Research Article

T cell engraftment in lymphoid tissues of human peripheral blood

lymphocyte reconstituted SCID mice with or without prior

activation of cells

COLLEEN OLIVE, CATHERINE CHEUNG AND MICHAEL C FALK

Renal Research Laboratory, Renal Dialysis Unit, Princess Alexandra Hospital, Woolloongabba, Brisbane, Queensland,Australia

Summary The reconstitution of severe combined immunode®ciency (SCID) mice with human PBL (Hu-PBL-SCID) was assessed using fresh unstimulated PBL and anti-CD3-stimulated PBL. Mice were reconstituted with

PBL by intraperitoneal injection of 1±2.5 ´ 107 PBL in PBS; controls received PBS. Successful engraftment ofhuman PBL in SCID mice was determined by measurement of human IgG in mouse sera, polymerase chainreaction (PCR) detection of human-speci®c HLA-DRb DNA in SCID periphery, and immunohistochemical

staining of mouse tissues (spleen, lymph nodes, thymus, liver and lung) with antibodies speci®c for human CD45and CD3. Human IgG was detected 1week after reconstitution in sera of all animals that received at least 1 ´ 107

PBL and continued to increase for 8weeks. Human-speci®c HLA-DRb DNA was detected in the majority ofmice 3weeks after reconstitution but not in controls. Moreover, immunohistochemical analysis of Hu-PBL-SCID

mouse tissues revealed the presence of human CD45+ cells in all tissues examined. CD3+ T cell engraftment wasobserved in lymphoid tissues irrespective of whether PBL had been activated prior to transfer or not.

Key words: Hu-PBL-SCID, lymphoid tissues, reconstituted, T cell engraftment.

Introduction

Severe combined immunode®ciency (SCID) mice have agenetic defect that results in the inability to rearrange T cell

receptor and immunoglobulin genes, and produce immu-nocompetent T cells and B cells, respectively.1 These micelack a functional immune system and have proved to besuitable hosts for the engraftment of human haematopoi-

etic stem cells and mature PBL.2,3 The functional capacityof the engrafted human immune cells and their potentialrole in generating humoral and cellular immune responses

in vivo, has been investigated. Long-term B cell antibodyproduction, including the production of xenoreactive anti-bodies,4 speci®c antibodies to recall antigen,5 and auto-

antibody production,6 has been demonstrated in SCIDmice repopulated with human PBL. In the case of T cells,those isolated from Hu-PBL-SCID chimeras appeared to be

in a state of unresponsiveness to conventional T cell stim-ulation, although this was reversible.7 More recent studieshave indicated that rejection of islet allografts in theHu-PBL-SCID recipient was mediated by functionally

competent donor graft in®ltrating T cells.8 In addition,anti-tumour immunity inHu-PBL-SCID to the development

of Daudi lymphoma has been shown to involve cd T cell

proliferation in response to Daudi.9 Together, these ®nd-ings indicate that the function of the human immune systemis retained, at least partially, in PBL-reconstituted SCID

mice. To this end, human-SCID chimeric animal modelshave been used to study a variety of human immune re-sponses, including those elicited in response to infection,10

self-antigens,6 tumours,9 and allografts.8

The successful engraftment of human immunocompetentcells into SCID mice is variable and may be attributed to a

number of factors including host NK activity,11 donor EBVstatus,3 and the activation status of the cells to be trans-ferred.12 Activation of human PBL with an anti-CD3monoclonal antibody has been shown to increase the inci-

dence of engraftment of human PBL into SCID lymphoidtissues, and resulted in T cell engraftment within the thy-mus.12 In vitro pre-activation of human T cells by mixed

lymphocyte tissue culture13 and human growth hormone,14

have also been shown to enhance the migration of humanT cells to SCID lymphoid tissues following reconstitution.

These studies have largely utilized FACS analysis of in vitroisolated cell populations to determine the levels of humancell engraftment in SCID mice.

