WO1995009235A9 - Immunodeficient mouse models of pathogenesis of human disease and efficacy and toxicity of disease treatments - Google Patents

Abstract

BTCD-hu2 chimeric mice for assaying the pathogenesis of human disease and efficacy and toxicity of disease treatments have been developed by implantation of human fetal thymus and human fetal liver tissue under the kidney capsule of T- and B-cell deficient mice, by engraftment of human fetal bone marrow cells into T- and B-cell deficient mice and combinations of both. The resulting chimeric mice contain human T-cells, human monocytes or combinations thereof in sufficient quantity in the mouse peripheral lymphoid compartment to support HIV-1 infection following intraperitoneal inoculation of HIV-1 into the mouse.

Description

IMMUNODEFICIENT MOUSE MODELS OP PATHOGENESIS OP HUMAN DISEASE AND EFFICACY AND TOXICITY OF DISEASE TREATMENTS

FIELD OF THE INVENTION

This invention relates to immunodeficient mice transplanted with human tissue and/or engrafted with human cells and methods of producing and utilizing the transplanted animals as models for the pathogenesis of human disease, such as AIDS.

This invention is directed to the development of a mouse chimeric for both human and mouse hematopoietic systems. The chimeric mouse of this invention has human fetal tissue and/or cells transplanted or engrafted therein which results in the re-population of the mouse peripheral lymphoid compartment with a sufficient number of human T-cells, human monocytes or combination thereof to support HIV infection following intraperitoneal inoculation of HIV-1 into the mouse. The chimeric mouse of this invention is capable of tolerating human tissue, which allows for the maturation and function of implanted human tissues and/or cells.

DISCUSSION OF THE PRIOR ART

The development of therapeutic agents for human use is extremely expensive and requires long periods of time. For many human diseases, the therapeutic drug and vaccine development processes have been greatly facilitated by the development of animal models that provide a window into the human pathophysiological disease processes as well as normal human physiological processes. However, for many other human diseases such as Acquired Immuno Deficiency Syndrome (AIDS) , there are no available small animal models that mimic the human pathophysiological disease processes.

The causative agent of AIDS, human immunodeficiency virus (HIV) , can only be propagated in cells from humans and certain non-human primates. In vivo , HIV only infects humans and chimpanzees. Humans of course, cannot be studied in a systematic fashion and chimpanzees are on the endangered species list. It is therefore, extremely difficult to study the pathogenesis of disease following infection. An animal model for AIDS would provide an opportunity to more closely study the human immune system and human hematopoietic system and provide insight into the post-infection events associated with HIV infection.

The human immune and hematopoietic systems have not been thoroughly dissected and defined primarily due to the lack of in vivo assays for normal immune system and hematopoietic cells. There have been numerous attempts over the past 30 years to transplant mice with normal human tissue. Direct transplants were generally unsuccessful. However, new methods of transplanting human tissue into immune deficient mice have been reported.

Immune deficient mice include scid, bnx, Rag-1 and Rag-2 mice. Homozygous scid mice (scid/ scid) were first identified fortuitously by Bosma in breeding experiments with CB-17 mice to develop immunoglobulin heavy chain mouse strains. Bosma, et al. (1983), Nature 301:527-530. Over the years a large number of experiments have indicated that the defect in immunoglobulin production is inherited and due to a mutation in a gene carried on mouse chromosome 16. Bosma, et al. (1991), Ann. Rev. Immunol. 9:323-350. The scid gene plays an important role in the lymphoid differentiation program and mutations in the gene prevent gene rearrangement and the subsequent production of mature T and B-cells. (Malynn, et al. (1988), Cell 54:453-460). All other hematopoietic lineages including natural killer (NK) cells and macrophages are normal. The mice have a very small thymus containing a few immature thymocytes. The action of this gene product is not restricted to normal lymphoid development since scid mice have a generalized radiation repair defect that renders the animals at least two times more sensitive to the effects of γ- radiation. Fulop, et al. (1990), Nature 347:479-482;

Biederman, et al. (1991), Proc. Natl. Acad. Sci. USA 88:1394- 1397. The radiation repair defect is manifested in all hematopoietic and non-hematopoietic lineages tested. In spite of these defects, homozygous scid mice (scid/ scid) have a normal lifespan and in all other respects appear normal although they are highly susceptible to infection. While scid mice are severely immune-deficient for both T and B-cells, they still possess a certain amount of natural immunity and non-lymphoid resistance systems such as macrophages. These activities are highly dependent on cytokines like interleukin- 2 (IL-2) and interferon (IFN) which are easily stimulated by bacterial or viral infection. Because their hematopoietic microenvironment is normal these mice are easily reconstituted with syngeneic cells with sublethal radiation conditioning. Fulop, et al. (1986), J. Immunol. 136:4438-4443. As a result they have provided important models for studying many aspects of lymphoid differentiation. Bosma, et al. Ibid. .

The immunodeficiency of homozygous Rag-1 and Rag-2 mice is similar to that of scid mice in that the homozygous Rag-1 and Rag-2 mutations interrupt genes involved in both T- and B- cell development and such mutations result in interference with maturation of both B- and T-cells. Momberts, et al., (1992), Cell, 68:869-877, produced mice homozygous for a mutation in the Rag-1 gene, which is thought to play a role in the regulation or catalysis of V(D)J recombination. The Rag-1 deficient mice do not have mature B- or T-lymphocytes, presumably as a result of loss of a common recombinase that is active in precursors of both B- and T-cells. Alt, et al., (1986), Immunol. Rev. 89:5-30.

Mice homozygous for the Rag-2 mutation fail to generate mature T- or B-lymphocytes as a result of a complete lack of ability to initiate the V(D)J recombination process, leading to a severe combined immune deficient phenotype. Shinkai, et al. (1992), Cell j£8:855-867.

The etiology of the congenital immunodeficiency of bnx mice differs from that of scid and Rag-1 and Rag-2 mice. The bnx strain was derived by crossing congenitally athymic nude mice with NK cell-deficient beige mice and LAK cell-deficient xid mice. Sharabi, et al., (1990), J. Exp. Med., 172:195. bnx mice also have a severe Ig deficiency due to the combined effect of a marked decrease in the numbers of helper T-cells and in the number of T-cell-independent B-cells. Dick, Jt (1991), Cancer Cells, 3.:39. One approach to the engraftment of human cells into immuno-deficient mice was first taken by McCune, et al. who surgically implanted human fetal thymus and lymph node under the renal capsule of scid mice and then injected fetal liver cells intravenously (IV). McCune, et al. (1988), Science 241:1632-1639. Two principle reasons for choosing fetal tissues were the concern that the production of a functional human immune system would lead to serious graft versus host disease (GVHD) and the idea that a human stromal environment must be provided because the murine environment would not support human lymphoid development. Indeed, human T-cells could initially be detected at a level of about 10% of the mononuclear cells in the peripheral circulation beginning 4 weeks after IV injection of fetal liver cells (10⁷ cells) . This level was maintained for 6 weeks after which no human cells could be detected. The rapid increase and then decline of T-cells implied that engraftment was occurring as a wave of T-cell differentiation. In most initial experiments the human fetal thymus was implanted several weeks prior to the injection of fetal liver cells, presumably to allow for vascularization. After injection of fetal liver the implanted thymus grew in size and developed many aspects of the normal architecture of a normal age-matched fetal thymus. By using different HLA typed donors for the thymus and fetal liver, evidence was presented suggesting that the fetal liver cells homed to the thymus, differentiated, and passed into the peripheral circulation.

In an effort to extend the durability of the transplant, more recent experiments have suggested that renal implantation of an intact piece of fetal liver adjacent to the fetal thymus improves the longevity of the xenograft by permitting the development of a novel human microenvironment. Namikawa, et al. (1990), J. Exp. Med. 172:1055-1063; Krowka, et al. (1991), J. Immunol. 146:3751-3756. In this system, only human T-cells were detected in the blood of the animals. After 1 month 1% of mice had detectable (0.2%) human cells. This proportion rose to a maximum of 50% positive animals containing 0.7 to 2.1% human cells from 4 months to 1 year. Krowka, et al., Ibid. . This suggests that there is a great deal of variability in the implantation methods or the cell viability from different donors. Phenotypically and functionally these T-cells were similar to human cells. Vanderkerckhove, et al. (1991), J. Immunol. 146:4173-4179. Low numbers of myeloid progenitors and mature cells were maintained within the implanted tissues, although none were found in the peripheral circulation or elsewhere.

In general SCID-hu constructs are made by surgical implantation of interactive human organ systems into the immunodeficient CB-17 scid mouse. One goal of any particular SCID-hu construct has been to provide an animal system that allows direct observation of normal and abnormal functions within the transplanted human tissue as a model of the same functions within intact human organs. However, in none of the previously reported SCID-hu mice transplanted with human liver or thymus tissue have human T-cells been maintained in significant numbers in the peripheral blood of the animals or in mouse lymphoid and hematopoietic organs, such as the lymph nodes, spleen or bone marrow. Consequently, previous SCID-hu mice have not been useful for peripheral blood studies in such diseases as AIDS, for example.

Because the peripheral blood of previous SCID-hu mice containing human thymus implants contained only very low levels of human lymphocytes (a mean value of 0.7% reported by Krowka, et al., Jjid.) it has been necessary to directly inoculate HIV-1 into the transplanted human tissue, i.e. into the thymus implant in order to achieve infection of the transplanted human tissue. This problem was partially overcome by developing a system for intravenous infection of SCID-hu mice. Kaneshima, et al. (1991) , Proc. Natl. Acad.

Sci. USA 88:4523. Intravenous inoculation of HIV into SCID-hu mice transplanted with human thymus, lymph node and connective tissue, resulted in the selective infection of human lymph nodes, however, human thymus and connective tissue were positive for human -globulin only. Furthermore, lymph node infection required a large input inoculum of HIV (120,000 TCID₅₀) . Clearly, therefore, an animal model system allowing peripheral inoculation of HIV for example, and which results in detectable levels of virus infected cells in peripheral blood and various organs would be of extreme importance for studying the pathophysiologic events following virus infection. None of the animal models decribed so far satisfy all of these requirements. The ability to study the normal developmental program of human hematopoiesis and the biological consequences of aberrant proliferation and differentiation of hematopoietic cells would be greatly facilitated by the establishment of an animal model and in vivo assay for human stem cells.

Congenitally, immune deficient mice have been successfully engrafted with human bone marrow by infusion of human bone marrow following sub-lethal irradiation of the mice (Dick, et al. (1991), Immunol. Rev., 124:25-43; Ka el-Reid, et al. (1988), Science, 242:1706-1709). Post-transplant treatment with human cytokines significantly increased bone marrow engraftment of irradiated scid mice with human myeloid and erythroid progenitors, as well as engraftment with B-cells after infusion with human bone marrow cells (Lapidot, et al. (1992), Science, 255:1137-1141) . However, continuous treatment with human cytokines was required for successful engraftment. An alternative approach to developing an animal model of human hematopoiesis involved transplanting human fetal bone into SCID mice (McCune, et al., (1991), Immunol. Rev., 124:45-62) . This latter approach resulted in the multilineage differentiation of human progenitor cells in implanted human bone for up to twenty weeks post-engraftment (Kyoizumi, et al., (1992), Blood, 79:1704-1711). Although significant levels of human hematopoiesis were observed in the mouse bone marrow of cytokine-treated SCID-hu mice and in the human bone implanted in SCID-hu mice, only minimal peripheral engraftment with human myeloid and lymphoid cells was observed in mice engrafted with either of these techniques (Kamel-Reid, et al., Ibid; McCune, Ibid). Consequently, these animals are only marginally useful for studying human hematopoiesis.

Hence, a need exists for an animal which is significantly engrafted with human lymphoid and myeloid precursor cells so as to repopulate the lymphoid and hematopoietic organs of the animal with human lymphoid and precursor cells and which does not require extraneous human cytokines. Such an animal will allow researchers to assess the factors involved in human hematopoiesis and to determine the effects of bone marrow transplantation and the effects of various drugs on hematopoietic cells. OBJECTS OF THE INVENTION

The present invention provides a murine B-cell and T-cell deficient (BTCD) mouse chimeric for human B-cells and T-cells and having human T-cells and/or monocytes in its peripheral lymphoid compartment in sufficient quantity to enable HIV-1 infection, for example, following intraperitoneal inoculation of HIV-1 into the mouse.

This invention also provides a chimeric mouse having and supporting a functional and viable human stromal microenvironment in which human hematopoietic cell maturation occurs.

This invention also provides a chimeric mouse having a high proportion of T-cells, B-cells and/or monocytes of human origin in its peripheral blood, spleen and lymph nodes. This invention provides a chimeric mouse capable of expressing human cytokines.

This invention provides a method for constructing chimeric mice capable of being infected with HIV-1 via intraperitoneal inoculation of HIV-1. This invention also provides a method for assaying the in vivo dissemination of HIV-1.

This invention also provides a screening method for determining the efficacy of anti-HIV-1 drug, anti-HIV-1 immunotherapy or hematopoietic drug or treatment. This invention provides a screening method for determining the toxicity of anti-HIV-1 drug, anti-HIV-1 therapy, hematopoietic drug or therapy or the efficacy of bone marrow transplantation.

This invention also provides a method of assessing the induction of the expression of human cytokines during disease progression in a chimeric mouse and during hematopoeisis.

This invention also provides a method for amplifying human bone marrow cells for transplantation in a human.

This invention provides a model for studying gene therapy in human stem and/or progenitor cells.

This invention also provides a method for generating human monoclonal antibodies. SϋMMARY OF THE INVENTION

The present invention is accomplished in connection with one aspect thereof by the construction of a chimeric mouse (hereinafter referred to as BTCD-hu2) having in its peripheral lymphoid compartment a sufficient number of human T-cells, monocytes or combination thereof which enables HIV-1 infection of the mouse following intraperitoneal inoculation of HIV-1 into the mouse. In a preferred embodiment, the BTCD-hu2 mouse has a proportion of T-cells of human origin in its peripheral blood in the range of from at least 5% to about 10%. In another preferred embodiment, the BTCD-hu2 mouse has transplanted under each kidney capsule human fetal thymus tissue interspersed with human fetal liver tissue. In another preferred embodiment, the BTCD-hu2 mouse is a BTCD-huCombo mouse (as hereinafter defined) . In yet another preferred embodiment, the bone marrow of the BTCD-hu2 mouse contains a proportion of CD45+ cells in the range of from at least about 10% to about 40% and the mouse peripheral lymphoid compartment contains at least about 5% to about 40% CD45+ cells. In another preferred embodiment, the BTCD-hu2 mouse also has human fetal spleen, intestine and lung tissues implanted therein.

In another aspect of the invention there is provided a method of constructing BTCD-hu2 mice capable of being infected with HIV-1 via intraperitoneal inoculation of HIV-1. The method includes implanting alternating pieces of human fetal liver and human fetal thymus tissue under at least one kidney capsule of the mice.

In another aspect of the invention there is provided a method of constructing BTCD-hu2 mice engrafted with human bone marrow cells. The method includes sublethally irradiating T- cell and B-cell deficient mice and inoculating pre-cultured human cells obtained from fetal bone marrow into the irradiated mice. In a preferred embodiment, the B-cell and T- cell deficient mice that are engrafted with human fetal bone marrow are BTCD-hu mice. In another preferred embodiment, the method provides for the engraftment of precultured adherent cells; In yet another aspect of the invention, there is provided a method for assaying the in vivo dissemination of HIV-1 in a BTCD-hu2 mouse, which includes the steps of a) inoculating a sufficient quantity of HIV-1 into the peritoneal cavity of a BTCD-hu2 mouse to cause HIV-1 infection and b) detecting HIV-1 in a human implant or human graft of the BTCD-hu2 mouse and in the peripheral lymphoid compartment of the BTCD-hu2 mouse.

