Engraftment of peripheral blood mononuclear cells from human T-lymphotropic virus Type 1 carriers in NOD/SCID/γcnull (NOG) mice

Abstract

The transmission of human T-lymphotropic virus Type 1 (HTLV-1) occurs mainly via breast-feeding, sexual intercourse and blood transfusions. After transmission, the HTLV-1 infection is predominantly maintained by cell-to-cell infection and clonal expansion; however, the details have not yet been clarified. To investigate how HTLV-1 infected cells act in an environment without an effective immune reaction, peripheral blood mononuclear cells (PBMCs) from asymptomatic HTLV-1 carriers were inoculated into nonobese diabetic/severe combined immunodeficient (NOD/SCID)/γc^null (NOG) mice, which have immunological dysfunctions of T- and B-lymphocytes and NK cells. Human mononuclear cells including both CD4+ and CD8+ T cells were found to have infiltrated into various organs, including the liver, kidney, spleen and lung, when the mice were sacrificed 1 month after inoculation. The copy numbers of HTLV-1 provirus detected in the tissue-infiltrating human cells were much higher than those in the original PBMCs from the carriers. The expression of HTLV-1 mRNA was demonstrated in the tissue-infiltrating cells by reverse transcriptase-polymerase chain reaction. Inverse-long polymerase chain reaction showed that the pattern of HTLV-1 proviral integration was different from that of the original carrier and that it varied among NOG mice inoculated with PBMCs from the same carrier. These results suggest the selective proliferation of particular clones of HTLV-1 infected cells in NOG mice. Alternatively, transmission and new integration of HTLV-1 from infected cells to noninfected cells might have occurred in an environment without an effective immune reaction. The NOG mouse is considered a good animal model for the patho-physiological study of HTLV-1 infection with immunodeficiency. © 2007 Wiley-Liss, Inc.

Human T-cell leukemia virus Type 1 (HTLV-1), the first discovered disease-causing human retrovirus to be isolated, is the etiologic agent of adult T-cell leukemia/lymphoma (ATL) and a progressive demyelinating disease also known as HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP).1, 2, 3 ATL occurs in 2–4% of HTLV-1 carriers after a long latent period, suggesting that additional factors participate in the development of ATL.4 The infectivity of the free virus is quite low, and it is thought that cell-to-cell infection via breast-feeding, sexual intercourse and blood transfusions is the major route of transmission.5, 6 It has been postulated that clonal proliferation of HTLV-1 infected cells likely is responsible for maintaining the proviral load level in a carrier.7, 8

Over the past 25 years, a variety of animal models of HTLV-1 infection have provided critical information about viral and host factors in ATL.9 The virus consistently infects rabbits, some nonhuman primates, and, to a lesser extent, rats. The squirrel monkey has also been successfully infected with HTLV-1. Mice and rats, particularly immunodeficient strains, are useful models in the assessment of tumor outgrowth and therapeutic invention strategies against lymphoma. Genetically altered mice, including both transgenic and knockout mice, offer important models for the testing of the role of viral and host factors in the development of ATL. At present, the nonobese diabetic (NOD)/severe combined immunodeficient (SCID) mice, such as NOD/LtSz-scid and NOD/Shi-scid mice are considered appropriate models for the analysis of human stem cells.10, 11, 12, 13 To eliminate NK cell activity, NOD/Shi-scid mice and the β2-microglobulin-deficient NOD/LtSz-scid (NOD/SCID/β2m^null) mice have also been developed.12, 13, 14, 15

A newly developed NOD/SCID/γc^null (NOG) mouse, which is double homozygous for the severe combined immunodeficiency mutation and interleukin-2Rγ (IL-2Rγ) allelic mutation (γc^null), was generated by 8 backcross matings of 57BL/6J-γc^null mice and NOD/Shi-scid mice. This mouse shows extremely high efficacy for the engraftment of human hematopoietic cells due to the immunological dysfunction of the multiple cytokine production capability, in addition to the functional incompetence of T, B and NK cells.16, 17, 18 When comparing NOD/SCID/γc^null mice, NOD/SCID/β2m^null mice and NOD/Shi-scid mice injected with anti-NK cell antibody, NOD/SCID/γc^null mice were reported to be the most efficient among 3 types for the engraftment of human hematopoietic cells.16 Recently, it was reported that both acute- and smoldering-type ATL cells engrafted efficiently in newborn NOD/SCID/β2m^null mice.19 Moreover, uninfected human peripheral blood mononuclear cells (PBMCs) and HTLV-1-producing cell line MT-2 have been inoculated into NOG mice to study the events in the early stages of HTLV-1 infection.20