In the present study we have compared the engraftment

of fresh unstimulated human PBL and anti-CD3-stimu-lated PBL in SCID mice. Engraftment was assessed inSCID mice by measurement of human IgG (HIgG) in

mouse sera, detection of human-speci®c HLA-DRb DNAin SCID periphery, and, in situ, by immunohistochemical

Correspondence: Colleen Olive, Renal Research Laboratory,

2nd Floor Renal Dialysis Unit, Princess Alexandra Hospital,

Cornwall Street, Woolloongabba, Qld 4102, Australia.

Email: <mdcolive@dingo.uq.edu.au>

Received 25 May 1998; accepted 2 July 1998.

Immunology and Cell Biology (1998) 76, 520±525

staining of mouse tissues with anti-human CD45 and CD3antibodies.

Materials and Methods

Mice

Approximately 8-week-old male C.B-17 scid/scid (SCID) mice were

obtained from a breeding colony (Walter and Eliza Hall Institute,

Melbourne, Victoria, Australia). Mice were housed under sterile

conditions at the Princess Alexandra Hospital animal facility for

the duration of the experiment. All mice were `non-leaky' SCID

and had mouse IgG levels <2lg/mL as assessed by inhibition

ELISA.

Preparation of PBL and SCID reconstitution

Approximately 100mL peripheral blood was obtained each from

two adult male healthy donors (donor A and donor B). Both do-

nors were EBV seronegative. The PBL were separated by density

gradient centrifugation over Ficoll-Paque (Pharmacia, North

Ryde, NSW, Australia), and washed three times in PBS. Cells were

counted and the cell viability was >95% by trypan blue exclusion.

For CD3 activation, donor B PBL at 2´ 106 cells/mL were incu-

bated overnight at 37°C, 5% CO2 in RPMI media containing

25mmol/L HEPES (Gibco BRL, Melbourne, Australia), 2mmol/L

glutamine, 10% foetal bovine serum, 100U penicillin, 100lg/mL

streptomycin (Trace Biosciences, Castle Hill, NSW, Australia), and

100 ng/mL anti-CD3 antibody (Dako, NSW, Australia). Cells were

washed in PBS and counted; cell viability remained >95%. Mice

were reconstituted with human PBL by intraperitoneal injection of

PBL in 0.5mL PBS. Control mice received PBS.

Measurement of human IgG in sera

Mice were bled periodically by retro-orbital eye bleed. The sera

were collected after centrifugation of blood and HIgG levels were

measured using an inhibition ELISA. Brie¯y, all ELISA plates

were coated with 0.5lg/well HIgG and then blocked for 1 h at

room temperature with PBS/1% BSA. Following washing of plates

with PBS/0.05% Tween-20, standards of known concentration of

HIgG or test sera from SCID mice were used to inhibit the binding

of peroxidase-conjugated rabbit anti-human IgA+IgG+IgM

(Dako, Botany, NSW, Australia); dilution 1:5000. o-Phenylenedi-

amine reaction products were detected by tetramethylbenzidine

(TMB) substrate (Dako) and absorbances at 450 nm were mea-

sured by an automated microplate reader.

Polymerase chain reaction detection of human HLA-DRbDNA in SCID periphery

Genomic DNA was prepared from mouse peripheral blood using a

one-step guanidinium hydrochloride extraction procedure.15 The

DNA was ampli®ed by polymerase chain reaction (PCR) in a

standard 50 lL reaction containing 10mmol/L Tris/HCl (pH8.4),

50mmol/L KCl, 1.5mmol/L MgCl2, 0.01% (w/v) gelatin,

0.2mmol/L of each deoxynucleotidetriphosphate (dNTP), 25 pmol

each of forward and reverse primer, and 1.25 units of Taq DNA

polymerase (Promega, Sydney, Australia). The following oligonu-

cleotide primers speci®c for human HLA-DRb were used,8

(forward 5¢-TTCTTCAATGGGACGGAGCG-3¢ and reverse

5¢-GCCGCTGCACTGTGAAGCTCTC-3¢). Polymerase chain

reaction ampli®cation was carried out for 1min each at 95°C, 68°C,and 72°C for 40 cycles. Human genomic DNA was included as a