In another aspect of the invention there is provided a screening method for determining the efficacy of an anti-HIV-1 drug or anti-HIV-1 therapy, which includes the steps of a) administering to an HIV-1 infected BTCD-hu2 mouse an anti-HIV- 1 drug or anti-HIV-1 therapy, b) assaying the peripheral lymphoid compartment of the HIV-1 infected mouse or human fetal tissue implant for the presence of HIV-1 and c) determining the efficacy of the drug or therapy on HIV-1 infection in said mouse.

In a related aspect of the invention there is provided a screening method for determining the toxicity of an anti-HIV-1 drug or anti-HIV-1 therapy, which comprises a) administering to an HIV-1 infected BTCD-hu2 mouse an anti-HIV-1 drug or anti-HIV-1 therapy, and b) assaying the peripheral lymphoid compartment or human fetal tissue implant of the HIV-1 infected BTCD-hu2 mouse for a change in the quantity or type of human T-cells, monocytes, or combination thereof and d) determining the toxicity of the drug or treatment.

In another aspect of the invention there is provided a method of detecting the expression of human cytokine genes in an HIV infected BTCD-hu2 mouse and determining the effects of HIV-1 infection on human cytokine gene expression in the chimeric mouse.

In another aspect of the invention there is provided a method of assessing the efficacy of bone marrow transplantation, which includes the steps of engrafting a B- cell and T-cell deficient mouse with human bone marrow cells and determining the degree of engraftment of the mouse.

In another aspect of the invention there is provided a method of amplifying human bone marrow cells prior to implantation of the bone marrow cells in a human patient, which includes the steps of engrafting a BTCD-huBM (as hereinafter defined) or BTCD-huCombo (as hereinafter defined) mouse with human bone marrow cells obtained from the patient or a donor, allowing the mouse to recover for about six to eight weeks and recovering human bone marrow cells from the mouse.

In still another aspect of the invention there is provided a method for generating human monoclonal antibodies which includes the steps of a) inoculating a BTCD-hu2 mouse, having transplanted therein human fetal spleen tissue, human fetal lymph node tissue or a combination thereof, and containing human B-cells and antigen presenting cells with a sufficient amount of antigen to cause the mouse to generate antigen-specific antibodies, b) collecting B-cells from the peripheral blood of said mouse;

c) optionally, transforming B-cells from step (b) with Epstein Barr Virus; d) fusing the B-cells from step (b) or step (c) with a permissive myeloma cell line, to thereby produce a hybridoma; e) isolating antibody producing cells produced in step (d) ; and f) purifying monoclonal antibody produced by the antibody producing cells of step (e) .

The term "BTCD-hu2 mouse (or mice)" is used herein to describe a mouse (or mice) deficient for murine B-cells and T- cells but which is chimeric for human B-cells, T-cells and/or monocytes.

The term "SCID-hu mouse (or mice)" is used herein to describe a BTCD-hu2 mouse of this invention which is genetically a SCID/SCID mouse (or mice) into which human fetal tissue has been implanted, resulting in a mouse having a sufficient number of human T-cells circulating in its peripheral lymphoid compartment to allow intraperitoneal inoculation of HIV-1 and subsequent HIV-1 infection of the mouse.

The term "SCID-BM mouse" (or mice) is used hereinafter to describe a BTCD-hu2 mouse (or mice) of this invention which is genetically a scid /scid mouse (or mice) having a sufficient number of human monocytes circulating in its peripheral ly phoid compartment to allow HIV-1 infection by intraperitoneal inoculation of HIV-1.

The phrase "peripheral lymphoid compartment" as used herein includes the lymph nodes, spleen, peripheral blood and peritoneal exudate cells of an animal.

The phrase "intraperitoneal inoculation" as used herein means inoculation into the peritoneal cavity and excludes inoculation into the transplanted material within the intraperitoneal cavity. The term "TCID₅₀" is used herein to mean the lowest dilution of supernatant of an HIV-1 tissue culture that is capable of infecting at least one half of appropriate quadruplicate tissue culture cells.

As used herein the phrase "substantial contact" means sufficient contact between at least two surfaces of implanted fetal tissue to provide an environment capable of sustaining implanted human fetal tissue in a BTCD-hu2 mouse during the natural life of the mouse. Contact between at least two surfaces is determined by the volume of tissue implanted per volume of kidney capsule.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a series of immunofluorescence profiles which were obtained by three-color flow cytometry of lymphocytes isolated from the peripheral blood (A and B) , spleen (C and D) and lymph nodes (E and F) of SCID-hu mice. Each of Figures IA, IB, IC, ID, IE and IF reflect the profiles obtained from analysis of 6 SCID-hu mice.

Figure 2 shows the immunofluorescence profiles obtained by three-color flow cytometry of mononuclear cells isolated by peritoneal lavage from the peritoneal cavities of SCID-hu mice. Figure 2A is the profile of the expression of human CD45+ and CD4+ and Figure 2B is the profile of the expression of CD45+ and CD8+. Each of Figures 2A and 2B reflect the profiles obtained from analysis of 3 SCID-hu mice. Figure 3 shows the results of immunofluorescence profiles obtained by three-color flow cytometry of pooled lymphocytes isolated from the peripheral blood, spleen and lymph nodes of SCID-hu mice analyzed for the expression of human CD4 (hatched boxes) , CD8 (solid bars) and the indicated TCR Vβ gene. The data are representative of that obtained from analysis of 3 SCID-hu mice.

Figure 4 is an autoradiograph of a Southern blot of PCR amplified DNA or cDNA. HIV-1 gag DNA was detected in SCID-hu mice infected with HIV-1 by intraimplant (Fig. 4A) or by intraperitoneal inoculation (Fig. 4B) . HIV-1 gag RNA was detected after intraimplant (Fig. 4C) or intraperitoneal inoculation (Fig. 4D) , HIV tat /rev mRNA was detected after intraimplant (Fig. 4E) and intraperitoneal inoculation (Fig. 4F) .

Figure 5 is an autoradiograph of a Southern blot of PCR amplified DNA from peritoneal exudate cells obtained from the peritoneal cavity of SCID-hu mice following HIV-1 infection one week (lane 1) or four weeks (lane 2) after intraperitoneal injection and probed with SK19.

Figure 6 is a photograph of ethidium bromide stained 1.5% NuSieve/0.5% agarose gels showing the amplification products of reverse transcriptase-polymerase chain reaction (RT-PCR) using various cytokine specific primers from various tissues of SCID-hu mice following HIV-1 intraperitoneal inoculation.

Figure 7 is a series of bar graphs showing human cytokine expression in SCID-hu mice and HIV-1 infected SCID-hu mice. The expression of tat/rev, 02-MG, TNF-α, TNF-/3, IL-2, IL-4 and IL-6 mRNA in the (A) human thymus/liver (hu-thy/liv) implant, (B) PBMC, (C) lymph nodes, and (D) spleens of SCID-hu mice (n=8) and HIV-1 infected SCID-hu mice (n=7) was determined by RT-PCR.

Figure 8 is a bar graph showing the percentage of human CD45+ cells in the bone marrow, lymph nodes, spleen and peripheral blood of SCID-huBM mice engrafted with uncultured or cultured human fetal bone marrow (HFBM) cells.

Figure 9 is a bar graph showing the effect of transplantation of adherent HFBM cells versus non-adherent HFBM cells on engraftment of irradiated SCID mice. The mean percentage of cells expressing human CD45+ are shown.

Figure 10 is a bar graph showing the temporal distribution of HFBM cells injected into irradiated SCID mice. The presence of human CD45+CD34+CD10- was assessed in the peripheral blood and bone marrow of BTCD-huBM mice by three- color flow cytometry and the data are expressed as the mean percentage of mice analyzed.

Figures 11A, 11B and 11C, 11D, HE, 11F, 11G 11H and HI are dot histograms of flow cytometric analysis of cells isolated from the bone marrow of a SCID mouse (HA, HD, HG) , a normal human fetus (HB, HE and HH) and a SCID-huBM mouse (HC, HF and 111) for the expression of human CD45 (HA, HB and HC) , CD34 (HD, HE and HF) and slgM (HG, HH, 111). Figures 12A, 12B, 12C and 12D are immunofluorescence profiles obtained by three-color flow cytometry of pure and mixed populations of human PBMC and SCID mouse bone marrow cells using anti-human CD45.

Figures 13A, 13B and 13C are graphs showing the percentage of human CD45+, CD45+CD34+CD10- CD45+CD34+CD10+, CD45+CD34-CD10+, CD45+CD20+sIgM and CD45+CD20+CDsIgM present in the (A) bone marrow, (B) spleen and (C) peripheral blood of individual BTCD-huBM mice. The mean percentage for each group is represented by a solid bar. Figure 14 is a graph showing the percentage of human

CD45+CD13+CD33+ cells in the bone marrow and peripheral blood of BTCD-huBM mice, two months after transplantation of irradiated SCID mice with pre-cultured HFBM cells. The mean percentage for each group is represented by a solid bar. Figures 15A and 15B are autoradiographs of the RT-PCR amplification products of RNA extracted from human fetal bone marrow cells (Fig. 15A) and mouse spleen cells (Fig. 15B) using species-specific primer pairs.

Figure 16 is a Western blot showing HIV-1 protein recognition by human IgG antibodies present in the serum of an HIV-1 infected BTCD-hu mouse also implanted with human spleen tissue.

DETAILED DESCRIPTION OF THE INVENTION

In the present invention, human/mouse chimerae are created by transplanting congenitally immunodeficient mice lacking both B-cells and T-cells (BTCD mice) , such as C.B-17 ' scid /scid mice, bnx mice, RAG-1 deficient or Rag-2 deficient mice with fetal human tissue and/or cells. The recipient animals lack functional B-cells and T-cells but are tolerant for xenografts. The resulting chimeric mice are referred to herein as BTCD-hu2 mice.

The BTCD-hu2 mouse is genetically deficient for murine B- cells and T-cells. The mouse is made chimeric for human immune system cells by transplanting human cells and/or tissue, such as, for example, human fetal thymus and fetal liver tissue, precultured or fresh human fetal bone marrow cells, or combinations thereof into the mouse. The BTCD-hu2 mouse is capable of sustaining the transplanted human tissue throughout the life of the mouse. The BTCD-hu2 mouse also contains a sufficient number of circulating HIV-1 infectable cells to allow HIV-1 infection of the transplanted human tissue and/or cells following peripheral inoculation of the mouse with HIV-1. The HIV-1 infectable cells present in the BTCD-hu2 mouse include human T-cells, and/or monocytes, the latter of which include circulating immature mononuclear phagocytes, macrophages, (i.e., highly differentiated mononuclear phagocytes) , or combinations thereof. The BTCD- hu2 mice of this invention may also contain human B-cells and/or other human immune system cells. The particular repertoire of human immune system cells present in the BTCD- hu2 mouse is dependent upon the specific human tissue and/or cells that have been transplanted into the mouse. The BTCD-hu mouse is genetically deficient for murine B- and T-cells but as a result of transplantation of human fetal liver and thymus tissue into the mouse kidney capsule the BTCD-hu mouse has significant numbers of circulating human B- and T-cells in the mouse peripheral compartment, spleen, lymph nodes and transplanted human tissue.

Another example of a BTCD-hu2 mouse is the BTCD-huBM mouse, which is obtained by transplanting and engrafting cultured human fetal bone marrow cells into a genetically murine B- and T-cell deficient mouse. Such transplantation results in significant levels of circulating human B-cells and monocytes in the mouse peripheral compartment and bone marrow.

Another example of a BTCD-hu2 mouse is the BTCD-huCombo mouse, which is obtained by transplanting and engrafting cultured human fetal bone marrow cells and transplanting human fetal liver and thymus tissue under the kidney capsule of a genetically murine B- and T-cell deficient mouse. The BTCD- huCombo mouse has significant numbers of circulating human B- and T-cells and human monocytes in its peripheral compartment. As noted above, the BTCD-hu2 mice of this invention can be infected with HIV-1 following peripheral inoculation. These mice can also be infected with other human pathogens due to the presence of human transplanted cells or tissues. The specific amount of pathogen necessary to cause infection of the human tissue can be determined by standard techniques for determining infectious dose or lethal dose. For example, the amount of HIV-1 necessary to cause infection via peripheral infection is generally in the range of from about 80 to 800 TCID₅₀ (tissue culture infectious dose) . The murine B-cell and T-cell deficient mice that can be transplanted with human cells and/or tissues as described above include, for example, homozygous scid, bnx, Rag-1 and Rag-2 mice, and may include other genetically B- and T-cell deficient mice. The BTCD-huBM mouse, according to the invention contains both human and mouse hematopoietic progenitor cells and the stromal microenvironment in which human stem cells, lymphoid and myeloid precursor cells can mature and function.

Several experimental animal systems have been established previously using either human bone marrow cells or human fetal bone transplanted into immunodeficient mice. Prolonged graft survival was observed in these systems, however, the number of detectable lymphoid cells or myeloid cells in the peripheral compartment of these mice was minimal. Moreover, significant engraftment of the mouse bone marrow with human cells only occurred if the mice received continuous treatment with human cytokines (Lapidot, et al., Ibid.). Consequently, previous animal systems have not reflected the complexity of hematopoiesis as it occurs in intact marrow nor have these animals provided an accurate model for the role of cytokines in hematopoiesis. Moreover, the low levels of human myeloid and lymphoid cells in the mouse peripheral compartment makes it virtually impossible to study the dissemination and function of these cells in the peripheral lymphoid tissue. A practical small animal model system for HIV-1 viremia and disease, in which candidate antiviral drugs can be evaluated rapidly for pharmacokinetics, toxicity, and antiviral efficacy, is also needed. The pandemic of the acquired immunodeficiency syndrome (AIDS) , caused by HIV-1, requires the rapid development of effective therapy and prophylaxis. Several candidate antiviral agents have been entered into clinical trials, and many more compounds have proven anti-HIV-1 activity in cultured cells. In most animal/retrovirus systems known to date, a period of clinically asymptomatic viremia precedes the development of overt disease. In general, animal systems can be used either as models for viremia, models for virus-induced disease, or both (Ruprecht, R.M. Intervirology (1989), 30:2-11). Thus far, no ideal animal model for human HIV-1 infection exists, and depending on the system used, only one or few aspects of HIV-1 viremia and disease in humans can be duplicated in animal models.

Testing candidate antiviral agents in inbred laboratory mice is desirable because a vast body of knowledge has been accumulated regarding murine genetics, immunology, cytokines, and metabolism, and because many immunological reagents and cytokines are available. Breeding can be accomplished easily because mice have a short generation time. The small size of these animals greatly facilitates drug studies, especially if only limited amounts of investigational compounds are available for analysis. This latter point is especially important for drug development from natural product sources. Furthermore, the small size of mice also assures relatively low housing costs.

The host range of HIV-1 is limited to humans, chimpanzees, and Gibbon apes. Neither murine cells nor mice can be infected by HIV-1. Using a human tissue transplantation approach, the present invention provides a murine model system wherein disseminated HIV-1 infection occurs for testing antiretroviral therapy, even though HIV-1 is unable to replicate in mice. In the present invention, chimeric BTCD-hu animals have been developed that permit prolonged survival and significant increase in the number of human T-cells in the peripheral blood of these mice, which can sustain HIV-1 replication in the mouse spleen, lymph node and peripheral blood, as well as in the thymic implant, following exposure to HIV-1. This mouse offers several uses and advantages over other animal model systems.