To examine how HTLV-1 infected human cells behave in an environment without an effective immune system, we investigated the engraftment and proliferation of PBMCs from HTLV-1 carriers inoculated into NOG mice. The phenotypes of mononuclear cells which infiltrated the various organs, proviral DNA loads of HTLV-1 and the expression of HTLV-1 mRNA were examined. The pattern of HTLV-1 proviral integration was compared between the NOG mice and original HTLV-1 carrier. It was shown that the NOG mouse is a good animal model for the patho-physiological study of HTLV-1 infection.

ATL, adult T-cell leukemia/lymphoma; FACS, fluorescence-activated cell sorter; HAM/TSP, HTLV-1-associated myelopathy/tropical spastic paraparesis; HTLV-1, human T-lymphotropic virus type 1; IL-PCR, inverse long-polymerase chain reaction; IL-2R, interleukin-2 receptor; NOD/SCID/γc^null (NOG), nonobese diabetic/severe combined immunodeficient; PBMCs, peripheral blood mononuclear cells; RT-PCR, reverse transcriptase-polymerase chain reaction; WBC, white blood cell.

Materials and methods

Subjects

Four HTLV-1 carriers and 2 healthy volunteers were recruited into this study after obtaining written informed consent. The study protocol was approved by the University of Miyazaki Institutional Review Board. The white blood cell counts of the HTLV-1-infected carriers were within the reference value without any abnormal cells. The geographies of the subjects are summarized in Table I. PBMCs were purified using HISTOPAQUE®-1077 density gradient centrifugation (10771-500ML, Sigma-Aldrich, St. Louis).

Table I. Geographies of Subjects and Inoculated NOG Mice

Subjects	Age (years old)	Sex	WBC (/μl)	Lymphocytes (%)	Inoculated mice
WBC, white blood cell.
Carrier 1	72	Female	4,500	17	NOG C1
Carrier 2	72	Male	7,000	61	NOG C2
Carrier 3	52	Female	3,700	44	NOG C3
Carrier 4	37	Female	6,700	58	NOG C4
Healthy volunteer 1	24	Male	not tested	not tested	NOG V1
Healthy volunteer 2	50	Male	not tested	not tested	NOG V2

Inoculation of human PBMCs from HTLV-1 carriers and healthy volunteers into the NOG mice

NOG mice were purchased from the Central Institute of Experimental Animals (Kawasaki, Japan). All mice were bred and maintained under specific-pathogen-free conditions in the Department of Bio-resources, Division of Biotechnology, Frontier Science Research Center, University of Miyazaki (Miyazaki, Japan). Female NOG mice, which were 8- to 10-weeks old, were used for the experiments. NOG C1, C2, C3 and NOG V1, V2 were inoculated with PBMCs from Carriers 1, 2, 3 and HTLV-1 negative Volunteers 1, 2, respectively. PBMCs of Carrier 4 were obtained twice with 1-month interval. PBMCs from the first bleed were inoculated into NOG C4-1. PBMCs from the second bleed of Carrier 4 were inoculated into 2 NOG mice (NOG C4-2 and C4-3).

A total of 5 × 10⁶ human PBMCs were injected intraperitoneally and intravenously to each mouse. The mice were sacrificed 4 weeks after inoculation. Their peripheral blood and organs (liver, kidney, spleen and lung) were supplied for the experiments. Mononuclear cells were isolated by HISTOPAQUE-1077 density gradient centrifugation. All samples obtained for molecular studies were preserved at −80°C until use. The experimental protocol was approved by the University of Miyazaki ethics review committee for animal experimentation.

Histopathological and cytological analyses

Tissues samples from harvested organs were fixed in 10% neutral-buffered formalin (Sigma-Aldrich, St Louis) and embedded in paraffin. Tissue sections (4-μm thick) were stained with hematoxylin-eosin (HE) for morphological studies, and serial sections were examined by immunohistochemistry using EnVision+kits (DakoCytomation, Kyoto, Japan) and primary antibodies against human CD3 (DakoCytomation), human CD4 (Nichirei Corp., Tokyo, Japan), human CD8 (DakoCytomation) and human CD79a (DakoCytomation). As the negative control for immunostaining, normal mouse IgG or rabbit serum were used instead of the primary antibodies.

