Cross-Reference to Related Application This application is a Continuation-in-Part of U.S. application Serial No. 933,789, filed November 24, 1986.

Background of the Invention

The invention described herein was made with the assistance of National Institutes of Health Grant No. 1-P01-HD19937-01A1. The Government has certain rights in this invention. Although the mechanisms and agents involved in the mammalian immune response reaction are not fully understood, techniques ^' to manipulate the immunologic response have great therapeutic potential. This is especially apparent in the care of pathological con- ditions which result in suppression of the normal immune responses and which are not readily amenable to treatment by conventional drug-based therapies. Such conditions include certain viral infections, and the immunosuppression caused by the administration of sπti- cancer drugs, drugs employed to treat autoimmune diseases, and those used to inhibit organ rejection following transplantation.

The capacity to respond to immunclogic stimuli rests primarily in the cells of the lymphoid syste~. During embryonic life, a stem cell develops, which differentiates along several different lines. For example, the stem cell may turn into a l-ymphoid stem cell which may differentiate to form at least two distinct lymphocyte populations. One population, called T lymphocytes, is the effector agent in cell- mediated immunity, while the other (B lymphocytes) is

the primary effector of antibody-controlled, or humoral immunity. The stimulus for B cell antibody production is the attachment of an antigen (Ag) to B-cell surface immunoglobulin. Thus, B cell populations are largely responsible for specific antibody CAb) production in the host. At times, and for certain Ags, B cells require the cooperation of T cells for effective Ab production.

Of the classes of T lymphocytes, T helper (Tj-,) cells arε antigen-specific cells that are involved in primary immune recognition and host defense reactions against bacterial, viral, fungal and other antigens. The T cytotoxic (T _c ) cells are antigen-specific effec¬ tor cells which can kill target cells following their infection by pathological agents.

While T helper (T^) cells are antigen- specific, they cannot recognize free antigen. For recognition and subsequent T, cell activation and pro¬ liferation to occur, the antigen must be presented to receptors or a receptor complex on the T^ cell together _^ with major histoco patibility complex CMHC) class II products, ^' or "leukocyte antigens". Thus, T^ cell recognition of pathogenic antigens is HC class II "restricted" in that a given population of T^ cells must be either autologous or share one or more of the restricting leukocyte antigen (LA) specificities expressed by the MHC of the host. Likewise, T _c cells recognize Ag in association with class I MHC LAs.

In the case of T^ cells, this function is per- formed by a limited number of specialized cells termed "antigen-presenting cellsj' (APC). It is now well- established that T helper (Tj-|) cells recognize pro¬ cessed soluble antigen in association with class II MHC LA, expressed on the surface of macrophages. Recently, other cell types, such -as resting and activated B

cells, dendritic cells, epidermal Langerhan's cells and human dermal fibroblasts, have also been shown to pre¬ sent antigen to T cells. Epstein-Barr virus trans¬ formed human lymphoblastoid B cells (LCL) have been shown to present tetanus toxoid, M . Leprae and Candida albicans to autologous antigen-specific T cells.

If a given T _n cell possesses receptors or a receptor complex which enable it to recognize the MHC-class II LA-antigen complex, it becomes activated, proliferates and generates lymphokines such as interleukin 2 (IL-2). The lymphokines in turn cause the proliferation of several types of "killer cells", including T _c cells and macrophagεs, which can exhibit antimicrobial and tumoricidal activity. After sti ula- tion subsides, survivors of the expanded T^ cells remain as memory cells in the body, and can expand rapidly again when the same antigen is presented. The importance of T cells in the recovery from acute viral infections has been well established for some viruses, in particular the myxovirus, (influenza), the pox- viruses (ectromslia and vaccinia), and the arenavirus, (lymphocytic choriomeningitis virus).

Despite the well-established importance of the defense mechanisms of the immune system to the well- being of the host, the therapeutic potential of these agents has not been realized. In a preliminary study, S. A. Rosenberg et al., New England 3. Med. , 313, 1485 (1985) reported that the systemic administration of autologous peripheral blood lymphocytes (PBL) which had been incubated with IL-2, along with additional IL-2, to patients with advanced cancer achieved cancer regression in 11 of the 25 patients treated. Rosenberg et al. termed the incubated PBLs "lymphokine-activated killer (LAK) cells" and reported that they were members of a cytolytic system which is distinct from that of

natural killer cells and cytotoxic T cells. However, subsequent studies employing this therapy have not con¬ firmed the promise of these early results.

Whatever the long-term results of this approach to "adoptive im unotherapy", Rosenberg et al. noted that " LtJ he major difficulty in the application of this approach to the treatment of human cancer has been the inability to generate sufficient numbers of autologous human cells with antitumor reactivity that could be used for systemic therapy." This is particu¬ larly problematic in the case of patients who are immunosuppressed due to infections, cancers, organ transplant or anti-neoplastic drugs and the like.

Likewise, numerous attempts have been made to isolate and maintain homogeneous populations of T _c or T _\ cells and to characterize them in terms of their antigen specificity and MHC restriction. These attempts usually involve the stimulation of mononuclear cells from a seropositive human or murine host with bacterial or viral preparations in combination with -non-prolifsrative APCs, such as irradiated autologous mononuclear ceils (MNCs). Proliferating polyclonal populations of Tj-, cells or T _c cells are cloned by limiting dilution to obtain homogeneous populations and then further proliferated and characterized by a variety of techniques. As noted by Rosenberg et al. in the case of cloned LAK cells, one of the major obstacles in cloning T lymohocytes is the limited availability of autologous, or alternatively, ailo- geneic MHC LA-matched MNCs, especially from clinical subjects.

To overcome this problem, APCs other than autologous MNCs have been employed as APCs. For example, T. Issekutz et al., J. Immunol. , 129, 1446 (1982) first disclosed that autologous Epstein-Barr

virus (EBV)-transformed LCL lines can present antigens associated with tetanus toxoid to tetanus-reactive polyclonal T cells and T cell clones. D. R. Kaplan et al., in Cellular Immunology, 88, 193 (1984) reported the production of three T _c cell clones by the prolif¬ eration of peripheral blood MNCs from a type A influenza immune donor. One of the clones proliferated in the presence of irradiated, virus-infected, autolog¬ ous MNCs or in the presence of irradiated, infected Epstein-Barr virus transformed allogeneic lymphoblas- toid cells (LCL).

B. G. Elferink et a_., Scand . Immunol. , 22, 585 (1985) further showed that autologous and allo¬ geneic Epstein-Barr Virus (EBV )-transformed LCL lines can present antigens associated with- . leorae bacili to M. leorae-reactive cloned T cell lines. A further paper by this group described the isolation of a T cell clone which may recognize only M^ leprae antigens. The cloning method used autologous EBV-transformed LCLs as APCs. The advantage of using EBV-transformed LCLs is that they may provide a continuous and unlimited source of APCs. ^' [ .B.A.G. Haanen et al., Scand . J . Immunol . , 23_, 101 (1986).]

Human cyto egalovirus (HCMV) is a large species-specific herpesvirus (DNA 1.5X10 ⁸ Da) which shares the properties of latency and reactivation with other members of this group. HCMV is ubiquitous (30-70% of adults are seropositive ) but the principal site(s) of latency of HCMV are uncertain, as are the molecular events involved in its reactivation. HCMV infection and reactivation can be asymptomatic, but are associated with appreciable morbidity and mortality in immunosuppressed subjects. For example, HCMV is the most common cause of opportunistic infection in bone marrow transplant patients. More than 80% of those who

develop HCMV pneumonia die despite current treatment modalities.

Whereas monoclonal antibody therapy may be helpful as a prophylactic treatment, it has been suggested that survival of immunosuppressed patients with serious HCMV infections may be dependent on the presence of HCMV-specific T cytotoxic cells. Quinnan et al., in New England J. Med. , 307, 7 (1982) have suggested that HCMV-specific cytotoxic T cell responses in these patients are associated with clinical recovery and cessation of viral excretion, whereas those with active infections who did not develop cytotoxic responses uniformly died.

R. C. Gehrz et 2l., Lancet, 2 , 844 (1977), dis- closed that infants with congenital HCMV infection have an antigen-specific defect in T helper cell prolifera¬ tion which is associated n L z persistence of reσli- cating viral infection for nonths to years αesoite the presence of HCMV-specific antibodies. Acquisition of HCMV-specific lymphocyte proliferative responses appears to be associateα with a diminution in viral excretion. Thus, it is likely that HCMV-specific T helper cells play a significant role in immune defense against this virus in i munocompetent hosts. In patients with reactivated HCMV infection, antibodies directed against HCMV antigens are not pro¬ tective or arε only useful in conjunction with appro¬ priate T cell responses. Peripheral blood lymphocytes (PBL) are also not likely to be useful as therapeutic agents, even if they could be obtained in sufficient quantities. The numbers of T helper cells or T cyto¬ toxic cells reactive with a particular antigen are extrε εly low and thus, sεlεctive expansion of antigen-specific T cells ix_ vitro will be required to obtain sufficient numbers for therapeutic purooses.

Moreover, therapeutic administration of allogeneic PBLs would activate T cells recognizing "foreign" leukocyte antigens, resulting in a mixed leukocyte culture reac¬ tion. This mixed leukocyte culture reaction may acti- vate undesirable non-specific immune responses, or alternatively, induce suppressor cells which might inhibit desired antigen-specific immune responses.

Autologous or allogeneic antigen-specific T cell lines are likely to express a desirεd thεrapεutic activity without attendant complications associated with the mixed leukocyte culture reaction. L. K. Bαrysiewicz et al., Eur. J. Immunology, 13 , 804 (1983) rεportεd thε gεnεration of short-tεrm polyclonal T{-, cell lines by the expansion of MNCs from seropositive subjects with soluble HCMV antigen in the presence of IL-2. When thε MNCs werε co-cultured on autologous HCMV-infected fibroblasts, polyclonal T _c cells were generated, which lysed HCMV-infεcted cells.

However, a need exists for improved methods to gεnerate and maintain populations of T^ cells and T _c cεlls which are specific to antigens associated with viruses such as HCMV ana other pathogenic agents. A further need exists for im unothεrapeutic methods basεd upon thε administration of antigen-specific T lympho- cyte populations of known biological activity to mam¬ malian subjects.

