Can cord blood cells support the cytokine storm in GvHD?

Cytokine & Growth Factor Reviews 11 (2000) 185±197

www.elsevier.com/locate/cytogfr

Mini Review

Can cord blood cells support the cytokine storm in GvHD? Shara B.A. Cohen a,*, Xiao Nong Wang b, Anne Dickinson b a

The Anthony Nolan Research Institute, The Royal Free Hospital, Pond Street, Hampstead, London NW3 2QG, UK The University Department of Haematology, School of Clinical and Laboratory Sciences, Royal Victoria In®rmary, Newcastle upon Tyne NE1 4LP, UK

b

Abstract Cord blood has a high number of proliferating hematopoietic progenitors and is therefore used as an alternative source of hematopoietic cells for allogeneic transplantation. In addition there is a wider availability of cord blood and a lower cost of procurement compared to bone marrow. However one of the most interesting immunological bene®ts of a cord blood transplant that has been proposed is the low severity of Graft versus Host Disease (GvHD). This review aims to address some of the immunological reasons why this may be the case by assessing the role of cord blood cytokines in the cytokine storm of GvHD. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: GvHD; Cytokines; Cord blood; Naive T cells

1. Introduction Allogeneic bone marrow transplantation is a treatment for a variety of haematological malignancies, and in recent years has even been used to treat autoimmune diseases and metabolic disorders [1,2]. However, although signi®cant advances have been made in histocompatability matching and conditioning regimens, graft-versus-host disease (GvHD) remains a major complication [3±5]. Cytokines are very important in the initiation and maintenance of acute GvHD and it is believed that the cytokine storm is central to the disease. 1.1. The cytokine storm The current dogma suggests that as donor T cells enter the blood of a mismatched host they are stimulated by alloantigen and produce a wide variety of cytokines which activates cells and enhances cellular * Corresponding author. Tel.: +44-171-284-8333; fax: +44-171284-8331. E-mail address: [email protected] (S.B.A. Cohen).

proliferation (Fig. 1). The pre-transplant conditioning regimens may enhance this e€ect, for example, radiation and/or chemotherapeutic agents may cause host cell damage and subsequent release of pro-in¯ammatory cytokines such as TNFa, IL-1 and IL-6 [6,7]. This causes the host cells to become activated and have an increased expression of HLA molecules. Consequently the host cells ability to be recognised as an allogeneic cell by the donor cells is increased. Donor T cells then rapidly expand with subsequent production of the proin¯ammatory Th1 type cytokines, such as IL-2 and IFNg. These Th1 cytokines cause further expansion of donor T cells and the activation of NK cells and macrophages [8,9]. NK cells, upon activation by IL-2 or IFNg, also produce large amounts of IFNg, which again in¯uences macrophage activation. This high level of cytokine production which occurs through the activation of T cells, macrophages and NK cells in acute GvHD (aGvHD) (the cytokine storm), can reach such a level with allogenic bone marrow transplantation as to cause mortality due to septic shock. However, the cytokine storm is more commonly involved (directly and indirectly) in the development of

1359-6101/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 9 - 6 1 0 1 ( 0 0 ) 0 0 0 0 4 - 6

186

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

damage to the GvHD target organs such as skin, gut and liver. 1.2. Role of host cells in the cytokine storm Recent studies have suggested that TNFa and IL10 cytokine gene polymorphisms of the host play a role in bone marrow transplant outcome. The genotype for polymorphism's associated with TNFa high production and IL-10 low production were studied in 49 allogeneic BMT recipients and results correlated with the severity of GvHD [10]. The widely studied TNFa-308 polymorphism did not show any signi®cant association, but the d3 homozygous allele of the TNFd microsatellite was preferentially associated with grade III/IV GvHD (7/11 patients), compared with its occurrence in 8/38 patients with grade 0/II GvHD ( p = 0.006). Alleles of the IL-101064 promoter region microsatellite polymorphism which possessed greater numbers of dinucleotide (CA) repeats also signi®cantly associated with more severe GvHD. 18/38 patients with grade 0-II GvHD pos-

sessed alleles with greater numbers (twelve or more) of dinucleotide repeats, compared with 9/11 cases with grade III-IV GvHD ( p < 0.02). Of the 38 patients with grade 0-11 GvHD, 3/38 had a both TNFd3/d3 and IL-10 [12±15] genotype compared with 6/11 patients with grade III-IV GvHD ( p < 0.001). However, there was no association of either the TNFd or IL-10 microsatellite polymorphism's with mortality. These results have been con®rmed in an independent study of 144 transplant patients and their donors, where the patient IL-10-1064 microsatellite alleles with greater dinucleotide repeats developed more severe grade III-IV GvHD ( p = 0.0045) [11]. Thus, patient cytokine gene polymorphism genotypes may in¯uence GvHD outcome by a€ecting cytokine activation during the pre-transplant conditioning regimens. These results were the ®rst to suggest a genetic predisposition to GvHD. Furthermore they suggest that the ability of host cells to make cytokine is equally important in the cytokine storm as that of donor cell cytokine production.

Fig. 1. The cytokine storm in graft versus host disease (GvHD). Initiation and maintenance of the in¯ammatory response in GvHD involves autocrine and paracrine cytokine networks. This ®gure illustrates the complexity of the cytokine storm by showing some of the cytokine/cellular interactions which occur when a CD4+ donor T cell recognises foreign (host) HLA. Allospeci®c recognition will also occur with donor CD8+ T cells recognising host alloantigen.

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

1.3. Cord blood transplantation Transplantation of human umbilical cord blood has recently been used as a successful alternative to allogeneic bone marrow for the haematopoietic reconstitution of children with various haematological disorders [12±14]. The results of cord blood transplants performed to date have suggested a reduced GvHD compared to allogeneic bone marrow transplantation [15]. The reason for this proposed reduction in GvHD is unknown. An in vitro skin explant model has been recently used to determine whether the transplanted cord blood cells are less immunogenic than transplanted adult cells. This model detects graft versus host (GvH) allor-

187

eactivity and can assess the ability of cord blood, compared to peripheral blood, to illicit a GvH response (GvHR) and to make cytokines [16,17] (Fig. 2). Using this model cord blood responder cells produce a GvHR, but the degree of histopathological damage and cytotoxic T lymphocyte precursor (CTLp) frequency is signi®cantly lower than with adult responder cells against the same third party skin [17] (Fig. 3). Furthermore, the CTLP frequency of the cord blood directly correlated with the degree of GvHR of the cells in the skin explant model (Fig. 3). Thus, as observed in the clinical setting, the potential for GvHD after cord blood transplants may not be absent, but just reduced in severity. Cord blood cells are routinely frozen pre-transplant