For example, the present method of generating test animals provides an animal that is useful in studying the pathophysiological events following HIV infection. It is crucial to evaluate therapeutic interventions directed to HIV- 1 in vivo using HIV-1 freshly isolated from patients. This is because HIV-1 functions differently following in vitro passage in tissue culture. This was dramatically demonstrated by the contrast between the potent effect of recombinant CD4 in blocking HIV-1 infection in the test tube and its minimal effect when given to patients (Daar E. S., and Ho D. D.

(1991), Amer. J. Med. , 90:225-265; Turner, S., et al. (1992), Proc. Natl. Acad. Sci. USA, 89:1335-1339; Moore, J.P. , et al. (1992), J. Virol. 66:235-43). Thus, in vivo screening of anti-HIV-1 therapies should enhance the targeting of development of drugs or drug combinations that will work in patients. The present mouse model for HIV infection mimics disseminated HIV-1 infection and thus, closely resembles actual infections seen in humans. Similar approaches can be used to assess therapies directed to HTLV-1 (and other retroviruses) .

The present animal model also provides a method for identifying and isolating unique cytokines important in the growth and maturation of human thymus and T lymphocytes. Involution of the thymus in later years has been associated with the decline of the immune system associated with age that may lead to cancer and autoimmune diseases. Unique human cytokines have been detected in vivo in the present BTCD-hu mice. This mouse should provide a means of assessing these novel thymic growth factors and may prove useful as a way of "recharging" the immune system in later life.

In the present invention, human/mouse chimerae were created by transplanting congenitally immunodeficient mice lacking both B-cells and T-cells (BTCD mice) such as C.B.-17 scid/ scid mice with human thymus and liver tissue. The recipient animals, lack functional murine T-cells and B-cells. In the case of recipient scid/scid, or SCID mice, the deficiency in B-cells and T-cells is due to a faulty VDJ recombinase mechanism. Schuler, et al., (1986), Cell, 46:963- 972; and Malynn, et al., (1988), Cell, 54:453-460. The resulting BTCD-hu mice provide both human hematopoietic progenitor cells and the stromal microenvironment in which human T-cell maturation occurs. These mice can be infected with HIV-1 intraperitoneally. Several experimental systems previously have been established using either fetal organ grafts or peripheral blood leukocytes from normal adult human donors. Prolonged graft survival was observed in both systems, however, the number of human T-cells in the peripheral blood was so low as to make the mice unusable for virus inoculation other than by directly inoculating virus into the transplanted tissue. As a result it was not possible to study the normal in vivo dissemination of virus into the various organs and physiological systems. Furthermore, the resultant HIV-1 infection was restricted to the human thymus and/or liver implant. Also, HIV-1 infection of these mice could only be accomplished using clinical isolates of HIV-1. Laboratory strains of HIV-1 failed to cause infection in these mice, despite the ability of laboratory strains of HIV-1 to infect humans, which has been demonstrated by the accidental HIV infection of at least two researchers with laboratory strains of HIV-1 (McCune, et al., (1991), Immunol. Rev. 124:45-61).

Previously human/mouse chimerae were created by transplanting human fetal tissues originating from liver, thymus, lymph nodes or bone as follows. Thymic and hepatic tissue were transplanted under the renal capsule, lymph nodes were transplanted into the mammary fat pads and human bone fragments were transplanted subcutaneously. However, as noted above, the resulting chimeric mice, i.e. containing both human thymus and hepatic tissue continuously produced human T-cells and the human tissue grew well, but the mean value of peripheral blood lymphocytes was only 0.7%. Krowka, et al., ' Ibid . Moreover, only very low levels of human lymphocytes were detected in mouse spleen or lymph nodes. These low levels of human T-cells in the peripheral blood, lymph nodes and spleen may be responsible for the failure of these researchers to obtain HIV infection in the chimeric mice by peripheral injection. The present invention provides a method for preparing BTCD-hu mice having significantly increased numbers of human peripheral blood T-cells and in which human T-cells are detectable in mouse spleen, lymph nodes and peritoneal cavity as well as in the transplanted human tissue, thus allowing for inoculation of HIV intraperitoneally, or by inoculation into the mouse rectum (peripherally) rather than by direct inoculation into transplanted human tissue. Consequently, HIV disseminates in the present BTCD-hu mice in a manner similar to the viral spread in HIV-infected humans. The present method involves the implantation of alternating human fetal thymic and human fetal liver tissue under the kidney capsule of BTCD mice, such as scid/εcid mice or homozygous Rag-1 deficient or Rag-2 deficient mice. The resulting BTCD-hu mice are physiologically distinct from recipient BTCD mice in that the BTCD-hu mice are capable of producing and maintaining circulating human T-cells. Moreover, the present BTCD-hu mice provide a functional and viable stromal microenvironment in which human T-cell maturation occurs. The implanted human tissue is supported by the mouse and is capable of growth and remains functional throughout the life of the BTCD-hu mouse. In general, implanted human tissue increases in size about 100 fold in the BTCD-hu mice.

Human fetal tissue for implantation is preferably derived from fetuses that are electively aborted at about gestational week 12 to about gestational week 22, preferably gestational week 17 to gestational week 21. Human fetal tissue is dissected free of surrounding connective tissue preferably within approximately 30 minutes of abortion. The specified organs, i.e. liver and thymus are washed and separated into ice cold Dulbecco's Physiologically Balanced Salts (PBS) and kept on ice from then on. Each organ is rinsed several times with ice cold sterile PBS. Intactness and sterility of the human fetal liver are critical. Only human fetal liver pieces with an intact capsule (including whole organ) are used in the implantation procedure in order to minimize contamination by organisms invading underneath a ruptured capsule or the stimulation of autolysis of liver and hematopoietic stem cells. The gallbladder is dissected away from the liver, carefully avoiding any spill of its contents. Regions towards the edge of the liver that have capsule surrounding the tissue on both sides are cut into slices, for example, 0.5cm X 0.5cm X 0.5cm, and as many pieces as required for the mice to be implanted are kept in PBS on ice until implantation. Liver slices are cut large enough for two kidneys to be implanted and small enough to facilitate the rapid diffusion of PBS into the tissue and metabolic products out of the tissue in order to minimize metabolic tissue damage. This procedure is similar to the perfusion of organ transplants in humans.

The human fetal thymus, in its capsule, is dissected away from the sternum and the heart. During this procedure the connective tissue capsule is left intact to reduce exposure of the thymus to a non-sterile environment. The thymus is then washed in ice cold, sterile PBS several times. Afterwards the capsule is dissected away from the thymus and the organ is cut into small pieces, such as 0.3cm³ pieces (as many as required for the mice to be implanted for the same reasons mentioned above for liver) along the grossly visible lines of thymic lobules, minimizing damage to the tissue. These pieces are kept on ice in PBS until implantation.

The human fetal thymus and liver pieces described above are cut on ice into pieces of from about 0.5 mm³ to about 2.0 mm³, preferably about 0.75 to about 1.25 mm³ and most preferably into 1 mm³ pieces. A plurality of pieces such as, for example, about 7 to about 12 pieces and preferably about 10 1mm³ pieces each of thymic and liver tissue are loaded into a cannula, preferably a 16 gauge cannula whose tip has been manually rounded and shortened. It is important to load liver and thymus alternately to maximize the contact interphase between the liver and thymus tissue, which is critical for survival and growth of the implant. Preferably, the total volume of fetal tissue implanted is in the range of about 0.5cm³ to about 1cm³, depending on the size of the particular mouse kidney capsule. Preferably, the amount of tissue implanted provides for contact between the alternating pieces of human fetal liver and kidney.

Human fetal spleen and/or human fetal intestine can also be similarly dissected and co-transplanted with human fetal liver and thymus tissue. In the case of fetal spleen co- transplanta-tion, a plurality of tissue pieces of from about 0.4 to 1 cm³ are transplanted to provide about 10 - 50xl0⁶ spleen cells. Similarly, intestine tissue is cut into pieces of about 3 x 5 cm and a plurality of pieces is co-transplanted with human fetal liver and thymus tissue.

The fetal tissue is transplanted into BTCD mice, preferably mice that are six to eight weeks old. The mice are anesthetized with any commonly used anesthetic drug, such as, for example, pentobarbital at an amount of about 40-80 mg/kg. An incision is made in the animal's right and left flanks and each kidney is exteriorized if both kidney capsules are to be implanted with tissue. Alternatively, a single kidney may be treated in this manner. Preferably, an incision of about 0.3mm to about 0.7mm, most preferably about 0.5mm is made at the tip of the lower kidney pole and the cannula inserted with its opening facing the kidney, after the kidney capsule is gently held up at the point of incision by an instrument, such as, for example, iris forceps. The cannula is then inserted in a gliding fashion further underneath the kidney capsule until the opposite kidney pole is reached. Here the opening of the cannula is turned to face first down and towards the kidney, and while angling the cannula end held in the hand past the midline of the kidney towards the dorsal side of the kidney, the inserted end of the cannula rises a short distance, for example, approximately 0.5mm underneath the kidney capsule. At this point the first tissue pieces are injected underneath the capsule. The injected pieces serve to block their own way towards the incision of the capsule due to the strain exerted on the capsule by the angled, inserted cannula. With the trocar kept fixed, the cannula is then pulled back a short distance, about 2mm, the opening turned up and towards the kidney while the angle of the cannula relative to the kidney is increased even further, to exert more strain on the capsule, thus preventing injected tissue pieces from being pushed out of the capsule incision. At this point the rest of the tissue is slowly injected, with several pieces of tissue gliding around the upper kidney pole and extending the capsule by about 3mm. While the last pieces are injected, the cannula is pulled out slowly while turning around its own axis to relieve the pressure and make room for more tissue. The capsule should collapse at the point of incision after the cannula is pulled out if the incision was made small enough and most of the tissue was injected at the opposite pole of the kidney. Preferably, both kidneys are implanted in that manner, resulting in a heterogeneous system having human liver and thymus tissue interspersed throughout. While this method of implantation is most preferable, any method for implanting interspersed human fetal liver and fetal thymus tissue may be employed. Regardless of the method of implantation, the total amount of each tissue that is transplanted is from about 10⁷ to about 10⁸ hematolymphoid cells, preferably from about 2xl0⁷ to about 5xl0⁷ cells under each kidney capsule.

The amount of time that lapses between obtaining human fetal tissue and transplantation has an effect on the success of the transplant. It is therefore, preferable that tissue be transplanted within 24 hours of pregnancy termination and most preferably, within 5 hours of termination. In general, the fresher the liver tissue, the more successful the implant. It is essential that tissue be maintained so as to minimize tissue damage, such as by maintaining the tissue on ice and perfusing with PBS at all times prior to transplantation. After implantation, the peritoneal layers are approximated with sutures and the wound is closed. Preferably, all surgical procedures are performed in a laminar flow hood using sterile technique. After surgery, the mice are preferably provided antibiotic prophylaxis and housed in an environment that can be monitored for mouse pathogens. The resulting BTCD-hu mice have detectable levels of human T-cells in the peripheral lymphoid compartment. The peripheral lymphoid compartment includes the mouse peripheral blood, lymph nodes, peritoneal cells and spleen. The BTCD-hu mice are capable of supporting long-term, multilineage human hematopoiesis, including differentiation of phenotypically normal and competent T-cells. The levels of human T-cells in the BTCD-hu mice are sufficient to support disseminated HIV-1 infection following intraperitoneal inoculation of HIV-1 into the mouse. The percentage of human T-cells present in the spleen and lymph nodes is significantly greater than that in the peripheral blood, indicating that there is a resident population of human T-cells in these organs, rather than merely cells present due to the presence of peripheral blood. Furthermore, there is a diversity of peripheral human T-cell population present in the resulting BTCD-hu mice including both human CD4+ and CD8+ cells. In general, BTCD-hu mice constructed in this manner contain peripheral blood lymphocytes containing at least about 3% to 10% human T-cells, preferably at least about 5% to about 7% human T- cells of the total number of peripheral blood lymphocytes. These human T- cells are composed of about 2 to 8% CD4+ T-cells and about 1 to 4% CD8+ T-cells. An amount of about 3% human T-cells in the peripheral blood is sufficient to allow HIV-1 infection of a BTCD-hu2 mouse by intraperitoneal inoculation.

The BTCD-hu mice obtained by the present method can be used as a model for HIV infection by either direct inoculation of the human thymus/liver implant with HIV-1 or by other routes, such as, for example, by inoculation of virus into the peritoneal cavity, or such as, into the rectum. For intraperitoneal infection an amount of HIV-1 in the range of a tissue culture infectious dose (TCID₅₀) of about 800 to about 8,000 TCID₅₀ is injected. Preferably, an amount of about 8000 TCID₅₀ HIV-1 in a volume of about 500 to about 1000 μl, preferably 800 μl is injected interperitoneally. Approximately one month following HIV-1 challenge it is possible to detect T-cell uptake of the virus and infection in the peripheral lymphoid organs and in the human tissue implant.

The present BTCD-hu mouse model for HIV-1 infection ' provides a significant advance in the study of AIDS in that it provides a much more natural model for studying the initial stages of HIV infection and dissemination than does direct injection into implanted tissue. It is known that following exposure of an individual to an inoculum of HIV-1, an infectious cycle is initiated that leads to systemic dissemination of virions and infected cells into lymphoid organs. McCune, (1991), Cell, 64:351-363. The subsequent disease course may be determined by the sites to which HIV-1 is seeded during the acute stage of infection. McCune, Ibid . ; Pantalleo, et al., (1993); New Eng. J. of Medicine, 328:327- 335. The generation of immune responses directed against HIV- 1 may result in a prolonged period of clinical latency. Pantalleo, et al., Ibid . However, a quiescent state of HIV-1 disease in the peripheral blood is often accompanied by ongoing active and progressive HIV-1 infection in lymphoid organs. Pantelleo, et al., Ibid . ; and Embretson, et al. (1993), Science, 362:359-362. Thus, examination of the peripheral blood of HIV-1 infected individuals does not detect the high degree of covert HIV-1 replication taking place in lymphoid tissues. Temin, et al., (1993), Nature, 362:292-293. Investigation of the pathophysiology of acute HIV-1 infection in humans is, therefore, restricted by the limited sampling of lymphoid organs during this stage. Thus, the present mouse model system for HIV-1 infection in which lymphoid organs and peripheral blood exhibit T-cell uptake of HIV-1 offers a window into the physiological events of the early stages of HIV-1 infection in vivo .

HIV-1 infected BTCD-hu mice can be assessed for HIV-1 infection by any known method. The titer of HIV-1 infected mononuclear cells present in the peripheral blood, spleen, thymic implant or lymph node of the BTCD-hu mice can be ascertained, for example, by isolating peripheral blood mononuclear cells (PBMC) from infected BTCD-hu mice and culturing titered numbers of cells in the presence of uninfected human phytohemagglutinin (PHA) -activated PBMC until such time as the p24 antigen content of the culture supernatant can be quantitated.

The presence of HIV-1 DNA and RNA in the human thymus/liver implants, and peripheral lymphoid compartment of the BTCD-hu mice infected by inoculation of HIV-1 into the peritoneal cavity can be assessed by polymerase chain reaction (PCR) assay or reverse transcriptase PCR (RT-PCR) , for example.