Brief Description of the Invention I. Production of T Cell Lines

A. T Helper Cells

The present invention is directεd to mεthods for the efficient production of a homogeneous popula¬ tion of T helpεr (Tf-,) cells which are specific for a viral antigen, such as an HCMV antigen. It is believed that the present methods provide the basis for the

efficient production of cloned T^ cells while using a minimum amount of antigen and lymphokine support, while maintaining both the antigenic specificity and functional activity of the clones. In its broadest aspect, the present method comprises:

(a) isolating a population of mononuclear cells (MNC) from the blood of a donor mammal, wherein the MNC population comprises T helper cells specific for a viral antigen, and auto¬ logous, non-proliferative, antigen-prεsεnting cells (APC);

(b) combining said isolated population of said MNCs with an amount of the viral antigen effective to cause the proliferation of a f helper cell which is specific for thε antigen;

(c) allowing said T nεlorr cell to proilfεrate for a period of time effective to allow non- proliferating MNCs -3 lose viability; and

(d) clonally expancing ne antigen-specific T helper ceil in the presence of an amount of non-proliferative antigen-presenting cells comprising a mixture of.(i) autologous MNCs, allogenεic MNCs or mixtures thereof; and (ii) autologous iymphoblast,oid cells (LCLs), allo¬ geneic LCLs or mixtures thereof; and an amount of said viral antigen effectivε to proliferate said cloned antigen-specific T helper cell, to yield said homogeneous population, and wherεin said LCLs arε prεsεnt in an amount εffective to increase the proliferation rate of the T helper cell over that caused by the MNCs.

Duriπg the selεctivε proliferation step (c), which may require 1-2 weeks, it can also be effective to add an additional amount of said viral antigen and the autologous, non-prolifεrative APCs, in order to cause the further proliferation of the T helper cell, so as to enhance thε production of a viablε polyclonal population of viral antigen-specific T _n cells prior to step (d). In step (d), antigen-specific monoclonal T^ cells may be derived _. by limiting dilution, or by single cell isolation by micromanipulation or flow cytometry. Antigen-spεcific polyclonal T cell lines, as well as monoclonal antigen-specific J cells, have potential therapεutic benefit. Thus, the term "T cεll line" as used herein may appropriately refer tα an expandεd population of polyclonal T cells which is uniquely reactive with a given antigen, or to a homo¬ geneous population of monoclonal T cells which has been derived by the expansion from a single progenitor T cell . Since the present invention comprises the pro¬ liferation of antigen-spεcific T^, cells, it also pro¬ vides a metjioc for proliferating a homogeneous popula¬ tion of T helper cells hich arε specific for a viral antigen comprising combining said homogεnεous popula- tion of T helper cells with an amount of the antigen and an amount of a mixture of non-prolifεrative-, antigen-presenting cells (APC) effective to cause the proliferation of thε population of T helper cells, wherein said mixture of APCs comprise (i) autologous or allogeneic lymphoblastoid cells (LCL) and (ii) auto¬ logous or allogeneic mononuclear cells (MNC), wherein the LCLs are present in an amount effεctivε to increase the prolifεration ratε o thε T hεlpεr cells over that caused by the MNCs. Preferably, the LCLs are derived from a continuous lymphoblastoid cell line, such as can

be produced by viral transformation, hybridoma tech¬ nology and the like. The APCs are rendered non- proliferative by techniques known to the art, e.g., by x-irradiation, treatment with chemical agents such as mitomycin C and the like.

Both non-proliferative MNCs and LCLs have been found to cause the proliferation of T _n cells when they are derived from the same host as the T^ cells or are allogeneic MNCs or LCLs which share one or more class II restricting antigens (abbreviation: "MHC

LA-matched"). It was surprisingly found that a mixture of LCLs and MNCs can synergistically present antigen to Tft cells, thus substantially increasing their prolifer¬ ation rate over that causεd by an εquivalεnt πumbεr of either type of APC when used alone.

For example, this effect is observed in thε case of a viral antigen such as an HCMV antigen, when a mixture of EBV-transformed autologous LCLs ana autolog¬ ous MNCs are employed as the APCs in a ratio of at least about 1:10. Preferably, the ratio of T _n cells to LCLs added is about 1:1-3 and thε ratio of Tj-, cells to MNCs is about 1:3-10.

Furthermore, only a limited density of LCL is required to substantially augment the expansion of T^ clones. This aspect of the present mεthoα is important from a practical point of viεw in that human antigen- specific T, clones can be expanded into large quanti¬ ties in a relatively short period of time using limited numbers of MNCs. For example, in. accord with the prε- sent method, HCMV-specific T^ clones could be expanded from l-2xlO ^s cells to 10* cells in 2 weeks by using l-2xl0 ⁷ MNC and 1x10 ^s LCL as APC.

It has also been found highly preferable to carry out proliferation steps (c) and/or (d) of the present method in the presence of an effectivε amount of interleukin-2 ("IL-2" or "TCGF").

A further aspect of the present invention com¬ prises increasing the proliferation rate of the viral antigen-specific T^ cells by conducting the initial proliferation step (steps a-c, hereinabove) or the subsequεnt expansion step (step (d), hereinabove) in the presεncε of an amount of monoclonal antibody which is specific for said viral antigen. In one embodiment of this aspect of the invention, the proliferation rate of a T _n cεll spεcific for an HCMV viral antigεn is incrεased by combining MNCs isolated from the blood of an HCMV-seropositivε donor mammal with a monoclonal antibody specific for an antigen present on an HCMV gene product such as a structural protεin. Thε mono¬ clonal antibody is εmployεd in combination with (a) an εffectivε prolifεration-sti ulating amount of said antigεn, and (b) an amount of autologous AF s sεlectεd from the group consisting of (i) non-proliferating MNCs, (ii) a non-proliferating LCL dεrivεd from a con¬ tinuous LCL, and (iii) ixturεs thereof, e.g. , mixtures which can cooperatively interact to further increase the proliferation rate of the T^ cells. Prefεrably, the autologous, non-proliferating antigen-presenting cells are selεctaα from thε group consisting of x- irradiatεd autologous MNCs, an x-irradiated Epstein- Barr Virus (EBV)-transformεd lymphoblastoid cεll iinε (LCL) and,mixturεs thεreof. Thεsε are the prefεrred APCs for use in step (b) of the present method, whether or not monoclonal antibody and/or IL-2 is employed to increase the proliferation rate. B. T Cytotoxic Cells

The prεsent method can also bε εmployεd to produce a homogeneous population of T cytotoxic (T _c ) cells which are specific for an HCMV antigen. This aspect of the prεsent invention comprises the steps of:

(a ¹ ) isolating a population of mononuclear cells (MNC) from the blood of a donor mammal, pre¬ ferably a human, wherein the MNC population comprises T _c cells specific for said viral antigen;

(b ¹ ) combining said isolated population of said MNCs with (i) an amount of autologous, non- proliferative, antigen-presenting cells (APC), e.g., irradiated MNCs, (ii) an amount of said

HCMV antigen, and (iii) an amount of inter- lεukin-2 (IL-2), effective to cause the pro¬ liferation of a T _c cell which is specific for said antigen;

(c ¹ ) allowing said T _c cell to orαlifεratε for a period of time effectivε to allow non- proliferating MNCs to lose viability; ana

(d ¹ ) clonally-exoanaing said antigen-spεci Ic T _c cεll in the presence of i) an 3~0'_.nt of non- ^' proiifεrativε, autologous, antigen-presenting cells; non-proliferative, allogeneic, antigen- presenting cells or mixturεs thεrεof, (ii) an amount of IL-2 and (iii) an amount of said

HCMV antigen, effective to proliferate said cloπeα antigεn-spεcifIc T _c cell, to yield said homogeneous population.

Steps (b ¹ ), (c ¹ ) or (c ^l ) are preferably carried out in the presence of a monoclonal antibody which is specific for an HCMV antigen, and which acts to increasε thε prolifεration ratε of said T _c cell.

As previously described for antigεn-specific T helper cell lines, polyclonal T _c lines reactivε with a

particular viral antigen can be expanded by continuous stimulation of T _c blasts exhibiting cytotoxic activity against target cells expressing the desired antigen. Homogeneous populations of virus-specific T _c clones derived from a single progeny T cell may be obtained by limiting dilution, or by single cell isolation by micromanipulation or by flow cytometry, followed by expansion according to step (d ¹ ).

Other preferred ernbodimεπts of this T _c cεll prolifεration method include (1) the use of whole HCMV- viral antigen or the use of HCMV-infectεd autologous or allogεnεic fibroblasts as a source of cεll-associatεd viral antigen, e.g., an HCMV immediate-early protein; (2) the addition of additional amounts of the viral antigen, IL-2 and the autologous, non-proliferative APCs during the coursε of stεp (c ¹ ) to causε furthεr prolifεration of said T _c cεll, and (3) thε use, in stεp (b ¹ ) or in stεp (d ¹ ), of antigen-presenting cells which further comprise non-proliferativε MNCs dεriveα - ^" rom a continuous autologous MNC line in an amount effective to increase the proliferation ratε of the T _c cells over that caused by the autologous ononuclear antigεn- prεsenting cells. This continuous, autologous MNC line prefεrably co prisεs a virus-transfor εd LCL, i.e., an EBV-transformed LCL.

The prεsent invention is also directed to a method for proliferating a homogeneous population of T _c cells which are specific for an antigen, such as a viral antigen, i.e., an HCMV antigen. Thε mεthod com- prises combining said homogeneous population of T cytotoxic cells with an amount of said antigen, an amount of interleukin-2 (IL-2) and an amount of a mix¬ ture of non-proliferative, antigen-presenting cells (APC) effective to cause the prolif ration of said population of T _c cells, whεrein said mixture of APCs

comprise allogeneic virus-transformed lymphoblastoid cells (LCL) and allogeneic mononuclear cells (MNC). The method further comprises combining said population of T _c cells with an amount of monoclonal antibody specific for said antigen, which monoclonal antibody acts to increasε the proliferation rate of said popula¬ tion of cells.

As used herein with respect to a pathological target such as a virus or an infectεd cεll, thε tεr "antigen ¹¹ refers to a compound such as a polypeptide, polypeptide complex, glycoprotein, nuclεic acid or the like, which εlicits an immune responsε. A pathological targεt antigεn ^' may bε a portion of thε pathological targεt itsεlf, e.g., a viral εnvεlooε glycoprotein, or it may be an antigen exprεssεc by a disεasεd tissuε such as nεoplastic tissuε or a virus-infacted cell. An "antigen-specific" T lymphocyte becomes activated in the presεncε of a single antigεn wnen it is orεsεnted by an antigεn-oresεnting cεll (APC), in— ssociation with a spεcific leukocyte antigen ( AJ exoressed by thε antigen-presenting cell. An antigen-soecifIc 7 _n ceil will proiifεratε in a suitaole mεαium in tne presence of the antigen for which it has soεci icity wnen said antigen is prεsεnr_ed to the T _h ceil oy an APC wnich also exprεsses the leukocyte antigen for Λ-nlcn the T _n cεll has specificity. As noteα hereinabove, in thε case of human T helper cells, thε spεcific leukocyte antigen will be a human MHC class II antigεn, whereas human T cytotoxic cells are spεcific for MHC class I antigens.

C. HCMV Antiαen-Soecific T Cell Lines

Preferred embodiments of the prεsent invention comprise homogeneous populations cf T _c or T^ cells, and methods for the production therεof, wherein a given ^" cell population is specific for an HCMV antigen.

