Reformation in chimeric antigen receptor based cancer immunotherapy: Redirecting natural killer cell

Natural killer (NK) cells are an important subset of lymphocytes which play a critical role in host immunity against cancers. With MHC-independent recognition, short lifespan and potent cytotoxicity, NK cells make a promising candidate for chimeric antigen receptor (CAR)-engineered cancer immunotherapy. Due to innate biological properties of NK cells, CAR-NK may outperform CAR-T therapy in terms of less side effects and more universal access, which may become a great reformation in CAR-based cancer immunotherapy. The CARs used in peripheral blood (PB) NK cells as well as NK cell line like NK-92 are the most important outfits defining antigenic specificity. The constructs of CARs used in NK cells from different sources vary, which all undergo generational optimization. The anti-tumor effects of CAR-NK have been validated in numerous preclinical trials for cancers, including hematologic malignancies and many solid tumors, which provide evidence for potential clinical application of CAR-NK. Additionally, this review concludes the challenges faced in the application of CAR-NK. Although CAR-NK is considered as one of the most possible “off-the-shelf” products, the improvement for the efficiency of expansion and transduction as well as the solution for underlying safety issues is still needed. Possible coping strategies for challenges and upgrades in techniques are also highlighted for future development in CAR-NK cancer immunotherapy.     

Encephalitogenic T cell clones with variant receptor specificity

We explored antigenic differences between guinea pig (GP)-basic protein (BP), rat (Rt)-BP, and respective peptides from the encephalitogenic region for Lewis rats by comparing the fine specificity of T lymphocyte lines and clones selected from animals primed with these Ag. Encephalitogenic T cell lines specific for GP-BP or Rt-BP predictably recognized the corresponding 72-89 and to a lesser degree the 72-84 (S55S) amino acid sequence. T cell lines selected from rats primed with GP-S55S responded preferentially to GP-S55S compared to other peptides. A T cell line raised to Rt-S55S, however, initially recognized the S55S and S72-89 peptides but were nearly unresponsive to the intact GP-BP or Rt-BP. T cell clones selected from the Rt-S55S line at that point had two distinct patterns of response: clones that recognized both of the BP and the S55S peptides adoptively transferred delayed-type hypersensitivity and experimental autoimmune encephalomyelitis. These clones also recognized residues 69-81 (S67) but not peptide S75-89. In contrast, T cell clones that responded only to synthetic peptides GP-S55S and Rt-S55S but not to the parent BP adoptively transferred delayed-type hypersensitivity but not disease in Lewis rats. The same clones failed to respond to either the S67 or the S75-89 sequences. These results demonstrate that the encephalitogenic Rt-S55S sequence houses a minimum of two T cell epitopes with differing specificities and functions. One epitope is immuno-dominant and resembles the encephalitogenic region of the intact BP molecule. The second non-encephalitogenic epitope is restricted to the S55S sequences and is not shared by the parent BP, the S67, or the S75-89 sequences. Both types of Rt-S55S-specific clones differ in fine specificity from encephalitogenic clones selected from GP-BP immunized rats, thus indicating that uniformity of T cell recognition of the encephalitogenic epitope is not an absolute condition for T cells to be encephalitogenic.     

Engineering T lymphocyte antigen specificity.

Targeting of immune cells by bispecific antibodies has proven to be a powerful tool for the investigation of cellular cytotoxicity, lymphocyte activation and induction of cytokine production, as well as to represent an innovative form of immunotherapy for the treatment of cancer. The hallmark of this approach is the use of the specificity of monoclonal antibodies to join target and immune cells by virtue of the dual specificity of bispecific antibodies for the two entities. More precisely, the bispecific antibody has two different binding sites, which are capable of recognizing tumor associated antigens on the one hand and lymphocyte activation sites on the other. This process of crosslinking results in the activation of the lymphocyte and triggering of its lytic machinery, as well as lymphokine production. A major advantage of this therapeutic modality is, that use is made of the normal cellular immune defence system and therefore is only associated with minor toxicity. The distinct lymphocyte populations, which can be used for adoptive immunotherapy and the various bispecific antibody preparations, as well as the chimeric immunoglobulin/T cell receptor construction, are the major topics of this review.     