positive PCR control; negative PCR controls contained all the

reaction components for PCR except template DNA. The PCR

products (5lL) were electrophoresed in a 1.25% agarose gel and

the length of the products assessed by comparison with molecular

size standards after staining with ethidium bromide. The speci®city

of each PCR product was con®rmed by Southern blotting

and hybridization with an internal HLA-DRb-speci®c probe

(5¢-CTGGAACAGCCAGAAGGAC-3¢) that was end-labelled

with [c32P]-ATP.

Immunohistochemistry

All surviving mice were killed 8weeks after reconstitution. Mouse

organs (spleen, lymph nodes, thymus, liver and lung) were taken

and ®xed in 10% formalin. Para�n-embedded SCID tissue sections

were stained with a mouse anti-human CD45 (leucocyte common

antigen) monoclonal antibody. Tissue sections were depara�nized,

incubated for 5min in 0.6% hydrogen peroxide/methanol to block

endogenous peroxidase, and then incubated for 15min at 95°C in

10mmol/L sodium citrate (pH6.0) target retrieval bu�er. After

washing for 5min in TBS/1% BSA, endogenous avidin and biotin

sites were blocked using a commercially available kit (Dako).

Tissue sections were blocked for 20min in 20% normal rabbit se-

rum before incubation for 30min at room temperature with anti-

human CD45 primary antibody; dilution 1:50 (Dako). Tissue

sections were washed in trisbu�ered saline (TBS)/1% BSA and then

incubated for 30min with rabbit anti-mouse biotinylated secondary

antibody; dilution 1:200 (Dako). Following washing in TBS/1%

BSA, positive cells were detected using StreptABComplex/horse-

radish peroxidase (HRP) and diaminobenzidine (DAB) according

to the manufacturer's instructions (Dako). The same protocol was

used for staining of SCID tissues with an anti-human CD3 poly-

clonal antibody; dilution 1:50 (Dako), except that slides were in-

cubated for 30min in target retrieval bu�er, blocked in 20%

normal swine serum and incubated with an anti-rabbit biotinylated

secondary antibody; dilution 1:300 (Dako). All antibodies and sera

were diluted in TBS/1% BSA. Positive controls for tissue staining

included human tonsil for human CD45 and CD3. Negative con-

trol staining of tissue sections was carried out using an irrelevant

isotype-matched antibody, instead of primary antibody. All tissue

sections were counterstained with Meyer's haematoxylin.

Results and discussion

Human IgG and HLA-DRb DNA in Hu-PBL-SCID

periphery

Human IgG levels were measured in PBL-reconstitutedSCID mice sera at several time-points following thetransfer of human PBL (Table 1). Human IgG was de-

tected at 1week after reconstitution, albeit at relativelylow levels, in animals that received at least 1´ 107 PBL.The levels of HIgG continued to rise over a period of8weeks in all PBL-reconstituted SCID, and reached levels

comparable to those detected in the original donor PBL;donor A, 2907lg/mL and donor B, 3160lg/mL. Thelong-term production of human antibodies in Hu-PBL-

SCID has previously been demonstrated.4±6 Together,these ®ndings indicate that the SCID environment sup-ports all the necessary immune factors required for B cell

T cell engraftment in Hu-PBL-SCID 521

function and antibody production. Very low levels ofHIgG were detected in SCID A4, suggesting that more

than 1 ´ 107 cells may be required for e�cient engraftmentof human PBL into SCID mice. However, more animalswould need to be examined to establish a correlation

between engraftment and cell number. The number ofcells to be transferred has previously been shown to be animportant factor in determining the success of repopula-

tion of human cells into SCID.3 Negligible levels ofHIgG, or no HIgG, were detected in PBS control mice,A5 and B5, at all of the time points tested. Of the 10 miceexamined, two died, one each at 6 and 8weeks after re-

constitution. These two SCID mice had relatively highHIgG levels in their sera when compared to the otheranimals in the group.

