Multimer technologies for detection and adoptive transfer of antigen-specific T cells

Abstract

Identification and purification of antigen-specific T cells without altering their functional status are of high scientific and clinical interest. Staining with major histocompatibility complex (MHC)-peptide multimers constitutes a very powerful method to study antigen-specific T-cell subpopulations, allowing their direct visualization and quantification. MHC-peptide multimers, such as dimers, tetramers, pentamers, streptamers, dextramers and octamers have been used to evaluate the frequency of CD8⁺ T cells, specific for tumor/leukemia-associated antigens as well as for viral antigens, e.g., CMVpp65 and EBV-EBNA. Moreover, MHC-peptide multimers have been used for rapid and efficient ex vivo isolation and expansion of T cells. A recent development in the field of MHC-peptide multimers led to the purification of CD8⁺ T cells specific for leukemia antigens. This might help to select leukemia-specific donor lymphocyte infusions (DLIs), thus allowing dissection of the noxious graft-versus-host disease (GvHD) from beneficial anti-viral and even anti-leukemic effects. This review covers different types of MHC-peptide multimers and their applications, as well as the impact that multimers might have on further development of DLIs.

Introduction

The development of novel flow cytometry techniques to detect antigen-specific T cells has improved our understanding of the kinetics and importance of T-cell responses, particularly to viral infections [1–3]. One of the most important innovations in flow cytometry has been in the fields of hardware, fluorochromes and data analysis tools for multi-parameter flow cytometry [4]. Major histocompatibility complex (MHC)-peptide multimers, such as dimers, tetramers, pentamers [5–7] (Fig. 1), streptamers [8] (Fig. 2), dextramers [9] and octamers [10], allow the visualization and isolation of antigen-specific T cells. Moreover, MHC-peptide multimers might constitute a powerful tool to specify donor lymphocyte infusions (DLIs), which are in clinical practice for patients after allogeneic stem cell transplantation (SCT). MHC-peptide multimers might help to purify anti-viral/anti-leukemic T-cell subpopulations, thus minimizing the risk of graft-versus-host disease (GvHD).

Fig. 1.

The structure of multimers. a The dimer complex. A three-chain complex molecule consisting of a recombinant heavy chain of MHC-Ig fusion chain, an immunoglobulin light chain disulfide bound to the heavy chain, and a non-covalently associated human β2 microglobulin molecule. b The tetramer complex. MHC class I molecules are folded with the peptide of interest and β2 microglobulin (β2m) and tetramerized by a fluorescence-labeled streptavidin. c The pentamer complex. Five MHC class I peptide complexes are multimerized by a self-assembling coiled coil domain

Fig. 2.

a The streptamer complex and the basic principle of the reversible staining technology. The fluorochrome can be substituted by a magnetic bead, thus rendering the streptamers to an option for magnetic-associated cell separation (MACS) through separation columns, even under GMP conditions for clinical application. b The ultimer complex. Six MHC class II peptide complexes are multimerized by a self-assembling coiled coil domain

Structure of MHC-peptide multimers

MHC-peptide dimers are based on class G immunoglobulins (IgG) as a molecular scaffold (Fig. 1a) [11]. The engineered form of the MHC complex provides specificity, and a multimer:IgG construct results in an increased avidity. This compensates for the relatively low affinity characteristic of monomeric MHC-TCR interactions.

MHC:IgG dimeric fusion proteins were constructed in 1993 [12]. Three extracellular domains of MHC molecules are fused to the N terminal of the variable region of the mouse heavy chain IgG1 through recombinant DNA technology. The expression vector containing the fusion protein gene is then co-transfected with genes for human b2-microglobulin (β2m) into a myeloma cell line deficient in immunoglobulin heavy chain, but retaining the expression of immunoglobulin light chain (lambda) [13]. Thereafter, the MHC can be easily labeled and loaded with the peptide of interest [14]. Furthermore, the hinge region in the immunoglobulin scaffold provides more flexible access to enable T-cell binding.