T cell receptor gene segment utilization by HLA-DR1-alloreactive T cell clones.

Transplantation of histoincompatible tissues leads to allograft rejection, which involves recognition of allogeneic MHC molecules by Ag-specific receptors expressed on T cells. The interaction of these molecules is highly specific yet poorly understood. We have investigated the relationship between TCR gene utilization and allo-MHC restriction patterns by using a one-way polymerase chain reaction to amplify the alpha- and beta-chain mRNA from a panel of 10 HLA-DR1-alloreactive T lymphocyte clones. Two previously unreported V alpha and five J alpha gene sequences were obtained. Although a few V alpha, V beta, and J alpha genes were utilized more than once, no correlation between TCR gene usage and DR1 alloreactivity was identified. At the sequence level, the presumed TCR alpha- and beta-chain CDR1 and CDR2 regions displayed limited diversity, whereas the CDR3 or junctional sequences were highly variable. Although most TCR probably interact with subtly different surface features of the DR1 alloantigen, we predict that TCR with similar CDR1 and CDR2 sequences would contact essentially identical regions of the DR1 molecule. The lack of sequence conservation in the junctional regions suggests that different endogenous peptides also may be recognized. Thus, alloreactive T cells may recognize not only allogeneic MHC molecules but perhaps also bound endogenous peptides.     

Fine specificity and antigen receptor expression among influenza virus-specific cytolytic T lymphocyte clones

Influenza virus stimulates a vigorous cytolytic T lymphocyte (CTL) response in the mouse that is directed to several virion polypeptides. This report examines the fine specificity of a panel of murine influenza-specific CTL clones restricted by MHC class I products of the H-2d haplotype. Ten of 22 A/JAPAN/305/57-specific CTL clones analyzed were directed to the A/JAPAN/305/57 hemagglutinin protein as detected by using target cells infected with a recombinant vaccinia virus containing hemagglutinin gene. Based on their fine specificity of hemagglutinin recognition, these clones defined four functional epitopes on the hemagglutinin. The remaining 12 cytolytic clones exhibited cross-reactivity for type A influenza viruses of the major human subtypes, and approximately 60% of these clones were directed to the nucleocapsid protein. KJ16-133 monoclonal antibody analysis of the utilization of the T cell receptor V beta 8 gene segment subfamily revealed that members of this V beta gene subfamily are expressed by both hemagglutinin- and nucleocapsid-specific MHC class I-restricted CTL (and by influenza-specific MHC class II-restricted T lymphocytes as well). These results suggest that CTL detect several distinct antigenic sites on the hemagglutinin. In addition, these results reveal no direct correlation between viral antigenic specificity and V beta gene expression by these virus-specific CLT clones.     

Coreceptor CD8-driven modulation of T cell antigen receptor specificity

The CD8 coreceptor modulates the interaction between the T cell antigen receptor (TCR) and peptide-major histocompatibility class I (pMHCI). We present evidence that CD8 not only modifies the affinity of cognate TCR/pMHCI binding by altering both the association rate and the dissociation rate of the TCR/pMHCI interaction, but modulates the sensitivity (triggering threshold) of the TCR as well, by recruiting TCR/pMHCI complexes to membrane microdomains at a rate which depends on the affinity of MHCI/CD8 binding. Mathematical analysis of these modulatory effects indicates that a T cell can alter its functional avidity for its agonists by regulating CD8 expression, and can rearrange the relative potencies of each of its potential agonists. Thus we propose that a T cell can specifically increase its functional avidity for one agonist, while decreasing its functional avidity for other potential ligands. This focussing mechanism means that TCR degeneracy is inherently dynamic, allowing each TCR clonotype to have a wide range of agonists while avoiding autorecognition. The functional diversity of the TCR repertoire would therefore be greatly augmented by coreceptor-mediated ligand focussing.     

