Investigating CTL Mediated Killing with a 3D Cellular Automaton

Cytotoxic T lymphocytes (CTLs) are important immune effectors against intra-cellular pathogens. These cells search for infected cells and kill them. Recently developed experimental methods in combination with mathematical models allow for the quantification of the efficacy of CTL killing in vivo and, hence, for the estimation of parameters that characterize the effect of CTL killing on the target cell populations. It is not known how these population-level parameters relate to single-cell properties. To address this question, we developed a three-dimensional cellular automaton model of the region of the spleen where CTL killing takes place. The cellular automaton model describes the movement of different cell populations and their interactions. Cell movement patterns in our cellular automaton model agree with observations from two-photon microscopy. We find that, despite the strong spatial nature of the kinetics in our cellular automaton model, the killing of target cells by CTLs can be described by a term which is linear in the target cell frequency and saturates with respect to the CTL levels. Further, we find that the parameters describing CTL killing on the population level are most strongly impacted by the time a CTL needs to kill a target cell. This suggests that the killing of target cells, rather than their localization, is the limiting step in CTL killing dynamics given reasonable frequencies of CTL. Our analysis identifies additional experimental directions which are of particular importance to interpret estimates of killing rates and could advance our quantitative understanding of CTL killing.     

Evaluation of CD8 T cell killing models with computer simulations of 2-photon imaging experiments

In vivo imaging of cytotoxic T lymphocyte (CTL) killing activity revealed that infected cells have a higher observed probability of dying after multiple contacts with CTLs. We developed a three-dimensional agent-based model to discriminate different hypotheses about how infected cells get killed based on quantitative 2-photon in vivo observations. We compared a constant CTL killing probability with mechanisms of signal integration in CTL or infected cells. The most likely scenario implied increased susceptibility of infected cells with increasing number of CTL contacts where the total number of contacts was a critical factor. However, when allowing in silico T cells to initiate new interactions with apoptotic target cells (zombie contacts), a contact history independent killing mechanism was also in agreement with experimental datasets. The comparison of observed datasets to simulation results, revealed limitations in interpreting 2-photon data, and provided readouts to distinguish CTL killing models.     

Estimating the In Vivo Killing Efficacy of Cytotoxic T Lymphocytes across Different Peptide-MHC Complex Densities

Cytotoxic T lymphocytes (CTLs) are important agents in the control of intracellular pathogens, which specifically recognize and kill infected cells. Recently developed experimental methods allow the estimation of the CTL''s efficacy in detecting and clearing infected host cells. One method, the in vivo killing assay, utilizes the adoptive transfer of antigen displaying target cells into the bloodstream of mice. Surprisingly, killing efficacies measured by this method are often much higher than estimates obtained by other methods based on, for instance, the dynamics of escape mutations. In this study, we investigated what fraction of this variation can be explained by differences in peptide loads employed in in vivo killing assays. We addressed this question in mice immunized with lymphocytic choriomeningitis virus (LCMV). We conducted in vivo killing assays varying the loads of the immunodominant epitope GP33 on target cells. Using a mathematical model, we determined the efficacy of effector and memory CTL, as well as CTL in chronically infected mice. We found that the killing efficacy is substantially reduced at lower peptide loads. For physiological peptide loads, our analysis predicts more than a factor 10 lower CTL efficacies than at maximum peptide loads. Assuming that the efficacy scales linearly with the frequency of CTL, a clear hierarchy emerges among the groups across all peptide antigen concentrations. The group of mice with chronic LCMV infections shows a consistently higher killing efficacy per CTL than the acutely infected mouse group, which in turn has a consistently larger efficacy than the memory mouse group. We conclude that CTL killing efficacy dependence on surface epitope frequencies can only partially explain the variation in in vivo killing efficacy estimates across experimental methods and viral systems, which vary about four orders of magnitude. In contrast, peptide load differences can explain at most two orders of magnitude.     