The present BTCD-hu mouse is also useful as a model system for studying the pathophysiology of in utero versus intrapartum vertical transmission of HIV-1 in humans. Although HIV-1 has been detected in fetal lymphoid organs (Mano, et al., (1991), AIDS Res. Hum. Retro. 7:337-341; Courgnaud, et al., (1991), AIDS Res. Hum. Retro. 7:337-341), it has been identified in the peripheral blood of less than half of HIV-1 infected newborns (Ehrnst, et al., (1991), Lancet 338:203-207; DeRossi, et al., (1992), AIDS 6:1117-1120; Krivine, et al., (1992), Lancet 339:1187-1189). It is possible that after intrauterine transmission, HIV-1 localizes to lymphoid organs where its presence escapes detection. The high degree of lymphocyte proliferation occurring in the neonatal thymus makes it an attractive environment wherein substantial replication of HIV-1 can occur (McCune, Ibid . ) . Various T-cell precursor populations in the thymus, including the immature "triple negative" CD3-CD4hi-CD8-T-cell precursors and the more mature CD4+CD8+ thymocytes are susceptible to HIV-1 infection (Schnittman, et al. (1992) , Proc. Natl. Acad. Sci. USA 87:7727-7731). Data obtained using the present mouse model indicate that human fetal thymus can become infected after peripheral exposure to HIV-1. Furthermore, after becoming infected with HIV-1 in the thymic environment, T- cells can migrate from the thymus and mediate peripheral dissemination of the HIV-1 infection. Taken together, these in vivo studies suggest that peripheral cells infected with HIV-1 in utero may home to the thymus where they can infect thymocytes with HIV-1 and thereby mediate subsequent infection of peripheral lymphoid tissues. Examination of peripheral blood immediately after birth may not detect the high degree of HIV-1 replication occurring in the thymus, spleen or lymph nodes. This is comparable to the dichotomy between the very active HIV-1 infection observed in lymph nodes and the low degree of HIV-1 infection seen in peripheral blood during the latent phase of HIV-1 infection in adults (Pantallelo, et al., (1993), Science 362:355-358). Therefore, the present BTCD-hu mice will provide a model for studying the effects of prenatal and postnatal anti-HIV interventions on the prevention of vertical transmission of HIV-1.

It has been demonstrated that cytokines such as TNF-α, TNF-/3, IL-2, IL-4 and IL-6 can modulate in vivo HIV-1 infection (Pantalleo, et al., (1993), New Eng. J. Med. , 328:327-335). IL-2 and IL-4 have been shown to synergistically promote HIV-1 replication in cultured thymocytes (Hays, et al. (1992), AIDS 6:265-272) and IL-6 secreted by thymic epithelial cells has been shown to up regulate HIV-1 replication in chronically infected cells (Schnittman, et al., Ibid . ) .

The present BTCD-hu mouse provides a model system that enables the in vivo assessment of the function of peripheral human T-cells and the role of cytokines in HIV-1 replication. Human cytokine mRNA can be detected in human tissue implants as well as the periphery compartment of the BTCD-hu mice of this invention. Since expression by human T-cells of mRNA for various human cytokines can be monitored in the present BTCD- hu mice, the use of such techniques as human-specific RT-PCR, for example, will provide a valuable model for exploring the in vivo role of cytokines in HIV-1 infection, or the role of cytokines in any retroviral infection under study in the BTCD- hu mouse system of this invention. The present BTCD-hu mouse provides an animal model for studying the efficacy and toxicity of drugs and other clinical interventions useful in the treatment of AIDS. For example, after intraperitoneal or direct transplant inoculation of the present BTCD-hu mouse with a sufficient amount of HIV to cause HIV-1 infection, an experimental drug or therapy can be administered to the infected animal. Moreover, the experimental drug or therapy can be administered prior to HIV- 1 inoculation of the present BTCD-hu mice. The subsequent disease course is determined by examining the degree of disseminated HIV-1 infection. Examination of the human fetal tissue transplant and various organs of the BTCD-hu mouse, such as, for example, the spleen, thymus or lymph nodes for the presence and quantity of HIV-1 will provide direct examination of the efficacy of the experimental drug or therapy. Examination of T-cell poeisis and/or monocyte poeisis will provide direct examination of the toxicity of the experimental drug treatment or therapy. The significant peripheral reconstitution of these BTCD-hu mice with human T- cells will permit the assessment of the toxic effects of experimental drugs or therapies on the production and dissemination of human T-cells.

Because the present BTCD-hu mice do not mount a murine immune response as a result of the B- and T-cell mutation, these mice may not be useful for studying the efficacy of AIDS vaccines. However, these mice provide human lymphocytes in the mouse peripheral lymphoid compartment in sufficient numbers to allow HIV-1 infection and are, therefore, useful for determining the efficacy and/or toxicity of anti-HIV-1 drugs or anti-HIV-1 therapies.

The present invention also provides a BTCD-hu2 mouse which has engrafted within the mouse bone marrow and lymphoid peripheral compartment human fetal bone marrow cells. This mouse, referred to herein as a BTCD-huBM mouse, contains in its peripheral lymphoid compartment B-cells, monocytes and macrophages, all of human origin. BTCD-huBM mice provide a functional and viable stromal microenvironment in which human hematopoiesis occurs. The implanted human cells are supported by the mouse and remain functional throughout the life of the BTCD-huBM mouse.

Human fetal bone marrow (HFBM) cells for implantation are preferably derived from fetuses that are electively aborted at about gestational week 12 to about gestational week 24, preferably gestational week 18 to gestational week 23. Human fetal bone marrow cells are obtained from the marrow cavities of the fetal bones, preferably fetal long bones, preferably within 5 hours of availability. The bone marrow cells are removed by lavaging the marrow cavities, for example, with phosphate buffered saline (PBS) followed by Ficoll-Hypaque density centrifugation. The interphase cells from the centrifugation are collected, washed in PBS, for example, and cultured at about 37°C, in the presence of about 5% C0₂, in appropriate medium, such as, for example, RPMI with added antibiotics. The cells obtained in this manner are cultured for about three to seven days, preferably about four days at which time the cells are harvested. Preferably, the adherent cells, i.e., those cells that stick to the culture plate, are harvested with the non-adherent cells, i.e. those cells that are free floating in the culture medium prior to harvesting. The adherent cells include human fetal bone marrow cells and stromal cells.

The harvested cells, which preferably include adherent cells, are transplanted into BTCD mice, preferably mice that are six to eight weeks old. The mice are subjected to sublethal irradiation about one hour prior to transplantation and then anesthetized with any commonly used anesthetic drug, such as, for example, pentobarbital at an amount of about 40 to 80 mg/kg. The amount of irradiation used depends on the genetic defect causing immuno-deficiency in the mouse. For example, scid /scid mice are preferably irradiated with about 400cGy since these mice are radiation-repair deficient, whereas, bnx and RAG-1 and RAG-2 mice can withstand up to about 800 cGy. The cultured cells are injected intravenously into the mice at a concentration of from about 1 x lθ⁶ to 1 x 10⁸, preferably 4 x 10⁷ cells in a total volume of about 500 ml.

Preferably, all procedures are performed in a laminar flow hood using sterile technique. After inoculation, the mice are preferably housed in an environment that can be monitored for mouse pathogens.

The resulting BTCD-huBM mice have significant engraftment of human cells in the mouse bone marrow. Furthermore, the mouse peripheral lymphoid compartment, which includes lymph nodes, peripheral blood, peritoneal cells and spleen, is also engrafted with human cells. The BTCD-huBM mice are capable of supporting long-term, multilineage human hematopoiesis, including differentiation of phenotypically normal and competent B-cells, macrophages and monocytes. However, these mice do not contain human T-cells.

The levels of engrafted human cells, which express human ^■ CD45, a human leukocyte common antigen that is present on human mononuclear cells at all stages of differentiation in the bone marrows of BTCD-hu2 mice is significantly higher than in the peripheral blood, indicating that there is a resident population of human cells in the bone marrow. In general, mice constructed in this manner contain at least about 10% to about 40% CD45+ cells in their bone marrow, preferably, at least about 25% CD45+ cells in the mouse bone marrow.

Mice engrafted in this manner also have a significant population of CD45+ cells in the peripheral blood and tissues of the peripheral lymphoid compartment. The levels of CD45+ cells in the peripheral blood of BTCD-huBM mice is in the range of from about 10% to about 50%, preferably at least 15%. Moreover, the level of human monocytes present in the BTCD- huBM mice is sufficient to support HIV-1 infection following intraperitoneal inoculation of the virus into the mouse even in the absence of human T-cells.

The subset of BTCD-hu2 mice of this invention that are chimeric for human bone marrow cells, i.e. the BTCD-BM mice above and the BTCD-hu Combo mice discussed below provide an excellent small animal model for studying human hematopoiesis, HIV-1 infection and drug efficacy and safety. These animals are chimeric for human hematopoietic cells and maintain significant levels of human cells in the mouse peripheral lymphoid compartment without the need for human cytokine supplementation. The significant numbers of human precursor cells observed in the bone marrow of these mice may be due to pre-culturing of the human fetal bone marrow cells prior to transplantation. It is possible that pre-culturing the fetal bone marrow cells increases the adherence between hematopoietic precursor cells and stromal cells such that both types of cells engraft together, thereby creating a human microenvironment for the engrafted human bone marrow cells. It is also possible that pre-culturing effects the stromal cells making them somewhat more adherent, such that they are more readily engrafted along with the bone marrow cells and may aid the bone marrow cells in engrafting. It is also possible that the mouse bone marrow can become engrafted with the human stromal precursor cells present in the adherent population, and provide species-specific maturation signals to the human progenitor cells. Although the presence of donor stromal cells in the bone marrow of transplanted recipients has been previously described (Keating, et al., (1982), Nature, 360:745-749) other investigators did not detect the presence of donor stromal cells in the recipient bone marrow after transplantation. It is possible that engraftment with stromal cells may depend upon the level of irradiation provided prior to transplantation and the number of stromal cells present in the infused bone marrow (Anklesaria, et al., (1987), Proc. Natl. Acad. Sci. USA, 84:7681-7685). It is also possible that transplantation of HFBM cells may uniquely lead to engraftment of the bone marrow with donor stromal cells even though transplantation with pediatric or adult bone marrow cells does not. This may be due to the presence in human fetal bone marrow of pluripotent bone marrow cells that can differentiate into hematopoietic precursors and stromal cells (Huang, et al. (1992) Nature, 360: 745-749) . Another possible explanation for the role of the adherent cells in enhancing the degree of engraftment is that this population is markedly enriched for human hematopoietic progenitor cells that adhere to stromal cells (Coulombel, et al., (1983),

Blood, 62: 291-297; Verfaillie, et al. (1990), J. Exp. Med. , 172:509-520) .

Although mature B-cells expressing slgM and slgD are detectable in the BTCD-huBM and BTCD-huCo bo mice, no human IgM or IgG is detected in the serum of BTCD-huBM mice or BTCD- huCombo mice. Since T-cells are detected in the peripheral blood of the BTCD-huCombo mice, the most likely cause for the inability of human B-cells in BTCD-huBM mice to secrete significant levels of IgM or IgG may be related to the inability of human T-cells to provide sufficient T-cell help to induce B-cell Ig secretion. An intrinsic deficiency in neonatal T-cell help is indicated by the observation that minimal serum levels of human IgM and IgG are observed in SCID mice injected intraperitonealy with human cord peripheral blood mononuclear cells, while significant serum levels of human IgM and IgG are detected in SCID mice injected intraperitonealy with a mixture of human cord blood B-cells and adult blood T-cells (Ueno, Y., et al., (1992) Scan . J . Immunol . 35, 415-419). It is also possible that the absence of secreted IgM and IgG in the present BTCD-huBM mice may be due to an intrinsic B-cell maturation block. However, co- implantation of BTCD-hu mice with fetal spleen tissue results in detectable levels of human HIV-specific antibodies following HIV infection of these mice, as shown in Figure 16. Reactivity of human IgG present in the HIV-infected BTCD-hu mice to the specific HIV-1 proteins gpl60, gpl20, p66, p51, gp41, p24 and pl7 is shown in Figure 16. Thus, BTCD-hu2 mice that are co-implanted with human fetal spleen tissue are useful for developing human monoclonal antibodies, to investigate the primary and secondary response of human lymphoid tissues to infection or antigenic stimulus and to test vaccines directed to human specific pathogens.

The maturation of human B-cells and monocytes occurs in the bone marrow of the BTCD-huBM and BTCD-huCombo mice of this invention in the complete absence of exogenous human cytokines. Furthermore, engraftment of the bone marrow is associated with reconstitution of the peripheral lymphoid compartment of these mice with human B-cell and monocytes. Thus, BTCD-huBM mice constructed in this fashion combined with species-specific reverse transcriptase polymerase chain reaction, (RTPCR) provide a valuable model for examining factors that stimulate or inhibit the in vivo maturation of human B-cells and monocytes. Furthermore, inasmuch as human T-cells are not detectable in BTCD-huBM mice, it should be possible to use them to examine the in vivo pathophysiology of isolated HIV-1 infection of human monocytes and thereby to evaluate the in vivo effectiveness of antiviral therapy on HIV-1-infected human monocytes. After exposure of a BTCD-huBM or BTCD-huCombo mouse to an inoculum of HIV-1, an infectious cycle is initiated that leads to systemic dissemination of HIV-1 and HIV-1 infected cells into lymphoid organs and bone marrow. The presence of HIV-1 in the bone marrow may be the etiology of hematological abnormalities seen in HIV-1-infected patients. Ineffective hematopoiesis may result from a direct effect of HIV-1 on stem/progenitor cells, stromal cells or mature cells present in the bone marrow. This effect may be mediated by infection of cells by HIV-1, a direct toxic effect of HIV-1 encoded proteins, or by the HIV-1-mediated suppression of stimulatory cytokines or induction of suppressive cytokines. These BTCD- huBM and BTCD-huCombo mice may be useful for studying the in vivo HIV-1 infection of human monocytes and B-cells. Since human hematopoiesis occurs in the mouse bone marrow and the peripheral lymphoid compartment. is populated with human T- cells, B-cells and monocytes, this mouse model should prove useful in studying the effect of acute and chronic HIV-1 infection on in vivo human hematopoiesis. Furthermore, the BTCD- huBM and BTCD-huCombo mice may be useful in screening the potential effectiveness of various interventions to reverse the negative effects of HIV-1 on hematopoiesis. The BTCD-hu2 mouse of this invention may also have transplanted therein human fetal thymus and human fetal liver tissue in addition to engrafted human fetal bone marrow cells. The resulting mouse, hereinafter referred to as a BTCD-huCombo mouse, has a full complement of human peripheral blood T- cells, B-cells, macrophages and monocytes, all of which are detectable in the mouse spleen, lymph nodes and peritoneal cavity, as well as in the transplanted human tissue. The BTCD-huCombo mouse may be prepared by the implantation of alternating human fetal thymic and human fetal liver tissue under the kidney capsule of a BTCD mouse and the engraftment of pre-cultured human fetal bone marrow cells in the bone marrow and peripheral compartment of the mouse. The transplantation and engraftment procedures are carried out separately. The order in which these procedures is done should not significantly effect the success of the operation. The procedure for implanting alternating human fetal thymus and human fetal liver tissue into a BTCD mouse is the same as that for constructing a BTCD-hu mouse. After a sufficient amount of recovery time is provided, such as, from about four to about six weeks, the BTCD-hu mouse is engrafted with pre-cultured fetal bone marrow cells in the same manner used to construct the BTCD-huBM mouse. The order of implanting human tissue and cells may be reversed and the tissues may be syngeneic or allogeneic to each other. The resulting BTCD-huCombo mice have detectable levels of human T-cells, B cells, macrophages and monocytes in the mouse peripheral lymphoid compartment. The BTCD-huCombo mice are capable of supporting long-term, ultilineage human hematopoiesis, including differentiation of phenotypically normal and competent T-cells, B-cells and monocytes. The levels of human T-cells and monocytes in the BTCD-huCombo mice are sufficient to support disseminated HIV-1 infection following intraperitoneal inoculation of HIV-1 into the mouse, as with the BTCD-hu mouse. The percentage of human T-cells present in the spleen and lymph nodes is significantly greater than that in the peripheral blood, indicating that there is a resident population of human T-cells in these organs, rather than merely transiently present due to the presence of these cells in the peripheral blood. Furthermore, as with BTCD-hu mice there is a diversity of the peripheral human T-cell population present in the resulting BTCD-hu mice, including both human CD4+ and CD8+ cells. In general, BTCD-huCombo mice constructed in this manner contain peripheral blood lymphocytes containing at least about 3% to 10% human T- cells, preferably at least about 5% to about 7% human T-cells. These human T-cells are composed of about 2 to 8% CD4+ T-cells and about 1 to 4% CD8+ T-cells. The percentages of B-cells, monocytes and macrophages in the peripheral compartment of BTCD-huCombo mice are similar to those observed in BTCD-huBm mice.