Editing T cell specificity towards leukemia by zinc finger nucleases and lentiviral gene transfer

The transfer of high-avidity T cell receptor (TCR) genes isolated from rare tumor-specific lymphocytes into polyclonal T cells is an attractive cancer immunotherapy strategy. However, TCR gene transfer results in competition for surface expression and inappropriate pairing between the exogenous and endogenous TCR chains, resulting in suboptimal activity and potentially harmful unpredicted antigen specificities of the resultant TCRs. We designed zinc-finger nucleases (ZFNs) that promoted the disruption of endogenous TCR β- and α-chain genes. Lymphocytes treated with ZFNs lacked surface expression of CD3-TCR and expanded with the addition of interleukin-7 (IL-7) and IL-15. After lentiviral transfer of a TCR specific for the Wilms tumor 1 (WT1) antigen, these TCR-edited cells expressed the new TCR at high levels, were easily expanded to near purity and were superior at specific antigen recognition compared to donor-matched, unedited TCR-transferred cells. In contrast to unedited TCR-transferred cells, the TCR-edited lymphocytes did not mediate off-target reactivity while maintaining their anti-tumor activity in vivo, thus showing that complete editing of T cell specificity generates tumor-specific lymphocytes with improved biosafety profiles.     

A new year and a new era for cell

Small regulatory RNAs can act by pairing with their target messages, targeting themselves and the mRNA for degradation; Lenz et al. (this issue of Cell) now report that multiple small RNAs are essential regulators of the quorum-sensing systems of Vibrio species, including the regulation of virulence in V. cholerae.     

Adaptive T cell immunotherapy in cancer

Impaired tumor-specific effector T cells contribute to tumor progression and unfavorable clinical outcomes. As a compensatory T cell-dependent cancer immunoediting strategy, adoptive T cell therapy(ACT) has achieved encouraging therapeutic results,and this strategy is now on the center stage of cancer treatment and research. ACT involves the ex vivo stimulation and expansion of tumor-infiltrating lymphocytes(TILs) with inherent tumor reactivity or T cells that have been genetically modified to express the cognate chimeric antigen receptor or T cell receptor(CAR/TCR), followed by the passive transfer of these cells into a lymphodepleted host. Primed T cells must provide highly efficient and long-lasting immune defense against transformed cells during ACT. Anin-depth understanding of the basic mechanisms of these living drugs can help us improve upon current strategies and design better next-generation T cell-based immunotherapies. From this perspective, we provide an overview of current developments in different ACT strategies, with a focus on frontier clinical trials that offer a proof of principle. Meanwhile, insights into the determinants of ACT are discussed, which will lead to more rational, potent and widespread applications in the future.     

Globo-A--a new receptor specificity for attaching Escherichia coli

Uropathogenic Escherichia coli strains designated as ONAP, based on their O negative A positive agglutination of human P1 erythrocytes, were shown to prefer the globo-A glycolipid as a receptor structure. The dependence on both the A terminal and the globoseries chain was confirmed by agglutination of human AP1, but not Ap or OP1 erythrocytes and by binding to the globo-A glycolipid on TLC plates. Neither Gal alpha 1----4Gal beta nor the A trisaccharide GalNAc alpha 1----3(Fuc alpha 1----2)Gal beta alone functioned as receptors. The bacteria thus appeared to recognize an epitope resulting from the combination of the terminal and internal structures.     

Naturally occurring cytotoxic T lymphocyte precursors with specificity for an Ig idiotype

The humoral response to the p-azobenzenearsonate hapten in the A/J mouse includes the major cross-reactive idiotype associated with anti-p-azobenzenearsonate (CRIA) found in all immunized mice. Limiting dilution cultures of non-immunized spleen cells of A/J mice with irradiated B hybridoma cells bearing the Ig idiotype, CRIA, in the presence of T cell growth factors developed cytotoxic activity against the CRIA-bearing hybridoma; in some wells this activity was completely abrogated by an anti-idiotype mAb specific for CRIA or by a univalent hapten antigen, tyrosine-p-azobenzenearsonate, indicating the existence of cytotoxic T cell precursors (CTL-P) specific for one or more idiotopes of CRIA in normal spleen cells. The CTL clones lysed targets in a H-2D-restricted manner and were cytotoxic for CRIA-bearing hybridoma lines, but not for CRIA-non-bearing, IgG1k-bearing hybridoma lines. These CTL-P were detected at a high frequency (1/4,500 to 1/10,000) in a spleen cell population of non-immunized, relatively aged A/J mice (16 to 30 wk of age), and at a lower frequency in spleen cells of younger A/J mice (8 wk of age). However, they were not detected in normal spleen cells of B10.A (CRIA-non-producer) mice at any age (less than 1/6 x 10(5)). Normal Ighd-congenic C.AL-20 mice (16 wk of age), that are CRIA producers had as a high frequency of the CTL-P as did A/J mice, whereas normal Ighb-congenic C.B-20 mice (CRIA-non-producers) had none. In the spleen cells of the CRIA-producers, cytotoxicity of the CTL-P developed only in cultures with small numbers of seeding cells. They were completely absent in cultures with greater numbers of cells; this may be due to the presence of suppressor cells of lower frequency but greater potency. In lymph node cells or PBL of relatively aged A/J mice, the CTL-P were also detected, but only in cultures containing higher cell numbers, and at low frequency (between 1/5 x 10(5) to 1/2 x 10(6)). In thymocytes of 8-wk-old A/J mice, they were occasionally detected at very low frequency (less than or equal to 1/1 x 10(6)), but were not present in the bone marrow cells at any age. These results demonstrate the high incidence of the generation of CTL-P specific for an autologous Ag, and indicate that CRIA on B cells may induce CTL specific for CRIA. However, the development of CTL-P may be inhibited by co-existent suppressor cells under normal conditions.     