When a T cell or a T cell clone is said to be antigen-spεcific, although thε composition comprising the antigen may be specified, the epitope, or specific antigenic site, on the antigenic composition, is not necessarily known. For example, three differεnt homo- geneous populations of T helper lymphocytes may be specific for one particular HCMV structural protein or glycoprotein. While they all respond to the same anti¬ gen, and thus, have the same antigen specificity, they may have differεnt spεcificitiεs at thε submolεcular level, meaning they may respond to differεrt regions of thε antigen, or eoitopes. Within this contεxt, dif¬ ferent HCMV-specific T lymphocytes have the same submolecular-specificity only if thεy rεcognizε thε samε εpitope on the same HCMV-associatεd antigεn. Thεy may still bε spεcific for the same antigen, however, if thε antigen has a plurality of epitopes as presented by APCs in diffεrεnt instancεs.

For example, the HCMV DNA genome is trans- cribed in sequential order, beginning with the restrictεd transcription of immediate-early genes encoding regulatory proteins required for subsequent expression of early genes. The major immediate-early gene (I-El) encodes a 68 kD regulatory protein which is εxprεssεd on thε membrane of infected cεlls 6-24 hours after onset of infection. HCMV-specific T cytotoxic cεlls arε thought to primarily recognize this immediate-early protein as part of their role in immune surveillancε to prεvεnt rεactivation of latent HCMV. Transcription of early genes precedes thε onset of DNA synthesis. Includεd among thε early gene products are virus-spεcific polyrnsrasεs and kinases nεcεssary for DNA rεplication. These enzymes do not appear to play an important role in immune respoπsεs. Howεver, the late HCMV genεs encode a variety of immunogεnic structural proteins and glycoproteins .

Included among these proteins are disulfide-bridged envelope glycopeptide complexεs, non-glycosylatεd envelope proteins, tegumεnt glycoprotεiπs bridging the viral nucleocapsid with the outer envelope; and matrix proteins forming the internal capsid structure of the virus.

The purification and characterization of a number of immunogenic structural glycopεptidε complexes and reduced glycopeptides from the HCMV envεlopε frac- tion, along with monoclonal antibociεs (MoAbs) spεcific thεreto, is fully disclosεd in U.S. application Sεrial No, 933,789, ^c iled Novεmber 24, 1986. the cisclosure of which Is incorporated by rεfεrence herein. The orimary glycopeptidεs, glycopeptide complεxεs ind MoAbs dεscribed therεin arε summarizεd in Tablε 1, below.

TABLE 1

HCMV Surface Glycopεptide Complexes and

Glycopeptides Immunoprecipitated by MoAbs

Glycopeptide Glycopeptides

Ion-Exchange HPLC Complex I muno- Immunoprecipitated

Peak* (MoAb Reac- prεcipitated After Reduction tivity) ( ol wt. , kD) (mol wt. , kD)

2 (9E10) ⁺ 93 50-52**

4. (9E10) 450 50-52**, 90, 116,

130, >200

U1C2, 9B7) 130, and >130 50-52, 93 ⁺⁺ , 130

2_ (none) 93 kD Glycopep- 93 kD Glycopep¬ tide not asso- tide*** not ciatεd in Di- Im unoprecipi- sulfidε-linked tatsd by any MoAb

Complex

Thε Detεrgεnt Extraction, H^LC, ^εouctlon, Im uno- prεcipitation and Ξlectropr.oresis Mεthodology employed to o tain these materials from a purified whole virus preparation (designated "Whole HCMV Antigen" in Example I, below) is fully set forth in B. Kari et al., 3 . Virology, 6_0_,_ 345 (1986), the disclosure of hich is incorpo'rated by rεfεrεncε herein.

Thε hybridomas producing these MoAbs have beεn deposited with In Vitro International, Linthicum, MD, and have been assigned the following access codes: Hb 2-29-9B7 ("9B7") = IVI-10117; Hb 2-15-9E10 ("9E10") = IVI-10118; Hb 1-48-41C2 ("41C2") = IVI-10119.

⁺⁺ gpA, **gpB, ***gpC.

The i munαgεnic glycopeptide complexes and glycopeptidεs llstεd on Table 1 arε useful to selεc- tivεly stimulatε the proliferation of T _f -, cells derived from thε blood of an HCMV-sεropositive donor. Like- wisε, monoclonal antibodiεs of known binding spεcifi- city such as 9E10, 4IC2 and 9B7 can bε used to ourify viral antigens. Thesε monoclonal antiooαles can also be usεd to augment antigen recognition by T _n cell clones, thereoy facilitating their exoansion in the presεπcε o f li itεα amounts of viral antigen.

II. Immunot eraov with ^τ Cell Lines

T cell recognition of different HCMV proteins is important in planning the proper course of immuno- therapy with the T cell lines. Thε T _n cells exempli¬ fied herein primarily rεcognize structural proteins of HCMV. They arε, thεreforε, likely to be important in the recognition of cell-free virus (i.e., in cases of acute viremia). Data in mice and humans show that cer- tain T cytotoxic cells rεcognize Immediatε-εarly pro- tεiπs, which arε εxprεssεd on thε surface of infectεd cεlls. Since thesε protεins are only produced after the viral genes are expressεd by the host cells, such T _c cells are important in prevεnting rεactivation of latεnt HCMV and their administration may be indicated for both prophylaxis and specific HCMV treatment.

The preferred T cell lines recognize viral antigens in association with human LA products. Therε- fore, therapeutic T cells must be either autologous or share one or more of the restricting specificities expressed by the patient. In the context of the transfer of isolated T cells to an individual as a means of therapy, allogeneic cells arε cεlls from a donor individual of the same speciεs as the recipient which are major histocompatibility complεx matchεd for an appropriatε human lεukocytε antigεn (HLA) class of antigens expressed by the recipient (abbreviation "MHC LA-matched"). Therεforε, whεn dεaling with a patient with a pathological infection such as chronic HCMV diseasε, onε can εxpand autologous antigεn-spεcific T cell lines ir_ vitro for subsequent T cell therapy via re-administration. Alternatively, when rapid access to T cell clone reagεnts is important to treat acute life- threatening infection, pra-expansion and administration of a purality of expanded allogεnεic therapεutic T cells of known viral antigen-spεcificity and/or HLA type can be employed. This technique is particularly important in the case of a patient who is im uno- supprεssεd.

'Thεrεforε, a further aspect of the present invention is dirεctεd to a method of using T lympho¬ cytes (T cells) for a thεrapeutic trεatmεnt of a mammal, such as a human patient, having a viral infec¬ tion, such as an HCMV infection. The method comprises treating said infectεd mammal with an amount of a ho o- geneous, clonally-expandεd, viral antigen-specific T cell population effective to elicit an increased immune reaction to the viral infεction, wherein said T cεll population is spεcific (a) for at lεast onε lεukocyte antigen (LA) present on thε surface of an aπtigen- prεsεnting cεll (APC) of said mammal, and (b) for an antigen of said virus.

Preferably, the clonally-expanded T cell popu¬ lation will consist essentially of T^ cells or T _c cells. As discussed hereinabove, it Is often prefer¬ able to administer a plurality (or "bank") of homoge- neous, clonally-expandεd, viral antigεn-specific T cell populations in order to (a) promote immune system recognition of antigens exprεssεd during diffεrεnt stages of infection and/or (b) to guaraπteε that an effective number of the total T cεll populations will be MHC- atched to thε recipient.

Therefore, the present method of T cεll Immu- notherapy comprises the administration of a plurality of T cell populations which comprise T cells having specificity (a) for a plurality of differεnt antigens of said virus, and (b) for at least one LA oresent on the surface of an APC of said mammal. Furthermorε, the presεnt εthod can also comprlsε thε aα inistration of a plurality of T cεll ocoulations, each comprising T cells having specificity (a) for at least—one antigen of said virus, and (b) for a oluraiity of different LAs present on the surfaces of APCs of a plurality of allo- typεs of ^' a single spεcies of said mammal, whεrein at lεast onε LA is LA-rεstriction-matchεd with saiα mammal. Polyclonal linεs as well as T cell clones may have significant therapeutic potential. A particular advantage of antigen-specific polyclonal T cell lines is that they include Tj-, and/or T _c cεlls which can rεcognize the target viral antigen in association with all restricting MHC LAs expressed by thε T cεll donor. The advantage of virus-specific Tj-, and T _c clonεs dεrivεd from a singlε progenitor T cεll is that thεy reprεsεnt a homogeneous population with cells of known antigen specificity, functional activity, and MHC LA- restriction specificity. Therefore, they arε likely to

εxhibit maximal therapeutic effect since all cells in the population exhibit the same functional activity. From a practical point of view, however, a much larger "bank" derived from discretε T cεll populations will be required if monoclonal T cells are to be used.

Since thε methods of the prεsεnt invεntion permit the administratien of a therapεutically- effective amount of T _n cells and/or T _c cells, the afflicted mammal may, but need not be, concurrently treated with exogenous lymphokines such as IL-2 (TCGF). Since the administration of large dosεs of IL-2 has bεεn associatεd with advεrse rεactioπs in some patients, the ability to enhance the immune rεspoπsε in the absence of IL-2 can provide a substantial i prove- mεnt in the efficacy of T cell therapy.

Representativε antigen-spεcific T cεll linεs which havε bεεn generatεd in accord with thε prεsent methods and which are representative of the populations which can be used to genεratε the T cell 'banks" discussed above are listεd in Tablε 2, oelow.

Table 2: T Cell Lines

HCMV IVI Access'

T Cell Designation Antigen Specificity Code

T _h a WRC-T3#3 gpA 10124 (SP-CN/LCL)

T _h a WRC-T2*41 gpA n/a

T _h a WRC-T2*88 64 kD matrix protein 10125

Tf- _j a WRC-T2*131 64 kD matrix protein n/a

T _c b SP-CN CA-1 I-E1 ^C 10126 T _c a SP-RK-5 structural protein 10127

^a clonal ^α polyclonal ^c Immediate-Early Protein (68 kD) d Depositεd with In Vitro Intεrnational, Lithicu , MD, in accord with thε Draft Patεnt and Tradεmark Office Deposit Policy for Biological Materials, SNA PCTJ, 32, 90 (1986).

Thε prεsεnt invention has been descriceα pri¬ marily In terms of the production of T _n ana T _c ceils which are spεcific for viral antigεns, anc tne treat¬ ment of viral Infections with homogeneous populations of thesε T cells. Howεvεr, It is bεlievεd that hcno- geneous populations of T^ and T _c cεlls which are spεci¬ fic for antigεns on a widε variεty of pathological targεts can bε prεparεd using thε prεsεnt mεthod, and that thεsε populations will be εffεctivε to augmεnt or stimulatε thε i unε rεsponse to a wide variεty of pathogens or pathological targets .