The construction of the BTCD-hu2 mice of this invention is successfully carried out by the present method in the absence of exogenous cytokines. The BTCD-hu2 mice of this invention support the transplanted tissues and/or cells throughout the natural life span of the mouse without treatment with human cytokines. Moreover, maturation of human B-cells and monocytes occurs in the bone marrow of BTCD-huBM and BTCD-huCombo mice in the absence of exogenous human cytokines.

The BTCD-hu2 mice of this invention have detectable levels of human cytokine gene expression. The BTCD-huBM and BTCD-huCombo mice have detectable levels of cytokine gene expression, such as, IL-3, IL-5, IL-6, IL-10, IL-7, LIF and M- CSF in the engrafted bone marrow. IL-7 has been shown to induce the proliferation of human B-cell precursors (Saeland, et al. (1991)), Blood, 78:2229-2238; Moreau, et al. (1993), Blood, 82:2396-2405) and to permit the in vitro growth of human B-cell precursors (Wolf, et al., (1991), J. Immunol., 147:3324). LIF stimulates the growth of human hematopoietic progenitor cells in culture (Verfaillie, et al., (1991), Blood, 77:263-270). The crucial role of LIF in the expansion of the hematopoietic precursor population has been indicated by the marked decrease in the number of stem cells observed in gene-targeted LIF-deficient mice (Escary, et al. 1993), Nature, 363:361-364). M-CSF is a member of a group of cytokines that induces progenitor cells in the bone marrow to proliferate and differentiate (Metcalf, D. , (1985), Science, 229:16). Thus, human cytokines that play important roles in different stages of human hematopoiesis are expressed by cells present in the bone of the BTCD-huBM and BTCD-huCombo mice of this invention. Moreover, the BTCD-hu2 mice of the invention have detectable levels of human cytokine gene expression in both the transplanted human tissue and the peripheral compartment of the mouse.

The BTCD-huBM and BTCD-huCombo mice constructed by the above-described processes provide valuable animal models for examining factors that stimulate or inhibit the in vivo maturation of human B-cells and monocytes. Moreover, because these chimeric mice contain significant levels of peripheral blood monocytes and macrophages, they are useful as animal models of HIV-1 infection, in the same manner as the BTCD-hu mouse. Furthermore, because T-cells are not detectable in BTCD-huBM mice, it is possible to utilize these mice to examine the in vivo pathophysiology of isolated HIV-1 infection of human monocytes and thereby to evaluate the effectiveness of antiviral therapy on HIV-1 infected monocytes. The BTCD-hu2 mice are useful as animal models for studying the efficacy of bone marrow transplantation. These chimeric animals are useful for assessing the effectiveness of transplant-ing various types of cells, such as stem cells, and pre-treatment of transplanted cells. Moreover, these animals are useful in assessing the safety of new hematopoietic drugs or treatments, such as, for example, assessing the effect of a test drug for protecting bone marrow from irradiation treatment or evaluating the toxicity of a new drug to human myelolymphoid lineage cells by determining the degree of engraftment on the mouse. The degree of engraftment correlates with efficacy of the pre-treatment of cells on transplantation.

In addition, these mice may provide a valuable model for assessing the effectiveness of in vivo gene therapy using human stem and precursor cells. For example, an attractive approach for treatment and prevention of HIV infection is by "intracellular immunization" (Baltimore, D. Nature 1988; 335:395-6). After transfection of cells with an expression vector coding for a protein or RNA capable of blocking the infectious cycle of HIV, the cell would be protected from HIV infection. Several stages during the replication of HIV are potential targets for molecular intervention (Mitsuya, et al., Science 1990; 249:1533-1544). For example, hammerhead ribozymes can be designed that specifically cleave HIV gag RNA and thereby markedly reduce viral replication (Sarver, et al., Science 1990:247:1222-1225). Recombinant retroviruses possessing amphotropic host ranges provide vector systems that permit the stable integration of DNA into the cellular chromosome in a stable and heritable fashion (Danos, O., and

Mulligan, R.C. Proc . Nat ' l . Acad . Sci . USA 1988;85:6460-6464). Intracellular vaccination of human stem/precursor cells with a retroviral vector would provide a continuous source of protected cells to populate the peripheral tissues. Successful retroviral mediated gene transfer into human CD34+ hematopoietic stem/progenitor cells from umbilical cord blood with subsequent stable expression in differentiated cells has been reported (Lu, et al., Exp . Med . 1993;178:2089-2096). The BTCD-hu2 mice of this invention are uniquely suited to be used as an in vivo system to assess such approaches to the treatment of HIV-infection.

Because of the close resemblance to normal human bone marrow differentiation in these mice, the BTCD-huBM and BTCD- huCombo mice of this invention can be used for in vivo screening of growth factors for the capacity of such factors to enhance production of myelolymphoid lineage cells. Such information can then be incorporated in the treatment of patients for lymphopenia or neutropenia which conditions often result in patients suffering from various and diverse disease processes, as well as from secondary effects of chemotherapy. Furthermore, those BTCD-hu2 mice of this invention which contain human bone marrow engrafted therein, can also be used either to amplify the numbers of autologous human bone marrow cells prior to transplantation into a patient in need thereof or alternatively, as a source of allogenic human bone marrow stem cells for transplantation into a patient in need thereof. The BTCD-huBM mouse is particularly suited for amplifying human bone marrow cells or providing a source of human bone marrow cells since these mice do not contain T-cells and thus, the problem of graft versus host disease which usually precludes the use of HLA-mismatched bone marrow transplantation in patients is avoided. Basically, BTCD-huBM mice can be engrafted with bone marrow cells from a patient or donor and after an appropriate period of recovery and growth, the mouse is sacrificed and the human bone marrow cells are recovered from the engrafted mouse bones by lavaging with sterile PBS, for example. The CD45+ cells are then purified from the obtained bone marrow cells. The human bone marrow cells, e.g., CD45+ can then be transplanted into the patient. Because BTCD-huBM mice contain human B-cells and antigen presenting cells in their peripheral lymphoid compartment they can also serve as a source for generating human polyclonal antibodies to any antigen. These mice can be inoculated with an antigen, such as, for example, HIV-gp 120 and after sufficient time, such as about 4 to 8 weeks, the antibodies can be collected and provided to a human patient in need thereof.

Human monoclonal antibodies may also be generated from vaccinated or actively infected BTCD-huBM or BTCD-huCombo mice by either fusing isolated Epstein Barr Virus (EBV) transformed human B-cells from such mice with appropriate myeloma cell lines (M.R. Posner, et al., J . Immunology, 146:4325-4332, 1991) or fusing non-EBV-transformed human B-cells from such mice with permissive human myeloma cell lines or mouse myeloma cell lines (D.F. Lake, et al., AIDS , 6:17-24, 1992). Antibody producing cells can then be isolated and the monoclonal antibody purified by routine methodologies.

EXAMPLE 1

Implantation of Human Thymic and Liver Tissue Into SCID Mice

SCID-hu mice were prepared by implanting human fetal thymic and liver tissue into the kidney capsules of scid/scid mice. Briefly, after the scid/scid mice were anesthetized with pentobarbital (40-80 mg/kg) , a 3 cm incision was made in the left and right flanks of the animal. The mouse kidney was held up with a he ostat and a 0.5 mm incision was made at the tip of each lower kidney pole. A cannula containing 10 1 mm³ pieces each of alternating human fetal thymus and liver (hu- thy/liv) obtained from the same donor were implanted with a 16 gauge cannula under both kidney capsules.

The human tissue was obtained from human fetuses that had been electively terminated at from 17-21 weeks of gestation. Only human fetal liver pieces with an intact capsule were used in the implantation procedure in order to minimize autolysis and contamination by organisms invading underneath a ruptured capsule. The fetal gall bladder was dissected away from the liver, carefully avoiding any spill of its contents. Regions towards the edge of the liver that had capsule surrounding the tissue on both sides were cut into slices (0.5cm X 0.5cm X 0.5cm) and as many pieces as mice to be implanted were kept in PBS on ice until implantation. Liver slices large enough for implantation of two kidneys were cut out. The human fetal thymus, in its capsule, was dissected away from the sternum and the heart. During this procedure the connective tissue capsule was left intact to reduce exposure of the thymus to a non-sterile environment. The thymus was then washed in ice cold, sterile PBS several times. Afterwards the capsule was dissected away from the thymus and the organ was cut into 0.3 X 0.3 X 0.3cm pieces (as many as mice to be implanted) along the grossly visible lines of thymic lobules, minimizing damage to the tissue. These pieces were kept on ice in PBS until implantation.

The human fetal thymus and liver pieces described above were cut on ice into 1 mm³ pieces and approximately 10 pieces each of liver and thymus tissue were loaded into a 16 gauge cannula, whose tip had been manually rounded and shortened. Liver and thymus were alternately loaded to maximize the contact interphase between liver and thymus. That is, there is substantial contact between the surfaces of implanted tissue to provide an environment capable of sustaining the implant during the natural life of the mouse.

The fetal tissue was implanted into male (6-8 wk old) mice within 5 hours of availability. The fetal gestational age was determined by foot length measurements. The cannula was inserted through the incision in the kidney in a gliding fashion further underneath the kidney capsule until the opposite kidney pole was reached. The opening of the cannula was turned to face first down and towards the kidney, and while angling the cannula end held in the hand past the midline of the kidney towards the dorsal side of the kidney, the inserted end of the cannula was raised approximately 0.5mm underneath the kidney capsule. At this point the first tissue pieces were injected underneath the capsule. The tissue pieces blocked their own way towards the incision of the capsule due to the strain exerted on the capsule by the angled, inserted cannula. With the trocar kept fixed, the cannula was then pulled back about 2mm, the opening turned up and towards the kidney while the angle of the cannula relative to the kidney was increased even further, to exert more strain on the capsule, thus preventing injected tissue pieces from being pushed out of the capsule incision. At this point the rest of the tissue was slowly injected, with several pieces of tissue gliding around the upper kidney pole and extending the capsule by about 3mm. While the last pieces were injected, the cannula was pulled out slowly while turning around its own axis to relieve the pressure and make room for more tissue. The capsule collapsed* at the point of incision after being pulled out since the incision was made small enough to allow this. Most of the tissue was injected at the opposite pole of the kidney. Both kidneys were implanted in that manner.

The peritoneal layers were approximated with 7-0 nylon sutures and the wound was closed with an Autoclip (Clay Adams, Parsippany, NJ) . All surgical procedures were performed in a laminar flow hood using sterile technique. After surgery, the mice were started on trimethoprim/sulfamethoxazole antibiotic (TMS; Schein Pharmaceutical Inc., Port Washington, NY) prophylaxis and were housed in bonnetted isolator cages (Lab Products, Inc. , Federalsburg, MD) in an environment that was monitored for mouse pathogens.

The resulting SCID-hu mice were assayed for the presence of human T-cells in the peripheral compartment.

Flow Cytometric Analysis . Mononuclear cells were harvested from the peripheral blood, spleens and lymph nodes of the SCID-hu mice and stained with PE-, FITC, or PerCP-conjugated mouse mAb to human CD4 (Leu 3a, Becton Dickinson, Mountain View, CA) , human CD8 (Leu 2a, Becton Dickinson), human CD3 (Leu 4, Becton Dickinson), or human CD45. TCR V gene expression was analyzed by staining mononuclear cells with PerCP-conjugated mouse mAb to human CD4 (Leu 3a, Becton Dickinson) , PE-conjugated mouse mAb to human CD8 (Leu 2a, Becton Dickinson), and FITC-conjugated mouse mAb to either TCR Vjβ2, V/35a, V/35b, V/?5c, V06, V/38, V312, Vj8l9 or Vα2 (T-Cell Diagnostics). Expression of human CD45, CD3, CD4 or CD8 or human CD4, CD8 and TCR V genes by lymphocytes present in the peripheral blood, lymph nodes and spleens of the SCID-hu mice were then assessed by three-color flow cytometric analysis using a FACScan cell analyzer with LYSIS- II software (Becton Dickinson) . Nonviable cells and unlysed red blood cells were gated out based on their forward and side scatter profiles, and lymphocyte gates were set on the basis of forward and side scatter profiles to correspond to gates set to control human (from healthy adult volunteer) lymphocytes. Cut off values for the quadrants were set after PE vs. FITC vs. PerCP emission was compensated for based on the analysis of single, double and triple staining of positive and negative control samples (human adult and CB-17 mouse mononuclear cells) and analysis of the appropriate mouse IgG isotype controls.

Three months after implantation, the percentage of lymphocytes expressing human CD3 and CD45 was 5.4% in the peripheral blood (Fig. IA) , 24.5% in the mouse spleen (Fig.

IC) and 36% in the mouse lymph nodes (Fig. IE) . Expression of human CD4 and CD8 by lymphocytes positive for human CD45 was examined and revealed that in the peripheral blood, 3.3% were CD4+ and 1.9% were CD8+ (Fig. IB), in the spleen, 15.7% were CD4+ and 6.2% were CD8+ (Fig. IF). The percentage of human T- cells present in the spleen and lymph nodes was significantly greater than that observed in the peripheral blood. This indicated that the human T-cells detected in the mouse lymphoid organs reflected a resident population and not cells from peripheral blood present in the tissue. To determine whether human T-cells were present in the peritoneal cavity of these SCID-hu mice, peritoneal exudate cells were harvested by peritoneal lavage with cold PBS. As determined by three-color flow cytometric analysis, 0.6% of the peritoneal exudate cells were human CD4+ cells (Fig. 2A) and 0.33% were human CD8+ cells (Fig. 2B) . The diversity of the peripheral human T- cells present in the SCID-hu mice was assessed by examining their expression of TCR V/3 genes with a panel of mAb that cover about 30% of peripheral human T-cells. As shown in Figure 3, a diverse population of human TCR VjS subsets were observed in the periphery of SCID-hu mice. Thus, construction of SCID-hu mice in this fashion results in the significant population of the peripheral lymphoid compartment with a broad spectrum of human T-cells. Implanted human tissue was found to have grown from about 1 mm³ to about 1.5 cm³. Microscopic examination revealed that the implanted tissue had well demarcated cortical and medullary regions which closely resembled those of human thymus. EXAMPLE 2

Infection of SCID-hu mice with HIV-1 . A clinical isolate of HIV-1, HIV-1₂₈ was obtained following co-culture of PBMC isolated from a 2 year old HIV-1 infected child with PHA-activated donor PBMC. The initial co- culture supernatant was harvested and co-cultured with PHA- activated PBMC to expand the quantity of HIV-1₂₈. The secondary co-culture supernatant was harvested and aliquots were frozen in liquid nitrogen. The tissue culture infective dose₅₀ (TCID₅₀) of the supernatant was determined by culturing titered dilutions of a thawed aliquot with phytohemagglutinin (PHA) activated donor PBMC (1.0 X 10⁶) in a total volume of 2.0 ml of RPMI 1640 with fetal calf serum (FCS) (19% v/v) and IL-2(32 units/ml). After two weeks of culture, the p24 antigen content of the culture supernatant was measured by using the HIV-1 p24 core profile ELISA assay (Dupont-NEN, Wilmington, DA) . The lowest dilution of supernatant that infected at least half of the quadruplicate cultures with HIV- 1 was taken as the end point or TCID₅₀. SCID-hu mice were infected either by direct injection of 300 TCID₅₀ of HIV-1₂₈ in a volume of 30 μl into one hu-thy/liv implant or by intraperitoneal injection of 8,000 or 800 TCID₅₀ of HIV-1₂₈ in a volume of 800 μl.