The HLA-DRb human-speci®c DNA was detected in themajority of SCID mice periphery at 3weeks post-reconsti-tution (Table 1; Fig. 1). There were no HLA-DRb PCR

ampli®cation signals detected in control SCID, or in SCIDA4, at any time point during the analysis. The integrity of

each genomic DNA sample was con®rmed by PCR am-pli®cation employing primers speci®c for mouse wheyacidic protein (not shown).

Human CD45+ cells in Hu-PBL-SCID lymphoid tissues

All surviving mice were killed at 8weeks post-reconstitu-tion. Upon macroscopic examination of mouse organs we

observed splenomegaly as characterized by an enlargedspleen in a few cases, whereas all other organs appearednormal. There was no evidence of tumour growth in any ofthe xenografted SCID. Animals showed no physical

symptoms of graft-versus-host disease; that is, hunchedback, ru�ed fur, diarrhoea or emaciation. Postmortemexamination of xenografted SCID showed no marked

lymphocytic in®ltration of the intestinal crypts in the few

Table 1 Human (H) IgG levels and HLA-DRb DNA detection in the periphery of SCID mice reconstituted with human peripheral blood

lymphocytes

SCID Cells 1-week HLA-DRb 3-week HLA-DRb 6-week HLA-DRb 8-week HLA-DRbHIgG DNA HIgG DNA HIgG DNA HIgG DNA

lg/mL lg/mL lg/mL lg/mL

A1 2.2´ 107 PBL 66.87 ± 1468 + 4320 NS 4009 +

A2 2.2´ 107 PBL 62.73 + 1072 + 2600 + NS NS

A3 2.0´ 107 PBL 23.01 ± 1112 + 3489 + Dead Dead

A4 1.0´ 107 PBL 1.86 ± 5 ± 14.13 ± 7.5 ±

A5 PBS control 1.41 ± 1.55 ± ND ± ND ±

B1 2.5 ´ 107/CD3 PBL 10.07 ± 775 + 2994 + 3679 +

B2 2.5 ´ 107/CD3 PBL 18.22 + faint 1250 + Dead Dead Dead Dead

B3 2.5 ´ 107/CD3 PBL 24.0 ± 928 ± 3503 + NS NS

B4 2.5 ´ 107/CD3 PBL 12.37 ± 893 + 4179 + 2822 +

B5 PBS control 0.77 ± 0.79 ± ND ± ND ±

Detection of HLA-DRb DNA PCR ampli®cation product is designated by `+' for positive and `±' for negative in those samples that were

tested.

NS, no sample; ND, not detected; SCID, severe combined immunode®ciency; PCR, polymerase chain reaction.

Figure 1 Polymerase chain reaction (PCR) detection of human-speci®c HLA-DRb DNA in the periphery of severe combined

immunode®ciency (SCID) mice reconstituted with human PBL. Genomic DNA was prepared from mouse peripheral blood obtained

3 and 6 weeks post-reconstitution of SCID mice with unstimulated PBL (SCID A1-A4) and anti-CD3-stimulated PBL (SCID B1-

B4). The SCID mice A5 and B5 were PBS controls. The PCR ampli®cation of mouse genomic DNA was performed with primers

speci®c for human HLA-DRb and the PCR products were detected by Southern blot hybridization with an internal oligonucleotide

probe. Human genomic DNA was used as a positive control (Pos) for PCR. Exposure of blots was 3 h, and 25min for 3-week and

6-week post-reconstitution, respectively. In the 6-week post-reconstituted SCID mice, there was no blood sample obtained for A1,

and SCID B2 died before sample collection.