Stochastic modeling of T cell receptor gamma gene rearrangement

The mechanisms controlling the recombination process of the gamma genes that encode the gamma chain of the antigen receptor of the gammadelta T lymphocytes are unclear. Based on experimental data on the recombination status of the two major TCR gamma genes expressed in V(gamma)4+ and V(gamma)1+ thymocytes, we tested the plausibility of three possible rearrangement mechanisms: (1) a time window mechanism according to which the two chromosomes are accessible to the recombination machinery during a defined period of time; (2) a feedback mechanism in which recombination stops shortly after the first in-frame rearrangement event anywhere in both chromosomes; and (3) a feedback mechanism with asynchronous chromosome accessibility, in which there is a first period when only one chromosome is accessible for recombination, followed by a second period when both chromosomes are accessible; shortly after the first in-frame rearrangement event, during any of these two periods, recombination will definitely stop. We model the time window mechanism using a pure probabilistic approach and the two feedback mechanisms using a continuous-time Markov chain formalism. We used maximum likelihood methodology to infer the goodness-of-fit of the models showing evidence for the last model, which best fits the data. Further analysis of this model suggests an evolutionary tradeoff between allelic and isotypic exclusion and the probability that a precursor differentiates into a mature gammadelta T lymphocyte.     

Peripheral T cell receptor gamma delta variable gene repertoire maps to the T cell receptor loci and is influenced by positive selection.

Although the mechanisms that determine TCR-alpha beta V gene repertoire are well studied, the genetic influences involved in TCR-gamma delta repertoire development are unclear. Unlike the TCR-gamma delta populations that localize in epithelial tissues, the circulating peripheral TCR-gamma delta V region repertoire is quite diverse. Previous studies have shown that three TCR-gamma chains and at least six TCR-V delta genes are expressed by splenic TCR-gamma delta cells. However, the relative frequency of individual gamma delta subsets among genetically diverse mice has not been determined. Therefore, the repertoire of TCR-gamma delta cells was examined using anti-TCR V region specific mAb against V gamma 2 and V delta 4 on TCR-gamma delta + cells from total splenocytes. We found that there was a strain-specific variation in TCR-gamma delta usage. The frequency of V gamma 2 expression in different strains varied from 54 to 12%, and the frequency of V delta 4 expression in different strains varied from 38 to 10%. However, the level of V delta 4 and V gamma 2 expression for an individual strain was highly consistent from experiment to experiment. F1 analysis between parental strains that differed in relative frequency of either V gamma 2+ or V delta 4+ cells revealed that high expression was genetically dominant, suggesting that positive selection events play a major role in the peripheral gamma delta repertoire. Variations in the levels of V gamma 2+ cells and V delta 4+ cells was not associated with Mls or MHC haplotype. Analysis of recombinant inbred strains revealed that high V delta 4 expression mapped to the TCR-gamma locus, while high V gamma 2 expression was influenced by the TCR-delta locus. Back-cross analysis confirmed that the TCR loci dominantly influenced the level of V delta 4+ cells and V gamma 2+ cells; however, there was clear evidence that multiple genes affect the TCR-gamma delta repertoire.     