Infection of the peripheral lymphoid compartment of SCID- hu mice with HIV-1 .

Direct Inoculation Into Implanted Tissue:

After direct inoculation of the hu-thy/liv implant with HIV-1, the SCID-hu mice were assessed for disseminated HIV-1 infection. Since hu-thy/liv was implanted in each kidney capsule of these SCID-hu mice, it was possible to assess whether HIV-1 directly injected into the hu-thy/liv implanted in one kidney capsule could be systematically disseminated and infect the other hu-thy/liv implanted in the opposite kidney capsule. HIV-1 was isolated by co-culture of thymocytes from both hu-thy/liv implants, the spleens and PBMC of 5 SCID-hu mice one month after direct HIV-1 inoculation into unilateral hu-thy/liv implants. The degree of HIV-1 infection present in the HIV-1 injected hu-thy/liv implant, the uninjected hu- thy/liv in the opposite kidney, the spleen and PBMC were determined by quantitative coculture. HIV-1 was isolated from as few as 320 thymocytes from both the injected and uninjected hu-thy/liv implants indicating the presence of over 3,125 TCID/10⁶ cells (Table 1 and Table 2) . HIV-1 was also isolated from as few as 3,000 splenocytes reflecting the presence of at least 333 TCID/10⁶ cells. In addition, HIV-1 was cocultured from PBMC obtained from the peripheral blood of these intraimplant injected SCID-hu mice. Thus, T-cells that become infected with HIV-1 in the hu-thy/liv implant can induce disseminated HIV-1 infection of SCID-hu mice constructed as described above.

Table 1

INFECTION OF SCID-hu MICE AFTER INOCULATION OF THE HUMAN thy/liv IMPLANT WITH HIV-1

HIV-1 CO-CULTURE RESULT

NUMBER OF

LYMPHOCYTES INJECTED UNINJECTED SPLEEN PBMC

ADDED TO CO-CULTURE IMPLANT IMPLANT

1 X 10⁶ + + + +

2 X 10⁵ + + + ND^*

4 x 10⁴ + + ND

8 x 10³ + + - ND

1 . 6 X 10³ + + - ND

3 . 2 X 10² + + - ND

Table 2

INFECTION OF SCID-hu MICE AFTER INOCULATION OF THE HUMAN thy/liv IMPLANT WITH HIV-1

HIV- -1 TITER (TCID/ 10⁶ CELLS )

MOUSE NUMBER INJECTED UNINJECTED SPLEEN BLOOD

IMPLANT IMPLANT

Tl >3 , 125 ND^* 5 ND

T2 >3 , 125 ND >2 ND

T3 >3 , 125 >3 , 125 333 >2

T4 >3 , 125 0 2 >1

T5 >3 , 125 >3 , 125 >2 >1

The data in each table are presented as TCID/10⁶ mononuclear cells and a ">" indicates that the co- culture was positive for the lowest number of cells added.

* Not done due to insufficient cell numbers

Three months after implantation under the renal capsule of scid mice, one hu-thy/liv implant in each of 5 SCID-hu mice was injected with 300 TCID₅₀ of HIV-1₂₈. One month later, the mice were killed, mononuclear cells were isolated from the injected hu-thy/liv implant, the uninjected hu-thy/liv implanted in the opposite kidney, spleen, and peripheral blood, extensively washed and then the indicated number of mononuclear cells were co-cultured with PHA activated PBMC (1 X 10⁶) . After 7 days of culture, an aliquot of the supernatant was harvested and assessed for the presence of p24 antigen. A positive value reflects the detection of greater than 100 pg/ml of p24 antigen in the co-culture supernatant.

Intraperitoneal Inoculation :

To examine the capacity of HIV-1 to infect SCID-hu mice by other routes, HIV-1 was inoculated into the peritoneal cavity of SCID-hu mice. One month after HIV-1 inoculation,

HIV-1 was isolated by co-culture from the hu-thy/liv implants, and spleens of 5 of 5 SCID-hu mice injected with 8,000 TCID₅₀, 1 of 2 SCID-hu mice injected with 800 TCID₅₀ and 0 of 2 SCID-hu mice inj ected with 80 TCID 50 ' The HIV-1 isolated was not residual virus from the initial inoculation since no HIV-1 was isolated by co-culture from the spleens of unimplanted SCID mice 1 month after injection with 8,000 TCID₅₀ of HIV-1₂₈. Assessment of the extent of HIV-1 infection by quantitative co-culture indicated that with over 3,125 TCID/10⁶ cells, the HIV-1 infection in the hu-thy/liv implant was comparable to that which occurred after intraimplant infection of SCID-hu mice (Table 2) . In addition, up to 25 TCID/10⁶ cells were present in the spleens of intraperitoneally injected SCID-hu mice. Thus, after peripheral inoculation with HIV-1, HIV-1 infected cells can migrate from the periphery into the hu- thy/liv implant and infect human T-cells present in the implant.

The results of HIV-1 infection of SCID-hu mice of this invention after intraperitoneal inoculation with HIV-1 are shown in Tables 3 and 4.

Table 3

HIV-1 INFECTION OF SCID-hu MICE AFTER INTRAPERITONEAL INOCULATION WITH HIV-1

HIV- -1 CO-CULTURE RESULT

NUMBER OF

LYMPHOCYTES PERIPHERAL

ADDED TO

IMPLANT SPLEEN BLOOD CO- CULTURE

1 X 10⁶ + + ND^*

2 X 10⁵ + + +

4 X 10⁴ + + ND

8 x 10³ + ND

1. 6 x 10³ + ND

3 . 2 X 10² + ND

* Not done due to insufficient cell numbers

SCID-hu mice were injected intraperitoneally with 8.0 X 10⁴ TCID_JO of HIV-1₂₈. One month later, the mice were killed, and mononuclear cells were isolated from the hu-thy/liv implant, spleen and peripheral blood of the infected SCID-hu mice. The cells were extensively washed and then the indicated number of mononuclear cells were co-cultured with PHA-activated PBMC (1 X 10⁶) . After 7 days of culture, an aliquot of the supernatant was harvested and assessed for the presence of p24 antigen. A positive value reflects the detection of greater than 100 pg/ml of p24 antigen in the co¬ culture supernatant.

Table 4

HIV-1 INFECTION OF SCID-hu MICE AFTER INTRAPERITONEAL INOCULATION WITH HIV-1

HIV-1 TITER (TCID/ 10⁶ CELLS )

SCID-hu MOUSE

NUMBER IMPLANT SPLEEN BLOOD

PI >3 , 125 0 ND^*

P2 >3 , 125 25 >5

P3 >3 , 125 >0 . 7 ND

P4 625 >0 . 7 ND

P5 >3 , 125 >2 >2

*Not done due to insufficient cell numbers. One month after intraperitoneal innoculation of SCID-hu mice with 8.0 x 10⁴ TCID₅₀ (PI, P2, P4 and P5) or 8.0 x 10³ TCID₅₀ (P3) of HIV-1₂₈, the mice were killed, mononuclear cells were isolated from the hu-thy/liv implant, the spleen, and the peripheral blood, extensively washed and, if sufficient cells were available, quantitative co-culture of the mononuclear cells with PHA-activated PBMC (1 x 10⁶) was performed. After 7 days of culture, an aliquot of the supernatant was harvested and assessed for the presence of p24 antigen. The coculture was considered positive if greater than 100 pg/ml of p24 antigen was detected in the supernatant. The data are presented as TCID/10⁶ mononuclear cells and a ">" indicates that the coculture was positive for the lowest number of added cells. HIV viral culture .

The titer of HIV-l infected mononuclear cells present in the peripheral blood, spleen, thymic implant or lymph node of the SCID-hu mice was determined. Five-fold dilutions of PBMC ranging from 1 X 10⁶ cells to 2 X 10² were cultured at 37°C in quadruplicate culture in 24 well culture plates with PHA- activated donor mononuclear cells (1.0 X 10⁶) in a total volume of 2.0 ml of RPMI 1640 with added FCS (10% v/v) and IL- 2 (32 units/ml) . After one to two weeks of culture, the p24 antigen content of the culture supernatant was measured as described above. The lowest number of added PBMC that infected at least half of the quadruplicate cultures with HIV- 1 was taken as the end point or TCID.

EXAMPLE 3 Infection of SCID-hu mice with various strains of HIV-l . A clinical isolate of HIV-l (HIV-1₅₉) obtained from an AIDS patient, or a laboratory strain of HIV-l (HIV-l,^ and HlV-ljy,) were intraperitoneally injected into SCID-hu mice at a dosage of S.OxlO⁴ TCID₅₀- One month later, the mice were killed, and mononuclear cells were isolated from the hu- thy/liv implant of each of the infected SCID-hu mice. The cells were extensively washed and the indicated number of thymocytes were co-cultured with PHA-activated PBMC (lxlO⁶) . After 7 days of culture, an aliquot of the supernatant was harvested and assessed for the presence of p24 antigen. The results are shown in Table 5.

Table 5

HIV-l INFECTION OF SCID-hu MICE AFTER INTRAPERITONEAL INOCULATION WITH DIFFERENT STRAINS OF HIV-l

HIV-l STRAIN

NUMBER OF LYMPHOCYTES HIV-1₅₉ HIV-I_MN HIV-I_RP

ADDED TO CO- CULTURE

1 X 10⁶ + + +

2 X 10⁵ + + +

4 X 10⁴ + - +

8 x 10³ + - +

1 . 6 x 10³ + - +

3 . 2 X 10² + — +

A positive value reflects the detection of greater than 100 pg/ml of p24 antigen in co-culture supernatant. Both clinical and laboratory strains of HIV-l were capable of infecting the SCID-hu mice, as shown in Table 5.

EXAMPLE 4

Detection of HIV-l DNA and RNA in HIV-l- infected SCID-hu Mice .

The presence of HIV-l DNA and RNA in the hu-thy/liv implants, and periphery of the SCID-hu mice infected by inoculation of HIV-l either into the implant or peritoneal cavity were assessed by PCR. Specifically, HIV-l DNA and RNA gag-encoded sequences and spliced tat/rev mRNA sequences were assessed by PCR. Mononuclear cells from the SCID-hu mice were lysed in guanidine isothiocyanate (4 M) buffer, cellular DNA and RNA were separated by cesium chloride (5.7 M) density gradient centrifugation and precipitated with ethanol. For DNA PCR, the HIV-l DNA (1 μg) was amplified for 35 cycles with a primer pair specific for the gag gene segment (SK38/39) , electrophoresed through 1.5% NuSieve/0.5% SeaKem agarose (FMC, Rockland, ME) gel containing ethidium bromide, and the amplified product was detected under ultraviolet light. HIV-l RNA was detected by PCR amplification of reverse transcribed RNA (RT-PCR) . After treatment of the cellular RNA with RNase- free DNase (Boehringer-Mannheim, Indianapolis, IN) , RNA (7 μg) in 1 μl of ddH₂0 was mixed with 4 μl of 5X buffer (250 mM Tris-HCl, pH 8.3, 375 mM KCl, 15 mM MgCl₂) , 2μl DTT (100 mM) , lμl of random hexamers (BRL-Gibco) and 5 μl mixed dNTPs (2 mM each) . Samples were mixed, heated to 65°C for 10 minutes, cooled on ice for 5 minutes and then 1 μl (200 units) of Superscript reverse transcriptase (BRL-Gibco) was added. This final reaction mixture was vortexed, briefly spun down, incubated at 37°C for 60 minutes and then placed on ice. HIV- 1 cDNA was amplified either with SK38/39 or a primer pair specific for tat /rev spliced mRNA sequences (TR-5/TR-3) . Specificity of the amplified product was confirmed by hybridization of a Southern blot of the amplified DNA and cDNA with a [γ^"32P]-ATP-labeled internal probe specific for the

SK38/39 product (SK19) or the TR-5/TR-3 product (TR-4) . A given sample was regarded as positive if PCR amplification resulted in DNA product of the predicted size that hybridized to the specific internal probe. Positive and negative controls were included in all runs and to prevent contamination, suggested guidelines for PCR quality control were followed. Krone, et al., (1990), AIDS, 3:517-540. For RT-PCR, the absence of residual DNA template was verified by the absence of an amplified product following PCR amplification of DNase-treated samples that had not been reverse transcribed.

The results are shown in Figure 4. HIV-l gag DNA was detected by SK38/39-primed PCR amplification in the hu-thy/liv implant, PBMC, spleen and lymph nodes of SCID-hu mice infected with HIV-l either by intraimplant injection (Fig. 4A) or by intraperitoneal inoculation (Fig. 4B) . In addition, HIV-l gag RNA was detected by SK38/39-primed RT-PCR in the hu-thy/liv implant, PBMC, spleen and lymph nodes of SCID-hu mice infected with HIV-l either by intraimplant injection (Fig. 4C) or by intraperitoneal inoculation (Fig. 4D) . The presence of tat /rev mRNA was detected by RT-PCR in the hu-thy/liv implant, PBMC, spleen and lymph nodes of SCID-hu mice infected with HIV-l either by intraimplant injection (Fig. 4E) or by intraperitoneal inoculation (Fig. 4F) . No HIV-l DNA or cDNA was detected in SCID-hu mice that had not been infected with HIV-l (data not shown) . The HIV-l DNA and cDNA detected was not from the initial inoculation since no HIV-l DNA and cDNA was detected in the spleens of unimplanted SCID mice 1 month after injection with 8,000 TCID_So of HIV-1₂₈ (data not shown). Since human CD4+ T-cells were present in the peritoneal cavity of these SCID-hu mice, these cells were examined to determine whether they became infected with HIV-l after intraperitoneal injection. Following intraperitoneal injection of HIV-1₂₈ (8,000 TCID₅₀) peritoneal exudate cells were harvested, extensively washed and the DNA was extracted. As shown in Figure 5, HIV-l DNA was detected by SK38/39-primed PCR amplification in the peritoneal exudate cells of SCID-hu mice either one week (lane 1) or four weeks (lane 2) after intraperitoneal injection. No HIV-l DNA was detected in peritoneal exudate cells harvested from unimplanted scid mice one week or four weeks after intraperitoneal injection with 8,000 TCID₅₀ of HIV-1₂₈. Thus, productive infection with HIV-l and active viral replication was occurring in the periphery of SCID-hu mice infected by inoculation of HIV-l either into the hu-thy/liv implant or the peritoneal cavity.