Fig. 2. Skin explant model for investigating the GvHR of Cord blood mononuclear cells (CBMC) compared to Peripheral blood mononuclear cells (PBMC). A. Cord blood or maternal blood mononuclear cells were used as responder cells against the same irradiated third party stimulator cells: in a one way MLC in the GvH direction. After seven days of incubation the supernatant was collected and used for the analysis of the cytokines TNFa, IFNg and IL-10. The responder cells were resuspended at 1  106 cells/0.2 ml autologous serum plus medium and added to skin explants from the third party stimulator. After three days of co-culture of responder cells with the skin explant, the explant was formalin ®xed and routinely stained by haematoxylin and eosin. The histopathology was then graded for GvHR (grades I±IV) (Fig. B±E) as described by Lerner et al. for GvHD [73]. The degree of histopathological damage re¯ects the degree and grade of GvHR, increasing with severity from background levels grade 0±1 to clinically relevant grades II±IV. Grades II±IV GvHR predicts often life threatening GvHD when the skin explant model is used in a HLA-identical setting [74]. B. Grade I skin GvHR showing mild vacuolisation of epidermal cells (haematoxyin and eosin x360). C. Grade II skin GvHR showing vacuolisation and dyskeratotic bodies. D. Grade III skin GvHR showing vacuolisation and subepidermal cleft formation. E. Grade IV GvHR showing subepidermal cleft formation and complete separation of dermis and epidermis.

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

Fig. 2 (continued)

188

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

whereas bone marrow is used fresh, and it has been proposed that it is this freezing process that reduces the cord blood potential for immunoreactivity and GvHD [18]. However, the results from the skin explant model suggest that the di€erence between cord blood cells compared to adult cells, in the transplant setting, is due to di€erences in cellular function and not a di€erence in cell processing. The skin explant study therefore provided a cellular basis for the possible use of HLA-mismatched unrelated cord blood cells as transplant material since it was the ®rst in vitro demonstration that a proportion of cord blood samples, unlike adult blood may be ``tolerant'' with up to six HLA mismatches. Thus the in vivo GvHD and in vitro GvHR are reduced with cord blood compared to bone marrow or adult cells, respectively. Due to the role of the cytokine storm in GvHD development, an obvious reason for this reduced GvHD and GvHR could lie in the reduced ability of cord blood cells to make cytokines. Therefore, many researchers have analysed the ability of cord blood cells to produce cytokines. Here we give an overview of this work and try to establish, from published data, whether cord blood lymphocytes have the ability to support the cytokine storm which is required to perpetuate GvHD. 2. Production of cytokines by cord blood cells In general cord blood cells produce less cytokine, have a lower bioactivity or a reduced frequency of

Fig. 3. A correlation between cord blood CTLP frequency and in vitro GvHR as measured using the skin explant assay. Cord blood CTLP frequency against third party cells was compared against the GvHR activity of the same cells using the skin explant assay (Fig. 2). High CTLP frequency correlated with low GvHR of grade I±II. Low CTLP frequency correlated with corresponding high GvHR of grade III±IV.

189

cytokine producing cells compared to adult cells. This has been reviewed by us elsewhere [18±20] and a summary of the literature to date is described in Table 1. We have highlighted below some examples of di€erences between adult and cord blood cell cytokine production with a focus on T cell and antigen presenting cell cytokine production. 2.1. T cells The most consistent results of cord blood T cell cytokine production have been for IL-2, IL-4 and IFNg. Regardless of the method of stimulation used, production of IL-2, IL-4 and IFNg by cord blood T cells is, in most studies, lower than adult T cells. The di€erences between adult and cord blood T cells in production of IL-2, IFNg and IL-4 has been demonstrated as reduced frequency of cytokine producing cells, reduced protein production, and di€erences in mRNA abundance and transcription [21±27]. However, cord blood T cells will make cytokine at similar levels to adult cells when stimulated optimally. For example, using an optimal concentration of PHA plus 0.5 mM A23187 (a calcium ionophore) enables adult cells to make 201 2 62 IU and cord blood cells to make 108217 IU of IFNg [28]. 2.2. Antigen presenting cells The cytokine production of antigen presenting cells in cord and adult blood has been extensively studied and there is a consensus that the cytokine production is reduced by cord blood antigen presenting cells, however IL-1 is an exception. In contrast to the other cytokines made by cord blood cells IL-1 message, the kinetics of protein production and the amounts of IL1 produced by monocytes within cord blood are equal to adult derived monocytes [29,30]. Similarly Lipopolysaccharide (LPS) activated macrophages from cord blood also have the same kinetics of IL-2 and IL-15 mRNA expression and the same translation rate as adult cells. However, although mRNA expression of IL-12 and IL-15 can be induced in both cord blood and adult mononuclear cells, protein accumulation of both cytokines is lower in cord blood compared to adult cells [31,32]. Thus, although the levels of IL-1 produced by cord blood antigen presenting cells is not reduced, the antigen presenting capability (through the production of other co-stimulatory cytokines) is reduced in cord compared to adult antigen presenting cells. 2.3. Cytokine production with allostimulation The majority of studies performed which analyse the ability of cord blood cells to make cytokine have been

190

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

Table 1 A comparison of cytokine production by cord blood (CB) and adult (AD) cells Cytokine

Cells

Stimulus

Assay

AD compared to CB

Reference

IL-1

APCa APC APC

LPSb LPS PHA+LPSg

ELISAc/mRNAd Bioassayf ELISA

AD=CBe AD=CB AD=CB

[29] [30] [83]

IL-2

T cells T cells T cells T cells T cells T cells T cells NK cells

Freezing/alloantigen PHA Superantigen Superantigen PMAi+Ij PMA+I PMA+I PMA+I

ELISA Bioassay Bioassay ELISA ICSk ICS Bioassay ICS

AD > CBh AD=CB AD > CB AD > CB AD=CB AD > CB AD > CB AD > CB

[84] [85] [86] [87] [76] [19,56] [88] [19]

IL-4

T T T T T

Superantigen PMA+I PMA+I Alloantigen PHA

Bioassay ICS RIA/mRNAl ELISA ELISA

AD > CB AD > CB AD > CB AD=CB AD=CB

[86] [56,76] [55] [89] [89]

IL-6

NK cells

No stimulation

ELISA

AD > CB

[90]

IL-8

APC

PMA/PHA2LPS

ELISA/mRNA

AD > CB

[91]

IL-10

MNCm

Allogenic MLCn

ELISA

AD=CB

[16]

IL-12

APC

LPS

ELISA/mRNA

AD > CB

[31]