Defined broadly, a pathological target is an entity within a mammalian body which is The cause of, or a result of, a disease which threatens the health and well-bεing of the mammalian host. The target may

be an entity which is foreign to the host such as a virus, bacterium, fungus, or the like, or it may be a product of a disease or pathological condition, such as a virus-infected cell, a nεoplastic cell, and the like. In specific embodimεnts of thε thεrapeutic treatmεnt of the present invention, the pathological target of the treatmεnt is occasionally a sεcondary infection which is more threatεning to the intendεd rεcipient bεcausε a primary illness has weakeπεd the recipient's ability to respond immunologically to the secondary infεction. Many aspects of the primary illness may weaken the intended recipiεnt's immunε responsε to the secondary infection, including therapeutic treatments which act to suppress aspects of the recipiεπt's immunε system. Such treatments include, . but are not limited to, radiation therapy, chemotherapy and the like.

In summary, the presεnt thεrapεutic εthod can employ allogeneic T cells from individuals of the same speciεs who are MHC LA- atchεd. This will enaole physicians and pharmacists to simply usε pre- formed dosagε forms of allogeneic T cells to treat a patient. Since T cεlls which arε not atchad are simply ineffec- tivε in aiding the patient, the dose may contain many cells, some of which are matched and will aid the patiεnt and somε of rf ^' nich arε not matchεd and will not aid the patient.

In this way, a singlε dosagε form may bε used to treat a number of patients sharing one or orε LAs exprεssεd by thε allogeneic therapeutic cells. In addition, the dose may contain both T _c and T _j -, cells which have different MHC class restrictions. Obviously, such a therapεutic trεatment doεs not requirε the preparation of doses on a case-by-casε basis. This means that the T cells can be expanded in large culture mediums, allowing for efficiεnciεs of

scale. In addition, the cells can be made i mεdiatεly available without requiring individualized collection, leukapheresis and culture.

Notwithstanding the foregoing advantages, however, the most promising feature of this new immunotherapy is the prospect that each dose will con¬ tain antigen-specific T cells. This reprεsents a sub¬ stantial advance over the Rosenbεrg εt al. mεthod which affords only non-specific enhancement of cytotoxic cells. In the present method, a dosage form may be provided which contains allogeneic T^ and T _c cells which are specific for a variety of antigεns which are all associatεd with a particular pathogεn. For instance, a unit dosage may contain a plurality of T^ and T _c cell clonεs which have been individually cultured and then mixed. The dosage form πav contain Th and T _c cells with a pre-selected numoer of LA rεstrictions, and also a pre-selecteα numoer of antigen specificities. If a particular pathogen is κnown to exprεss 10 or 20 known epitopes on a variety o ^~ anti¬ gens, structural or otherwise, the αose may contain appropriate T cells having specificity for eacn.

Treatment methodologies will normally involve the parenteral administration of a theraoeutically effective amount of autologous or allogeneic T cells in a suitable liquid vehicle, such as a physiological salt solution. One such protocol is that provided by 3. A. Rosenberg et al. , New England 3 . Mεd., 315, 1485 (1985), but the number of cells delivεred per dose, and the number of doses administerεd will vary widely, depεnding upon factors dεtεrmined by the clinician, including the pathology under treatmεnt, the physique and the physical condition of the patient and other factors. However, due to the enhanced efficiency of the proliferation method of thε prεsent invεntion, it

ay be possible to remove substantially fewer MNCs from the patient, and to return substantially morε autolog¬ ous antigen-specific T cells to the patient, than called for by the Rosenberg et al. methodology.

Dεtailεd Description of thε Invεntion Thε invεntion will bε further dεscribεd in accord with thε following detailed examples.

EXAMPLE I

1. Preparation of Viral Antiαεns

Human primary fibroblast culturεs grown in Dulbecco's Modified Eagle's Medium (DMEM) plus 10% fetal bovinε serum werε infected with Towne strain HCMV at an multiplicity of infection (MOI) of 1-5. At 3-4 days post infection [ ³ κJ glucosamine (5 uCi/ l, 22 Ci/mM, A εrsham, Arlington Hεights, IL) or [ ^3S s] ethionine (5 uCi/ml, 109 Ci/mM, DuPont/NEN, Boston, MA) was acded as a marker during subsequent purification, for certain of the expεrimεnts described hereinbelow.

Cεlls and cellular debris were removεd from the medium by low-speed centri ugation at 7500 x g for 20 min. Virus was collected from the supernatant by cεntrifugation at 48,200 x g for l hr. The viral pellet was resuspenα ^' ed in Tris NaCl buffer (50 M trishydroxymethylamino mεthanε-HCl, pH 7.4, and 150 mM NaCl) and layered onto 20-60% sucrose gradients made with the same buffer. Velocity sedimentation was done at 131,300 x g for 1 hr. Gradients werε collεctεd and monitorεd for optical dεnsity at 280 nm. A major pεak of optical dεπsity which bandεd at 43% sucrose was collectεd. Thεsε fractions werε dilutεd with Tris NaCl buffεr and virus collεctεd by centrifugation at 131,300 x

g for 2 hrs. This purified whole virus preparation was designated "Whole HCMV Antigen".

In order to solubilize membrane consti¬ tuents, the whole HCMV antigen preparation was re- suspended in 2-4 ml TN buffer (Tris-NaCl buffer (50 mM Tris hydrochloride, pH 7.5, 10 mM NaCl, 2 mM phenyl ethylsulfonyl fluoride)) containing 1% Triton-X 100 (TX-100). After incubation at ambient tεmperaturε for 60 minutes, this solution was layerεd onto a linear 20-60% sucrose gradient for rate zonal cεntrifugation at 120,000 g for 60 minutes. The TX-100 solubilizεd matεrial remained at the top of the gradiεnt, whilε thε HCMV nuclεo- capsids were banded by velocity seαimεntation. Altεrπativεly, detergent extracts of Towne strain HCMV wεrε prepared using 1.0% Nonidet P-40 (NP-40, Sigma Chεmical Company). In this mεthod, 10-15 mg of thε purified whoiε HCMV antigεn werε susoεndεd in 3-5 ml of TN buffεr containing 1% NP-άO. Detergent-virus suspensions were stirred for one hour at room te oεraturε. The extracted proteins were separatεd frd insoluble protεins by nigh- speed centrifugation for 1 hour. Thε εxtract obtainεd by εithεr of thεsε εthods was dεsignatε "HCMV dεtεrαεnt εxtract".

Whεrεas HCMV-spεcific T nεlpεr cells appear to recognizε HCMV protεins and glycoproteins contained within the whole HCMV antigen prepara¬ tions, HCMV-specific cytotoxic T ceils rεcognize primarily HCMV-encodεd protεins εxpressεd on the surface of infectεd cells. In particular, the major immediatε-εarly (I-El) protein, which is εxprεssεd on thε cell membrane within 6-24 hours of infection, appears to bε important in expansion of HCMV-specific cytotoxic T cεlls.

Therεforε, autologous fibroblasts or allogeneic fibroblasts sharing one or more class I lεukocytε antigεns recognized by the T cell donor were infected with Towne strain HCMV at a ultipli- city of infection of 5-10 for 3 hours. In some cases, the cells were infected in the presence of cyclohexamide (50 μg/ml). After removal of cyclo- hεxamidε, Actinomycin D (5 μg/ml) was addεd to prε- veπt further DNA transcription, thereby allowing for selεctivε εxprεssion of I-E gεnεs of HCMV.

Infεcted fibroblasts not treated with cyclohexamide and Actinomycin D exprεssεd both I-E and late gεnε products of HCMV. Thεsε prεparations wεrε designated "cell-associated viral antigen". Herpεs si plεx virus type I (HSV-I) was partially purified as dεscribed by R. C. Gehrz et al., Lancet , 2_, 844 (1977) and heat-inactivatεd at 56°C for 1 hr. to yiεld "HSV Antigεn".

EXAMPLE II

1. Generation of HCMV-Soecific T Helper Cell and T Cytotoxic Cell Lines

HCMV-specific T cell blasts were preparεd in 25 cm ² upright tissue culture flasks by stimulating 10 x 10 ^s mononuclear cells (MNC) at a concentration of 1 x 10 ⁶ MNC/ml with 10 μg of heat-inactivated whole HCMV antigεn (56°C, 1 hr) suspεndεd in RPMI 1640 mεdium supplemented with 15% heat-inactivatεd HCMV- seronegative human serum (PHS). After 7-10 days at 37°C in 5% C0 atmosphεrε, the HCMV-specific blasts were either further expanded in bulk culture, or cloned by limiting dilution in 96 well U-bottomεd tissuε culturε plates at 0.3 cell/well in the pre¬ sence of x-irradiated (5,000 R) autologous MNC

TRR (MNC ) as feεdεr cεlls, wholε HCMV antigεn at a

concent.ration of 1 μg/ml (for Tj-, cells) or cell- associated antigen at a fibroblast:MNC ratio of 1:20 (for T _c cells), and 10-20% interleukin-2 (IL-2) (TCGF, Biotest, Frankfurt, Germany). There- after, cεlls werε refed εvery 3-4 days with fresh IL-2-containing media. Whole HCMV antigen and autologous x-ir-radiated feeder cells were added to the media every 7-14 days.

Growing cells were fed and transferred to 96 well flat-bottomed tissue culture plates. Grow¬ ing clones were then transferred to 24 well plates for further expansion and then reseeded into 25 cm ² tissue culture flasks for large scale production. Expanded clones were subcloned by a second limiting dilution to ensurε clonality. Clonεs wεrε dεveloped from 4 diffεrεnt HCMV-sεropositivε donors yiεlding morε than 100 individual T cεll lines.

2. Charactεrization of the HCMV-T^ and T _c Cεll Liπεs and Clonεs

All of thε T _n and T _c linεs and clones employed in the Examples hereinbelow were characterizeα as to phenotype, prolifεrative rεsponsεs, IL-2 production and cytotoxic activity as follows:

A. PhεnotvQε Analvsis

T cell lines wεrε analyzεd for expression of CD phenotypic determinants using the mono¬ clonal antibodiεs 0KT3 (total T cεlls), 0KT4 (hεlper/inducer T cells ^" ), and 0 T8 (cytotoxic/ suppressor T cells) (Ortho Pharmaceuticals Inc., Raritan, NO) using an indirect immuno- fluorescencε assay. Fluoresceπcε was detected eithεr by fluorεscεncε microscopy using a Zeiss flourescent microscope or flow cytometry

using the EPICS sp cell 541 (Coulter Corp., Hialeah, FL). Polyclonal T _j -, lines wεrε pre- dp inantly CD3+4+8-; all T ₎ -, clonεs εxpressed thε CD3+4+8- phenotype. T _c lines were prε- dominantly CD3+4-8+; T _c clonεs wεre CD3+4-8+.