Mononuclear cells obtained from the liver, spleen and lung of the NOG mice were washed 2 times with phosphate-buffered saline (PBS) and incubated for 30 min with antibodies, as described below. Fluorescein isothiocyanate (FITC)-labeled anti-human CD4 and phycoerythrin (PE)-labeled anti-human CD8 monoclonal antibodies (Becton Dickinson, San Jose, CA), or control isotype-matched murine immunoglobulin G (IgG) were used for immunocytological analysis using a FACS Calibur (Becton Dickinson).

Quantification of human cells and HTLV-1 provirus

The chromosomal DNA was isolated from the tissues and cells by sodium dodecyl sulfate (SDS) -protease K digestion at 56°C, followed by phenol-choroform extraction and ethanol precipitation.21

HTLV-1 proviral copy numbers (i.e., proviral DNA load) were measured by real-time PCR using LightCycler® DX 400 (Roche Diagnostics, Germany), as described previously.22 Quantitative real-time PCR was performed for human albumin DNA, mouse GAPDH DNA, and HTLV-1 provirus. The primers and the probe for the human albumin gene were as follows (Sigma-Aldrich Japan, Tokyo, Japan): the forward primer Alb-S (5′-GCTG TCATCTCTTGTGGGCTGT-3′), the reverse primer Alb-AS (5′-AAACTCATGGGAGCTGCTGGT-3′), and the FAM-labeled albumin TaqMan probe (5′-FAM-CCTGTCATGCCCACACAAAT CTCTCC-TAMRA-3′).23 TaqMan Gene Expression Assays (Mm 99999915_g1, Applied Biosystems Japan, Tokyo, Japan) were used for the primers and the probe for the mouse GAPDH gene. The primers and the probe for the pol region of HTLV-1 were as follows: the forward primer (5′-AACCAATTCATTCAAACATCTGACC-3′: positions 3735-3759), the reverse primer (5′-GCTTTCACAGGAGCCAATGG-3′: positions 3877-3858), and the FAM-labeled probe (5′-FAM-TGTTCCTATCTTACTCCACCACAGTCACCGA-TAMRA-3′: positions 3767-3797). Quantitative PCR was performed in a duplicate manner. No samples from NOGC-1, no tissue samples from NOGC-2 except peripheral blood, and no peripheral blood samples from NOG C4-2/C4-3 were tested for HTLV-1 proviral DNA loads due to an insufficient specimen amount (Fig. 2).

Detection of HTLV-1 mRNA by reverse transcriptase-polymerase chain reaction

Total cellular RNA was extracted from organ samples using TRIzol® Reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions.24 About 5 μg of total RNA was reverse-transcribed using M-MCV Reverse Transcriptase (Invitrogen) using random primers. The following primers were used for RT-PCR. PX-1 (5′-AGGGTTTGGACAGAGTCTTC-3′: positions 7335-7354) and PX-2 (5′-AAGGACCTTGAGGGTCTTAG-3′: positions 7590-7571) for tax/rex mRNA; Pol-F1 (5′-GGG CCCCCTGACTTGTCCA-3′: positions 2805-2823) and Pol-R1 (5′-TTGCCGAATGGGCTGCAGGA-3′: positions 3050-3031) for pol mRNA; GAPDH-118F (5′-GTCGGAGTCAACGGATTTG GTCG-3′: positions 118-140) and GAPDH-640R (5′-CATGGACTGTGGTCATGAGTCC-3′: positions 640-618) for GAPDH mRNA(Sigma-Aldrich Japan). The PCR products were electrophoresed on 1.2% agarose gel, and visualized by ethidium bromide staining. The RNA sample from an HTLV-1 positive cell line, HUT102, was used as a positive control at a 1:100 dilution.

Analysis of the pattern of HTLV-1 proviral integration by inverse long PCR

To analyze HTLV-1 proviral integration, IL-PCR was used to amplify the genomic DNA adjacent to the integration sites of the HTLV-1 provirus, as described previously.7 Briefly, the genomic DNA was digested with EcoRI, self-ligated by T4 ligase, and then digested with MluI. Long PCR amplification of the resultant DNA was performed using the rTth DNA polymerase (Applied Biosystems, Foster City, CA). The primers used in this analysis were primer 1 in the U5 region of the long-terminal repeat (LTR) (5′-TGCCTGACCCTGCTTGCTCAACTCTACGTCTTTG-3′: positions 556-589) and primer 2 in the U3 region of the LTR (5′-AGT CTGGGCCCTGACCTTTTCAGACTTCTGTTTC-3′: positions 8345-8378). The PCR products were electrophoresed on 1.2% agarose gel, and visualized by ethidium bromide staining. All assays were performed in triplicate.