IL-15

APC

LPS

Immunoblot/mRNA

AD > CB

[32]

TNFa

APC T cells NK cell NK cells MNC

LPS PMA+I No stimulation PMA+I Alloantigenic MLC

Bioassay ICS ELISA ICS ELISA

AD > CB AD > CB AD > CB AD > CB AD > CB

[29] [19,56] [90] [19] [16]

IFNg

T cells T cells T cells T cells T cells T cells T cells T cells NK cells NK cells NK cells NK cells NK cells MNC

PHA PHA PHA+A23187 PHA2IL-2 Alloantigen Superantigen PMA+I PMA+I No stimulation IL-12 PMAÿI PMA+I K562 cells Alloantigenic MLC

ELISA Bioassay Bioassay ELISA ELISA ELISA RIA/mRNA ICS ELISA ELISA/mRNA ELISA/mRNA ICS ELISA ELISA

AD > CB AD > CB AD > CB AD > CB AD=CB AD > CB AD > CB AD > CB AD > CB AD=CB AD > CB AD > CB AD > CB AD > CB

[89,92,93] [85] [28] [92] [89] [87] [55] [56,76] [90,94] [94] [94] [19] [95] [16]

TGFb1

APC

PHA

ELISA/mRNA

AD > CB

[91]

GMCSF

APC

PMA/PHA

mRNA

AD > CB

[96]

BCGF

T cells

PGA+PMA

ELISA

AD > CB

[88]

a

cells cells cells cells cells

APC; Antigen Presenting Cell. LPS; Lipopolysaccharide. c ELISA. Enzyme linked immunoadsorbant assay (a quick and sensitive immune binding assay which uses detection and capture cytokine speci®c antibodies to quantitate cytokine levels [75]. This method does not assess cytokine bioactivity. b

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

191

d mRNA analysis. The advantages of mRNA assays are that (a) they are relatively simple, (b) they are molecule speci®c and (c) multiple cytokines can be assayed from the same sample. However, the type of cell which is producing the cytokine cannot be de®ned. In addition the presence of mRNA does not demonstrate the release of functional protein [75]. e AD=CB; levels of cytokine derived from adult and cord blood cells were equal. f Bioassay. Bioassays quantitate cytokine levels. The assay uses cells which proliferate speci®cally to the cytokine being analysed and therefore its measures of cytokine biological activity. Thus, bioassays measure active protein. However, the cell lines used may not be speci®c for one cytokine or may drift in sensitivity [75]. Furthermore, the sample being analysed may contain factors which are suppressive to the growth of the cell line. g PHA; Phytohaemagglutinin. h AD > CB; levels of cytokine were greater when derived from adult compared to cord blood. i PMA; Phorbol-12-myristate-13-acetate. j I; Ionomycin. k ICS; Intracellular cytokine staining allows the measurement of the percentage of cells which are capable of producing a given cytokine following stimulation [76,77]. Cells are stimulated with mitogen in the presence of substances which cause intracellular accumulation of newly synthesised proteins. After stimulation cells are stained with a ¯uorescent extracellular marker, such as anti-CD3, ®xed and permeabilized to allow a ¯uorescent anti-cytokine antibody inside the cell. The samples are analysed on a cyto¯uorimeter. This technique analyses the frequency of cytokine producing cells and allows the investigator to determine which cell is making which cytokine. However, this technique will not quantitate the level of cytokine being made. l RIA; Radioimmunoassay. m MNC; Mononuclear cells. n MLC; Mixed Lymphocyte Culture.

after mitogen or superantigen stimulation (see Table 1). A more physiologically relevant model to investigate the capacity of cytokine production by cord blood cells is the initiation of a GvHR in an in vitro allogeneic system. A one way mixed lymphocyte culture (MLC) in the GvHR direction was used to analyse the ability of cord blood responder cells to make cytokine (as described in Fig. 2). After allogeneic stimulation cord blood mononuclear cells produced lower levels of TNFa and IFNg than peripheral blood mononuclear cells (4055 2 551 versus 2196 2 338 pg/ml and 79.6 2 11.1 versus 46.9 2 9.41 IU/ml, respectively). These di€erences were statistically signi®cant ( p = 0.007 and 0.031, respectively). Thus, similar to when cord blood cells are stimulated with mitogen or superantigen, in the allogeneic setting, cord blood cells produce less cytokine than adult cells. However, the production of all cytokines analysed was not reduced, since the IL-10 production by cord blood mononuclear cells and peripheral blood mononuclear cells was comparable in the MLC (14122226 versus 1231254 pg/ml) (Fig. 4).

antigenic naiveteÂ. Indeed it has been shown that naive T cells from the adult periphery, similar to cord blood T cells, produce lower levels of IFNg than memory T cells, whilst IL-4 expression is virtually absent in naive adult and cord blood T cells [38±40]. However, this is not the case for IL-2. Naive and memory adult T cells are able to produce similar amounts of IL-2 [21], whereas naive T cells from cord blood produce lower levels. Furthermore, CD45RA+ T cells derived from cord blood stimulated with Phorbol-12-myristate-13acetate (PMA) plus Ionomycin have a reduced frequency of IL-2 producing cells compared to the same cells derived from adult populations [41]. In addition, CD2 plus CD28 stimulation of adult CD4+ CD45RA+ cells increases the IL-2 mRNA to a greater extent than CD4+CD45RA+ cells derived

3. Why is cytokine production reduced in cord blood cells? 3.1. The majority of cord blood cells are naive Although there is some con¯icting literature, cord blood T cells have been generally reported to be phenotypically and functionally immature. This has been demonstrated in terms of CD45 isoform expression [33±35], and in the expression of other cell surface markers such as CD38 [33,34,36,37] and CD29 [36,37] (see Table 2). Thus the reduced expression of IFNg and IL-4 by cord blood T cells may simply re¯ect their

Fig. 4. Cytokine production by cord blood mononuclear cells (CBMC) and peripheral blood mononuclear cells (PBMC) in a one way allogeneic MLC. Cytokine production by CBMC and PBMC were stimulated against the same third party in an MLC as shown in Fig. 2. The supernatants were collected after seven days of incubation and assayed by ELISA for IFNg IU/ml, TNFa (pg/ml) and IL-10 (pg/ml).