8. Lymphocyte Proliferation.

T cell lines and clones wεrε rεsted in tissue culture media in the abseπcε of TCGF over- night, and then restimulatεd for 72 hrs with εithεr wholε HCMV antigεn or HSV antigen to detεrmiπe the specificity of proliferative rεsponsε of T^ clonεs. All T^ lines and clonεs dε onstratεd positivε proliferative responses to HCMV, whereas responses to HSV werε similar to tissuε culturε εdia back¬ ground control. Thus, all T _n lines and clonεs wεre HCMV-specific. T _c cell clones proliferated poorly to HCMV antigen in the ^■ absencε of IL-2.

C. ^' Interleukin-2 (IL-2) Production.

Th clonεs wεre stimulated with whole HCMV antigen, and the supernatants harvεstεd at 24 hrs and assayed for IL-2 activity using thε urinε CTLL-20 (IL-2 dependent) cell line. All T _n clones were shown to produce IL-2, as demonstrated by thε survival and growth of CTLL-20. T _c clonεs wεrε not tεstεd for IL-2 production.

D. Cytotoxic Activity.

All T^ clones wεre tested for cytotoxic activity against the NK target cell line, K562; autologous uninfected and CMV- and HSV- infected fibroblasts as target cells. No NK

or virus-spεcific cytctoxic activity was observed with any of the Tjη clones.

All T _c clones werε testεd for cytotoxic activity against thε NK targεt cεll lines,

K562; autologous and unmatched or partially matched allogεnεic uninfectεd and HCMV-, HSV-, and influenza-infεctεd fibroblasts as target cells. No anti-NK activity was observed with any of the T _c clones. All HCMV-specific T _c clones exhibitεd cytotoxic activity against autologous and MHC LA-matched .HCMV-infεcted fibroblasts, but not against allogeπεic unmatched HCMV-infεctεd fibroblasts. No cyto- toxic activity was observed to any target cεlls infεctεd with HSV or influenza virus. Thus, the T _c clonεs described herein exhibit HCMV-spεcific cytotoxic activity restricted by MHC class I antigens, anc therefore exhibit the characteristics of virus-specific cyto¬ toxic T cells.

EXAMPLE III Proliferative Response of HCMV-Specific T _h Clones to HCMV. Antigεns

Seventeen HCMV-T^ clones obtained from donor WRC werε characterized extensively as to pheπotypε, prolifεrative responses, IL-2 production and cytotoxic activity. All clones werε CD3+4+8-; prolifεrativε to HCMV but not HSV (as described in Example II (2) above), all produced IL-2 when stimulated with HCMV but not HSV, and none exhibitεd cytotoxic activity. Thus, all clonεs wεrε considered to be.T hεlpεr cεlls. All clonεs also exhibited class II MHC restriction specifi- city, as dεtεrminεd by blocking with anti-class II

monoclonal antibodies and rεactivity to wholε HCMV antigen presented by autologous or allogeneic antigen presenting cells sharing the appropriate Dw restriction determinant expressεd in association with DR, DQ or DP.

For studies of T cell reactivity to purified HCMV antigens, x-irradiatεd autologous moπonuclεar cεlls were addεd as a sourcε of antigan-prεsεnting cεlls in a 3-day restimulation culture. One μCi per well of tritiated thymidinε was addεd for thε final 16 hours of culture. Culturεs wεre harvested on glass fiber filters and counted in a liquid scintillation spectrophotometer. Thε results of thesε assays arε summarizεd on Table 3, bεlow.

TABLE 3 Reactivity of HCMV-Specific Th Clones to Whole HCMV Antigen, HCMV Detergent Extract (Envelope Glycoproteins), Precipitate from Detergent Extract (Internal Proteins), and HPLC-Purified HCMV Glycoproteins.

Proliferative Response of HCMV-T _n Clone in Counts per Min. (CPM)

Clone + MHC ¹ ™ -. ^• CMV Antiqen

Clone Whole HCMV Towne

Clone Clone + HCMV Detergent Capsid

No. Alone MNC ^IRR Antiqen Extract Ppt. HPLC-2UR* HPLC-3UR* HPLC-4UR*

WRC-T2#69 98 11/ 73,282 4,855 6,186 178 142 161

WRC-T3+3 67 1,011 55,080 9,311 3,355 1,194 19,706 25,387

WRC-T3*4 111 475 51,185 15,632 1,580 163 365 155

WRC-T3*16 87 142 121,030 4,553 154 142 308 155

WRC-L6 114 142 63,867 5,669 4,433 213 118 427

WRC-L7 241 1,079 70,484 29,131 7,439 2,338 585 520

WRC-L8 124 150 33,977 10,120 2,742 800 189 204 1

WRC-L10 325 981 70,773 32,569 10,307 10,321 74,902 31,063

WRC-L14 217 315 68,558 39,805 3,158 389 1,073 .277

WRC-L25 399 64 89,847 36,149 7,994 614 678 330

WRC-L43 181 670 102,787 16,722 16,389 404 496 1,495

WRC-T3*8 48 1,834 20,297 33,736 15,641 229 172 299

WRC~T2#41 103 480 6,200 3,880 262 1,346 4,215

WRC-L31 122 268 31,497 45,766 26,927 276 114 206

WRC-L34 52 319 12,477 15,077 8,762 196 187 253

WRC-L35 49 134 5,658 22,965 2,682 148 120 211

WRC-L15 94 131 4,108 1,639 98 163 140

♦Purified, unreduced glycopeptide complexes derived from HPLC peaks 2(2UR), 3(3UR) and 4(4UR) as shown on Table 1. HPLC-2UR, UR contains primarily gpB, reactive wit MoAb 9E10j 2UR also contains an im unologically distinct gpC, which is non-reactive with MoAb's reactive with either gpA or gpB. HPLC-3UR contains primarily gpA reactive with MoAb 41C2.

The data summarized on Table 3 demonstrate that all 17 clones werε rεactivε to wholε Towne HCMV antigen, displaying from 4108 counts/minute to 121,030 counts/minute. These results suggest that the immuno- dominant determinant(s) recognizεd by thεsε HCMV-T^ clones are exprεssgd by structural proteins and/or gly¬ coproteins from the virion.

All clones also rεacted significantly to Triton X-100 extracts of Towne HCMV, although to a lessεr extent than that to whole Towne viral particles. Thus, it appears that the majority of T^ clones either rεcognizε primarily protεins or glycoproteins includεd within the envelope of the virus, or proteins that associate sufficiently with thε viral εnvεlopε to bε rεcαvεred in the detεrgεnt extract. Thε lowεr responses compared to whole Townε antigεn may rεflεct inhibitory effects of residual dεtεrgent, or a loss of immunogεnicity duε to structural altεrations rεsulting from thε εxtraction procεαure. Of intεrεst, all T _n clonεs also showed a low but significant proliferativε rεsponsε to prεcipitatεs of viral antigεn following detergent extraction. While it is likely that some residual envelope glycoproteins and tegument protεins are included within the precipi- tate, it is also possible that nucleocapsid proteins, which presumably comprise the dominant protein in the precipitatε, may also εxprεss antigεnic determinants recognized by T helper cells.

Thε T cεll clonεs wεrε then stimulated with purified glycoprotein complexεs obtainεd by anion exchange HPLC -sfractionation of unrεducεd dεtεrgεnt extract of whole HCMV antigen (see Examplε I (1)).

As set forth in B. Kari et al., J. Virology, 60, 345 (1986), and in Table 1, herεinabove, glycopro- tein complεxεs contained within peaks 2UR and 4UR

includε a glycoprotein complex recognizεd by a monoclo¬ nal antibody (9E10). This antibody is cross-rεactive with several other viruses, and is capable of neutral¬ izing HCMV in the absence of complement. In contrast, 5 a separate and distinct group of glycoprotein complexεs are isolated to a high degrεε of purity in peak 3UR, and are recognized by a separate class of monoclonal antibodies which uniquely react with HCMV, and neutra¬ lize HCMV efficiently only in the presence of comple- 0 mεnt.

Following rεduction, thε principal glycopro- tεins dεrivεd from thε 3UR complεx havε molεcular wεights of 130,000; 90,000; and 50,000-52,000 kD; thε two glycoproteins derived from the 2UR/4UR complex

15 recognized- by the 9E10 monoclonal antioody have molecu¬ lar weights of 93,000 and 50,000-52,000 kD. A third glycoprotein has beεn dεmonstratεd in pεax 2UR which is distinct from those glycoproteins recognized by the 9E10 monoclonal antibody. Therefore, therε arε at

20 lεast thraε major glycoprotεin complexεs containεd within thε εnvεlope of HCMV that have been iαεntifiεd as incorporating in excess of 95% of tne total glyco¬ proteins within the HCMV viral envelopε. Sincε these HPLC methods appear to yield glycoprotεin complexes of

25. >95% purity, it was of interest to α ^' emonstratε thε pattern of T^ reactivity to thesε individual glycopro¬ tεin complexes.

As shown in Table 3, of the sevεntεen T _j -, clones testεd for rεactivity with HPLC-purifiεd glyco-

30 protεin complexes, two responded to peaks 3UR and 4UR but not to pεak 2UR. A third T^ clonε rεspondεd to all thrεe HPLC peaks. It is likely that these clones are primarily responding to the major glycoprotein complex which is also recognizεd by monoclonal antibodies which

35 detεct thε glycoprotein complex principally isolated in

peak 3UR. It is likely that tailing of peak 3UR into peak 4UR accounts for thε apparεnt cross-rεactivity . This specificity was confirmed for T^ clone WRC-T3#3, WRC-T2#41, and WRC-LIO, which proliferatεd specifically 5 to glycopeptide gpA produced by m-RNA translation of the gpA gene i__ vitro (data not shown).

Of particular interεst, thε majority of clonεs (14/17) do not appear to recognize determinants exprεssεd on any of thε major εnvεlopε glycoprotεin

10 co plεxεs, as rεprεsεntεd by matεrials 2UR, 3UR and 4UR. Thus, thε immunodominant HCMV antigenic detεr- minant rεcognizεd by thεsε clonεd T hεlper cells may residε εithεr in a nonglycosylatεd membrane protein or in an internal protein. If so, this may be similar to

15 influenza virus, in which antibody recognition involves primarily .the surface glycoproteins (i.e. , hεmagglutinin ) , whεrεas cytotoxic T ly pnocytεs arε known to recognizε primarily intεrnal nucleocapsid pro¬ teins.

20 Table 4 summarizes data regarding the specifi¬ city of HCMV-specific T _π clones reactive with indivi¬ dual HCMV (glyco)proteins . Clonεs wεrε gεnεratεd by initial stimulation of MNCs from seropositivε αonor WRC with whole HCMV antigen to select for T^ reactive with

25. structural proteins and glycoproteins of HCMV.