Predominant T cell receptor gene elements in TNP-specific cytotoxic T cells.

H-2b class I-restricted, TNP-specific CTL clones were obtained by limiting dilution cloning of either short term polyclonal CTL lines or spleen cells of TNP-immunized mice directly ex vivo. Sequence analyses of mRNA coding for TCR alpha- and beta-chains of 11 clones derived from CTL lines from individual C57BL/6 mice revealed that all of them expressed unique but clearly nonrandom receptor structures. Five alpha-chains (45%) employed V alpha 10 gene elements, and four of those (36%) were associated with J beta 2.6-expressing beta-chains. The alpha-chains from these four TCR, moreover, contained an acidic amino acid in position 93 of their N or J region-determined sequences. Clones isolated directly from spleen cells carried these types of receptors at lower frequency, 27% V alpha 10 and 19% J beta 2.6, indicating that bulk in vitro cultivation on Ag leads to selection for these particular receptors. However, even in TNP-specific CTL cloned directly ex vivo, V alpha 10 usage was increased about fivefold over that in Ag-independently activated T cells in H-2b mice (4 to 5%). The selection for V alpha 10/J beta 2.6-expressing cells was obtained repeatedly in other TNP-specific CTL lines from C57BL/6 mice but not in FITC-specific CTL from the same strain or in TNP-specific CTL lines from B10.BR (H-2k) or B10.D2 (H-2d) mice. We conclude from this (a) that the selection for V alpha 10/J beta 2.6+ T cells is driven by the complementarity of these receptors to a combination of TNP and MHC epitopes and (b) that predominant receptor structures reflect the existence of a surprisingly limited number of T cell-relevant hapten determinants on the surface of covalently TNP-modified cells.     

Vaccine and antibody-directed T cell tumour immunotherapy

Clearer evidence for immune surveillance in malignancy and the identification of many new tumour-associated antigens (TAAs) have driven novel vaccine and antibody-targeted responses for therapy in cancer. The exploitation of active immunisation may be particularly favourable for TAA where tolerance is incomplete but passive immunisation may offer an additional strategy where the immune repertoire is affected by either tolerance or immune suppression. This review will consider how to utilise both active and passive types of therapy delivered by T cells in the context of the failure of tumour-specific immunity by presenting cancer patients. This article will outline the progress, problems and prospects of several different vaccine and antibody-targeted approaches for immunotherapy of cancer where proof of principle pre-clinical studies have been or will soon be translated into the clinic. Two examples of vaccination-based therapies where both T cell- and antibody-mediated anti-tumour responses are likely to be relevant and two examples of oncofoetal antigen-specific antibody-directed T cell therapies are described in the following sections: (1) therapeutic vaccination against human papillomavirus (HPV) antigens in cervical neoplasia; (2) B cell lymphoma vaccines including against immunoglobulin idiotype; (3) oncofoetal antigens as tumour targets for redirecting T cells with antibody strategies.     

Identification of noncanonical melanoma-associated T cell epitopes for cancer immunotherapy

The identification of tumor-associated T cell epitopes has contributed significantly to the understanding of the interrelationship of tumor and immune system and is instrumental in the development of therapeutic vaccines for the treatment of cancer. Most of the known epitopes have been identified with prediction algorithms that compute the potential capacity of a peptide to bind to HLA class I molecules. However, naturally expressed T cell epitopes need not necessarily be strong HLA binders. To overcome this limitation of the available prediction algorithms we established a strategy for the identification of T cell epitopes that include suboptimal HLA binders. To this end, an artificial neural network was developed that predicts HLA-binding peptides in protein sequences by taking the entire sequence context into consideration rather than computing the sum of the contribution of the individual amino acids. Using this algorithm, we predicted seven HLA A*0201-restricted potential T cell epitopes from known melanoma-associated Ags that do not conform to the canonical anchor motif for this HLA molecule. All seven epitopes were validated as T cell epitopes and three as naturally processed by melanoma tumor cells. T cells for four of the new epitopes were found at elevated frequencies in the peripheral blood of melanoma patients. Modification of the peptides to the canonical sequence motifs led to improved HLA binding and to improved capacity to stimulate T cells.