top of page

Samuel Silverstein, M.D.

Emeritus John C. Dalton Professor of Physiology & Cellular Biophysics, and Professor of Medicine

Structure and functions of polymorphonuclear (PMN), mononuclear phagocytes, lymphocytes, platelets and endothelial cells (EC) in innate immunity and diseases.



Cellular Immunotherapy: My colleagues and I have developed a quantitative, experimentally verified model for cellular immunotherapy of neoplastic and infectious diseases. The model incorporates an equation derived previously to describe neutrophil bactericidal activity in three dimensional fibrin gels in vitro and in sub-cutaneous bacterial infections in vivo. As with neutrophils, we have found that the concentration of antigen-specific human and mouse CD8+T-cells determines their efficacy in killing cognate antigen–expressing target cells in two dimensional tissue culture plates, in three dimensional collagen-fibrin gels, and in established melanomas in vivo.

Using limiting dilution assays we have found that only 2-2.8% of CD8+T-cells are cytolytically active, and that the cytolytically active CD8+T-cells account for all tumoricidal activity of the entire CD8+T-cell population. Cytokines, such as IL-15 and IL-21 and adjuvants such as anti-CD40, in combination with IL-2, increase the fraction of cytolytically active CD8+T-cells in a population. Quantitative analyses of these cytokine-activated CD8+T-cells show that CD8+T-cell activation in the presence of specific cytokines and/or adjuvants increases not only the percentage of cytolytically active CD8+T-cells in a population, but also the specific cytolytic activity of each T-cell. Thus, activation of CD8+T-cells in the presence of cognate antigen produces an increase in the number of antigen-specific CD8+T-cells, while activation of CD8+T-cells in the presence of antigen plus adjuvant/cytokines produces increases in both the number and quality of the cytolytic CD8+T-cells. Thus we have been able to separate, quantitatively, the effect of antigen vs. adjuvant on the quantity and quality of CD8+T-cells.

The three dimensional collagen-fibrin gel assays we employ mimic precisely the efficiency with which antigen-specific CD8+T-cells kill cognate antigen-expressing target cells in mouse spleen in vivo. Accordingly, they can be used as a “standard” for the cytolytic potential of any preparation of tumor antigen-specific CD8+T-cells. By comparing the tumoricidal activity of a given CD8+T-cell preparation in vitro with its tumoricidal activity in vivo, we can obtain a measure of the aggregate immunosuppressive activity of the intra-tumoral environment. Our studies show it is about 50% for B16 mouse melanoma.

The equation that models CD8+T-cell killing of cognate antigen-expressing target cells (bt = b0 e-kpt + gt, in which bt is the concentration of antigen-expressing target cells at time t, b0 is the concentration of antigen-expressing target cells at t = 0, k is an experimentally derived value that defines the rate of killing of antigen-expressing target cells in ml/cognate antigen-specific CD8+T-cell/min, p is the concentration of cognate antigen-specific CD8+T-cells, and g is the growth rate of antigen-expressing target cells/min), enables us to calculate the critical intra-tumoral T-cell concentration (CTC) of total and cytolytically active antigen-specific CD8+T-cells that must be achieved to control tumor growth. For B16 melanomas in syngeneic mice these concentrations are 3 x 106 total and 6 x 104 cytolytically active antigen-specific CD8+T-cells/g tumor.

Our plans for the future are to isolate populations of cytolytically active CD8+T-cells using available high throughput single cell analysis and isolation technologies, and to use these cells to answer the following questions :

1. Is expression of cytolytic activity a stochastic or clonal property of CD8+T-cells?

2. What are the molecular characteristics of cytolytically active CD8+T-cells that distinguish them from their cytolytically inactive siblings?

3. Can we identify a plasma membrane marker that will enable us to identify and select the cytolytically active cells in a CD8+T-cell population? Alternatively, can we identify a plasma membrane marker that will enable us to identify and selectively remove the cytolytically inactive cells in a CD8+T-cell population?

4. Can we identify methods to induce 10-50% of CD8+T-cells in a population to express cytolytic activity?

5. Will populations of antigen-specific CD8+T-cells containing 10% or more cytolytically active antigen-specific CD8+T-cells produce sterilizing immunity against melanomas in mice and humans, as our calculations and studies suggest?

Selected Publications


Li, Y., Karlin, A., Loike, J.D., and Silverstein, S.C. 2004. Determination of the critical concentration of neutrophils required to block bacterial growth in tissues. J Exp Med. 200(5):613-22.

Loike, J.D., Shabtai, D.Y., Neuhut, R., Malitzky, S., Lu, E., Husemann, J., Goldberg, I.J., and Silverstein, S.C. 2004 Statin inhibition of Fc receptor-mediated phagocytosis by macrophages is modulated by cell activation and cholesterol. Arterioscler Thromb Vasc Biol. (11):2051-6.

Wyss-Coray , T., Loike, J.D., Brionne, T.C., Lu, E., Anankov, R., Yan, F., Silverstein, S.C., and Husemann, J. 2003. Adult mouse astrocytes degrade amyloid-beta in vitro and in situ.
Nat Med. (4):453-7


Husemann, J., Loike, J.D., Anankov, R., Febbraio, M., Silverstein, S.C. 2002. Scavenger receptors in neurobiology and neuropathology: their role on microglia and other cells of the nervous system. Glia. (2):195-205. Review


Li, Y., Karlin, A., Loike, J.D., Silverstein, S.C. 2002. A critical concentration of neutrophils is required for effective bacterial killing in suspension. Proc Natl Acad Sci U S A. (12):8289-94.

Berger, M., Budhu, S., Lu, E., Li, Y., Loike, D., Silverstein, S.C., Loike, J.D. 2002. Different G(i)-coupled chemoattractant receptors signal qualitatively different functions in human neutrophils. J Leukoc Biol. May;71(5):798-806.

bottom of page