192

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

cord blood. However this di€erence can be removed by increasing naive cord blood cell stimulation, for example by the addition of PMA to CD2 plus CD28 stimulated cells [42]. Thus, the ability of the naive cells within cord blood to make some cytokines is reduced compared to the equivalent cells in the adult. However, naive cord blood cells can be stimulated to have a cytokine production equivalent to that of adult naive T cells. This raises the question of whether the in vitro cytokine analysis that we have discussed re¯ects events occurring in vivo. Indeed it has been suggested that the phenotype of CB cells noted in some in vitro experiments may be bypassed in vivo when there are more physiological stimuli available [15]. 3.2. Suppressor activity of cord blood cells It has been documented that some populations of cord blood cells possess suppressor activity, which may explain the reduced cytokine production by cord blood cells. 3.2.1. T cells Cord, but not adult T cells, have the potential to mature into high IL-10, low IL-2 producing cells [15]. These cells, previously isolated from autoimmune dis-

ease targets [43,44], could down-regulate the immune response, since IL-10 can cause a long lasting antigen speci®c unresponsiveness against alloantigen [45]. Thus, cord blood has the ability to mature into T suppressor cells. These cells could play a role in the reducing cord blood T cell proliferative responses and reducing cord blood T cell cytokine production. An increased proportion of CD8+CD28ÿ cells within cord blood, (which in the adult, has been associated with suppressor function [46]) has also been described [47]. Risdon et al. postulated that these suppressor cells were, at least in part, responsible for the long lasting unresponsiveness and lack of proliferation of cord blood lymphocytes upon rechallenge with alloantigen. In addition they could play a role in reducing cord blood cell cytokine production. 3.2.2. Alternatively activated macrophages Alternatively activated macrophages protect organs from unwanted in¯ammatory or immune reactions [48]. These macrophages produce anti-in¯ammatory cytokines such as IL-1 receptor antagonist [49,50] and not the pro-in¯ammatory cytokines IL-1, TNFa, IL-6, IL-12 and macrophage in¯ammatory protein (MIP)-1a [51,52]. They therefore possess suppressor activity. Because they have been found in the placenta it has been suggested that placental macrophages are the

Table 2 Expression of naive and memory markers on Adult (AD) and Cord Blood (CB) T cellsa Antigen Naive markers CD45RAb

CD38c

Memory markers CD45ROb

CD29d

AD compared to CB expression of antigen

Reference

CB > AD CB > AD CB > AD CB > AD CB > AD No di€erence CB > AD CB > AD CB > AD CB > AD CB > AD

[37] [35] [33] [34] [36] [97] [98] [36] [37] [33] [34]

AD > CB AD > CB AD > CB AD > CB AD > CB AD > CB

[35] [33] [34] [97] [36] [37]

a Summary of the results from di€erent studies indicating whether the surface expression of di€erent antigens is greater in cord blood compared to adult blood (CB > AD) or vice versa (AD > CB). b After activation, CD45RA+ T cells down regulate CD45RA expression and gain high density expression of CD45RO [78±80]. c The function of CD38 is unknown, but it has a characteristic expression pattern. Mature thymocytes, cord blood T cells and activated adult T cells express high levels of CD38, naive and resting memory cells from the adult express low levels of CD38 [24,81,82]. CD38 is a marker of immaturity only when coexpressed with CD45RA. d CD29 is a marker of memory T cells.

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

prototype of naturally occurring suppressor macrophages [53]. Thus, these macrophages may cross the placenta to the cord blood where they could in¯uence the function of the other cell populations, including inhibiting the ability of cord blood cells to produce cytokines. 3.3. Lack of a lymphocyte activation factor within cord blood sera The literature has many examples of di€erences in the levels of cytokines produced between cord and adult cells (see Table 1), di€erences in cytokine concentrations are also seen in the sera [19]. In addition, the levels of other soluble factors (non-cytokine) within cord and adult serum are di€erent [19] and these di€erences in sera could be responsible for some of the phenotypic and functional di€erences noted between adult and cord blood cells [54]. Indeed incubating adult cells in cord blood sera (mimicking the foetal microenvironment) reduces the frequency of cytokine producing adult cells [19] to a frequency equivalent to that of cord blood [22,23,25,27,55,56]. These results suggest that the foetal microenvironment may in¯uence the cells' ability to make cytokines, keeping the neonatal T cells in a non-responsive state. 3.4. Intracellular signalling Bertotto et al. [57], suggested that the defective response of cord blood T cells is due to an intrinsic T cell derangement which associated with a failure in transmembrane transduction of the activation signals provided by anti-CD3 stimulation. However Lewis et al. [21] proposed that the di€erences are due to events downstream of Ca2+ signalling and protein kinase C (PKC) activation, and that reduced transcription results from alterations in the interaction of transcription factors with the regulatory regions of the cytokine genes. Wherever the ``defect'' is, it does not seem to be a general defect of signalling, since although there is a decreased expression of cytokine mRNA and protein levels by mononuclear cells from cord compared to adult blood after activation, regulation of TCR, CD8 and p56lck in cord blood and adult cells is equivalent [23]. 3.5. Cytokine message transcription is low in cord blood cells The decreased cytokine production by neonatal cells could be due to diminished transcription of cytokine genes and mRNA accumulation [55]. The expression of IFNg, IL-4, TNFa, GM-CSF, IL-15 and TGFb message are all lower in cord blood T cells compared

193

to adult T cells regardless of the method of stimulation [22,26,55,58±61]. Our own unpublished data support these studies. Cord blood mononuclear cells stimulated by PHA with or without IL-2 over a period of 24 to 96 h produced little or no TNFa and IFNg mRNA, although measurable levels of these cytokines were produced in the supernatants. However, CD34 positive cells from cord blood produced mRNA to growth factor GCSF and GCSF receptor. The instability of the mRNA could therefore be a factor of di€erentiation of the T cells in cord blood, as well as the intrinsic nature of the cytokine/growth factor involved. 4. The response of cord blood cells to cytokines Although analysis of the ability of cells to make cytokine, and the frequency of cytokine producing cells is important, the ability of the cells to respond to cytokine is equally important in the cytokine storm. 4.1. The e€ect of cytokines on cord blood cell function Recent studies show that cord blood may have the capacity to be immunomodulated to a greater extent than adult blood. For example, although IL-10 inhibits the production of TNF and IFNg to the same extent for both adult and cord blood (Fig. 5A) IL-10 inhibits the cytotoxic capacity of cord blood to a greater extent (Fig. 5B). Thus, IL-10 inhibits the alloreactivity of cord blood cells to a greater degree than adult cells [16]. These results suggest that, if immunosuppressive cytokines such as IL-10 are produced in vivo pre or post transplantation, they may be more ecient in downregulating a cytokine storm produced by a cord blood graft than a bone marrow graft. We have suggested above that cord blood GvH alloreactivity may be more eciently downregulated post transplant than the more mature cells present in bone marrow. Further studies [62] have demonstrated that cord blood cytotoxicity and cytokine production are more susceptible to cyclosporin induced inhibition than adult blood, again suggesting that cord blood cells have a greater capacity to be immunosuppressed than adult cells. 4.2. Expression of cytokine receptors on cord blood Expression of the IL-2 receptor a subunit (CD25) has been analysed by several groups with con¯icting results, showing increased [63], decreased [23,35,64,65] and similar levels [47] of CD25 expression by cord blood compared to adult T cells. The reasons for these con¯icting results are unclear and may represent methodological di€erences. However, expression of the IL-