Subcloning of the amplified fragments of IL-PCR from the liver of NOG C4-2 were subjected to sequencing assay according to the protocol of the Big Dye Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) using ABI Prism 310 DNA Sequencer (Applied Biosystems). Based on DNA sequence analysis of the flanking region of the provirus, it was possible to identify the genomic DNA sequence adjacent to the integration sites of HTLV-1 provirus.25 Integration site-specific primers were designed based on their sequences. One-tenth of the volume of the primary IL-PCR products was subjected to an additional amplification. The forward primer for this PCR was HTLV9013F (5′-AGCCCATCCTATAGCACTCTC-3′; positions 9014-9034). The reverse primers for the integration site-specific PCR were as follows: 5′-GTGATGTGTGCGTTCAACTCAC-3′ for clone1, 5′-TGAAGCCTGCCAGTGGATATTC-3′ for clone2, 5′-GTTTGAA ACCCGGCACATTCCT-3′ for clone3, 5′-TGAACATGAATA GTCATCTGCAAGCA-3′ for clone4, 5′-ATCGGTCTCTGACC AATCCTGA-3′ for clone5.

Results

Efficient engraftment of human PBMCs in NOG mice

All mice grew normally without piloerection or weight loss for a month. No tumors or lymph node swelling was found when the mice were sacrificed. Infiltration of mononuclear cells was seen in various organs of all of the tested NOG mice on microscopic examination. Eight mice in total were used and engraftment of human PBMCs was confirmed in all of these mice. Representatively, HE and immunohistochemical staining in lungs derived from the NOG mice inoculated with PBMCs from carriers and healthy volunteers are shown in Figure 1. The infiltration of mononuclear cells was also observed in liver, kidney, lung and spleen in all mice inoculated with PBMCs from carriers and healthy volunteers (Fig. 1, Table II). The atypia of mononuclear cells in the NOG mice inoculated with PBMCs from HTLV-1 carriers seemed to be more severe than that in the NOG mice inoculated with PBMCs from healthy volunteers. The degree of infiltration of mononuclear cells in the tissues tended to be more severe in the NOG mice, which were inoculated with PBMCs from HTLV-1 carriers, than that in the NOG mice with PBMCs from healthy volunteers.

Figure 1. Immunohistological analysis of the lungs of NOG mice inoculated with peripheral blood mononuclear cells (PBMCs) from HTLV-1 carriers (lane a) and noninfected healthy volunteers (lane b). Magnification ×400. HE, hematoxylin-eosin staining.

Table II. Infiltration of Mononuclear Cells in Various Organs of NOG Mice

Mice	Tissues of NOG mice
	Liver	kidney	spleen	lung
+ + +: >20%, + +: 10∼20%, +: <10%, −: no infiltration.
NOG C1	+	+ ∼ + +	+ +	+ ∼ + +
NOG C2	+ ∼ + +	+	+ +	+ + ∼ + + +
NOG C3	+	+	+ +	+
NOG C4-1	+ + +	+ +	+ + +	+ + +
NOG V1	+	+	+	+
NOG V2	+	+	+ +	+

Immunohistochemical and cytological analysis of mononuclear cells infiltrated to organs of NOG mouse

Immunohistochemical staining using anti-CD3, anti-CD4, anti-CD8 and anti-CD79a antibodies was conducted on the tissue specimens of the lungs of the NOG mice inoculated with PBMCs from healthy volunteers and carriers (Fig. 1). T cell population was evident in the NOG mice inoculated with PBMCs from both the healthy volunteers and carriers. B cells stained with anti-CD79a antibody were hardly seen, especially in case of the mice inoculated with PBMCs from HTLV-1 carriers. Infiltration of both CD4+ and CD8+ cells was observed, however, CD4+ cell infiltration seemed to be dominant.

