Cytomegalovirus Stimulation of T-Cells Can Trigger a Secondary Natural Killer Cell Response

Nonyerem Nwaneri

The immune system is a complex network of various cells and organs that defend the body against infection by pathogens. Cells in the immune system have the ability to recognize something as either self or nonself, and they try to get rid of anything that is viewed as an invader. Viruses are a major class of pathogens. Upon first encounter with a virus, the host defense consists of an immediate "innate" response, followed by the development of a highly specific "adaptive" response. Innate immune responses involve cells, such as macrophages and natural killer (NK) cells. Macrophages respond to structures that signify the presence of a pathogen [1], while NK cells can be activated by cytokines produced by macrophages, or triggered by molecules on the surface of host cells following their infection [1,2]. The adaptive or "acquired" immunity is coordinated by helper T cells, which direct other cells' functions when the immune response is triggered. First antigen-presenting cells ingest and break down the invader molecules and display the viral antigens on their surface for T-cells to read. Reading T- cells activate B-cells, which then produce and secrete antigen-specific binding proteins or antibodies. An antibody will bind specifically with one antigen and not to others. The human body produces antibodies because, given the high concentration of infectious agent that is needed to cause disease, macrophages could not manage the infection on their own. However, antibodies outnumber the invader molecules and can help to break them down more effectively.

In cases where T-cells do not function properly as coordinators of this whole process, other body cells cannot perform their regular functions. Thus, the body is open to attack by opportunistic infections.

An opportunistic infection (OI) is an infection that takes advantage of the "weakness" of the immune system. For example, Human Immunodeficiency Virus (HIV) selectively targets T-cells and destroys them. As a person's T-cell count falls, he or she is at higher risk of getting an OI, especially as it falls below 200. Cytomegalovirus (CMV) is a virus in the herpes family that acts as an OI causing serious illness in people who have a T-cell count of less than 50. CMV infection can cause a variety of syndromes and disorders in different parts of the body, including the eyes, intestines, lungs and brain. CMV retinitis can cause blurred vision, spots or "floaters" and eventually blindness.

There is accumulating evidence that the innate and acquired arms of the immune response have important regulatory interactions with each other. For example, innate responses to pathogens may activate antigen-presenting cells so that these cells can function optimally to stimulate the subsequent acquired T-cell response. However, in chronic infections such as CMV, innate responses may be important for host defense even after a robust adaptive response is established. It is therefore of interest to evaluate the potential interrelationship of the innate and acquired arms of the immune system following stimulation with CMV antigens. Here, we report the results of a series of experiments that show that stimulation of CD4 T-cells with CMV antigens can trigger NK "innate" cell responses.

Fig. 1: Intracellular cytokine staining results for 4 donors. Subjects 1-3 are CMV seropositive, subject 4 is CMV seronegative. PBMC were incubated with control antigen (top panel) or soluble CMV antigen (bottom panel) for 19 hours. Brefeldin A was added for the last 6 hours of culture. Cells were permeabilized and stained with anti-IFN-g PE and anti-CD4 FITC. Small lymphocytes were gated electronically. Subjects 1-3 demonstrate CMV-specific IFN-g production by CD4+ lymphocytes, and marked IFN-g production by CD4- lymphocytes. Subject 4 displays no CMV-specific IFN-g production over background.

Materials and Methods

Cell preparation: Peripheral blood mononuclear cells (PBMC) were obtained from healthy donors. In some experiments, CD4+ or CD56+ cell subsets were isolated by immunomagnetic bead separation. Cells were magnetically labeled with CD4 or CD56 microbeads and passed through a separation column placed in a magnetic field. The magnetically labeled CD4+ T-cells were retained in the column while the unlabelled CD4 negative (CD4-) T-cells were washed through for collection. Positively selected CD4+ or CD56+, and CD4 or CD56 depleted populations were thus obtained. Cells that were labeled with a positive marker signify an actual T-cell (CD4+) or natural killer cell (CD56+). Cells that are unlabelled are negative and are considered depleted, non T-cell, non-NK cell populations.

CMV antigen: The soluble CMV antigen utilized was a CMV-infected tissue culture. Control antigen was similarly obtained from uninfected tissue culture. In every experiment, cultures with control antigen were set up in parallel with those containing CMV antigen.

Fig. 2: Kinetics of IFN-g production by CD4+ and CD4- lymphocytes occurred during the first 6 hours and increased until 18 hours and declined slightly at 24 hours. CD4- lymphocytes produced a strong IFN-g response at 18 hours that persisted at 24 hours. No significant IFN-g response was observed following incubation of cells with control antigen.

Elispot Assay: Resulting brown spots were counted with a dissecting microscope, each spot representing IFN-g accumulation by a single cell.