30

35

TABLE 4 Specificity of HCMV-Specific Th Clones Reactive with Individual HCMV (Glyco)proteins PrQliferative Responses Clone + MNC* -<- Antigen ³

Clone Whole

WRC _h Clone + HCMV ^• HPLC-Purifled

Clones Alone MNC* ^C Virion qpΛ 6 __ Prg _^ t_e i_n HSV Adeno

T3 #3 405 805 35,666 19,706 ^~ N.D ^" . 683 211

T2 #41 871 746 12,410 6, 158 N.D. 584 701

T2 #88 667 162 28,163 112 . 29_j_284 173 114

T2 #131 271 288 28,036 N.D. l iΛ ^{i:35 273}

L33 522 901 5,366 188 488 300 791

L42 341 238 13,360 666 317 225 197 1

a. Proliferative response measured as uptake of ³ H-thymidine in CPM; the following antigens were used:

(i) Whole CMV Virus: Towne strain CMV virus purified from cell culture supernatant on sucrose gradients.

(ii) Envelope glycoprotein gpA: peak 3UR purified by anion exchange HPLC of detergent extract of Towne CMV.

(iii) 64 kD matrix protein: partial purification by reverse phase/gel filtration HPLC.

(iv) HSV, Adenoi whole virus purified from supernatants of infected cell cultures b. WRC Th clones obtained by initial stimulation with whole Towne virus; 18% of Th clones were gpA specific; 33% of T clones were specific for HPLC-purified 64 kD protein, c. MNC* = Irradiated autologous MNC as APC ^' ' -\ d. N.D. = Not determined

Eighteen pεrcent of T^ clones tested werε specific for gpA as described in detail in Example 1.1. Reprεsεntativε εxamples includε WRCT3#3 and WRCT2+41. Thirty-thrεe percent of the HCMV-specific T _h clones

' ^■ 5 were found to be reactive with whole virus as well as a partially-purified matrix protein of molecular weight 64,000 daltons. This protein was obtained by reverse phase HPLC and gel filtration, according to a modifica¬ tion of the method of B. R. Clark, 3. Virol., 49_, 279

10 (1984). Whole Towne HCMV obtained by centrifugation of thε supεrnatant of the HCMV-infεcted fibroblasts was solubilized in 6 M guanidinium chloride and run on rεvεrsε phase HPLC with a C-18 column. Proteins adsorbed to the column werε εlutεd with acεtonitrilε

15 and dεtected by monitoring thε column εffluεnt at 214 nanomεtεrs. Pεaks collεcted were pooled and the 64 kD protein was identifiεd by SDS-PAGE.

Thε 64 kD protεin was further purified from co-elutεd proteins by size-εxclusion chromatography 0 using TSK 4000 and TSK 3000 SEC columns linkεd in series. An aggregatε form of thε 64 kD protεin was isolatεd as dεtεrminεd by SDS-PAGE using this method.

Clones WRC-T2/88 and WRC-T2/131 are reprεsεn¬ tativε of HCMV-spεcific Th clonεs which proliferated in 5 response to the 64 kD matrix protein. Of interεst, . _* many of the clones (i.e., WRC-L33 and WRC-L42) were non-reactivε with εithεr the major eπvεlopε glycopro¬ tεin, gpA, or thε abundant 64 kD intεrnal matrix pro- tέin. Thus, it is apparεnt that HCMV-specific T helper 0 cells exist to a variety of structural HCMV proteins.

EXAMPLE IV

Cvtctoxic Responses of HCMV-Scecific T _c Lines and Clones to HCMV Antigens 5 Polyclonal and monoclonal T _c cells were assayed for cytotoxicity against the NK targεt cεll,

K562; and autologous or allogεnεic fibroblasts infected with HCMV, HSV or influenza virus. The following specific methodologies were used:

A. Preparation of Target Cells

1. HCMV-Infectεd Human Skin Fibroblasts (SF) Human diploid skin fibroblasts (SF) were obtained from skin biopsiεs of adult donors and nεwborn forεskins. Cεlls wεrε passagεd at lεast 6 ti εs, and frozεn at l°C/minutε in

Dulbεcco's Modified Eagle Medium (DMEM) con¬ taining 10% fetal calf sεrum (FCS) and 10% dimεthylsulfoxidε (DMSO) and storεd in liαuid nitrogεn. Bεforε usε, cεlls wεrε tnawεd in a 37°C watεr bath, washεd with media, and seeded onto 25 cm ² tissue culture flasκs. 5F .-.ere grown to confluencε, infected with HCMV at a multiplicity of infεction of C.2, ana incu¬ bated until 70-90% of the monolayer demonstratεd cytopathic effect. Ceils were then labellεd with Na2Crθ ₄ (375 μCi in 2 ml mεdium) overnight at 37°C in an atmosphere of 5% CO2. The cells werε washεd three times, and resuspended In RPMI in 10% FCS.. Lfninfεctεd ⁵ⁱ Cr-labεllεd SFs wεrε used as controls.

2. HSV- and Influεnza-infεctεd Skin Fiύroolasts An SF monolayer in a 25 cm ² flask was infected with HSV or influenza at a multiplicity of infection of 5-10 and incubated overnight. T-he cεlls wεrε labεllεd thε samε way as HCMV-infεctεd SF. ^'

K562

The erythroleukemia cell line K562 was used as a source of target cells to measure NK activ¬ ity. K562 cells (1 x 10 ⁶ ) werε suspended in 1 ml medium and 375 μCi of Na2Crθ in a 25 cm ² flask overnight at 37°C in an atmospherε of 5%

C ⁿ 2). Thε cεlls wεrε rεsuspεndεd in RPMI supplεmεntεd with 10% FCS after threε washings.

SF Which Sεlectivelv Exσrεss I-E HCMV Proteins

SFs wεrε infεctεd with HCMV at a multiplicity of infection of 5-10 for threε hours in the presencε of ^' cyclohεxamidε (15 μg/ml). After removal of cyclohεxamidε, Actinomycin-D (5 μg/ml) and Na ₂ Crθ4 ( ³⁷⁵ ^ ^ci ^ wεrε added to the culture. The cεlls wεrε thεn washεd thrεε ti εs and rεsuspended in RPMI in 10% FCS as target cells. The cytotoxicity as.say was per- formed undεr Actinomycin-D to prevent DNA transcription. Uninfected SFs treated with cyclohexamidε and Actinomycin D in parallel werε usεd as control.

B. Cytotoxicity Assay

Three thousand target cells in 100 μl were added to each well of V-bottomed icrotiter plates. Serial dilutions of effector cells in 100 μl werε thεn addεd to thε targεt cells to yield effεctor-to-target ratios of 50:1, 25:1, and 12:1. Thε plates werε centrifugεd at 50 g for 5 minutes at 37°C in an atmosphere of 5% C0 ₂ for 10 hours. One hundred μl of supernatants wεrε harvεstεd and countεd in a gamma countεr (Beckman Gamma 9000). Controls consisted of maximum lysis and spontaneous release obtained by

adding 100 μl of Triton-X-100 in medium respectively to target cells. Calculations are done according to the following formula:

% ⁵¹ Cr release=Experimental CPM-spontanεous releasε CPM

Maximum CPM-spontanεous release CPM Experimental counts per minute represented cells from wells with effector cells.

Specific release (SR) = (% ^sl Cr release HCMV- infected targets) - (% ^sl Cr release uninfectεd targεts). Rεsults wεrε εxprεssed as the mean from triplicate or quadruplicate wells. Spontaneous relεasε from both infected and uninfεctεd SF targεt cεlls was less than 40%.

1. Cytotoxic Activity of HCMV-Soecific Polyclonal T _c Cell Linεs Stimulatεd with Townε HCMV-Infected Autologous Skin Fibroblasts (SF)

Mononuclεar cells from seropositivε donor SP-CN wεrε stimulatεd for 7-10 days in bulk culturε with autologous skin fibroblasts (SF) infected for 20 hours with Towne HCMV. T cεll blasts wεrε restimulated wεεkly with HCMV-infected fibroblasts, autologous irra- diatεd MNC as fεεdεr cεlls and IL-2. Rεsulting polyclonal CD8+ T cεll linεs were εvaluatεd for cyto¬ toxic activity against uninfected and HCMV-infectεd autologous fibroblasts and the NK target cell line K562. The results of this study are presεnted on Table 5, below.

TABLE 5 Cytotoxic activity of SP-CN cell lines stimulated with Towne HCMV-infected autologous skin fibroblasts (SF) Lysis of target cells

HCMV

HCMV 20 hr.- CH-act.D. Infected

Effector Uninfected Infected treated CH-act.D. Cells E:T ratio' Autologous SF Autologous SF Autologous SF Autologous SF K56

SP-CN CA line 1 50:1 9.1 57.5 11.2 28.9 56. 25: 1 5.1 44.4 9.7 25.3 46. 12:1 5.2 40.8 4.8 23.1 36.

SP-CN CA line 2 50:1 6.7 40.7 6 . 6 20.5 31. 25:1 4.6 42.4 6.3 19.2 26. 12:1 4.3 27.9 -3.7 21.6 8.

a. Effector:target cell ratio b. Values expressed as mean specific ^5l Cr release from quadruplicate wells

As can be seεn from thε data presented on Table 5, both cell lines exhibited cytolytic activity against autologous HCMV 20 hour-infected fibroblasts expressing immediate-early and late viral proteins; and against CMV-infεctεd autologous fibroblasts that had been treated with cyclohexamidε (CH) and Actinomycin D (act. D) to selectivεly εxprεss thε immεdiatε-εarly proteins of HCMV. Both lines also exprε≤sεd signifi¬ cant lεvεls of NK-like cytotoxic activity.

2. Cytotoxic Activity of HCMV-Specific T _c Cell Lines and Clonεs Stimulatεd with Sucrosε Gradient- Purified Towne HCMV Viral Particles

CD8+ T _c cell lines and clones werε obtained from seropositivε donor SP-RK by initial stimulation of MNC with sucrose gradient-purified Towne HCMV viral particles, followed by repeated stimulation with HCMV viral particles, autologous irradlateα MNC as fεedεr cεlls, and IL-2. Thrεe CD8+ lines and tne CD3÷ 5P-RK clone *3 lysed HCMV-infected autologous fibroblasts but not allogeneic unmatcheα HCMV-infectec fiorcolasts or autologous fibroblasts Infεctεd with εithεr HSV or influεnza virus as shown in Table 6, Dεlow.