To confirm the results of histoimmunochemical analysis, cell surface phenotypes of the mononuclear cells isolated from the organs of the NOG mice were examined by flow cytometry using a FACS Calibur. This analysis showed both CD4+ and CD8+ T cells in the liver, spleen and lung. Furthermore, the percentage of CD4+ T cells (44.0–81.0%) was higher than that of CD8+ cells (12.9–36.6%). These results support the findings of immunohistochemical staining of lung specimens.

Proviral DNA loads of HTLV-1 in NOG mice

To quantify the number of human cells, mouse cells, and HTLV-1 provirus present in the tissue of NOG mice, human albumin DNA, mouse GAPDH DNA and HTLV-1 provirus were quantified using real-time PCR. The percentage of human cells in the tissues varied from 13% (kidney in NOG C3) to 92% (spleen in NOG C4-1). More than 50% of cells in the peripheral blood, spleen and lung in all of the NOG mice were considered to have been derived from human. The number of inoculated human cells in each mouse was only 5 million and much fewer than the number of the infiltrated cells in NOG mice. These data suggest that the inoculated human cells proliferated in NOG mice and infiltrated into various organs.

HTLV-1 proviral DNA copy numbers in 10⁵ human cells in each tissue of NOG mice are shown in Figure 2. The proviral loads per 10⁵ human cells were much higher in all NOG mice than those in the PBMCs of the derived carriers. The number of HTLV-1 proviral DNA copies was almost equal to or greater than 10⁵ in the peripheral blood of NOG C3, and several organs of NOG C4-1 and C4-2. These results suggest that almost all the human cells in these organs were infected with HTLV-1. Alternatively, more than 1 copy of HTLV-1 provirus may exist in 1 human cell infiltrated into these NOG mice organs.

Figure 2. HTLV-1 proviral DNA loads (HTLV-1 copy numbers per 10⁵ human cells) in the various organs of NOG C2, C3 and C4 (-1,-2 and -3) and in the peripheral blood mononuclear cells (PBMCs) of Carrier 2, 3, and 4. PBMCs of Carrier 4 were obtained twice with 1-month interval. PBMCs from the first bleed were inoculated into NOG C4-1. PBMCs from the second bleed of Carrier 4 were inoculated into 2 NOG mice (NOG C4-2 and C4-3).

Detection of HTLV-1 mRNA in NOG mice by RT-PCR

It is known that HTLV-1 tax plays an important role in the proliferation of infected cells.26, 27, 28 Therefore, RT-PCR analysis was performed to detect HTLV-1 mRNA in the organs of NOG C3 and C4-1. The expression of both tax/rex and pol mRNA was detectable by RT-PCR using 30 or 35 cycles of amplification (Figs. 3a and 3b). These data suggest that the proliferation of HTLV-1-infected cells was associated with a certain level of expression of HTLV-1 mRNA. However, the levels of mRNA expression did not seem high, their being equivalent to that of an RNA sample of HUT102 that was diluted 100 times and used as a positive control.

Figure 3. The expression of tax/rex mRNA and pol mRNA detected by reverse transcription polymerase chain reaction (RT-PCR). (a) tax/rex mRNA, (b) pol mRNA. RT-PCR (30, 35 cycles) with and without reverse transcription reaction were assayed in NOG C3 (liver, kidney), NOG C4-1 (liver, kidney) and NOG V2 (liver). Amplification of GAPDH in nonsaturating conditions was used to normalize RNA samples. The RNA sample from a cell line, HUT102 was used at a 1:100 dilution as a positive control.

Detection of HTLV-1 proviral integration pattern by IL-PCR

The pattern of HTLV-1 proviral integration in the various organs did not show any significant difference in NOG C4-1. However the pattern did differ from that in peripheral blood of Carrier 4 (Fig. 4a). Then, the patterns of HTLV-1 proviral integration in the livers of NOG C4-1, -2 and -3, which were inoculated with PBMCs derived from same Carrier 4, were compared. The pattern of HTLV-1 proviral integration in NOG C4-1, -2 and -3 was not only different from that of Carrier 4, but also different among 3 mice (Fig. 4b). Therefore, the selective proliferation of certain clones of HTLV-1 infected cells is considered to have occurred in each of the NOG mice after the inoculation of PBMCs from Carrier 4. Alternatively, a new infection of HTLV-1 might have occurred from infected cells to noninfected cells in NOG mice, which resulted in the different pattern of HTLV-1 proviral integration.