Intracellular cytokine staining (ICS): This is a technique to detect single cell cytokine production by flow cytometry and is used to evaluate cytokine (i.e. IFN-g) responses made by CD4 T-cells after stimulation with CMV or control antigen. To ensure that the cytokine produced was not secreted by the cell, Brefeldin A was used to inhibit the extracellular transport of cytokines. In order to analyze cells on a 3-color flow cytometer, they need to be stained with antibodies. Cells were stained with fluorescent-conjugated antibodies to IFN-g, CD4, CD56, CD3 or CD8. Specific subsets of T-cells were gated electronically and cytokine expression was analyzed.

Labeling of cells: A fluorescent dye was used to label cells without affecting cellular functions. In selected experiments sorted CD4+ cells were stained with the florescent dye, then added back to the CD4 depleted population. As a result, we have two T-cell populations that are distinct from each other. This technique is important in testing the first hypothesis.

Results

Comparison of ELISPOT and ICS techniques for the detection of CMV-specific IFN-g production: A preliminary experiment was performed to compare two quantitative techniques for evaluating lymphocyte responses to specific antigen: the ELISPOT and intracellular cytokine staining (ICS). Following overnight (19 hours) incubation with soluble CMV antigen, we observed strongly positive ELISPOT responses in 3/3 CMV-seropositive donors and a negative response in one CMV seronegative. No response was observed following incubation with the control antigen. ICS was performed following the addition of Brefeldin A for the last 6 hours of a 19-hour incubation. As expected, IFN-g secretion by CD4+ lymphocytes was observed in the 3 CMV seropositive donors. Unexpectedly, a strong IFN-g response was observed in CD4- T-cells (see Fig. 1). An Elispot assay was used to verify that CD4 T-cells were necessary for the stimulation of IFN-y secretion, and that NK cells were not directly stimulated by CMV antigen. When we performed an ELISPOT assay with cells that had been depleted of CD4 T-cells, CMV antigen elicited no response. These findings prompted a series of experiments to elucidate the nature of the CD4 negative cells that produced IFN-g following stimulation with soluble CMV antigen.

Kinetics of the CMV response: First, we performed a kinetic experiment in order to observe IFN-g production by CD4+ and CD4- T-cells in response to CMV stimulation. Of cells from a CMV-seropositive donor. Following the addition of CMV antigen, Brefeldin A was added for the last 5 hours of 6, 12, 18, and 24 hour cultures. Cells were harvested at these time points and ICS was performed. IFN-g production by CD4+ lymphocyte occurred during the first 6 hours and increased until 18 hours and declined slightly by 24 hours. (see Fig. 2). CD4- cells first produced a strong IFN-g response at 18 hours and this response persisted at 24 hours. Staining with CD3 antigen revealed that all the CD4 lymphocytes were CD3+, and that the CD4-IFN-g + cells were CD3- (data not shown). These results support the preliminary experiment and thus indicate that CMV antigen stimulated an immediate IFN-g response by CD4+CD3+ lymphocytes, followed after 12 hours by IFN-g production by CD4-CD3- lymphocytes.

These observations suggested two alternative hypotheses: (1) CD4 T-cells decline slightly following antigenic stimulation, which was previously, reported [3]. (2) Antigen-specific stimulation of CD4+ T-cells results in the secondary activation of a CD4- (i.e., non-T-cell) population.

Fig. 3: CMV antigen-stimulated IFN-g production by CFSE stained CD4+ lymphocytes and CFSE unstained CD4- lymphocytes. CD4+ cells were isolated by positive selection, irreversibly stained with CFSE, and added back to the CD4 depleted population. Cells were harvested and analyzed following 24 hours of antigen stimulation.

Tracking originally CD4+ cells by florescent staining: In order to test hypothesis 1, above, we used a labeling technique to identify CD4+ cells throughout the 24-hour period. To do this cells were sorted by immunomagnetic bead selection into CD4+ and CD4- populations. The CD4+ cells were then stained with a florescent marker and then added back to the non-stained (CFSE-) cells. The reconstituted population was stimulated by CMV antigen. After 24 hours, stained and non-stained cells produce IFN-y following CMV stimulation (see Fig. 3). Therefore, this result showed that the IFN-g response by CD4 negative cells was not due to a decline of CD4 T-cells, allowing us to reject hypothesis 1, above.

Examining the secondary activation of a non T-cell: Next, we set out to test the alternative hypothesis, that initial stimulation of T-cells leads to secondary triggering of NK cells. To do this, we performed a kinetic experiment to study IFN-g production by the CD3+ and CD3- T-cells that were simultaneously stained for CD56, a marker of natural killer cells. Since we are interested in the regulatory interaction of the innate and acquired arms of the immune system, it was appropriate to pinpoint this secondary non T-cell activation to a natural killer cell. Natural killer (NK) cells are part of the innate response. Analysises of CD3+ population (top panel) again showed immediate IFN-g production by CD3+CD56-, which peaks at 12 hours and is reduced by 24 hours (see Fig. 4). The CD3- population (non - CD4+CD3+ T-cell population) started to produce IFN-g at 18 hours and sustained its production at 24 hours. This response was made by CD56+ natural killer cells. Therefore, these results show that an antigen-specific T-cell response can result in activating natural killer cells.