Dynamics of immune escape during HIV/SIV infection

Several studies have shown that cytotoxic T lymphocytes (CTLs) play an important role in controlling HIV/SIV infection. Notably, the observation of escape mutants suggests a selective pressure induced by the CTL response. However, it remains difficult to assess the definite role of the cellular immune response. We devise a computational model of HIV/SIV infection having a broad cellular immune response targeting different viral epitopes. The CTL clones are stimulated by viral antigen and interact with the virus population through cytotoxic killing of infected cells. Consequently, the virus population reacts through the acquisition of CTL escape mutations. Our model provides realistic virus dynamics and describes several experimental observations. We postulate that inter-clonal competition and immunodominance may be critical factors determining the sequential emergence of escapes. We show that even though the total killing induced by the CTL response can be high, escape rates against a single CTL clone are often slow and difficult to estimate from infrequent sequence measurements. Finally, our simulations show that a higher degree of immunodominance leads to more frequent escape with a reduced control of viral replication but a substantially impaired replicative capacity of the virus. This result suggests two strategies for vaccine design: Vaccines inducing a broad CTL response should decrease the viral load, whereas vaccines stimulating a narrow but dominant CTL response are likely to induce escape but may dramatically reduce the replicative capacity of the virus.     

Memory CTL-hybridoma: a model system to analyze the anamnestic response of cytolytic T lymphocytes

We have previously described monoclonal CTL-hybridomas growing continuously in culture in the absence of a known antigenic stimulus or growth-promoting factor(s), exhibiting specific anti-H-2Db killing activity in vitro after a distinct 2-hr lag period before the onset of lysis. Here we provide evidence indicating that the hybridoma Md 26.15 represents "memory" CTL, and we examine the mechanism whereby they respond to antigen/mitogen. The brief period (2 to 3 hr) required for stimulation suggests that expression of potentiated effector function(s) after stimulation requires no cell replication. The frequency of effector hybridoma cells capable of specific binding to target cells before and after stimulation is the same, as determined in a direct hybridoma-target cell conjugation test, although cytolytic activity is markedly enhanced. On the other hand, kinetic and cell dispersion assays indicate that the potentiation of cytolytic activity can be attributed to shortening of the lag period of killing as well as to a small enhancement in the recycling ability of cytolytic effectors. We suggest that shortening of the lag period involves activation of potential effectors, which is manifested at a post-binding stage of effector-target interaction. The CTL hybridoma described herein provides a unique model system for studying CTL reactivation.     

A General Functional Response of Cytotoxic T Lymphocyte-Mediated Killing of Target Cells

Cytotoxic T lymphocytes (CTLs) kill virus-infected cells and tumor cells, and play a critical role in immune protection. Our knowledge of how the CTL killing efficiency varies with CTL and target cell numbers is limited. Here, we simulate a region of lymphoid tissue using a cellular Potts model to characterize the functional response of CTL killing of target cells, and find that the total killing rate saturates both with the CTL and the target cell densities. The relative saturation in CTL and target cell densities is determined by whether a CTL can kill multiple target cells at the same time, and whether a target cell can be killed by many CTLs together. We find that all the studied regimes can be well described by a double-saturation (DS) function with two different saturation constants. We show that this DS model can be mechanistically derived for the cases where target cells are killed by a single CTL. For the other cases, a biological interpretation of the parameters is still possible. Our results imply that this DS function can be used as a tool to predict the cellular interactions in cytotoxicity data.     

A General Functional Response of Cytotoxic T Lymphocyte-Mediated Killing of Target Cells

Cytotoxic T lymphocytes (CTLs) kill virus-infected cells and tumor cells, and play a critical role in immune protection. Our knowledge of how the CTL killing efficiency varies with CTL and target cell numbers is limited. Here, we simulate a region of lymphoid tissue using a cellular Potts model to characterize the functional response of CTL killing of target cells, and find that the total killing rate saturates both with the CTL and the target cell densities. The relative saturation in CTL and target cell densities is determined by whether a CTL can kill multiple target cells at the same time, and whether a target cell can be killed by many CTLs together. We find that all the studied regimes can be well described by a double-saturation (DS) function with two different saturation constants. We show that this DS model can be mechanistically derived for the cases where target cells are killed by a single CTL. For the other cases, a biological interpretation of the parameters is still possible. Our results imply that this DS function can be used as a tool to predict the cellular interactions in cytotoxicity data.     