TABLE 6

Cytotoxic activity of SP-RK T cell lines and clone stimulated with sucrose gradient-purified Towne HCMV viral particles

Lysis of target cells

Influenza HCMV 20 h:r.-

HCMV 20 hr.- HSV Virus Uninfected Infected 1 1vlHC

Effector Uninfec :ted Infected infected Infected MHC Class I Class I Cells E:T ratio SP-RK SF SP-RK SF SP ^■ -RK SF SP-RK SF Unmatched SF Unmatcheid SF K5 SP-RK line 7 50:1 0.6 20.4 -1.6 4.1 6.8 2.5. 13

25:1 1.6 19.0 0.0 4.4 8.3 3.0 11

12:1 1.8 15.7 -0.9 0.7 2.5 1.9 11

SP-RK line 1 . 50:1 2.2 34.4 -3.7 5.7 5.6 2.8 11

25:1 2.0 33.1 -0.4 6.7 6.3 . 4.9 9

12:1 -1.6 28.5 -7.2 4.6 5.1 1.8 6

SP-RK line 4 50:1 3.0 21.3 -2.9 7.0 6.3 2.9 18

25:1 3.4 22.3 0.8 7.2 6.3 8.9 18

12:1 2.8 18.6 0.5 0.6 5.1 ^■ 1.9 19

SP-RK clone 3 50:1 1.3 33.4 2.5 5.2 7.5 1.1 17

25:1 2.8 33.0 0.9 7.5 7.7 7.3 18

12:1 2.3 36.6 •0.1 6.5 .8.8 11.3 11

These lines expressεd little cytotoxic activ¬ ity against the K562 NK target cell. Thus, these T cell lines and clone are charactεristic of HCMV- specific, MNC class I LA-rεstrictεd T cytotoxic cεlls which recognize an antigεn(s) εxprεssεd by wholε viral antigεn <(i.ε., structural protεins).

EXAMPLE V Augmentation of Antiαen-Inducεd Prolifεration of

HCMV-Spεcific T _h Clonεs Using Polyclonal

Anti-Sεrum from SεroDositivε Oonors and

HCMV-SDεcific Monoclonal Antibodiεs

1. Monoclonal Antibody. HCMV-T _n clonε WRC-T3*3 was used at a concentration of 1.5 x 10"* cells pεr wεll as a source of responder cεlls. Autologous x-irradiatεd MNC (ID ⁵ ) wεre addεd as a sourcε of antlgεn-prεsenting cεils. An optimal dilution of whole HCMV antigen or 0.1 ug/ml of HPLC-purifiεd 3UR was added as a source of HCMV-specific antigenic stimulation. Serial dilu¬ tions of monoclonal antibodies 4IC2, which recog¬ nizes the immunodominant glycoprotein A (gpA) or a control monoclonal antibody 35F10, wnich rεcognizεs a nonglycosylatεd protεin designated protein D were added at concεntratlons ranging from 10 down to 0.01 μg/ml protεin per well. The results of this expεrimεnt arε sum arizεd in Tablε 7, bεlow.

TABLE 7

MoAb Directed Against HCMV Glycoprotein A (gpA)

Augments thε Prolifεrativε Rεsponsε of gpA-Spεcific

T _n Clonε WRC-T3+3

Proliferative Rεsponsε of WRC-T3*3

MoAb MoAb 41C2 (anti-qpA) MoAb 35F10 (anti-pD) Concentration Whole HCMV Whole HCMV (μq protein/well) 3UR Aq ^b 3JR ____ 10 19,6673 5,674 17,577 2,349

3 32,526 8,774 18,195 3,508

1 34,713 12,528 18,160 ^' 4,691

0.3 41,996 14,271 16,324 6,371

0.1 36,300 8,143 16,618 4,835 0.03 31,486 9,252 19,435 4,349

0.01 14,406 8,442 17,327 3,676

^a Counts pεr minute.

^D Whole HCMV viral antigεn (strain AD169) usεd at a 1:1000 dilution.

The data summarizεd in Tablε 7 dεmonstratε that monoclonal antibody 41C2 significantly augmεntεd thε prolifεrative response of the T^ clone when 3UR or whole HCMV antigen was used for antigenic stimulation. In contrast, no significant augmentation was observed with the monoclonal antibody directεd against protεin D using either 3UR or whole HCMV antigen. Thus, it appεars that monoclonal antibodiεs dirεctεd against an immunogenic glycoprotein complex which specifically stimulates a particular T _j -, clone, is effective to enhancε thε prolifεrative responsε.

Six HCMV-T)-, clones shown to be responsive to whole HCMV antigεn wεrε thεn stimulatεd with wholε HCMV

antigεn plus serial dilutions of either monoclonal antibody 41C2 (to gpA) or monoclonal antibody 35F10 (to protein D). Seropositive and sεronεgativε wholε sεra were also tεstεd undεr thε samε conditions. Thε rεsults of these expεriments are summarizεd in Table 8, below.

TABLE 8

MoAb Rεactivε with HCMV Glvcoorotεin A (qpA)

Auqments thε Prolifεrativε RεsDonsε of

HCMV- -Spεcif.ic Th Clonεs to Wholε HCMV Antigεn

Proliferativε Rεspoπsε of HCMV-T _h Clonεs ⁰

Culturε WRC- WRC- WRC- WRC- WRC- Conditions ³ T3+8 T2*4i T2*69 L10 T3*3

Clone alone 167 453 274 121 237

_MNC ,IIRRRR _{259 2?3 20Q} 2,995 185 and HCMV Ag 19,778 3,076 5,751 3,936 20,734

MoAb_ 1C2 (antI-goA)

10 ug/ml 25,913 6,948 9,565 16,128 25,292

3 ug/ml 26,275 8,627 11,321 17,224 19,370 l ug/ml 22,073 7,815 9,577 17,183 20,599

0.3 μg/ml 21,558 6,867 8,010 14,807 20,489

0.1 μg/ml 16,677 4,100 5,210 11,456 21,482

0.03 μg/ml •17,026 4,202 6,283 8,063 16,002

0.01 μg/ml 16,315 3,294 5,403 7,670 18,069

MoAb 35F10 (anti-pD)

10 μg/ml 18,587 6,737 4,683 7,584 18,840

3 μg/ml 19,100 5,061 6,616 9,488 18,967

1 μg/ml 20,471 5,903 6,382 7,376 17,799

0.3 μg/ml 18,916 8,001 4,826 8,342 18,605

0.1 μg/ml ^' 18,842 3,542 5,404 7,570 . 18,518

0.03 μg/ml 17,439 3,550 6,321 9,226 18,993

0.01 μg/ml 18,248 2,964 5,971 7,262 22,511

Seropositive Whole Serum (Donor CR)

3.3 lambda/ml 51,068 11,412 17,616 16,345 24,918

Seronegative Wholε Sεruπi (Donor SF)

3.3 lambda/ml 27,506 6,784 8,103 15,464 20,445

^a 1.5X10 ¹ * T _n /wεll + 10 ^s autologous MNC ^IRR /wεll + wnolε

HCMV antigεn. ^D Counts pεr minute.

Surprisingly, monoclonal antioody 41C2, directed against gpA, aug entεd the proliferativε response of all six Tj-, clones to whole HCMV antigen, despitε thε fact that only 3 of thε 6 clonεs respond to HPLC-purified gpA in the absence of antibody (see Table 7). The monoclonal antibody dirεctεd against protεin D did not augmεnt the responsε of any of the clones tested. Whole serum from the seropositivε donor augmεnted the response of all clones, wherεas that from

thε seronegativε donor did not. These data suggest that the mechanism of augmentation does not necessarily involve a direct interaction betwεεn thε antibody. and thε specific viral protεin or glycoprotein recognized by the T^ clone.

EXAMPLE VI Presεntation of HCMV Antioεn to T Hεloεr Cεll Clonεs bv MNC, LCL or Mixtures Therεof 1. Epstein-Barr Virus (EBV)-transformεd Lymphoblastoid Cell Lines (LCL).

An autologous LCL from serooositivε donor SP-CN (SP-CN LCL) was εstablishεd by thε mεthod of Sugdεn and Mark, 3 . Virology, 23, 503 (1977) and grown in RPMI 1640 mεdium (GIBCO, Grand Island, NY) supple¬ mented with 10% heat-inactivatεd fetal calf serum (FCS). AMG LCL (DR 2,2/Dw 2,2) and NHS LCL (DR 3,8) werε gifts from Dr. F. Bach (IRC, University of Minnesota, Minneapolis, MN).

Generation of HCMV-SDecific T Ceil Clones.

HCMV-specific T cell clones werε gεnεrated from a seropositive donor (SP-CN). Briefly, mononuclear cells (MNC) wεrε culturεd at I0 ⁵ cells/ml in RPMI 1640 mεdium supplε εnted with 15% hεat-inactivatεd HCMV-sεronεgativε human sεrum (PHS) and 1 μg/ml wholε HCMV antigεn. Aftεr 7-10 days, T cεll blasts were isolated and cloned by limiting dilution at 0.3 cells/well in round-bottom wells containing 0.2 ml RPMI-15% PHS medium with 10,000 autologous x- irradiatεd (5,000 R) MNC, ^C 1 μg/ml wholε HCMV anti¬ gen (T-owne strain) and 15% IL-2 (TCGF, Biotest, Frankfurt, Wεst Gεrmany ^" ). Plating εfficiencies of 2% to 12% were observed. . The clones werε expanded in culture mεdium containing 10-20% IL-2, and

rεstimulatεd with autologous x-irradiatεd MNC/LCL mixture and 1 μg/ml whole HCMV antigen every 1-2 weeks.

3. Proliferation Assay.

HCMV-specific cloned T _n cells wεrε assayεd 7-14 days a ^' ftεr restimulation. The microcultures wεrε set up in round-bottom plates (Flow Lab., Inc., VA) with a total volume of 0.2 ml/well containing 10 ¹ * HCMV-spεcific T^ cεlls, various concεntrations of x-irradiated autologous MNC and/or LCL and whole HCMV antigen (Towne strain) or HSV antigen. The cultures werε incubatεd for 3 days, and 1 μCi/wεll ³ H-thymidine (New England Nuclear, Boston, MA) was added to each well for the final 16-24 hours of culture. Wells were harvestεd with an automatic cell harvεster, cells collectεd on glass fibεr filter paper, and radioactivity measured in a Beck an 5801 liquid scintillation spectrometer.-

4. LCL Presεnts HCMV Antigεn to HCMV-Spεcific T _h Clones.

A panel of HCMV-specific T^ clonεs gεnεratεd from a sεropositivε donor (SP-CN) wεrε tεstεd for HCMV-spεcific proliferative activity using either x-irradiated autologous MNC (10 ^s /wεll) or x- irradiatεd autologous LCL (10 ^1* - 3 x lOVwell) as APC. HSV antigen was included as an antigεn spεc- ificity control. Table 9 shows thε rεsults obtained using seven clones.