Tetramers were first described by Altman et al. [15] in 1996 for direct visualization and quantification of antigen-specific cytotoxic T cells. A tetramer, a protein with four subunits (tetrameric) (Fig. 1b), is based on recombinant class I molecules that have been biotinylated. These molecules are folded with the peptide of interest and tetramerized by a fluorescence-labeled streptavidin molecule, which can bind up to four biotin molecules. Tetramers can bind up to three T-cell receptor (TCR) molecules at once due to the tetrahedral disposition of the complex. MHC-peptide tetramer technology is a single-cell assay that can be used to quantify both CD8⁺ and CD4⁺ T-cell responses. Various studies have shown the MHC-peptide tetramer technique to be exquisitely antigen specific and highly sensitive [16, 17]. Wooldridge et al. [17] have published an extensive review with describing ways to improve the use of tetramers.

MHC-peptide pentamers contain five class I MHC-peptide complexes that are multimerized by a self-assembling coiled coil domain (Fig. 1c). All five MHC-peptide complexes face in the same direction resulting in a very high avidity. Each MHC-peptide pentamer also comprises up to five fluorescent or biotin tags for bright and efficient labeling.

Neudorfer et al. [14] generated a reversible MHC-peptide multimer, designated streptamer, for the detection and purification of human cytotoxic T lymphocytes (CTL) directed against tumor-associated and viral antigens. Streptamers consist of peptide-loaded HLA-Strep-tagIII molecules and Strep-Tactin polymers (Fig. 2a). HLA-Strep-tagIII is a fusion protein between one HLA monomer and two Strep-tagII sequences sequentially arranged by a short linker [18]. Strep-tagII comprises eight amino acid residue peptides (Trp–Ser–His–Pro–Gln–Phe–Glu–Lys) showing a strong binding affinity for an engineered streptavidin derivative called Strep-Tactin. The complex HLA/Strep-tagIII can be dissociated from Strep-Tactin by the addition of biotin. The dissociation of the HLA/Strep-tagIII from the CTL occurs in a spontaneous manner by another round of washing. The specific CTL can be transferred to the patient without carrying any residual streptamers.

Batard et al. [9] have developed a multimer based on a dextran backbone carrying a MHC-peptide, streptavidin and multiple fluorescein, named dextramer. Dextramers can identify, enumerate and track antigen-specific CTLs in the peripheral blood with up to four independent specificities.

The more the MHC-peptide complexes binding to TCRs, the more is the avidity. Guillaume et al. [10] selected CTLs with an octameric MHC-peptide complex. However, this octamer activates CTLs for Fas dependent apoptosis.

Class II MHC-peptide multimers (Fig. 2b) are the analog to MHC-pentamers. Ultimers are aligned in a planar configuration, optimized for best access of the TCRs on the cell surface to the MHC binding groove. Ultimers are linked to a core structure through their peptides, resulting in an independent alignment of each MHC molecule to its complementary TCR. Such characteristics improve the binding quality.

Recent innovations in the field of multimers

Very recent studies using MHC-peptide multimers labeled with four streptavidin colors, in parallel, detect up to 15 different T-cell specificities in a single sample; moreover, 63 specificities can be detected using six colors [19]. Acquiring so much information from one patient sample is crucial when access to material is restricted. These innovations present a major breakthrough, which will overcome the greatest limitation of these reagents, the restriction of analysis to one specific MHC-peptide combination. Furthermore, Schumacher’s group have established MHC-peptide multimers labeled with quantum dots. Their experiments suggest that eight different T-cells specificities can be detected within a single sample using six quantum dots [20].

Hadrup et al. have developed engineered MHC-peptide multimers to express cleavable peptide ligands. These constitutive MHC-peptide multimers are now available for a series of common HLA molecules and allow a quick generation of hundreds of HLA-peptide multimers of different specificity in a short time, in every laboratory that makes use of the new technology, allowing for high throughput epitope discovery [21, 22].

After labeling with MHC-peptide multimers, antigen-specific T cells can be enriched via magnetic beads. By this process, the detection limit can go down to 1 × 10⁶ for CD8⁺ T cells or 2.5 × 10⁶ for CD4⁺ T cells, which is far beyond the lower limit of detection for conventional staining approaches [23–25].