Monoclonal cytotoxic T lymphocyte hybridomas capable of specific killing activity, antigenic responsiveness, and inducible interleukin secretion

We have recently described the production of cytotoxic T lymphocyte (CTL) hybridomas that grow continuously in culture, exhibiting constitutive, allospecific (anti-H-2b) killing activity. We now report on the response of these monoclonal CTL hybridomas to specific antigen (H-2Db) and to mitogenic lectins. Both specific antigen and T cell mitogens enhance hybridoma-mediated specific target cell killing. In addition, stimulated, but not unstimulated hybridoma cells secrete considerable amounts of IL 2 into the culture medium. Repeated cloning of the hybridomas provides strong evidence that both killing activity and IL 2 secretion can be attributed to one cell. Unfractionated Con A supernatants, containing IL 2 and other factors known to influence T cell responsiveness, or IL 2-containing media of stimulated hybridomas affect neither the growth nor the lytic activity of the hybridomas. Anti-LFA-1 monoclonal antibody, a potent inhibitor of CTL and CTL hybridoma-mediated target cell lysis, abolishes antigen- or mitogen-induced IL 2 secretion by the CTL hybridomas. Involvement of a single hybridoma receptor in antigen recognition (afferent and efferent) and in initiating IL 2 secretion is proposed. The CTL hybridomas displaying retarded killing activity before the antigenic or mitogenic stimulation appear to represent an intermediate stage in CTL differentiation, reminiscent of "memory" CTL.     

TCR-independent killing of B cell malignancies by anti-third-party CTLs: the critical role of MHC-CD8 engagement

We previously demonstrated that anti-third-party CTLs (stimulated under IL-2 deprivation against cells with an MHC class I [MHC-I] background different from that of the host and the donor) are depleted of graft-versus-host reactivity and can eradicate B cell chronic lymphocytic leukemia cells in vitro or in an HU/SCID mouse model. We demonstrated in the current study that human allogeneic or autologous anti-third-party CTLs can also efficiently eradicate primary non-Hodgkin B cell lymphoma by inducing slow apoptosis of the pathological cells. Using MHC-I mutant cell line as target cells, which are unrecognizable by the CTL TCR, we demonstrated directly that this killing is TCR independent. Strikingly, this unique TCR-independent killing is induced through lymphoma MHC-I engagement. We further showed that this killing mechanism begins with durable conjugate formation between the CTLs and the tumor cells, through rapid binding of tumor ICAM-1 to the CTL LFA-1 molecule. This conjugation is followed by a slower second step of MHC-I-dependent apoptosis, requiring the binding of the MHC-I α2/3 C region on tumor cells to the CTL CD8 molecule for killing to ensue. By comparing CTL-mediated killing of Daudi lymphoma cells (lacking surface MHC-I expression) to Daudi cells with reconstituted surface MHC-I, we demonstrated directly for the first time to our knowledge, in vitro and in vivo, a novel role for MHC-I in the induction of lymphoma cell apoptosis by CTLs. Additionally, by using different knockout and transgenic strains, we further showed that mouse anti-third-party CTLs also kill lymphoma cells using similar unique TCR-independence mechanism as human CTLs, while sparing normal naive B cells.     

Mathematical models of cytotoxic T-lymphocyte killing

By killing infected host cells, cytotoxic T lymphocytes (CTL) mediate an important defense mechanism against viruses and other intracellular pathogens. Quantitative aspects of this killing process have been studied for several decades in vitro. More recently, methods have been developed to measure the timescales of CTL killing in vivo. Here, we review the estimates of kinetic rates involved in CTL killing which were obtained in these studies, and elaborate on the differences between them.     