522 C Olive et al.

gut mucosa biopsies taken for analysis, although somescattered CD45+ cells were detected (not shown). Previous

studies of human T cell engraftment in SCID mice haveshown that splenomegaly is a prominent feature of thesechimeric animals and results from extensive murine extra-

medullary splenic haematopoiesis.12±14 In addition, while agraft-versus-host reaction has been demonstrated in some

studies in the Hu-PBL-SCID,3,12,16 this appears not to be amajor health problem.

Immunohistochemical staining for human CD45 in ourHu-PBL-SCID revealed the presence of CD45+ cells in theliver and lung, and in particular there was extensive human

CD45 cellular in®ltration into the spleen, lymph nodes andthymus, which, surprisingly, was equally pronounced in

Table 2 Immunohistochemical staining of human CD45 and CD3 in Hu-PBL-SCID tissues

SCID CD45+ cells CD3+ T cells

Spleen Lymph

nodes

Thymus Liver Lung Spleen Lymph

nodes

Thymus Liver Lung

A1 +++ +++ ++++ +++ ++ +++ ± + ++ ++

A2 ++++ ++++ ++++ +++ +++ ++ ++ +++ ++ +++

A3 +++ ++++ ++++ +++ + +++ ++++ ++++ +++ +

A4 ± ± ± ± ± ± ± ± ± ±

A5 ± ± ± ± ± ± ± ± ± ±

B1 ++++ ++++ +++ +++ ++++ + ± ++++ +++ ++

B2 +++ NS NS +++ ++++ +++ NS NS +++ ++++

B3 +++ +++ ++++ +++ ++++ +++ +++ ++ ++ +

B4 ++ ++++ +++ +++ ++++ ++ ++++ ++++ ++ +++

B5 ± NS ± ± ± ± NS ± ± ±

Staining of either CD45 or CD3 was scored by two independent observers as follows: ±, no staining; +, scattered staining; ++, weak

localized staining; +++, intense localized staining; ++++, intense staining throughout. NS, no sample; SCID, severe combined im-

munode®ciency.

Figure 2 Immunohistochemical staining of human (Hu)-PBL-severe combined immunode®cient (SCID) mouse tissues for human

CD45+ cells. Para�n-embedded tissue sections of (a) spleen, (b) lymph nodes, (c) thymus, (d) liver and (e) lung were obtained from a

SCID mouse (A3) 8 weeks following intraperitoneal injection of 2 ´ 107 unstimulated PBL. Tissue sections were stained with anti-

human CD45 and positive cells detected with diaminobenzidine (DAB) substrate (brown colour). Massive in®ltration of human

CD45+ cells were observed in the spleen, lymph nodes and thymus; numerous CD45+ cells were also detected in the liver whereas a

few scattered CD45+ cells were observed in the lung. Note that in this case CD45+ cells localized within the peri-arteriolar sheaths in

the spleen. Magni®cation ´ 100; ´ 250 in (b) and (c).

T cell engraftment in Hu-PBL-SCID 523

mice that received fresh unstimulated PBL and those thatreceived CD3-stimulated PBL (Table 2). In contrast, we

observed that the level of CD45 cell staining was greater inthe lungs of animals that were reconstituted with CD3-stimulated PBL because all of these mice showed intense

staining of CD45+ cells throughout the bronchioles, alveoliand lung parenchyma (Table 2). There were no otherapparent di�erences observed between those animals that

received unmanipulated PBL and those that received CD3-stimulated PBL because all animals, except SCID A4,showed evidence of human CD45+ cell engraftment. These®ndings indicate that abrogation of host NK activity by

anti-asialoGM-1, which has previously been shown to en-hance the engraftment of human PBL into SCID,11±16 is notnecessary for the establishment of chimeric Hu-PBL-SCID.