194

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

Fig. 5. The inhibitory e€ect of IL-10 on the cytokine production and cytotoxic activity of cord blood mononuclear cells (CBMC) and peripheral blood mononuclear cells (PBMC). A. The presence of exogenous IL-10 in a primary one way MLC inhibited IFNg and TNFa production in both CBMC and PBMC (responder cells) to a similar extent. B. The target cells (stimulation cells) were labelled with 15Cr and tested for lysis by e€ector cells (responder cells) generated from a primary MLC using paired CBMC or PBMC in the presence or absence of IL-10. IL-10 appeared to reduce the speci®c cytotoxicity of CBMC responder cells (78.2%) to a greater extent than PBMC (43.5%).

2R b and g subunits are both reduced on cord blood compared to adult mononuclear cells [65±67]. In addition, with CD2 or CD2 plus CD28 stimulation, adult CD4+CD45RA+ cells upregulate the expression of CD25 (the IL-2 receptor a chain) more than CD4+CD45RA+ cells derived from cord blood, but this di€erence disappears with the addition of PMA [42]. Decreased expression of the TNFa, IL-4 and IL-6 receptors by cord blood compared to adult T cells have also been demonstrated [35,65]. However, the IL4 receptor is expressed and is functional on cord blood B cells since whole mononuclear cells from cord and adult blood can make IgE to the same extent with IL4 stimulation [68]. The IFNg receptor is also expressed on cord blood B cells since addition of IFNg to mononuclear cells from cord and adult blood stimulated with IL-4 increases secretion of IgE by cord [69]. The IFNg receptor is also on cord blood NK cells since IFNg has a positive dose response e€ect on the cord blood NK activity [70]. Thus, many cytokine receptors have a reduced expression on cord blood T cells, but not cord blood B or NK cells, compared to adult cells. We could therefore assume that perpetuation of the T cell responses in GvHD after a cord blood transplant, and thus maintenance of the cytokine storm, may be inhibited by an inability of the donor (cord blood) T cells to respond to cytokine.

5. Conclusion A critical aspect of aGvHD is the cytokine storm, which involves the production of, and response to, cytokines by both donor and host cells (see Fig. 1). Comparisons made between cells within adult and

cord blood show di€erences in cytokine gene expression, the production of cytokines and the frequency of cytokine producing cells. T cell, macrophage and NK cell production of anti-in¯ammatory cytokines are reduced in cord blood compared to adult blood (Table 1). Furthermore, the ability of cord blood cells to respond to cytokine is reduced, due to the low expression of many cytokine receptors. Thus, the literature implies that cord blood cells cannot support the cytokine storm. Nevertheless, there are incidences of patients who have had a cord blood transplant and still develop severe GvHD [71]. Recent data has suggested that these patients have a genetic predisposition to developing the disease since TNF and IL-10 polymorphisms within the host have been correlated with GvHD incidence and severity in 1±4 HLA antigen mismatched cord blood transplants [72]. Thus, to summarise, cord blood cells have a reduced ability to both make and respond to cytokines and may not be able to maintain an in¯ammatory response. Although this could explain the reduced incidence and severity of GvHD with a cord blood transplant there are incidences of strong GvHD. In light of current evidence we suggest a high severity of GvHD after a cord blood transplant is due to the ability of the host cells to make cytokine and this may be due to the genetic predisposition of the recipient.

Acknowledgements The research conducted by Drs Wang and Dickinson was supported by grants from the Leukaemia Research Fund, The European Commission (Bio4-980236) and the Tyneside Leukaemia Research Fund. The cord blood research was also speci®cally by the Hayward Foundation. Dr Cohen is supported by the Anthony Nolan Bone Marrow Trust. The authors

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197

wish to thank and acknowledge Dr L Sviland, Norway, for the histopathological interpretation of the skin explant assays.

References [1] Tabbara IA. Allogeneic bone marrow transplantation. South Med J 1996;89(9):857±68. [2] Ikehara S. Bone marrow transplantation for autoimmune diseases. Acta Haematol 1998;99(3):116±32. [3] Hows JM, Yin JL, Marsh J, et al. Histocompatible unrelated volunteer donors compared with HLA nonidentical family donors in marrow transplantation for aplastic anaemia and leukaemia. Blood 1986;68(6):1322±8. [4] Marks DI, Cullis JO, Ward KN, et al. Allogeneic bone marrow transplantation for chronic myeloid leukaemia using sibling and volunteer unrelated donors. A comparison of complications in the ®rst 2 years. Ann Intern Med 1993;119(3):207±14. [5] McGlave P, Bartsch G, Anasetti C, et al. Unrelated donor marrow transplantation therapy for chronic myelogenous leukaemia: initial experience of the National Marrow Donor Program. Blood 1993;81(2):543±50. [6] Antin JH, Ferrara JL. Cytokine dysregulation and acute Graft versus Host Disease. Blood 1992;80:2964±8. [7] Ferrara JL. Cytokine dysregulation as a mechanism of graft versus host disease. Curr Opin Immunol 1993;5(5):794±9. [8] Via CS, Finkelman FD. Critical role of interleukin 2 in the development of acute graft-versus-host disease. Int Immunol 1993;5:565±72. [9] Fowler DH. Donor CD4-enriched cells of Th2 cytokine phenotype regulate graft versus host disease without impairing allogeneic engraftment in sublethally irradiated mice. Blood 1994;84(10):3540±9. [10] Middleton PG, Taylor PRA, Jackson G, Proctor SJ, Dickinson AM. Cytokine Gene polymorphisms associating with severe acute graft-versus-host disease in HLA-identical sibling transplants. Blood 1998;92(10):3943±8. [11] Cavet J, Middleton PG, Segall M, Noreen H, Davies SM, Dickinson AM. Recipient tumour necrosis factor a and interleukin 10 gene polymorphisms associated with early mortality and acute graft versus host disease severity in HLA matched sibling bone marrow transplants. Blood 1999;94(11):1±7. [12] Bogdanic V, Nemet D, Kastelan A, et al. Umbilical cord blood transplantation in a patient with Philadelphia chromosome-positive chronic myeloid leukaemia. Transplantation 1993;56(2):477±9. [13] Pahwa RN, Fleischer A, Than S, Good RA. Successful hematopoietic reconstitution with transplantation of erythrocytedepleted allogeneic human umbilical cord blood cells in a child with leukaemia. Proceedings of the National Academy of Sciences of the Unites States of America 1994;91(10):4485±8. [14] Gluckman E, Rocha V, Boyer-Chammard A, et al. Outcome of cord-blood transplantation from related and unrelated donors. Eurocord Transplant Group and the European Blood and Marrow Transplantation Group. N Engl J Med 1997;337(6):373±81. [15] Madrigal JA, Cohen SBA, Gluckman E, Charron DJ. Does cord blood transplantation result in lower graft-versus-host disease. It takes more than two to Tango. Human Immunology 1997;56:1±5. [16] Wang XN, Proctor SJ, Dickinson AM. Cytokine production and IL-10 regulation of cord blood cells following allogeneic stimulation (Abstract). Immunology 1998;95(suppl.1):126. [17] Wang XN, Sviland L, Adebo J, et al. Cellular alloreactivity of