TABLE 9: PROLIFERATIVE ACTIVITY OF HCMV-SPECIFIC T _n CELL CLONES USING EITHER IRRAΪJTAIEI) AUTOLOGOUS MNC OR LCL

T _h CELLS.+ MNC ^IRR3 (CPM) T _h CELLS + LCL ^IRR (CPM)

Th Clone --- +T0WNE AG ^C -H1SV ^C —- +T0WNE AG +HSV

SP-CN/T3-#9 65 ± 6 ^d 8,114 ± 2,045 NO 379 ± 105 21,370 + 3,504 306 + 148

SP-CN/T2-#41 182 ± 79 9,377 ± 671 154 + 85 154 + 32 1,458 ± 379 ND

SP-CN/T2-*34 763 ± 250 10,634 + 506 403 + 31 205 + 158 4,749 + 156 ND n

SP-CN/T3-*16 132 ± 44 10,296 ± 1,468 244 + 136 206 + 60 16,738 ± 1,656 413 ± 84 o^

SP-CN/T3-#8 61 + 7 33,507 + 2,342 252 + 343 150 + 3 1,434 + 245 141 ± 49

SP-CN/T5-*43 82 + 20 35,384 ± 4,301 502 ± 230 ND 1,020 + 104 449 + 67

SP-CN/T2-*131 278 + 367 68,307 + 3,562 78 + 7 119 + 29 1,529 + 216 ND

a x-irradiated autologous MNC (5,000 rads) were added at 10 ^s MNC/well. b χ-irradiated autologous MNC (8,000 rads) were added at lO ¹ * 3x10 ^* */well. ^c Heat-inactivated (56°C/l hr.) whole HCMV antigen (Towne) or HSV antigen was added at 2 μg antigen/well. d The data is expressed as (mean + 1 S.D.) CPM from triplicate wells.

All of the Tη clones listed on Table 9 proliferatεd well in response to HCMV antigen but not HSV anti¬ gen when MNC were added to the cultures as APC. The magnitude of proliferation ranged from 8114 to 68,307 cpm. Most of these clones also exhibited

HCMV antigen-specific proliferativε responses whεn LCL wεrε usεd as APC with a rangε of 1,020 to 21,370. Clonεs SP-CN/T2-131 and SP-CN/T5-43 pro¬ liferated well to HCMV in the presεncε of irradi- atεd autologous MNC, but poorly in the presence of irradiated autologous LCL.

On the other hand, clones SP-CN/T3-16 and SP-CN/T3-9 proliferatεd well to HCMV in the pre- sεncε of irradiatεd autologous LCL, although their proliferation to HCMV in thε prεsεnce of irradiated MNC was no bεttεr than thε other five clones. Thus, thεrε is no apparεnt correlation between pro¬ lifεrative response to HCMV using MNC as APC and that using LCL as APC.

It is well known that antigen-spεcific T hεlpεr (T _n ) cells recognize antigen associatεd with class II MHC products on thε surfacε of APC (P. Erb εt al., J. Exper. Med., 142, 460 (1975)). We studied the MHC restriction of some of these T _n clones using LC as APC, and found all of the clones testεd are class II restricted. LCL from other MHC class II-matched donors also prεsεnted HCMV antigen well to the two high-rεsponding clones (SP-CN/T3-16 and SP-CN/T3-9), b ^' ut poorly to the five low- responding clones. For εxample, SP-CN LCL (autologous, DR2,4/DW2,4) as well as AMG LCL (DR2,2/DW2,2) prεsεnted HCMV antigεn to clonε SP-CN/T3-9, whεreas NHB LCL (DR3,8) did not present HCMV to the same clone.

5. Augmenting Effect of LCL on Proliferativε Rεsponsεs of T _n Cells to MNC and HCMV.

Facing limited supplies of autologous MNC as APC to stimulate HCMV-specific T^ cells and inefficient antigen presentation by autologous LCL when used alone, combinations of MNC and LCL were employed as APC and feeder cells during reactivation of the T^ clones. Surprisingly, the combination of LCL and MNC showed a synergεstic effect on proliferative response as illustrated in Table 10, below.

TABLE 10 PROLIFERATIVE RESPONSES OF T HELPER CELL CLONE SP-CN/T5-43 USING AUTOLOGOUS WRC MNC AND/OR

LCL AS APC

ANTIGEN PRESENTING CELLS ³ H-THYMIDINE INCORPORATED

(CELLS/WELL) (ΔCPM)a

MNC ^IRR LCL ^IRR

— — 73 ± 34 10" 101 ± - 186 3x10" -106 ± 43

10 ⁵ 179 + 117

1 100** --- 6,048 ± 33

10* 10" 13,280 ± 4,019

10* 3x10* 8,923 ± 952

10" 10 ⁵ 5,089 ± 1,293

3x10* 21,541 ± 895

3x10" 10* 43,336 ± 10,489

3x10" 3x10" 33,730 ± 4,593

3x10" 10 ⁵ 16,732 ± 1,304

1 100 ^3s - —-—-— 58,890 ± 4,121

10 ⁵ 10" 77,493 ± 8,728

10 ^s 3x10* 71,544 + 1,938

10 ⁵ 10 ^s 47,641 ± 2,313

a The proliferation assays werε sεt up by using 10* T cells/well, 10*-10 ^s MNC ^IRR and/or LCL ^IRR /wεll and 2ug/ ell whole Towne HCMV antigen. ³ H-thymidine incorporation of clone SP-CN/T5-43 plus SP-CN LCL ^IRR (10*-10 ⁵ /wεll) was in thε rangε of 59-92.

³ H-thymidinε incorporation of clone SP-CN/T5-43 plus SP-CN LCL ^IRR at LCL concentrations of 10*/well, 3xl0*/well and 10 ⁵ /wεll wεrε 629? 1,279; and 2,466; rεspεctively.

LCL alone as APC at a range of 10* LCL/well to 10 ⁵ LCL/wεll did not stimulatε any significant T cell proliferative responsε to whole HCMV antigen above background. Howevεr, thε addition of LCL at 10*

LCL/wεll and 3 x 10* LCL/wεll did augmεnt the 3-day proliferative responsε of Tj-, cεlls to MNC plus wholε HCMV substantially, although 10 ⁵ LCL/wεll sεε ed to suppress the proliferativε rεsponsεs, probably duε to εxcessive cεll dεnsity in the culturε.

Thε same cultures werε εxpandεd In the pre- sεncε of 15% IL-2 for 12 days and cεll counts wεrε pεr- formεd on day 7 and day 12 aftεr the oeginning of thε culture. Thε rεsults of thεsε εxperl εnts arε sum- marizεd In Table II, below.

TABLE 11

NUMBER OF SP-CN/T5-43 CELLS EXPANDED AFTER RESTIMULATION

WITH HCMV ANTIGEN IN THE PRESENCE OF AUTOLOGOUS MNC AND/OR

_{LCL AS APC} a

T _h CELLS RECOVERED

ANTIGEN PRESENTING CELLS NNUUMMBBEERR OOFF CCEELLLLSS RATIO OF T _n CELLS RECOVERED

_IRR (CELLS/WELL) _JRR R REECCOOVVEERREEDD ( (xxllOO ⁵⁵ )) T _h CELLS STARTED

MNC ^" LCL D DAAYY 77 D DAAYY 1122 DAY 7 DAY 12

< <00..11 < <00..11

10* < <00..11 < <00..11

3x10* < <00..11 < <00..11

10 ⁵ < <00..11 < <00..11

10* < <00..11 0 0..88 __ 3 I 10* 10* 0 0..44 8 8..88 1 1 2 299 c

10* 3 -*x<_•1i0n ¹ * 0n.8Q 1 18β.4Λ 3 c 611

I 10* 10 2 6.4 7 21.

3x10* <0.1 1.6 5

3x10* 10* 7.2 46 24 153

3x10* 3x10* 3.2 79 11 263

3x10* 10 ^s 2.4 20 8 61

10 ⁵ 2.4 12.8 8 43

10 ⁵ 10* 8 186 27 620

10 ⁵ 3x10* 10.4 NA 35 NA

10 ⁵ 10 ⁵ 8 78 27 260

a The cultures were set up in parallel with those of Table 6. Cells from triplicate wells under the same conditions were combined on day 3 and expanded b in the presence of IL-2 for 12 days.

Not analyzed. ^• ..

As shown In Table 11, when no MNC were pre¬ sent, T _n cells were not expandεd εven when both LCL and whole HCMV antigen werε added to the culturε. A com¬ bination MNC and wholε HCMV antigεn lεd to a 3- to 43-fold incrεase in T cell numbers during 12 days of culture depending on the amount of MNC that were involved. When LCL werε added along with MNC and HCMV antigen, a 29- to 620-fold Incrεasε in T cell numbers was observed over thε sa ε pεriod of time under the same tissue culture conditions, clearly indicating that LCL were augmenting the expansion of activated T _n cells.

Aliquots of these cultures werε plated in 96-well flat-bottom microtiter platεs on day 7 at 0.2 ml/wεll in triplicatε. Aftεr ovεrnight pulsing with L ³ HJ -thymidine, a similar augmenting effect cf LCL on cell proliferation as measurεα by incorporatεd L ³ Hj -thymidine was observεd as summarized on Table 12, below.

134 + 21

--- io* 423 + 440 3x10* 599 + 262

10 ^s 837 + 170

10* 1,080 + 72

10* 10* 0e 6>,,,779u10-1o08 + 7/ /oo8oo6 i Q ¹ * 3x10* 8 8°,, T7722)^66 + + + 1 1i,,r00m5555R n

10* 10 ^s 3 3,,660066 + + 1 14433

3x10* 3,456 + 122

3x10* 10* 27,063 + 3,488

3x10* 3x10* 24,998 + 1,610

3x10* IO ⁵ 13,483 + 355

10 ⁵ --- 22,369 + 1,590

10 ⁵ 10* 58,315 + 2,159

10 ⁵ 3x10* 52,110 + 3,232

IO ⁵ 10 ^s 37,876 ± 3,533

a The cells were taken from cultures shown in Table 7 on day 7 and added to ffllaatt--bloottttoomnieedd wweellllss aatt 00..22 mmll//wweellll iinn ttrriipplliiccaattee,. The cultures were labeled with [. ³³ HHj]--tthymidine for 18 hrs. before harvesting.

Thesε results confirmed the feasibility of using both MNC and LCL as feeder cells along with anti¬ gen in activation and expansion of T _h. clones. The extent of Th _, cell expansion is proportional to the con- centration of MNC ^IRR in the range of iO* to 10 ^s

MNC/well, but not proportional to the concentration of LCL ^IRR . The optimal ratios of LCL to cloned T _h cεlls arε in thε rangε of about 1:1 to 3:1.

Samples of antigen-prεsεnting LCL cεll lines SP-CN/T3-43 (Access Code: IVI-10122); and SP-CN/T5-43 (Access Code: IVI-10123) have also been dεpositεd with In Vitro Intεrnational, Linthicum, MD, in accord with the Draft PTO Deposit Policy for 3iological Materials, BNA PTCJ, 3_2_, 90 (1986). Cultures of these deposited cεll linεs will bε madε availablε to the public upon the grant of a patεnt basεd upon thε present application. It is to bε understood that the availability of a deposit does not constitute a license to practice tne suojεct invention in derogation of patεnt rights granteα by tne United States Govεrnment.

While certain representations embodiεd arε described herεin for thε purposes of Illustration, it will bε apparent to those skilled in thε art that modi- fications thεrein may be madε without departing from the spirit and scope of the invention.

We wish to thank Children's Hospital, St. Paul, Minnesota, for generous support and encouragement in this work.