A representative example of staining of tissues from a Hu-PBL-SCID is given in Fig. 2. In contrast, no humanCD45+ cells were detected in the control mice. In the

spleens of some SCID, we found that human CD45+ cellslocalized within the peri-arteriolar sheaths of the whitepulp, whereas there was no speci®c localization of CD45+

cells observed within the lymph nodes and thymus. These

®ndings support the migration speci®cally of human cellsinto Hu-PBL-SCID tissues. To rule out the possibility ofcross-reactivity of the anti-human CD45 antibody with

mouse leucocytes, we also carried out immunohistochemi-cal staining of C57/BL6 mouse spleen with an anti-mouseand anti-human CD45 antibody. Numerous murine

CD45+ cells were detected in the C57/BL6 spleen whereasstaining for human CD45+ cells was negative (not shown).

Human CD3+ T cells predominate in Hu-PBL-SCID

lymphoid tissues

To determine the extent of CD3+ T cell in®ltration in ourSCID chimeras, we performed immunohistochemical stain-ing of SCID tissues with an anti-human CD3 antibody.

CD3+ T cells were detected in the liver and lung, andpredominated in the spleen, lymph nodes, and thymus(Table 2). We found no major di�erences between animals

that received unmanipulated PBL and those that receivedCD3-stimulated PBL regarding human CD3 T cell engraft-ment in lymphoid tissues. Interestingly, CD3+T cell staining

was observed in the thymus when animals received unstim-ulated PBL. These ®ndings indicate that T cell migrationfrom the peritoneal cavity and entry into the thymus can

occur in SCID mice repopulated with unstimulated PBL.A representative example of staining of tissues from aHu-PBL-SCID is given in Fig. 3. T cell entry into the thymusof Hu-PBL-SCID has only previously been demonstrated

following activation of PBL,12 or treatment with growthhormone,14 indicating that di�erent homing signals may berequired for T cell entry into di�erent lymphoid tissues.

There have been many reports of the engraftment ofSCID mice with human PBL; however, successful engraft-

Figure 3 Immunohistochemical staining of human (Hu)-PBL-severe combined immunode®cient (SCID) mouse tissues for human

CD3+ T cells. Para�n-embedded tissue sections of (a) spleen, (b) lymph nodes, (c) thymus, (d) liver and (e) lung, were obtained from

a SCID mouse (A3) 8 weeks following intraperitoneal injection of 2´ 107 unstimulated PBL. Tissue sections were stained with anti-

human CD3 and positive cells detected with diaminobenzidine (DAB) substrate (brown colour). Engraftment of human CD3 T cells

was observed in the spleen, lymph nodes, thymus, and liver whereas fewer T cells were detected in the lung. Magni®cation ´ 100;

´ 250 in (b) and (c).

524 C Olive et al.

ment, or enhanced engraftment, has previously beenachieved largely with some form of manipulation of the

human PBL, irradiation of the host, or by removal of hostNK cells.8,11±14 Unlike previous studies,12 we found nodi�erence in the engraftment of SCID mice by human PBL

when using CD3-activated as opposed to unstimulated PBLas assessed by immunohistochemistry. Moreover, the de-tection of human speci®c HLA-DRb DNA by PCR, to-

gether with measurement of serum HIgG, allows anassessment of the level of human PBL engraftment at� 3 weeks post-reconstitution, and does not require theanimals to be killed.

Our data showed CD45 and in particular T cell engraft-ment in mouse lymphoid tissues of Hu-PBL-SCID, irre-spective of whether PBL had been activated prior to transfer

or not. Furthermore, T cell engraftment in the SCID thymusoccurred following the transfer of unstimulated PBLindicating that this model may be suitable in further

functional studies relating to the human immune response.

Acknowledgements

This work was carried out in the Department of SurgeryResearch Laboratories at the Princess Alexandra Hospital.We thank the Renal Research Trust Fund for their con-

tinued support, and the Mayne Bequest Fund (Universityof Queensland). We are grateful to Dr Michael Walsh(University of Queensland Medical School Pathology) for

mouse tissue sectioning, and Ms Anne-Louise Bullock(Department of Medicine) for kindly providing the mousewhey acidic protein PCR primers.