By amino acid exchange, the affinity of MHC multimers to the CD8⁺ molecule can be engineered. Hereby, it is possible to selectively detect only high-affinity T cells or increase affinity in such a way that even low-affinity T cells can be specifically stained. This enables function and tissue distribution of highly promising T cells for immunotherapy to be determined [26].

Apparently, cross-linking of the TCR by the HLA multimers induces apoptosis in the selected CTLs. Bouquié et al. [27] have solved this problem with the generation of a new multimer comprising immunomagnetic beads coupled to a monoclonal antibody specific for the AviTag peptide and coated with HLA-peptide monomers bearing the non-biotylated AviTag. Such multimers have a reduced apoptosis rate in the selected population presumably because the cross-linked TCRs are too distant from each other to trigger any signaling, thus minimizing the loss of reactive clones, which is a matter of importance when populations are rare. This multimer might be used in clinical practice.

Application of MHC-peptide multimer technology

Since Altman et al. established a practical design for the molecular tools that tag T cells in an antigen-specific manner in 1996 [15], MHC-peptide multimers have rapidly become the gold standard for T-cell analysis and manipulation, enabling the enumeration, isolation and stimulation of T cells of known antigen specificity [8].

MHC-peptide tetramers were originally used for T-cell activation. Thereafter, they were used successfully to analyze antigen-specific T-cell populations ex vivo [28]. MHC-peptide tetramers permit both detection and isolation of antigen-specific T cells present at low numbers in peripheral blood [29–31].

Recently, MHC-peptide multimers have been extended to evaluate functional aspects of T cells. Tetramer staining can be combined with intracellular detection of cytokines (e.g., IFN-γ), chemokines (MIP-1α), and cytotoxins (perforin/granzymes) following antigen/mitogen stimulation in vitro to assess the functional status of tetramer-positive T cells [32]. In addition, the tetramer-positive T cells can be sorted and expanded in in vitro cell culture for any further functional analysis [18, 23, 33, 34]. Also, tetramers are used to track in vivo T-cell clones of adoptively transferred T cells due to their specific and stable binding to the TCR [15] and to determine percentages of different T-cell populations [35, 36]. The use of tetramers may help to answer other immunological questions. Using kinetic analysis of tetrameric complex binding to TCR, Savage et al. [37] demonstrated that over the course of an immune response, CTLs have higher affinity and longer duration of binding to the class I MHC-peptide complex. Immunomonitoring using multimers has become an integral part of clinical vaccination trials. The Cancer Immunotherapy (CIMT) Interlab Trial has attempted to standardize this technique at the European level [38, 39]. Also, this technology provides information on the phenotype of clonal CTLs and addresses other questions, such as whether clonal CTLs have the same cytokine production profile [40].

Comparison of different MHC-peptide multimers

Pentamers include five MHC-peptide complexes in the same orientation, which results in an enhanced avidity interaction with the TCR when compared to MHC-peptide tetramer technology, which relies on MHC-peptide complexes held in a tetrahedral configuration. Pentamers allow up to five fluorescent labels, yielding brighter signals than tetramers [5]. Streptamers are as sensitive as conventional tetramers regarding the detection of trace cell amounts. Regarding the isolation of T cells for clinical application, reversible streptamers may be favored over tetramers and pentamers. Following streptamer dissociation, antigen-specific CTLs remain functionally active [14]. In our study [41], all three MHC-peptide multimer technologies can be used yielding similar results. The lowest background signals were obtained by tetramer technology (Fig. 3).

Fig. 3.