Two-mutation models for bone cancer due to radium, strontium and plutonium

Data from beagle experiments and radium dial painters were used to derive two-mutation carcinogenesis models for bone cancer induced by the bone-seeking radionuclides radium, strontium and plutonium. For all data, the model fits indicate that at low doses both mutation rates depend linearly and equally strongly on dose rate. For the high-LET alpha-particle emitters, a cell killing term reduces the second mutation rate at high dose rates. In all cases, the combined effect of both mutation rates is a linear-quadratic dose-effect relationship for cancer at low doses. This behavior may lead to experimental data that could be mistaken as showing a threshold below which no cancers are induced. Derived parameters such as toxicity ratios and tumor growth times compare well with values reported in the literature. Furthermore, results for plutonium indicate that rapid burial of the nuclide in the growing bones of juvenile beagles leads to a significant reduction of its toxicity, as was suggested previously. The results for radium in beagles compare well with those for humans and suggest that the models derived for strontium and plutonium in beagles may be translated to humans. The significant model parameters for the accurate animal data could then also be used to fit human epidemiological data.     

Estimating Costs and Benefits of CTL Escape Mutations in SIV/HIV Infection

Mutations that allow SIV/HIV to avoid the cytotoxic T lymphocyte (CTL) response are well documented. Recently, there have been a few attempts at estimating the costs of CTL escape mutations in terms of the reduction in viral fitness and the killing rate at which the CTL response specific to one viral epitope clears virus-infected cells. Using a mathematical model we show that estimation of both parameters depends critically on the underlying changes in the replication rate of the virus and the changes in the killing rate over time (which in previous studies were assumed to be constant). We provide a theoretical basis for estimation of these parameters using in vivo data. In particular, we show that 1) by assuming unlimited virus growth one can obtain a minimal estimate of the fitness cost of the escape mutation, and 2) by assuming no virus growth during the escape, one can obtain a minimal estimate of the average killing rate. We also discuss the conditions under which better estimates of the average killing rate can be obtained.     

HDACi phenylbutyrate increases bystander killing of HSV-tk transfected glioma cells

Malignant glioma patients have a dismal prognosis with an urgent need of new treatment modalities. Previously developed gene therapies for brain tumors showed promising results in experimental animal models, but failed in clinical trials due to low transfection rates and insufficient expression of the transgene in tumor cells, as well as low bystander killing effects. We have previously shown that the histone deacetylase inhibitor 4-phenylbutyrate (4-PB) enhances gap junction communication between glioma cells in culture. In this study, we demonstrate an activation of recombinant HSV-tk gene expression, and a dramatic enhancement of gap junction-mediated bystander killing effect by administration of the HSV-tk prodrug ganciclovir together with 4-PB. These findings that 4-PB potentiates "suicide gene" expression as well as enhances gap junctional communication and bystander killing of tumor cells justify further testing of this paradigm as an adjunct to suicide gene therapy of malignant gliomas.     

Cytotoxic T lymphocytes induce caspase-dependent and -independent cell death in neuroblastomas in a MHC-nonrestricted fashion

The MHC class I- restricted processing and presentation pathway is frequently nonfunctional in tumor cells; therefore, the direct targeting of tumor cells by CTLs may be difficult, if at all possible, to achieve. We used neuroblastoma (NB), which represents a striking example of a tumor with an impaired MHC class I pathway, as a model to study bystander effects of activated T lymphocytes on tumor cells. We found that NB cell lines are susceptible to killing by differentiated CD8(+) CTL clones in a MHC class I-nonrestricted manner that involves two programs of cell death distinguished on the basis of different kinetics, sensitivities to caspase inhibitors, and cytokine-blocking reagents. The "early" death exhibited characteristic features of apoptosis, whereas the "delayed" caspase-independent death exhibited features associated with necrosis and was partially inhibited by TNF-alpha-blocking and prevented by overexpression of Bcl-2 or Bcl-x(L). Our data reveal a previously unappreciated complexity of death pathways induced in tumor cells by immune activation and suggest that redirecting nonspecific effector CTLs to even a small proportion of NB cells or activating CTLs in a tumor's proximity may have therapeutic effects in patients with NB.     

Poxvirus-encoded serpins do not prevent cytolytic T cell-mediated recovery from primary infections.