EXAMPLE 5

Intestinal Inoculation of HIV-l into SCID-hu Mice . The quantity of HIV-l in an initial co-culture supernatant was expanded as in Example 1. HIV-l was inoculated into the rectum of a SCID-hu mouse at dosage of 8000 TCID₅₀. One month after HIV-l inoculation, the mouse was sacrificed and the extent of HIV-l infection was assessed by PCR as described in Example 4.

HIV-l DNA was detected in the implanted human thymus and liver tissue.

EXAMPLE 6

Detection of Human Cytokine Gene Expression by RT PCR. The pattern of in vivo human cytokine expression of the human cells present in SCID-hu mice was assessed. To ensure that the PCR amplification product was of human origin, primers were designed so that the nucleotide sequence of the 3 ' end was complementary to a human cytokine cDNA sequence that was not present on the mouse cytokine cDNA. Derivation of the PCR product from mRNA was insured by designing primer pairs that yielded an amplification product that spanned exon- exon junctions. After reverse transcription of mRNA extracted from the hu-thy/liv implant, PBMC, spleen and lymph nodes of the SCID-hu mice, cDNA (7 μg) was amplified by PCR with human- cytokine specific primers for 60 cycles of denaturation at 94°C for 1 minute, annealing at 65°C for 1 minute and extension at 72°C for 1 minute. The presence of the target mRNA was indicated by the presence of an amplification product of the predicted size following fractionation of the PCR products by electrophoresis and ethidium bromide staining. The primer pairs of each cytokine were selected from published DNA sequences based on previously described guidelines. Saiki, R.K. 1990, Amplification of genomic DNA. In PCR Protocols, M. A Innis., D.H. Gelfand, J.J. Sninsky, T.J. White, editors, Academic Press, San Diego, CA. 13-20. The nucleotide sequences for 5* and 3¹ primers respectively were: /?2-microglobulin (02-MG) , TCTGGCCTTGAGGCTATCCAGCGT (SEQ ID NO:l) and GTGGTTCACACGGCAGGCATACTC (SEQ ID NO:2); IL-2, CATTGCACTAAGTCTTGCACTTGCACTTGTCACA (SEQ ID NO:3) and ATTGCTGATTAAGTCCCTGGGTCTT (SEQ ID NO:4); IL-4, CTCACAGAGCAGAAGACTCTGTGC (SEQ ID NO:5) and AAGCCCGCCAGGCCCCAGAGGTTCCT (SEQ ID NO:6); IL-6, TACATCCTCGACGGCATCTCAGCCC (SEQ ID NO:7) and CTGGTTCTGTGCCTGCAGCTTCGTCAGC (SEQ ID NO:8); TNF-α CGCTCCCCAAGAAGACAGGGGGCC (SEQ ID NO:9) and

GATGGCAGAGAGGAGGTTGACCTTGGT (SEQ ID NO: 10); TNF-/3, CCCAGGGGCTCCCTGGTGTTG (SEQ ID NO: 11) and GTGGGTGGATAGCTGGTCTCCCTGGG (SEQ ID NO:12). The amplification product for each primer pair was confirmed by demonstrating hybridization of an internal probe to the predicted PCR amplification product following Southern blotting. Human specificity for each primer pair was verified by demonstrating that amplification of the predicted product did not occur after RT-PCR or RNA extracted from mouse mitogen-activated mononuclear cells. All samples were analyzed by RT-PCR for the presence of mouse or human β2- microglobulin to verify the integrity of the sample mRNA and the efficiency of subsequent reverse transcription. Positive and negative controls were included in all runs and suggested guidelines for PCR quality control were followed as described above.

The detection of human 02-MG mRNA indicated that human cells were present throughout the peripheral lymphoid compartment of an HIV-l infected SCID-hu mouse, however differences in human cytokine mRNA expression by human cells present in different regions were observed. In a representitive gel shown in Figure 6, analysis by RT-PCR of one HIV-1-infected SCID-hu mouse detected the expression of TNF-α mRNA in the hu-thy/liv implant and spleen, TNF-/3 mRNA in the hu-thy/liv implant, spleen and PBMC, IL-2 mRNA in the hu- thy/liv implant and spleen, and IL-4 and IL-6 mRNA in the hu- thy/liv implant. To assess the effect of HIV-l infection on cytokine expression in SCID-hu mice, the expression of human mRNA encoding TNF-α, TNF-/3, IL-2, IL-4 and IL-6 in the hu- thy/liv implant, PBMC, spleens and lymph nodes of 8 SCID-hu mice was compared with that of 7 HIV-l infected SCID-hu mice. In the 7 HIV-1-infected mice, active HIV-l infection was demonstrated by the detection of tat/rev mRNA in 7 out of 7 of the hu-thy/liv implants, 7 out of 7 spleens, 6 out of 7 PBMC and 6 out of 7 lymph nodes (Figure 7) . Expression of TNF-α, TNF-/3 and IL-2 mRNA was increased in the PBMC (Fig. 7B) , lymph nodes (Fig. 7C) and spleens (Fig. 7D) of the HIV-l infected SCID-hu mice. For example, whereas TNF-α, TNF-/3 and IL-2 mRNA was detected in 1 of 8, 5 of 8 and 1 of 8 SCID-hu mice respectively, they were detected in 5 of 6, 6 of 6 and 3 of 6 HIV-l infected SCID-hu mice, respectively. Taken together, this data suggest that in vivo HIV-l infection stimulates cytokine production by human T-cells. EXAMPLE 7

Construction of SCID-huBM mouse .

Human fetal bone marrow (HFBM) cells were obtained from 18 to 23 gestational week fetuses after the elective termination of pregnancy by lavaging the marrow cavities of the fetal long bones with PBS within 8 hours of availability followed by Ficoll-Hypaque density centrifugation. The interphase cells were collected and washed twice in PBS, counted and either resuspended in PBS at 8 x 10⁷ cells/ml (uncultured-HFBM cells) or cultured in RPMI 1640 with added FCS (10% v/v) , penicillin/streptomycin (100 U/ml) , and Gentamicin (500 ug/ l) at 4 x 10⁶ cells/ml at 37°C (cultured- HFBM cells) . After 4 days of culture, the adherent and non- adherent cultured bone marrow cells were harvested, washed twice in PBS, counted and resuspsended at 8 x 10⁷ cells/ml.

Alternatively, nonadherent cells were harvested by collecting the cells obtained after the culture flasks were washed with ice cold PBS and the adherent cells were obtained by gently scraping the culture flask with a cell scraper (Costar, Cambridge, MA) .

Six to eight week old C.B-17 scid/scid mice were exposed to sublethal irradiation with 400 cGy and within one hour were anaesthetized with Pentobarbital (40-80 mg/kg) and than injected intravenously with 4 x 10⁷ uncultured HFBM cells or 4 x 10⁷ cultured HFBM cells in a total volume of 500 μl. The resulting BTCD-huBM mice were assayed for the presence of mononuclear cells in the peripheral lymphoid compartment. The mice were sacrificed eight weeks after construction and the cells present in the bone marrow, lymph nodes, spleen and peripheral blood were assessed for their expression of the human leukocyte common antigen, CD45, which is present on human mononuclear cells at all states of differentiation.

For three-colored flow cytometric analysis, cells were stained with PerCP-, PE-, and FITC- conjugated mouse mAb.

PerCP-conjugated mouse mAb to human CD45, FITC-conjugated mAb to human CD10, or CD20, and PC-conjugated mAb to human CD34 was obtained from Becton Dickinson (Mountain View, CA) ; FITC- conjugated mAb to human CD13 and PE-conjugated, affinity purified (minimal crossreactivity) Fab'2 fragment to human Fc5μ-IgM, and FITC-conjugated, affinity purified (minimal crossreactivity) Fab'2 fragment to human Fc-IgG were obtained from Jackson Immunoresearch Laboratories, Inc. (West Grove, PA) ;

Flow Cytometric Analysis .

At indicated times, SCID-huBM mice were killed and mononuclear cells were obtained from the peripheral blood, lymph nodes and spleen. Bone marrow cells were obtained by lavage of the mouse femurs with ice-cold PBS-NaN₃. All cell suspensions were washed twice, counted and resuspended at 1 x 10⁶ cells/ml in PBS-NaN₃. Cells were treated with the indicated antibodies for 30 min. at 4°C and 10,000 events analyzed by FACS. Control HFBM cells were used to set gates for lymphoid (gate Rl) or myeloid (gate R2) cells. Cut off values for the quadrants were set after compensation for PE versus FITC versus PerCP emission based on the analysis of single, double and triple staining of positive and negative control samples (human fetal bone marrow and C.B-17 mouse bone marrow, respectively) as well as appropriate FITC, PD or PerCP labeled isotype controls. All the antibodies used were species-specific and minimally cross-reactive as determined by performing the appropriate control experiments. As shown in Figure 1, pre-culturing HFBM cells significantly increased the engraftment of the mouse bone marrow with human cells (p < .0005). Furthermore, while a high degree of engraftment (> 10% human CD45+ cells) was detected in the bone marrow of all of the SCID mice (n=ll) infused with cultured HFBM, no human CD45+ cells were detected in the bone marrow of 8 of 15 SCID mice transplanted with uncultured HFBM. The greater extent of bone marrow engraftment observed after injection of cultured HFBM cells was paralleled by a higher degree of peripheral reconstitution. A significantly higher percentage of human CD45+ cells was detected in the spleens (p < 0.0066), lymph nodes (p < 0.0001) and the peripheral blood (p < 0.0008 ) of irradiated SCID mice that received cultured HFBM cells than in the spleens, lymph nodes and peripheral blood of irradiated SCID mice that were infused with uncultured HFBM cells. Thus, the successful engraftment of the bone marrow and peripheral lymphoid compartments of SCID mice with CD45+ human cells was markedly increased by pre- culturing HFBM cells prior to infusion.

To investigate the role of human stromal precursor cells on the engraftment of SCID-BM mice, cultured HFBM cells were separated into an adherent layer enriched for stromal cells precursors and non-adherent cells depleted of stromal cell precursors. Irradiated SCID mice were intravenously injected either with non-adherent cultured HFBM cells or with the adherent and non-adherent cultured HFBM cells. As shown in Figure 9, depletion of the adherent population from cultured HFBM cells markedly decreased the capacity of the cultured HFBM cells to reconstitute the bone marrow of the SCID-huBM mice. This result suggests that human stromal precursor cells may play a role in reconstitution of SCID-huBM mice with human hematopoietic cells.

The rate of migration of the human cells to the mouse bone marrow was assessed by examining the distribution of human CD45+ cells and CD34+CD10+ precursor cells 2 hours and 48 hours after intravenous infusion of irradiated SCID mice with cultured HFBM cells (Figure 10) . Minimal change was observed in the percentage of lymphocytes in the peripheral blood of SCID-huBM mice expressing human CD45+ cells detected at two hours (5.67%) or 48 hours (5.10%) after injection. In contrast, from 2 hours to 48 hours after infusion, the population of human CD45+ cells in the bone marrow increased from 0% to 35%. In addition to the temporal difference between repopulation of the peripheral blood and the bone marrow compartment with human cells,there was also a qualitative difference. Whereas only minimal numbers of immature CD34+CD10- precursor cells were detected in the periphery of the SCID-huBM mice at any time point, by 48 hours after injection 2.33% of the lymphocytes in the bone marrow were immature CD34+ CD10- precursor cells. These observations suggest that human precursor cells may selectively migrate to the mouse bone marrow following peripheral injection.

To investigate human hematopoiesis in the SCID-huBM mice, the expression of surface antigens associated with the maturational state of human B-cells in the bone marrow and peripheral lymphoid compartment was assessed by flow cytometry.

A representative dot histogram of the flow cytometric analysis of cells isolated from the bone marrow of a SCID mouse, a normal human fetus, and a SCID-huBM mouse for the expression of human CD45, CDIO, CD20 and slgM is shown in Figures HA-llI. The absence of background staining of SCID mouse cells by three-color flow cytometry using labeled mouse monoclonal antibodies to human CD45 and CDIO (Figure HA) , CDIO and CD34 (Figure HD) and CD20 and slgM (Figure HG) illustrates the high degree of specificity of the three-color flow cytometric assay used for the analysis of surface markers expressed by human cells in the BTCD-huBM bone marrow (Figure HC, F an I) . Using artificial mixtures of different numbers of human PBMC mixed with SCID mouse bone marrow cells, it was ascertained that discrete populations of human cells as low as 0.4% are detectable by three-color flow cytometry using anti- human CD45 (Figure 12) .

The maturational state of human B-cells during the process of B-cell lineage commitment and differentiation in human fetal bone marrow can be sequentially divided based on expression of CD34, CDIO, CD20 and slgM into the most immature cells CD34+CD10-, followed by stage I- CD34+CD10+; stage II- CD34-CD10+CD20-sIgM-; and stage III-CD10+CD20+sIgM+ (Labien, et al., (1990), Leukemia, 4:354-358; Loken, et al. (1987), Blood, 70:1316). The later maturation of human B-cells was examined by assessing the relative number of CD20+sIgM- and CD20+sIgM+ B-cells because human fetal bone marrow cells differ from adult bone marrow cells in that CD10 expression is not lost during the maturation of pre-B-cells (Vlllablanca, et al. (1990), J. Exp. Med., 172:325-334; Uckun, F.M. , (1990), J. Leuk. Biol., 481:138-148) . Figure 13 shows the distribution of CD45+CD34+CD10-, CD45+CD34+CD10+, CD45+CD20-l-sIgM- and CD45+CD20+sIgM+ human lymphocytes in the bone marrow, peripheral blood and spleens of ten SCID-huBM mice. Two months after transplantation, a significant population of precursor CD34+CD10- cells were detected in the bone marrows (1.24% ± 0.37) of the SCID-huBM mice and a smaller population of precursor CD34+CD10- cells were detected in the spleens (0.22% ± 0.009). In addition, whereas in the bone marrow there was a predominance of pre-B-cells (8.66 ± 1.63) over immature/mature B- cells (4.18 ± 0.72), in the peripheral blood there were more immature/mature B-cells (8.13 ± 1.88) than pre-B-cells observed (1.55 ± 0.72) . To determine how B- cell maturation in the SCID-huBM mice compared to normal fetal bone marrow, six individual fetal (20-23 g.w.) bone marrow samples were analyzed for the expression of CD45, CD34, CD10, CD20, and slgM and the mean percentage ( ±SEM) of CD23+ CD10-, CD34+, CD10+, CD34-CD10+, CD20+ slgM-, and CD20+ sIgM+ cells was 2.83 ± 1.01, 13.83 ± 2.07, 56.5 ± 4.68, 21.83 ± 3.07 and 22.33 ± 1.99, respectively. Thus, the relative distribution of B-cells at different maturational states in the SCID-huBM mice was similar. To assess the engraftment of human myeloid lineage cells in the SCID-huBM mice, human cells in the bone marrow and peripheral blood of SCID-huBM mice in the myeloid gate were examined by flow cytometry for the expression of the myeloid associated antigens, CD13 and CD33. Two months after transplantation, human CD13+ CD33+ cells constituted 5.49 ± 1.76 of the cells in the mouse bone marrow (n=ll) and 2.03 ± 0.55 cells (n=6) in the peripheral blood of the SCID-huBM mice (Figure 14) . Thus, in addition to being reconstituted with human B-cells,the SCID-huBM mice were reconstituted with human monocytes as well.

EXAMPLE 8

Detection of human cytokine gene expression by RT-PCR .

The pattern of in vivo human cytokine expression by HFBM cells or by the human cells present in SCID-huBM mice was assessed by using the method of Example 6.