References

1 Schuler W, Weiler IJ, Schuler A et al. Rearrangement of anti-

gen receptor genes is defective in mice with severe combined

immune de®ciency. Cell 1986; 46: 963±72.

2 Kamel-Reid S, Dick JE. Engraftment of immune-de®cient mice

with human hematopoietic stem cells. Science 1988; 242: 1706±

9.

3 Mosier DE, Gulizia RJ, Baird SM, Wilson DB. Transfer of a

functional human immune system to mice with severe com-

bined immunode®ciency. Nature 1988; 335: 256±9.

4 Williams SS, Umemoto T, Kida H, Repasky EA, Bankert RB.

Engraftment of human peripheral blood leukocytes into severe

combined immunode®cient mice results in the long term and

dynamic production of human xenoreactive antibodies. J. Im-

munol. 1992; 149: 2830±6.

5 Nonoyama S, Smith FO, Ochs HD. Speci®c antibody pro-

duction to a recall or a neoantigen by SCID mice reconstituted

with human peripheral blood lymphocytes. J. Immunol. 1993;

151: 3894±901.

6 Tighe H, Silverman GJ, Kozin F et al. Autoantibody produc-

tion by severe combined immunode®cient mice reconstituted

with synovial cells from rheumatoid arthritis patients. Eur. J.

Immunol. 1990; 20: 1843±8.

7 Tary-Lehmann M, Saxon A. Human mature T cells that are

anergic in vivo prevail in SCID mice reconstituted with human

peripheral blood. J. Exp. Med. 1992; 175: 503±16.

8 Shiroki R, Poindexter NJ, Woodle ES, Hussain MS, Moha-

nakumar T, Scharp DW. Human peripheral blood lymphocyte

reconstituted severe combined immunode®cient (hu-PBL-

SCID) mice. A model for human islet allograft rejection.

Transplantation 1994; 57: 1555±62.

9 Malkovska V, Cigel F, Storer BE. Human T cells in hu-PBL-

SCID mice proliferate in response to Daudi lymphoma and

confer anti-tumour immunity. Clin. Exp. Immunol. 1994; 96:

158±65.

10 Mosier DE, Gulizia RJ, Baird SM, Wilson DB, Spector DH,

Spector SA. Human immunode®ciency virus infection of hu-

man-PBL-SCID mice. Science 1991; 251: 791±4.

11 Barry TS, Jones DM, Richter CB, Haynes BF. Successful

engraftment of human postnatal thymus in severe combined

immune de®cient (SCID) mice: Di�erential engraftment of

thymic components with irradiation versus anti-asialo GM-1

immunosuppressive regimens. J. Exp. Med. 1991; 173: 167±

80.

12 Murphy WJ, Conlon KC, Sayers TJ et al. Engraftment and

activity of anti-CD3-activated human peripheral blood lym-

phocytes transferred into mice with severe combined immune

de®ciency. J. Immunol. 1993; 150: 3634±42.

13 Armstrong N, Cigel F, Borcherding W, Hong R, Malkovska V.

In vitro preactivated human T cells engraft in SCID mice and

migrate to murine lymphoid tissues. Clin. Exp. Immunol. 1992;

90: 476±82.

14 Murphy WJ, Durum SK, Longo DL. Human growth hormone

promotes engraftment of murine or human T cells in severe

combined immunode®cient mice. Proc. Natl Acad. Sci. USA

1992; 89: 4481±5.

15 Jeanpierre M. A rapid method for the puri®cation of DNA

from blood. Nucleic Acids Res. 1987; 15: 9611.

16 Murphy WJ, Bennett M, Anver MR, Baseler M, Longo DL.

Human-mouse lymphoid chimeras: Host-vs.-graft and graft-

vs.-host reactions. Eur. J. Immunol. 1992; 22: 1421±7.

T cell engraftment in Hu-PBL-SCID 525