Figure 4. Detection of HTLV-1 proviral integration pattern in Carrier 4 and NOG C4 (-1,-2 and -3) by inverse-long polymerase chain reaction (IL-PCR). (a) Triplicate analysis of the peripheral mononuclear cells (PBMCs) of Carrier 4 and various organs of NOG C4-1 (liver, kidney, spleen and lung) by IL-PCR. HUT102 cells were used as the HTLV-1 positive control. MWM: molecular weight marker. (b) Triplicate analysis of the PBMCs of Carrier 4 and the liver from the NOG mice (C4-1 and C4-2, C4-3) by IL-PCR. N: PBMCs from a healthy volunteer as an HTLV-1 negative control. P: HUT102 cells were used as the HTLV-1 positive control. MWM: molecular weight marker.

To examine these possibilities, IL-PCR products from NOG C4-2 were subcloned and the genomic sequences of adjacent sites of provirus were identified. Five clones were obtained and their genomic sequences were analyzed. The HTLV-1 proviruses were confirmed to be integrated into the human chromosome only and not into the mouse chromosome (Table III). Then, we designed the primers, which were specific for the sequence of adjacent human DNA. Using these primers and a primer for the LTR of HTLV-1, integration site-specific PCR was performed to examine whether cells with these 5 proviral integration sites exist in the PBMCs of Carrier 4 and in the livers of NOG C4-1, and -3 (Fig. 5). Clones 1 and 5 were confirmed to exist in Carrier 4. Clone 4 was detected in both NOG C4-2 and C4-3, but not in the others. As it is difficult to assume that HTLV-1 integrated coincidently into the same site of the genome of the human cells in NOG C4-2 and C4-3 by new infection, it seems likely that Clone 4 was derived from Carrier 4. The number of HTLV-1 infected cells belonging to Clone 4 may be small in the PBMCs of Carrier 4, and the sensitivity of PCR may not have been enough to detect them. On the other hand, Clones 2 and 3 were detected in NOG C4-2 only. Accordingly, there is a possibility that the proviral integrations of Clones 2 and 3 were the result of cell-to-cell infection occurring in NOG C4-2.

Figure 5. Integration site-specific polymerase chain reaction (PCR) in the peripheral mononuclear cells (PBMCs) of Carrier 4 and in the liver of NOG C4 (-1,-2 and -3). Clones 1-5: PCR was performed using integration site-specific reverse primer 1-5 for each clone, respectively. C: PBMCs from Carrier 4; 1, 2, 3: Genomic DNA from NOG C4-1, C4-2, C4-3; N: PBMCs from healthy volunteer; P: Subcloned plasmid (clone 1-5) of IL-PCR products of NOG C4-2.

Table III. Integration Sites of HTLV-1 Provirus of Clones of HTLV-1-Infected Cells in NOG C4-2

Clone	Integration sites of chromosome		Gene
1	7	q11.21	–
2	16		–
3	8	q12	–
4	4	q22.3	–
5	12	q13.3	PTGES3 (unactive progesterone receptor 23kD)

Discussion

There have been several reports of the successful engraftments of HTLV-1 transformed cell lines and ATL cells in NOD/SCID mice.10, 11, 12, 13, 14, 15, 19 However, there have been no reports of the engraftment of PBMCs from asymptomatic HTLV-1carriers in NOG mice. We inoculated PBMCs from asymptomatic carriers into NOG mice and analyzed the HTLV-1 infected cells 4 weeks after inoculation. Human cells, mainly T-lymphocytes were found to proliferate and infiltrate into various organs and the peripheral blood of NOG mice in the present study. The PBMCs from HTLV-1 noninfected healthy volunteers were also found to be engrafted in the NOG mice. The infiltration of CD4 positive T-cells with atypia tended toward dominance in cases of NOG mice inoculated with PBMCs from HTLV-1 carriers, although the number of subjects was small and caution is necessary in interpretation of these results. In addition, small numbers of B cells were observed in the NOG mice inoculated from healthy volunteers. Engraftment of not only T cells but also B cells was reported when NOG mice were inoculated with mobilized human hemopoietic stem cells.29, 30 However, B-cells were not found in the NOG mice inoculated with PBMCs from HTLV-1 carriers. There is a possibility that HTLV-1 infected T-cells had a growth advantage and proliferated more than HTLV-1 noninfected B-cells in the case of the NOG mice inoculated with PBMCs from HTLV-1 carriers. HTLV-1 provirus was found only in the human cells, and the proviral DNA loads of infiltrated cells were much higher than those in the PBMCs of original HTLV-1 carriers, whose PBMCs was inoculated into NOG mice. These results also suggest that HTLV-1 infected T-lymphocytes proliferated and infiltrated into various organs of NOG mice.