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

195

human cord blood cells detected by T-cell frequency analysis and a human skin explant model. Transplantation 1998;66(7):903±9. Cohen SBA, Perez-Cruz I, Fallen P, Gluckman E, Madrigal JA. Analysis of the cytokine production by cord and adult blood. Human Immunology 1999;60:331±6. Perez-Cruz I, Bogunia-Kubik K, Cohen SBA. Cord blood cytokines. In: Cohen SBA, Gluckman E, Rubinstein P, Madrigal JA, editors. CB characteristics: role in stem cell transplantation. London: M Dunitz, 2000. p. 73±95. Fallen P, Cohen SBA. Cord blood T cells. In: Cohen SBA, Gluckman E, Rubinstein P, Madrigal JA, editors. Cord blood characteristics: role in stem cell transplantation. London: M Dunitz, 2000. p. 39±59. Lewis DB, Yu CC, Meyer J, English BK, Kahn SJ, Wilson CB. Cellular and molecular mechanisms for reduced interleukin 4 and interferon-gamma production by neonatal T cells. J Clin Invest 1991;87(1):194±202. Sautois B, Fillet G, Beguin Y. Comparative cytokine production by in vitro stimulated mononucleated cells from cord blood and adult blood. Exp Hematol 1997;25(2):103±8. Pirenne-Ansart H, Paillard F, De Groote D, et al. Defective cytokine expression but adult-type T-cell receptor, CD8, and p561ck modulation in CD3- or CD2-activated T cells from neonates. Pediatr Res 1995;37(1):64±9. Bon®ll M, Akbar AN, Salmon M, Robinson M, Burford G, Janossy G. Immature CD45RA(low)RO(low) T cells in the human cord blood. I. Antecedents of CD45RA+ unprimed T cells. J Immunol 1994;152(12):5613±23. Wilson CB, Westall J, Johnston L, Lewis DB, Dower SK, Alpert AR. Decreased production of interferon-gamma by human neonatal cells. Intrinsic and regulatory de®ciencies. J Clin Invest 1986;77(3):860±7. Lewis DB, Larsen A, Wilson CB. Reduced interferon-gamma mRNA levels in human neonates. Evidence for an intrinsic T cell de®ciency independent of other genes involved in T cell activation. J Exp Med 1986;163(4):1018±23. von Freeden U, Zessack N, van Valen F, Burdach S. Defective interferon gamma production in neonatal T cells is independent of interleukin-2 receptor binding. Pediatr Res 1991;30(3):270±5. Kesson AM, Bryson YJ. Uptake of extracellular Ca2+ is a requirement for production of interferon-gamma by cord blood mononuclear cells. Journal of Interferon Research 1991;11(2):81±6. Weatherstone KB, Rich EA. Tumor necrosis factor/cachectin and interleukin-1 secretion by cord blood monocytes from premature and term neonates. Pediatr Res 1989;25(4):342±6. Glover DM, Brownstein D, Burchett S, Larsen A, Wilson CB. Expression of HLA class II antigens and secretion of interleukin-1 by monocytes and macrophages from adults and neonates. Immunology 1987;61(2):195±201. Lee S, Suen Y, Chang L, et al. Decreased interleukin-12 (IL-12) from activated cord versus adult peripheral blood mononuclear cells and upregulation of interferon-gamma, natural killer, and lymphokine-activated killer activity by IL-12 in cord blood mononuclear cells. Blood 1996;88(3):945±54. Qian JX, Lee S, Suen Y, Knoppel E, van de Ven C, Cairo M. Decreased Interleukin-15 from activated cord versus adult peripheral blood mononuclear cells and the e€ect of interleukin-15 in upregulating anti-tumor immune activity and cytokine production in cord blood. Blood 1997;90(8):3106±17. Beck R, Lam-Po-Tang P. Comparison of cord blood and adult blood lymphocyte normal ranges: a possible explanation for decreased severity of graft versus host disease after cord blood transplantation. J Immunol Cell Biol 1994;72:440±4. D'Arena G, Musto P, Cascavilla N, et al. Flow cytometric

196

[35]

[36]

[37] [38] [39] [40] [41] [42] [43] [44] [45] [46]

[47]

[48] [49] [50]

[51]

[52]

[53]

