal blood, respectively. Prof G D’Agnolo, chairman + of the Laboratory of Cell Biology, is gratefully acknowledged for tinct subpopulation of human CD 4 helper/inducer T lympho- the continuous support of the study. cytes defined by anti-CD45R antibodies derived sequentially from a differentiation pathway that is regulated by activation dependent post-thymic differentiation. J Immunol 1988; 141: 1464–1470. References 21 Pirenne H, Aujard J, Eljaafari A et al. Comparison of T cell functional changes during childhood with the ontogeny of 1 Schatz DG, Oettinger MA, Schlissel MS. V(D)J recombi- Cdw29 and CD45RA expression on CD+4 T cells. Pediatr nation: molecular biology and regulation. Ann Rev Immunol Res 1992; 31: 81–86. 1992; 10: 359–383. 22 Shiratsuchi H, Tsyuiguchi I. Tubercolin PPD reactive T cells 2 LaRocque R, Robinson MA. Diversity in the human T cell in cord blood lymphocytes. Infect Immun 1981; 33: 651–657. receptor beta chain. Hum Immunol 1996; 48: 3–11. 23 Munk ME, Kaufmann SHE. Human cord blood receptor anti- 3 Wei S, Charmley P, Robinson M, Concannon P. The extent of body cell response to protein antigens of Paracoccidoides bra- human germline T-cell receptor V␤ gene segment repertoire. ziliensis yeast forms. Immunology 1995; 84: 98–104. Immunogenetics 1994; 40: 27–36. 24 Miles EA, Wagner JA, Jones AC et al. Peripheral blood 4 Arden B, Clark SP, Kabelitz D et al. Human T-cell receptor mononuclear cell proliferative responses in the first year of variable gene segment families. Immunogenetics 1995; 42: life in babies born to allergic parents. Clin Exp Allergy 1996; 455–500. 26: 780–788. 5 Alam A, Lule´ J, Coppin H et al. T-cell receptor of the ␤-chain 25 Holt PG. Primary allergic sensitization to environmental anti-

Bone Marrow Transplantation Reduced T cell receptor repertoire in cord blood E Alfani et al 89 gens: perinatal T cell priming as a determinant of responder T-cell differentiation in vitro. Br J Haematol 1998; 103: phenotype in adulthood. J Exp Med 1996; 183: 1297–1301. 1198–1205. 26 Trivedi HN, HayGlass KT, Gangur V et al. Analysis of neo- 36 Choi Y, Kotzin B, Herron L et al. Interaction of Staphylo- natal T cell and antigen presenting cell functions. Hum Immu- coccus aureus toxin ‘superantigen’ with human T cells. Proc nol 1997; 57: 69–79. Natl Acad Sci USA 1989; 86: 8941–8945. 27 Campana D, Janossy G, Coustan-Smith E et al. The 37 Yamamoto K, Masuko-Hongo K, Tanaka A et al. Establish- expression of T cell receptor-associated proteins during T cell ment and application of a novel T cell clonality analysis using ontogenesis in man. J Immunol 1989; 142: 57–66. single-strand conformation polymorphim of T cell receptor 28 Doherty PJ, Roifman CM, Pan S et al. Expression of the messenger signals. Hum Immunol 1996; 48: 23–31. human T cell receptor V␤ repertoire. Mol Immunol 1991; 28: 38 Maniatis T, Fritsch EF, Sambrook J. Molecular Cloning: A 607–612. Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory 29 Bonati A, Zanelli P, Ferrari S et al. T-cell receptor ␤-chain Press: Cold Spring Harbor, NY, 1989. gene rearrangement and expression during human thymic 39 Yamamoto K, Sakoda H, Nakajima T et al. Accumulation of ontogenesis. Blood 1992; 79: 1472–1483. multiple T cell clonotypes in the synovial lesions of patients 30 Schelonka RL, Raaphorst FM, Infante D et al. T cell receptor with rheumatoid arthritis revealed by a novel clonality analy- repertoire diversity and clonal expansion in human neonates. sis. Int Immunol 1992; 11: 1219–1223. Pediatr Res 1998; 43: 396–402. 40 Blumberg RS, Yockey CE, Gross GG et al. Human intestinal 31 Garderet L, Dulphy N, Douay C et al. The umbilical cord intraepithelial lymphocytes are derived from a limited number blood ␣␤ T-cell repertoire: characteristics of a polyclonal and of T cell clones that utilize multiple V␤ T cell receptor genes. naive but completely formed repertoire. Blood 1998; 91: J Immunol 1993; 150: 5144–5153. 340–346. 41 Landor M. Maternal-fetal transfer of immunoglobulins. Ann 32 Orita M, Iwahara H, Kanazawa H et al. Detection of polymor- Immunol 1995; 74: 279–283. phisms of human DNA by gel electrophoresis as single-strand 42 Antin JH, Ferrara JL. Cytokine dysregulation and acute graft- conformation polymorphisms. Proc Natl Acad Sci USA 1989; versus-host disease. Blood 1992; 80: 2964–2968. 86: 2766–2770. 43 Madrigal JA, Cohen SBA, Gluckman E et al. Does cord blood 33 Orita M, Suzuki Y, Sekiyama T et al. Rapid and sensitive transplantation result in lower graft-versus-host disease? It detection of point mutation and DNA polymorsphisms using takes more than two to tango. Hum Immunol 1997; 56: 1–5. the polymerase chain reaction. Genomics 1989; 5: 874–879. 44 Roncarolo MG, Vaccarino E, Saracco P et al. Immunologic 34 Shen D-F, Doukhan L, Kalams S et al. High-resolution properties of cord blood. In: Broxmeyer HE (ed). Cellular analysis of T-cell receptor ␤-chain repertoires using DNA Characteristic of Cord Blood and Cord Blood Transplan- heteroduplex tracking: generally stable, clonal CD8+ expan- tation. AABB Press: Bethesda, MD, 1998, pp 67–79. sion in all healthy young adults. J Immunol Meth 1998; 215: 45 Boﬁll M, Akbar AN, Salmon M et al. Immature CD45RA- 113–121. lowROlow T cells in the human cord blood. J Immunol 1994; 35 Sanchez M, Alfani E, Visconti G et al. Thymus-independent 152: 5613–5623.

Bone Marrow Transplantation