Miyazato et al. reported that human uninfected PBMCs were inoculated into NOG mice and those PBMCs were infected with HTLV-1 by the following inoculation of HTLV-1 producing cells (MT-2) into NOG mice.20 Transcripts of the tax gene were undetectable in 2 of 3 mice, while the remaining mouse showed a low level of expression. There was an increase of tax gene transcription after 24 hr of culture in vitro. In the present study, HTLV-1 tax/rex and pol mRNA were detected in vivo by RT-PCR in the NOG mice inoculated with HTLV-1 carriers, although the level was not very high. The difference in the results between these 2 studies may be due to the difference between the viral expression pattern of newly infected cells (in Miyazato's study) and that of long-term infected cells (in our study). Moreover, the HTLV-1 proviral DNA loads in spleen in the Miyazato study were low and less than 1% of cells were infected. In contrast, the HTLV-1 proviral DNA loads in the livers and kidneys tested for mRNA in the present study were high, and 40–80% of cells were infected. Therefore, the HTLV-1 mRNA signals may have been easier to detect in the present study. Alternatively, there is a possibility that greater expression of tax mRNA in the present study may have contributed to the increased number of HTLV-1 infected cells in the infiltrated organs. However, the viral signal was not detected in the serum samples of NOG mice (data not shown). Therefore, viral particles may not have been produced in the NOG mice, although viral expression in the cells was detected by RT-PCR. This situation is similar to the condition of HTLV-1 asymtomatic carriers, because the HTLV-1 viral expression can be detected in PBMC by RT-PCR but the virus particles themselves were not detectable.31, 32

The pattern of HTLV-1 proviral integration was shown not to differ among the various organs in NOG mice. However, this pattern was differed greatly from that of the original asymptomatic carrier. Moreover, when 3 different NOG mice were inoculated with PBMCs from same carrier, their HTLV-1 proviral integration patterns were found to be different. These data raise the question of whether this difference is due to the expansion of a certain clone of HTLV-1 infected cells, which exist in the original carrier, or if it is due to a new infection from cell to cell in NOG mice. The HTLV-1 proviral DNA loads of several organs were shown to have more than 1 copy per human cell. This result suggests the latter possibility. When we tested whether 5 integration sites of the HTLV-1 provirus in NOG C4-2 mouse existed in the PBMCs of original carrier and 2 other mice (C4-1 and C4-3) inoculated with PBMCs from the same carrier, 3 were present in the carrier or in the other mice. However, 2 integration sites of HTLV-1 provirus were not found. Therefore, it is possible that a new infection from cell to cell occurred in NOG mice. If a new infection of HTLV-1 occurred, there may be a new integration of HTLV-1 provirus into the genome of the cells, which already had an integration of HTLV-1 provirus. Therefore, it is possible that there were several clones with more than 2 copies of proviral integration in NOG mice. If this is true, the pattern of HTLV-1 proviral integration detected by IL-PCR does not always indicate the accurate clonality of HTLV-1 infected cells in the NOG mice in this case. In addition, because IL-PCR of NOG C4-1, -2 and -3 showed several very strong bands, the expansion of particular clones of HTLV-1 infected cells may have occurred at the same time. Further studies are necessary to determine whether a new infection of HTLV-1 among human cells actually occurred in the NOG mice when they are inoculated with PBMCs from HTLV-1 carriers.

In conclusion, HTLV-1 infected cells from asymptomatic carriers were shown to proliferate and infiltrate various organs of NOG mice. The selective proliferation of particular clones of HTLV-1 infected cells and/or the transmission of HTLV-1 from infected cells to noninfected cells may have occurred in an environment without an effective immune reaction. Recently, it was reported that immunosuppressive treatment for the prevention of graft rejection after living-donor liver transplantation may have induce the development of ATL in HTLV-1 carriers.33 Therefore, it is important to know the patho-physiology of HTLV-1 infected cells under immunosuppression. NOG mouse is considered to be a good animal model for understanding it.

Acknowledgements

We thank Ms. I. Kobayashi, Ms. K. Yuji, and Dr. H. Nomura (University of Miyazaki) for their technical assistance.