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197 characterisation of human umbilical cord blood lymphocytes: immunophenotypic features. Haematologia 1998;83(3):197±203. Han P, Hodge G, Story C, Xu X. Phenotypic analysis of functional T-lymphocyte subtypes and natural killer cells in human cord blood: relevance to umbilical cord blood transplantation. Br J Haematol 1995;89(4):733±40. Berry SM, Fine N, Bichalski JA, Cotton DB, Dombrowski MP, Kaplan J. Circulating lymphocyte subsets in second- and thirdtrimester fetuses: comparison with newborns and adults. Am J Obstet Gynecol 1992;167(4):895±900. Rabian-Herzog C, Lesage S, Gluckman E, Charron D. Characterisation of lymphocyte subpopulations in cord blood. J Hematother 1993;2(2):255±7. Lewis DB, Prickett KS, Larsen A, Grabstein K, Weaver M, Wilson CB. Restricted production of interleukin 4 by activated human T cells. Proc Natl Acad Sci USA 1988;85(24):9743±7. Sanders B, Andersson J, Andersson U. Assessment of cytokines by immuno¯uorescence and the paraformaldehyde-saponin procedure. Immunological Reviews 1991;119:65±93. Akbar A, Salmon M, Janossy G. The synergy between naive and memory T cells during activation. Immunol Today 1991;12(6):184±8. Ito M, Koide W, Sakurai M. Changes in intracellular cytokine levels in newborn and adult lymphocytes induced by HSV-1. Journal of Medical Virology 1998;56(2):145±50. Hassan J, Reen DJ. Cord Blood CD4+CD45RA+ T cells achieve a lower magnitude of activation when compared with their adult counterparts. Immunology 1997;90(3):397±401. Cohen SBA, Katsikis PD, Chu CQ, et al. High IL-10 production by the activated T cell population within the rheumatoid synovial membrane. Arthr and Rheum 1995;38:946±52. Brookes S, Cohen SBA, Price EJ, et al. T cell clones from Sjogrens syndrome salivary gland produce high levels of IL-10. Clin Exp Immunol 1996;103:268±72. Groux H, Bigler M, de Vries J, Roncarolo M. Interleukin-10 induces a long term antigen-speci®c anergic state in human CD4+ T cells. J Exp Med 1996;184:19±29. Damle NK, Mohagheghpour N, Hansen JA, Engleman EG. Alloantigen-speci®c cytotoxic and suppressor T lymphocytes are derived from phenotypically distinct precursors. J Immunol 1983;131(5):2296±300. Risdon G, Gaddy J, Horie M, Broxmeyer HE. Alloantigen priming induces a state of unresponsiveness in human umbilical cord blood T cells. Proc Natl Acad Sci USA 1995; 92(6):2413±7. Mues B, Langer D, Zwadlo G, Sorg C. Phenotypic characterisation of macrophages in human term placenta. Immunology 1989;67:303±7. Fenton MJ, Buras JA, Donnelly RP. IL-4 reciprocally regulated L-1 and IL-1 receptor antagonist expression in4human monocytes. J Immunol 1992;149:1283±8. Schebesch C, Kodeja V, Muller C, Hakij N, Orfanos CE, Goerdt S. Alternatively activated macrophages actively inhibit proliferation of peripheral blood monocytes and CD4+ T cells in vitro. Immunology 1997;92:478±86. Cheung DL, Hart PH, Vitti GF, Whitty GA, Hamilton JA. Contrasting e€ects of interferon-gamma and interleukin-4 on the interleukin-6 activity of stimulated human monocytes. Immunology 1990;71:70±5. Bonder CS, Dickensheets HL, Finaly-Jones JJ, Donnelly RP, Hart PH. Involvement of IL-2 receptor gamma chain (gammac) in the control by IL-4 of human monocyte and macrophage proin¯ammatory mediator production. J Immunol 1998;160:4048±56. Chang M-DY, Pollard JW, Khalili H, Goyert SM, Diamond B. Mouse placental macrophages have a decreased ability to present antigen. Proc Natl Acad Sci USA 1993;90:462±6.

[54] Cohen SBA, Morgan CL, Perandin F, Perez-Cruz I, Martinez B, Madrigal JA. Cord blood serum causes reduced lymphocyte responses in comparison to adult serum. Human Immunology 2000;61(2):111±4. [55] Lewis D, Yu C, Meyer J, English K, Kahn S, Wilson C. Cellular and molecular mechanisms for reduced IL-4 and IFNg production by neonatal T cells. J Clin Invest 1991;87:194±202. [56] Chalmers IMH, Janossy G, Contreras M, Navarrete C. Intracellular cytokine pro®le of cord and adult blood lymphocytes. Blood 1998;92:11±8. [57] Bertotto A, Gerli R, Lanfrancone L, et al. Activation of cord T lymphocytes. II. Cellular and molecular analysis of the defective response induced by anti-CD3 monoclonal antibody. Cell Immunol 1990;127(2):247±59. [58] Cairo MS, Suen Y, Knoppel E, van de Ven C, Nguyen A, Sender L. Decreased stimulated GM-CSF production and GMCSF gene expression but normal numbers of GM-CSF receptors in human term newborns compared with adults. Pediatr Res 1991;30(4):362±7. [59] English BK, Hammond WP, Lewis DB, Brown CB, Wilson CB. Decreased granulocyte-macrophage colony-stimulating factor production by human neonatal blood mononuclear cells and T cells. Pediatr Res 1992;31(3):211±6. [60] Chang M, Suen Y, Lee S, et al. Transforming growth factorbeta 1, macrophage in¯ammatory protein-1 alpha, and interleukin-8 gene expression is lower in stimulated human neonatal compared with adult mononuclear cells. Blood 1994;84(1):118± 24. [61] Qian JX, Lee S, Suen Y, Knoppel E, van de Ven C, Cairo MS. Decreased interleukin 15 from activated cord versus adult peripheral blood mononuclear cells and the e€ect of interleukin 15 in upregulating antitumor immune activity and cytokine production in cord blood. Blood 1997;90:3160±77. [62] Wilcox EC, Wang XN, Oliver K, Dickinson AM. Increased sensitivity of the cord blood alloresponse to cyclosporin A. Blood 1999;94(10 suppl. 1):572a Abstract. [63] Lucivero G, Dalla Mora L, Bresciano E, Loria MP, Pezone L, Mancino D. Functional characteristics of cord blood T lymphocytes after lectin and anti-CD3 stimulation. Di€erences in the way T cells express activation molecules and proliferate. Int J Clin Lab Res 1996;24(4):255±61. [64] Amlot PL, Tahami F, Chinn D, Rawlings E. Activation antigen expression on human T cells. I. Analysis by two-colour ¯ow cytometry of umbilical cord blood, adult blood and lymphoid tissue. Clin Exp Immunol 1996;105:176±82. [65] Zola H, Fusco M, Macardle PJ, Flego L, Roberton D. Expression of cytokine receptors by human cord blood lymphocytes: comparison with adult blood lymphocytes. Pediatr Res 1995;38(3):397±403. [66] Saito S, Fujii M, Saito M, Kato Y, Moriyama I, Ichijo M. IL-2 receptor expression and function on human cord blood mononuclear cells following PHA and anti-CD3 antibody stimulation. Journal of Reproductive Immunology 1989;16(1):31±42. [67] Saito S, Morii T, Umekage H, et al. Expression of the interleukin-2 receptor gamma chain on cord blood mononuclear cells. Blood 1996;87(8):3344±450. [68] Pastorelli G, Rousset F, PeÁne J, et al. Cord blood B cells are mature in their capacity to switch to IgE-producing cells in response to interleukin-4 in vitro. Clinical and Experimental Immunology 1990;82(1):114±9. [69] Peleman R, Delespesse G. In vitro synthesis of human IgE by neonatal lymphocytes: enhancing e€ect of interferon-gamma. Cellular Immunology 1990;129(2):299±309. [70] Ueno Y, Miyawaki T, Seki H, et al. Di€erential e€ects of recombinant human interferon-gamma and interleukin 2 on natural killer cell activity of peripheral blood in early human development. J Immun 1985;135(1):180±4.