Previous observations that the highly conserved poxvirus-encoded serpins inhibit cytotoxic activities of alloreactive CTL via granule and/or Fas-mediated pathways was taken to indicate their involvement in immune evasion by poxviruses. We now show that interference with 51Cr release from target cells by ectromelia and cowpoxvirus is limited to alloreactive but not MHC-restricted CTL. The data are in support of the paramount importance of CTL and its effector molecule perforin in the recovery from primary ectromelia virus infection and question the role of serpins in the evasion of poxviruses from killing by CTL. Further analysis of poxvirus interference with target cell lysis by alloreactive CTL revealed that suppression primarily affects the Fas-mediated, and to a lesser extent, the granule exocytosis pathway. Serpin-2 is the main contributor to suppression for both killing pathways. In addition, inhibition of lysis was shown to be both target cell type- and MHC allotype-dependent. We hypothesize that differences in TCR affinities and/or state of activation between alloreactive and MHC-restricted CTL as well as the quality (origin) of target cells are responsible for the observed phenomenon.     

Epitope-dependent avidity thresholds for cytotoxic T-lymphocyte clearance of virus-infected cells

Cytotoxic T lymphocytes (CTLs) are crucial for immune control of viral infections. "Functional avidity," defined by the sensitizing dose of exogenously added epitope yielding half-maximal CTL triggering against uninfected target cells (SD(50)), has been utilized extensively as a measure of antiviral efficiency. However, CTLs recognize infected cells via endogenously produced epitopes, and the relationship of SD(50) to antiviral activity has never been directly revealed. We elucidate this relationship by comparing CTL killing of cells infected with panels of epitope-variant viruses to the corresponding SD(50) for the variant epitopes. This reveals a steeply sigmoid relationship between avidity and infected cell killing, with avidity thresholds (defined as the SD(50) required for CTL to achieve 50% efficiency of infected cell killing [KE(50)]), below which infected cell killing rapidly drops to none and above which killing efficiency rapidly plateaus. Three CTL clones recognizing the same viral epitope show the same KE(50) despite differential recognition of individual epitope variants, while CTLs recognizing another epitope show a 10-fold-higher KE(50), demonstrating epitope dependence of KE(50). Finally, the ability of CTLs to suppress viral replication depends on the same threshold KE(50). Thus, defining KE(50) values is required to interpret the significance of functional avidity measurements and predict CTL efficacy against virus-infected cells in pathogenesis and vaccine studies.     

Development of a Novel In Silico Docking Simulation Model for the Fine HIV-1 Cytotoxic T Lymphocyte Epitope Mapping

Introduction

Class I HLA''s polymorphism has hampered CTL epitope mapping with laborious experiments. Objectives are 1) to evaluate the novel in silico model in predicting previously reported epitopes in comparison with existing program, and 2) to apply the model to predict optimal epitopes with HLA using experimental results.

Materials and Methods

We have developed a novel in silico epitope prediction method, based on HLA crystal structure and a peptide docking simulation model, calculating the peptide-HLA binding affinity at four amino acid residues in each terminal. It was applied to predict 52 HIV best–defined CTL epitopes from 15-mer overlapping peptides, and its predictive ability was compared with the HLA binding motif-based program of HLArestrictor. It was then used to predict HIV-1 Gag optimal epitopes from previous ELISpot results.

Results

43/52 (82.7%) epitopes were detected by the novel model, whereas 37 (71.2%) by HLArestrictor. We also found a significant reduction in epitope detection rates for longer epitopes in HLArestrictor (p = 0.027), but not in the novel model. Improved epitope prediction was also found by introducing both models, especially in specificity (p<0.001). Eight peptides were predicted as novel, immunodominant epitopes in both models.

Discussion

This novel model can predict optimal CTL epitopes, which were not detected by an existing program. This model is potentially useful not only for narrowing down optimal epitopes, but predicting rare HLA alleles with less information. By introducing different principal models, epitope prediction will be more precise.     