Specificity of staining by streptamers and tetramers. CD8⁺ T lymphocytes from an HLA-A2⁺ CMV-seropositive healthy donor were counter-stained with streptamers and tetramers and analyzed by flow cytometry. Dots in the upper right quadrant display such double-positive T cells. Similar frequencies of CD8⁺/multimer⁺ T cells could be detected by both technologies; however, more unspecific binding was observed by using streptamers (left dot blot, lower right quadrant)

Many studies support MHC-peptide multimers as a potential modality for diagnostic and therapeutic treatment, including the selection of cells for adoptive transfer [14, 15, 42, 43]. Neudorfer et al. [14] demonstrated that the streptamer-based sorting technique has both high purity of peptide-specific T cells and complete reversibility. The functional status of streptamer-sorted T cells is preserved, which may be of particular relevance in a situation where small populations of MHC-peptide multimer-sorted T cells are directly transferred to a conditioned patient. Streptamers are detached from the T cells prior to transfer, avoiding the immune responses directed against the MHC-peptide multimer reagents. Bead-conjugated streptamers will be a further development to the currently used tetramer-based isolation methods [44], because streptamer-coated beads can be detached following isolation, which is in agreement with the good manufacturing practice (GMP) guidelines. The MHC-streptamer technique will facilitate the development of adoptive immunotherapies with antigen-specific T-cell populations, an approach which can be expanded from viral to tumor antigens.

Potential clinical application of MHC-peptide multimers: DLIs

Patients with leukemia, who are at high risk of relapse despite chemotherapy, might receive allogeneic SCT. Thereafter, an anti-leukemic reaction may become effective due to the graft-versus-leukemia (GVL) effect. However, the graft can reject not only leukemia cells, but also other tissue in the recipient in a disease known as GvHD (Fig. 4). To avoid GvHD, Marmont et al. [45] depleted T cells from the transplant. In this setting, an increased leukemia relapse rate was observed. This underlines the crucial role of T lymphocytes in GVL. Such finding drove the idea to treat chronic myeloid leukemia (CML) and acute myeloid leukemia (AML) with DLIs [46].

Fig. 4.

GVL effect after HSCT and DLIs/Ag-specific CTLs. After chemotherapy and irradiation, HSC are infused for the re-population of the bone marrow. However, the graft exerts not only a GVL effect, but potentially also causes GvHD. In case of relapse, either DLIs or antigen-specific CTLs are infused with a good outcome against leukemia cells due to the GVL reaction. LAA leukemia-associated antigen, HD healthy donor, HSC hematopoietic stem cell, DLIs donor lymphocyte infusions, CR complete remission, MRD minimal residual disease, AML acute myeloid leukemia

DLIs are an approach to direct an adoptive immune response against leukemia cells into recipients. Nevertheless, a better response of DLIs can be seen in myeloid leukemia rather than in lymphoid leukemia. Kolb et al. [47] suggested that this may be due to the presence of myeloid leukemia-derived dendritic cells that present leukemia antigens to donor T cells.

Although leukemia cells have mechanisms to evade DLIs attack, once the CD8⁺ T cells are primed, any further encounter with the specific target will result in its lysis [48, 49]. In this regard, streptamers could be used for adoptive transfer of primed specific DLIs.

Virus-associated DLIs as a paradigm for antigen-specific DLIs

Recipients of allogeneic hematopoietic stem cell transplantation (HSCT) can experience reactivation of some latent-stage viruses, such as Epstein–Barr virus (EBV) and cytomegalovirus (CMV) [44, 50]. CMV infection is the most frequent of all. Twenty years ago, CMV reactivation occurred in 80% and developed into a disease in 40% of the CMV-seropositive allogeneic recipients [51]. This scheme has improved in the last years with the use of drugs to control the virus. However, viruses become resistant to drugs. HSV was found to be resistant to acyclovir in five of seven patients in a study by Langston et al. [52]. Even more, CMV induces or increases GvHD after allogeneic transplantation [53] and in return GvHD can aggravate CMV disease [54].