Isolation of RNA. Long bones of BTCD-huBM mice constructed as in Example 7 were frozen in liquid nitrogen, wrapped in aluminum foil, crushed with a hammer and refrozen. Crushed tissue was solubilized in quandidine isothiocyanate buffer solution (4 M Guanidine isothiocyanate/0.IM Tris-HCl/l% 2-mercaptoethanol/0.5% Sarcosyl) and vortexed for 2 min. The nucleic acids were extracted with an equal volume of phenol:chloroform, precipitated with 70% ethanol at -70°C and resuspended in guanidine isothiocyanate buffer solution. The RNA was obtained by cesium chloride (5.7M) density gradient centrifugation, precipitated twice with ethanol, resuspended to a concentration of 1 μg/ml in double distilled-deionized, DEPC treated water, and then stored frozen (-70°C) .

RNA (7μg) in 7 μl of ddH₂0 was added to 4 μl of 5X buffer (250 mM Tris-HCl, pH 8.3/375 mM KCl/15 mM MgCl₂) , 2 μl DTT (100 mM) , 1 μl of random hexamers (BRL-Gibco) and 5 μl mixed dNTPs (2mM each) . The samples were mixed, heated to 65°C for 10 minutes, placed on ice for 5 minutes, 1 μl (200 units) of Superscript® reverse transcriptase (BRL-Gibco) was added, the reaction mixture was vortexed, briefly spun down, incubated at 37°C for 60 minutes and then placed on ice. After reverse transcription of total RNA extracted from the HFBM cells or the bone marrow cells of the SCID-huBM mice, cDNA was amplified by PCR with human cytokine specific primers for 60 cycles of denaturation at 94°C for 1 minute, annealing at 65°C for 1 minute and extension at 72°C for 1 minute. To ensure that the PCR amplification product was of human origin, the primers were designed so that the nucleotide sequence of the 3' end was complementary to a human cytokine cDNA sequence absent on the mouse cytokine cDNA. The primer pairs were designed to yield an amplification product that spanned exon- exon junctions to ensure that the PCR amplification product was derived form mRNA. The presence of the target mRNA was indicated by the presence of an amplification product of the predicted size following fractionation of the PCR products by electrophoresis and ethidium bromide staining. The primer pairs for each cytokine were selected from published DNA sequences based on previously described quidelines. The nucleotide sequences for 5• and 3 • primers respectively were B2-microglobulin (β2-MG) , TCTGGCCTTGAGGCTATCCAGCGT (SEQ ID NO: 1) and GTGGTTCACACGGCAGGCATACTC (SEQ ID NO: 2); IL-3, CCTTTGCCTTTGCTGGACTTCAAC (SEQ. ID NO: 13) and CAGTCAACCGTCCTTGATATGGATTGG (SEQ ID NO: 14);

IL-4, CTCACAGAGCAGAAGACTCTGTGC (SEQ ID NO: 5) and AAFCCCGCCAGGCCCCAGAGGTTCCT (SEQ ID NO 6); IL-5, TTGCTAGCTCTTGGAGCTGCC (SEQ ID NO: 15) and CTTGCAGGTAGTCTAGGAATTGGTTTACT (SEQ ID NO: 16) ; IL-6, TACATCCTCGACGGCATCTCAGCCC (SEQ ID NO: 7) and CTGGTTCTGTGCCTGCAGCTTCGTCAGC (SEQ ID NO: 8); IL-7, CTGTTGCCAGTAGCATCATCTGATTGTG (SEQ ID NO: 17) and CTTGCGAGCAGCACGGAATAAAAACAT (SEQ ID NO: 18); IL-10, CTCCTGACTGGGGTGAAGGGCCAGCCCA (SEQ ID NO: 19) and AGTCGCCACCCTGATGTCTCAGTTTCGT (SEQ ID NO: 20) ; LIF, AACAACCTCATGAACCAGATCAGGAGC (SEQ ID NO: 21) and ATCCTTACCCGAGGTGTCAGGGCCGTAGG (SEQ ID NO: 22); M-CSF, TTGGGAGTGGACACCTGCAGTCT (SEQ ID NO: 23) and CCTTGGTGAAGCAGCTCTTCAGCC (SEQ ID NO: 24) ; GM-CSF, CGGCGTCTCCTGAACCTGAACCTGAGTAGA (SEQ ID NO: 25) and GTTCTCTTTGAAACTTTCAAAGGTGATAGTC (SEQ ID NO: 26) ; and SCF, TCCTATTTAATCCTCTCGTCAAAACTG (SEQ ID NO: 27) and TCTAAAGAATTCTTCAGGAGTAAAGAGC (SEQ ID NO: 28) . The specificity of the PCR amplification was confirmed by demonstrating that the amplification product for each primer pair hybridized to an internal probe complementary to the predicted PCR amplification product following Southern blotting. The specificity of each primer pair for human cDNA was verified by demonstrating that no amplification of the predicted product occurred after RT-PCR of RNA extracted from mouse tissue that expressed the corresponding mouse cytokine. All samples were analyzed by RT-PCR for the presence of mouse or human B2- microglobulin to verify the integrity of the sample mRNA and the efficiency of subsequent reverse transcription. Positive and negative controls were included in all runs and suggested guidelines for PCR quality control were followed. The specificity of the reaction is shown in a representative gel of the amplification products obtained after RT-PCR analysis of RNA extracted from HFBM cells and mouse spleen cells with species-specific primer pairs (Figure 15) . Amplification products of the predicted sizes were visualized after RT-PCR with human-specific primer pairs for human IL-4, IL-7, LIF, M- CSF and SCF of RNA extracted from HFBM cells but not after RT- PCR of RNA extracted from mouse cells (Figure 15A) despite the presence of the corresponding mouse cytokine mRNA in the mouse cells (Figure 15B) .

Because the SCID-huBM mice of Example 7 did not receive exogenous human cytokines and cytokines play an important role in the regulation of hematopoiesis, it was investigated whether or not endogenous production of human cytokines associated with the regulation of human hematopoiesis occurred in the bone marrow of the SCID-huBM mice. The expression of human cytokine mRNA was evaluated by RT-PCR with human mRNA- specific cytokine primers as described above. The results are shown in Table 6.

Table 6. Human Cytokine Gene Expression in the Bone Marrow of SCID-huBM Mice

SCID-huBM Mouse

# 1 # 2 # 3 # 4 # 5 # 6 # 7 # 8 # 9 β2 -m + + + + + + +

IL-1 - - + + - ND ND

IL-3 - - - - - +

IL-4 + + + - ND +

IL-5 - - - - ND -

IL-6 - - - - + +

IL-7 + + + + + +

IL-10 - - - - ND -

LIF + + + + + + +

G-CSF - + — + ND —

3M-CSF - -

M-CSF + + + + + + + ND -

SCF — + - - - - + + +

Two months after transplantation with cultured HFBM, mononuclear cells were isolated from the bone marrow of 9 SCID-huBM mice and mRNA expression of the indicated human cytokine was asessed by RT-PCR. The predicted ethidium bromide stained amplification product for the mRNA of the indicated cytokine was either detected (+) , not detected (-) or RT-PCR was not done (ND) . As shown in Table 6 and Figure 15, the expression of mRNA for IL-7, LIF and M-CSF was detected by RT-PCR in the bone marrows of 9 of 9, 9 of 9, and 7 of 8 SCID-huBM mice examined, respectively. Thus following infusion of HFBM cells into irradiated SCID mice, the HFBM cells engraft in the murine bone marrow and express cytokine genes associated with continued human hematopoiesis.

EXAMPLE 9

Construction of BTCD-huCombo mice .

BTCD-huCombo mice were prepared by implanting six to eight week old scid/scid with human fetal thymus and liver tissue under the kidney capsules as in Example 1 and after a brief recovery of four to six weeks, the implanted mice were sublethally irradiated and inoculated with pre-cultured fetal bone marrow cells as in Example 8. BTCD-huCombo mice were also prepared by first constructing BTCD-huBM mice as in Example 8 and then implanting synergenic human fetal thymus and human fetal liver tissue, which had been implanted temporarily into another BTCD mouse, under their kidney capsules following a four week recovery from the first procedure.

The BTCD-huCombo mice constructed in this manner were analyzed at eight weeks, sixteen weeks and three months later for the expression of human cell surface antigens by lymphocytes in the mouse peripheral blood. The results are shown in Table 7 and Table 8. The data are provided as percentage positive.

Table 7

EFFECT OF IMPLANTATION OF HUMAN thy/liv INTO SCID-huBM MICE ON PERIPHERAL BLOOD

TRANPLANTED MOUSE 8 WEEKS 16 WEEKS TISSUE NUMBER CD45 CD4 CD8 CD19 CD45 CD4 CD8 CD19

Human 44-1 3.1 3 0.2 0 6 4.6 0.4 0 thy/liv 44-2 1.3 1 0.1 0 1.1 0.7 0.4 0 32 11 2.5 16

HFBM+Human 44-3 5 0 0 6 55 13 3 40 thy/liv 45-2 3.3 0 0 3

The data indicate that BTCD-huCombo mice are repopulated with human B- and T-cells. Table 8

LONG-TERM EFFECT OF IMPLANTATION OF HUMAN thy/liv INTO SCID-huBM MICE ON PERIPHERAL BLOOD

MOUSE 3 MONTHS NUMBER CD45 CD3 CD4 CD8 CD19 CD14

63b2 89 43 15 34 1

63bl 10 13

63b3 6 14

65-1 34 14 13 4 0

65-2 25 8 6 8 0

The data in Table 8 indicate that BTCD-huCombo mice are repopulated with human B-cells (CD19) and T-cells (CD3, CD4 and CD8) three months after implantation and engrafting.

Claims

What is claimed is:

1. A BTCD-hu2 mouse having transplanted therein human fetal tissue, human fetal cells or a combination thereof and having a sufficient number of human T-cells, monocytes or combination thereof in the mouse peripheral lymphoid compartment to allow HIV-l infection of said mouse following intraperitoneal inoculation of HIV-l into said mouse.

2. The BTCD-hu2 mouse according to claim 1 wherein said mouse is a BTCD-hu mouse.

3. The BTCD-hu2 mouse according to claim 1 wherein said mouse is a BTCD-huBM mouse.

4. The BTCD-hu2 mouse according to claim 3 wherein the bone marrow of said mouse contains a proportion of CD45+ cells in the range of from at least about 10% to about 80% and said mouse contains at least about 10% to about 40% CD45+ cells.

5. The BTCD-hu2 mouse according to claim 1 wherein said mouse is a BTCD-huCombo mouse.

6. The BTCD-hu mouse according to claim 2 wherein the proportion of T-cells in the peripheral blood of said mouse that is of human origin is in the range of at least about 5% to about 7%.

7. A method of constructing BTCD-hu2 mice capable of being infected with HIV-l intraperitoneal inoculation of HIV- 1, which method comprises implanting alternating pieces of human fetal thymus and human fetal liver tissue under at least one kidney capsule of T- and B-cell deficient mice.

8. The method according to claim 7 which comprises implanting said alternating pieces of human fetal thymus and human fetal liver tissue in substantial contact with one another within the mouse kidney capsule.

9. The method according to claim 8 further comprising the steps of a) anesthetizing B-cell and T-cell deficient mice, b) making an incision in the left and right flanks of said anesthetized mice to allow access to each kidney capsule of said anesthetized mice, c) making an incision in each lower kidney pole of said anesthetized mice, and d) injecting through said incision in each lower kidney pole alternating pieces of human fetal thymus and human fetal liver tissue under each kidney capsule.

10. The method according to claim 8, further comprising the steps of sublethally irradiating the T-cell and B-cell deficient mice and inoculating pre-cultured human cells obtained from fetal bone marrow into said irradiated mice.

11. The method according to claim 10 wherein said precultured human cells are adherent cells.

12. A method of constructing BTCD-huBM mice capable of being infected with HIV-l via intraperitoneal inoculation of HIV-l and capable of maintaining engrafted human cells without an exogenous source of human cytokines, which method comprises the steps of sublethally irradiating B-cell and T-cell deficient mice and inoculating said irradiated mice with pre- cultured human fetal bone marrow cells.

13. The method according to claim 12 wherein said pre- cultured cells are adherent cells.

14. A method for assaying the n vivo dissemination of HIV-l in a BTCD-hu2 mouse which comprises a) inoculating a sufficient quantity of HIV-l into the peritoneal cavity of a BTCD-hu2 mouse to cause HIV-l infection and b) detecting HIV-l in a human implant or human graft in said mouse and in the peripheral lymphoid compartment of said mice.

15. A screening method for determining the efficacy of an anti-HIV-1 drug or anti-HIV-1 therapy, which comprises the steps of a) administering to an HIV-l infected BTCD-hu2 mouse an anti-HIV-1 drug or anti-HIV-1 therapy, b) assaying the peripheral lymphoid compartment or human fetal tissue implant of said HIV-l infected mouse for the presence of HIV-l and c) determining the efficacy of the drug or therapy on HIV-l infection in said mouse.

16. A screening method for determining the toxicity of an anti-HIV-1 drug or anti-HIV-1 therapy, which comprises a) administering to an HIV-l infected BTCD-hu mouse an anti-HIV-1 drug or anti-HIV-1 therapy, b) assaying the peripheral lymphoid compartment or human fetal tissue implant of said HIV-l infected BTCD-hu2 mouse for a change in the quantity or type of human T-cells, monocytes, or combination thereof, and c) determining the toxicity of the drug or treatment toward human cells and/or tissues.

17. A method of assessing the expression of human cytokine genes comprising a) intraperitoneally inoculating a BTCD-hu2 mouse with an amount of human retrovirus sufficient to cause retroviral infection and b) determining the effects of HIV-l infection on human cytokine gene expression in said mouse.

18. A screening method for determining the toxicity of a hematopoietic drug or treatment; which comprises the steps of a) administering to a BTCD-huBM or BTCD-huCombo mouse a hematopoietic drug or therapy, b) assaying the peripheral lymphoid compartment, human bone marrow graft, or combination thereof of said mouse for a change in the quantity or type of human B-cells, monocytes, macrophages, T-cells or combinations thereof, and c) determining the toxicity of the drug or treatment.

19. A method for amplifying an amount of human bone marrow cells prior to implantation of said bone marrow cells in a human patient in need thereof, which comprises the steps of a) engrafting in a BTCD-huBM mouse or BTCD-huCombo human bone marrow cells obtained from the patient or a donor; b) allowing the mouse to recover from the procedure in step (a) for about six to about eight weeks; c) recovering human bone marrow cells from the recovered mouse.

20. A method for generating human monoclonal antibodies which comprises the steps of a) inoculating a BTCD-hu2 mouse having transplanted therein human fetal spleen tissue, human fetal lymph node tissue or a combination thereof and containing human B-cells and antigen presenting cells with a sufficient amount of antigen to cause said mouse to generate antigen-specific antibodies; b) collecting B-cells from the peripheral blood of said mouse; c) fusing the collected B-cells with a permissive myeloma cell line, to thereby produce a hybridoma; d) isolating antibody producing cells produced in step (c) ; and e) purifying monoclonal antibody produced by the antibody producing cells of step (d) .

21. The method according to claim 20 further comprising the step of transforming B-cells from step (b) with Epstein Barr Virus and in step (c) fusing the transformed cells with a permissive myeloma cell line.

22. A method of assessing the efficacy of bone marrow transplantation which comprises the steps of a) pre-treating purified human bone marrow cells with a hematopoietic drug or treatment; b) engrafting the cells from step (a) into a BTCD- hu2 mouse; and c) determining the degree of engraftment in said mouse.