S.B.A. Cohen et al. / Cytokine & Growth Factor Reviews 11 (2000) 185±197 [71] KoÈgler G, Wilke M, Middleton PG, et al. Cytokine gene polymorphism of the recipient in¯uences the development of acute GvHD in HLA mismatched cord blood transplants. Blood 1999;94(10 suppl. 1):639a Abstract. [72] Kogler G, Middleton PG, Wilke M, et al. In¯uence of cytokine gene polymorphism of the recipient on acute GvHD after cord blood transplantation, submitted 2000. [73] Lerner KG, Kao GF, Storb R, Buckner CD, Clift RA, Thomas ED. Histopathology of graft vs host reaction (GvHR) in human recipients of marrow from HLA-matched sibling donors. Transplant Proc 1974;6:367±71. [74] Dickinson AM, Hromadnikova I, Sviland L, et al. Use of a skin explant model for predicting GvHD in HLA-matched bone marrow transplants e€ect of GvHD prophylaxis. Bone Marrow Transplant 1999;24:857±63. [75] Marini J, Cohen SBA. T cells and cytokines. In: Balkwell F, editor. Cytokines: a practical approach, 3rd ed., Oxford University Press, in press. [76] Anderson U, Anderson J, Lindors A, Wagner K, Moller G, Heusser C. Simultaneous production of interleukin 2, interleukin 4 and interferon-g by activated human blood lymphocytes. European Journal of Immunology 1990;20:1591±6. [77] Schauer U, Jung T, Krug N, Frew A. Measurement of intracellular cytokines. Immun Today 1996;17(7):305±6. [78] Pinto L, Covas M, Victorino R. Loss of CD45RA and gain of CD45RO after in vitro activation of lymphocytes from HIVinfected patients. Immunology 1991;73(2):147±50. [79] Wallace DL, Beverley PC. Phenotypic changes associated with activation of CD45RA+ and CD45RO+ T cells. Immunology 1990;69(3):460±7. [80] Yamada A, Kaneyuki T, Hara A, Rothstein D, Yokoyama M. CD45 isoform expression on human neonatal T cells: expression and turnover of CD45 isoforms on neonatal versus adult T cells after activation. Cell Immunol 1992;142(1):114±24. [81] Lightstone EB, Marvel J. CD45RA is detected in all thymocyte subsets de®ned by CD4 and CD8 by using three-colour ¯ow cytometry. Immunology 1990;71(4):467±72. [82] Alessio M, Roggero S, Funaro A, et al. CD38 molecule: structural and biochemical analysis on human T lymphocytes, thymocytes, and plasma cells. J Immunol 1990;145(3):878±84. [83] Sautois B, Fillet G, Beguin Y. Comparative cytokine production by in vitro stimulated mononuclear cells from cord blood and adult blood. Experimental Hematology 1997; 25:103±8. [84] van de Mast B, van Beelen E, Claas FHJ. The in¯uence of cryoperservation on cytokine secretion pro®les of peripheral blood mononuclear cells: consequences for immunological monitoring, Ph.D. thesis. In: Thesis P, editor. The immunological e€ects of blood transfusion in humans. Chap. 6 1998. [85] Miyawaki T, Seki H, Taga K, Sato H, Taniguchi N. Dissociated production of interleukin-2 and immune (gamma)

[86]

[87]

[88] [89] [90] [91]

[92] [93] [94]

[95]

[96]

[97]

[98]

197

interferon by phytohaemagglutinin stimulated lymphocytes in healthy infants. Clinical and Experimental Immunology 1985;59(2):505±11. Takahashi N, Imanishi K, Nishida H, Uchiyama T. Evidence for immunological immaturity of cord blood T cells. Cord blood T cells are susceptible to tolerance induction to in vitro stimulation with a superantigen. J Immun 1995;155(11):5213±9. Kruse A, Neustock P, Reuter M, Kirchner H. T-cell surface molecule expression and interferon-gamma production in human cord blood. Journal of Interferon Research 1993;13(3):221±5. Watson W, Oen K, Ramadahin R, Harman C. Immunoglobulin and cytokine production by neonatal lymphocytes. Clinical and Experimental Immunology 1991;83(1):169±74. Trivedi HN, Hayglass KT, Gangur V, Allardice JG, Embree JE, Plummer FA. Analysis of neonatal T cell and antigen presenting cell functions. Human Immunology 1997;57(2):69±79. Racadot E, Schaal J, Van Lemmens P, Billot M, Herve P. Immunological characteristics of umbilical cord blood cells: phenotypic and functional analysis. J Hemathother 1993;2(2):251±3. Chang M, Suen Y, Lee SM, et al. Transforming growth factorbeta 1, macrophage in¯ammatory protein-1 alpha, and interleukin-8 gene expression is lower in stimulated human neonatal compared with adult mononuclear cells. Blood 1994;84(1):118± 24. Chin TW, Ank BJ, Strom SR, Stiehm ER. Enhanced interferon production and lymphokine-activated cytotoxicity of human placental cells. Cell Immun 1988;113(1):1±9. Kruse A, Rink L, Rutenfranz I, Kolanczyk B, Kirchner H. Interferon and lymphokine production by human placenta and cord blood. Journal of Interferon Research 1992;12(2):113±7. Lau A, Sigaroudinia M, Yeung M, Kohl S. Interleukin-12 induces interferon-gamma expression and natural killer cytotoxicity in cord blood mononuclear cells. Pediatr Res 1996;39(1):150±5. Nair MP, Schwartz SA, Menon M. Association of decreased natural and antibody-dependent cellular cytotoxicity and production of natural killer cytotoxic factor and interferon in neonates. Cellular Immunology 1985;94(1):159±71. Buzby J, Lee S, Van Winkle P, DeMaria C, Brewer G, Cairo M. Increased granulocyte-macrophage colony-stimulating factor mRNA instability in cord versus adult mononuclear cells is translation-dependent and associated with increased levels of A+U-rich element binding factor. Blood 1996;88(8):2889±97. Mills K, Gross T, Varney M, et al. Immunological phenotype and function in human bone marrow, blood stem cells and umbilical cord blood. Bone Marrow Transplant 1996;18(1):53± 61. Harris DT, Schumacher MJ, Locascio J, et al. Phenotypic and functional immaturity of human umbilical cord blood T lymphocytes. Proc Natl Acad Sci USA 1992;89:10,006±10.