The adoptive transfer of prophylactic transfusions of virus-specific CD8⁺ T cells assures a protective method of treatment [55] and virus-reactive T cells can be easily selected from seropositive donors by streptamers [14]. Furthermore, the infusion of unselected allogeneic donor lymphocytes can result in an antiviral response and also in GvHD. The enrichment and expansion of virus-specific CD8⁺ T cells can both eliminate GvHD and induce an anti-viral effect, as demonstrated by many groups [56–58]. However, this procedure is expensive and requires long-term T-cell culture. A new approach to solve this problem is the use of HLA-peptide tetramers directly to select antigen-specific T cells from healthy seropositive donors and transfer them directly into the recipient without any further manipulation [59]. Cobbold et al. [44] adopted this system with CMV-specific T cells, and found considerable expansion in vivo after adoptive transfer using the HLA-peptide tetramer staining.

Furthermore, CMV-specific CTL activity can be measured by tetramer-binding to cells that express IFNγ, which increased after transfer in the first few months [44]. Remarkably in this study, a reduction in CMV viremia was clearly observed in all patients and no CMV disease was seen in any patient. Additionally, no side effects were observed during the administration of CMV-specific CTL.

A desirable GVL effect through DLIs in the absence of GvHD: the future starts now

A burning question is how to avoid GvHD, but maintain the GVL effect through DLIs. In some cases, GVL has a positive correlation with the severity of GvHD. However, GVL has been observed in the absence of GvHD; therefore, both effects can be separated [55]. Mandigers et al. [60] used DLIs as a treatment for indolent lymphoma after allogeneic SCT with very promising results and limited GvHD.

One strategy to achieve GVL without GvHD is to initiate DLIs with 10⁶ T cells/kg body weight and to increase the dose until GVL is reached with a suggested average interval of 48 days between two doses. This treatment diminishes 15% of the treatment-related mortality in comparison with higher initial doses [61]. Transfection of T cells with a “suicide gene” prior to DLIs opens the option to kill such T cells by activating them either with a drug, which is non-toxic to mammalian cells [62] or the apoptosis program in cells [63], in case of GvHD occurrence. Nowadays, GVL effect might be improved minimizing the risk of a GvHD with MHC-peptide multimer technology. In analogy to DLIs specific for viral antigens, production of leukemia-specific DLIs is feasible on a GMP level.

Since the advent of MHC-peptide multimer technologies, big steps have been taken in medical research: from Cobbold’s [44] successful adoptive transfer of CMV-specific CTL using HLA-tetramer selection, without any side effect, to Neudorfer’s approach using dissociated streptamers for the treatment of patients with cancer or infectious diseases. Molldrem et al. [64] used tetramer technology to analyze a peptide derived from proteinase 3 (an epitope termed PR1)-specific CTLs in CML. They found a positive correlation in the number of PR1-specific CTLs and the disappearance of the tumor after treatment. Such findings are supported by Melenhorst et al. [26].

Currently, our group is performing adoptive transfer of CMV-specific CTL into AML patients using streptamers technology [65]. Primarily, the streptamer-based DLIs should be obtained from the stem cell donor, as in conventional DLIs. However, under certain circumstances, e.g., when the donor is not available any longer, a third party donor might be considered for the approach, which of course bears a higher risk of GvHD. First experience with third party donors in this setting demonstrates that this approach is feasible even without the development of GvHD. Promising results have been obtained so far. A clearance of the CMV antigenemia and an overwhelming increase in effector CTL in the host has been reached [66]. We are now adopting this approach to leukemia-associated antigens such as WT-1 and RHAMM. Employing the streptamer technology, a high pure fraction of WT-1-specific CTL was obtained from healthy donors and, in a further step, infused into AML patients. Such CTLs conserve their cytotoxic properties and, in an expected way, will be able to eliminate any residual leukemia blast.

Conclusion

Development of MHC-peptide multimer technology represents an important milestone in the field of immunology and a major technical breakthrough in the analysis of T-cell responses, as MHC-peptide multimers allow the labeling and/or sorting of T cells according to their antigen specificity. The MHC-peptide multimer technology has considerably improved our ability to monitor antigen-specific T-cell responses and facilitates the development of adoptive T-cell transfer regimens for the treatment of patients with cancer or infectious diseases. Furthermore, MHC-peptide multimer-based selection of leukemic antigen-specific DLIs might solve the problem of dissecting GVL from GvHD and constitutes an efficient therapy to eradicate leukemic cells.