Fig. 4: Kinetics of IFN-g production by CD3+CD56-, and CD3-CD56+ lymphocytes following stimulation with CMV antigen. PBMC were stimulated with CMV for the indicated times. The top panel shows an early IFN-g response by CD3+ lymphocytes that are CD56-. The bottom panel shows a delayed IFN-g response by CD3-CD56+ lymphocytes.

Discussion

The presented results demonstrate that a secondary CD4 T-cell response against CMV can lead to the triggering of a non-specific innate immune response (an NK cell response). These findings are consistent with the observation that a patient lacking NK cells was subject to recurring herpes virus infections [3] and suggest that, natural killer cell responses may have an important role in host defense even after an acquired T-cell response has been mounted to infection. This complements the established view that innate immune responses provide immediate host defense following exposure to pathogen and set the stage for the development of an optimal acquired immune response.

It has been shown previously that bacterial superantigens can activate NK cells when added to cultured PBMC [4,5]. The latter studies suggested that the mechanism for NK activation was a paracrine effect on NK cells of IL-2 and IFN-g cytokine production by T-cells. While we did not demonstrate this our study, the kinetic relationships that we observed are consistent with a paracrine mechanism.

As shown, IFN-g production by cells following antigen stimulation is frequently used as a measure of antigen-specific T-cell responses in clinical studies. Our data show that variations in NK cell responsiveness may confound the interpretation of such studies unless the role of different lymphocyte subsets and the kinetics of the immune response are taken into consideration.

In summary, we have demonstrated that CD4+ T-cell responses elicited by CMV result in the activation of NK cell responses. Therefore, even during the activation of an acquired T-cell response, a secondary NK response may be recruited to more effectively control herpes cytomegalovirus and other opportunistic infections. This secondary NK cell recruitment may also be necessary to restore the weakened immune system into a robust system that can now fight against opportunistic infections.

Key Terms

Antigen (Ag): proteins specific to particular microorganism. The antigens act as an identity card that allows our immune system to recognize invader organisms that need to be eliminated.

Antigen Presenting Cells (APC): Cells that ingest pathogens and display the antigen's degraded components on the surface. Dendritic cells, macrophages, and B cells are major antigen presenting cells.

Brefeldin A (BfA): inhibits transport of cytokines through the Golgi apparatus and blocks secretion.

Clusters of Differentiation (CD_): groups of single-cloned antibodies that identify the same cell-surface receptors, abbreviated by CD followed by an arbitrary number.

CD3: associated with T-cell receptor, required for surface expression.

CD4: coreceptor for MHC II molecules found on T helper cells.

CD56: coreceptor for natural killer cells (NK cells).

Cytokines: proteins that affect the behavior of cells. Cytokines of lymphocytes are termed interleukins and act on specific receptors on the cells they affect.

IFN-g: macrophage activator, expressed on T-cells and Natural Killer cells.

Interleukin-2 (IL-2): a cytokine produced by T-cells, CD25 and CD122 receptors.

Paracrine effect: Cellular signaling in which a factor secreted by a cell affects other cells in the local environment.

Peripheral Blood Mononuclear Cells (PBMC): lymphocytes and monocytes isolated from peripheral blood through Density Centrifugation.

Superantigens: Microbial antigens that have an extremely potent activating effect on T-cells that bear a specific variable region.

T-cells: Lymphocytes that develop and differentiate in the thymus.

Acknowledgements

Haslett, Patrick AJ and Kaplan, Gilla. This study was conducted in the Steinman Laboratory of Cellular Physiology and Immunology at the Rockefeller University, NewYork.

References

[1] Medzhitov R, Janeway CJ, Jr. Review Articles: Advances in Immunology: Innate Immunity. N Engl J Med 2000; 343:338-344.

[2] Colonna M, Nakajima H, Cella M. Inhibitory and activating receptors involved in immune surveillance by human NK and myeloid cells. J Leukoc Biol 1999; 66:718-722.

[3] Biron CA, Byron KS, Sullivan JL. Severe herpes virus infections in an adolescent without natural killer cells. N Engl J Med 1989; 320:1731-1735.

[4] Hengel H, Brune W, Koszinowski UH. Immune evasion by cytomegalovirus--survival strategies of a highly adapted opportunist. Trends Microbiology 1998; 6:190-197.

[5] Viola, A. and Lanzavecchia, A., 1996. T-cell activation determined by T-cell receptor number and tunable thresholds. Science, 273: 104-6.

Last modified 9/10/02