Increased PRAME antigen-specific killing of malignant cell lines by low avidity CTL clones, following treatment with 5-Aza-2��-Deoxycytidine

The cancer testis antigen Preferentially Expressed Antigen of Melanoma (PRAME) is overexpressed in many solid tumours and haematological malignancies whilst showing minimal expression in normal tissues and is therefore a promising target for immunotherapy. HLA-A0201-restricted peptide epitopes from PRAME have previously been identified as potential immunogens to drive antigen-specific autologous CTL responses, capable of lysing PRAME expressing tumour cells. CTL lines, from 13 normal donors and 10 melanoma patients, all of whom were HLA-A0201 positive, were generated against the PRAME peptide epitope PRA(100-108). Specific killing activity against PRA(100-108) peptide-pulsed targets was weak compared with CTL lines directed against known immunodominant peptides. Moreover, limiting dilution cloning from selected PRAME-specific CTL lines resulted in the generation of a clone of only low to intermediate avidity. Addition of the demethylating agent 5-aza-2'-Deoxycytidine (DAC) increased PRAME expression in 7 out of 11 malignant cell lines including several B lineage leukaemia lines and also increased class I expression. Pre-treatment of target cells was associated with increased sensitivity to antigen-specific killing by the low avidity CTL. When CTL, as well as of the target cells, were treated, the antigen-specific killing was further augmented. Interestingly, one HLA-A0201-negative DAC-treated line (RAJI) showed increased sensitivity to killing by clones despite a failure of expression of PRAME or HLA-A0201. Together these data point to a general increased augmentation of cancer immunogenocity by DAC involving both antigen-specific and non-specific mechanisms.     

Modelling the dynamics of LCMV infection in mice: II. Compartmental structure and immunopathology

In this study, we develop a mathematical model for analysis of the compartmental aspects and immunopathology of lymphocytic choriomeningitis virus (LCMV) infection in mice. We used sets of original and published data on systemic (extrasplenic) virus distribution to estimate the parameters of virus growth and elimination for spleen and other anatomical compartments, such as the liver, kidney, thymus and lung as well as transfer rates between blood and the above organs. A mathematical model quantitatively integrating the virus distribution kinetics in the host, the specific cytotoxic T lymphocyte (CTL) response in spleen and the re-circulation of effector CTL between spleen, blood and liver is advanced to describe the CTL-mediated immunopathology (hepatitis) in mice infected with LCMV. For intravenous and "peripheral" routes of infection we examine the severity of the liver disease, as a function of the virus dose and the host's immune status characterized by the numbers of precursor and/or cytolytic effector CTL. The model is used to predict the efficacy of protection against virus persistence and disease in a localized viral infection as a function of the composition of CTL population. The modelling analysis suggests quantitative demands to CTL memory for maximal protection against a wide range of doses of infection with a primarily peripheral site of virus replication without the risk of favoring immunopathology. It specifies objectives for CTL vaccination to ensure virus elimination with minimal immunopathology vs. vaccination for disease.     

The Kinematics of Cytotoxic Lymphocytes Influence Their Ability to Kill Target Cells

Cytotoxic lymphocytes (CTL) have been reported to show a range of motility patterns from rapid long-range tracking to complete arrest, but how and whether these kinematics affect their ability to kill target cells is not known. Many in vitro killing assays utilize cell lines and tumour-derived cells as targets, which may be of limited relevance to the kinetics of CTL-mediated killing of somatic cells. Here, live-cell microscopy is used to examine the interactions of CTL and primary murine skin cells presenting antigens. We developed a qualitative and quantitative killing assay using extended-duration fluorescence time-lapse microscopy coupled with large-volume objective software-based data analysis to obtain population data of cell-to-cell interactions, motility and apoptosis. In vivo and ex vivo activated antigen-specific cytotoxic lymphocytes were added to primary keratinocyte targets in culture with fluorometric detection of caspase-3 activation in targets as an objective determinant of apoptosis. We found that activated CTL achieved contact-dependent apoptosis of non-tumour targets after a period of prolonged attachment – on average 21 hours – which was determined by target cell type, amount of antigen, and activation status of CTL. Activation of CTL even without engagement of the T cell receptor was sufficient to mobilise cells significantly above baseline, while the addition of cognate antigen further enhanced their motility. Highly activated CTL showed markedly increased vector displacement, and velocity, and lead to increased antigen-specific target cell death. These data show that the inherent kinematics of CTL correlate directly with their ability to kill non-tumour cells presenting cognate antigen.