University of Minnesota
MICaB Graduate Program
MICaB Home | U of M Medical School | Graduate School

Unit's home page.

MICaB Faculty

Michael Farrar
Michael A. Farrar, Ph.D.


Department of Laboratory Medicine and Pathology

Washington University, 1993, Ph.D.

612-625-0401 office
612-625-3608 lab

Research Interests:

Signal transduction and lymphocyte development

Research in my lab explores how signal transduction pathways regulate lymphocyte development and activation. In addition, we are exploring how deregulation of these signaling pathways leads to the development of cancer. 

Figure 1

Fig. 1.  Tregs (stained in green) develop in the medulla (stained in blue and magenta) of the thymus.  We are currently trying to understand how this specific microenvironment promotes Treg development. 

Lymphocytes develop from hematopoietic stem cells and go through characteristic stages of differentiation that results in the formation of functional, mature T or B cells. These stages are carefully regulated by the action of growth factor and cytokine receptors, as well as the clonotypic T cell and B cell antigen receptors. A key question is how signaling pathways downstream of these receptors regulate various aspects of B and T cell maturation.  One major research focus in the lab is how cytokine signaling and co-stimulatory molecules govern the development of regulatory T cells (Tregs) in the thymus (Fig. 1). Specifically, we have discovered that the transcription factor STAT5 is required for the development of regulatory T cells (Tregs). Tregs are required to prevent autoimmunity; mice and humans that lack Tregs develop multiple autoimmune syndromes. We have demonstrated that mice that lack STAT5 also fail to develop Tregs. Conversely, constitutive activation of STAT5 in Tregs leads to a large increase in this cell type. More recently we have demonstrated that TCR and cytokine signaling cooperate to drive regulatory T cell development in a two- step model of regulatory T cell development.  In this model TCR/CD28 co-stimulatory molecules lead to the development of Treg progenitor cells, which express components of the IL2R complex.  The Treg progenitors can then rapidly be converted into mature Foxp3+ Tregs by stimulation with IL2, which activates STAT5 (Fig. 2).  Our current work is focused on identifying which cells in the thymus produce the IL2 needed for Treg development and what induces these cells to produce IL2. 

Figure 2

Fig. 2.  Two step model of Treg development.

Finally, we seek to determine whether Tregs with enhanced STAT5 signaling act as more efficient suppressors of autoimmune disorders such as diabetes or inflammatory bowel disease.

Fig 3

Fig. 3.  Our current working model to describe how Flt3/Ras signals initiate an IL7R/STAT5 developmental program that underlies B cell differentiation.

The second major research focus in the lab is how cytokine pathways regulate B cell development and leukemia (Fig. 3).  This is being done using a variety of approaches including the use of mice lacking the genes required for normal B cell development and mice expressing dominant negative and activated Ras, Raf and Stat5 transgenes. Using these approaches we have discovered that activation of the transcription factor STAT5 is sufficient to largely restore normal B cell, but not T cell, development in interleukin-7-receptor deficient mice. In addition, we also have evidence that the STAT5 pathway may play an important role in the process that regulates whether a hematopoietic stem cell will develop into a B or T cell (lineage commitment). We are currently trying to identify STAT5 gene targets involved in these processes.

Figure 4

Fig. 4.  Stat5 activation combined with loss of one allele of either Pax5 or Ebf1 leads to highly penetrant B cell leukemia.

As part of our studies of STAT5 signaling in normal B cell development we discovered that mice expressing a constitutively active form of STAT5 (Stat5b-CA) develop a form of progenitor B cell leukemia that resembles human acute lymphoblastic leukemia.  The penetrance with which Stat5b-CA mice develop leukemia is only ~1-2%.  However, crossing these mice to mice with defects in pre-BCR signaling or which lack one copy of the genes encoding the transcription factors Ebf1 or Pax5 results in leukemia as quickly as 8 weeks after birth in 75-100% of the mice (Fig. 4).   We are currently using a novel genetic screen involving sleeping beauty transposon based mutagenesis to identify other genetic defects that cooperate with STAT5 activation to initiate B cell leukemia.

In analyzing these mice we discovered that Ebf1 and Pax5 do not act as typical tumor suppressor genes, in which both alleles of the tumor suppressor gene must be deleted to reveal an effect.  Rather, Ebf1 and Pax5 appear to be part of a tumor suppressor gene network involving multiple transcription factors.  These transcription factors normally function to promote B cell development and suppress differentiation into other cell lineages.  However, they also act to induce the expression of tumor suppressor genes and repress the expression of potential oncogenes. 

Fig 5

Fig. 5.
  Ebf1 and Pax5 are part of a self-reinforcing network of transcription factors that play an important role in directing normal B cell differentiation and preventing the development of B cell leukemia.

In this model (Fig. 5) defects in two alleles of any of the genes that make up the network appear to be sufficient to result in a loss of tumor suppressor function to promote B cell development and suppress differentiation into other cell lineages.  However, they also act to induce the expression of tumor suppressor genes and repress the expression of potential oncogenes.  In this model (Fig. 5) defects in two alleles of any of the genes that make up the network appear to be sufficient to result in a loss of tumor suppressor function

Finally, we are also examining whether genetic defects observed in our mouse model of leukemia are important in human leukemia.   We have observed that the STAT5 activation is increased in a substantial fraction of B cell acute lymphoblastic leukemias from human patients (B-ALL).  Importantly, we found that in at least one subset of B-ALL the magnitude of STAT5 activation was inversely correlated with overall patient survival (Fig. 6).  We are currently exploring the factors that result in differences in STAT5 activation in human leukemia in the hope of identifying potential targets that may useful for future pharmaceutical intervention.

Figure 6

Fig. 6.  The overall level of STAT5 phosphorylation, which correlates with STAT5 activation, predicts whether patients with BCR-ABL+ leukemia will respond to current treatment or not.


Selected Recent Publications:

  • M.A. Farrar and L.M. Heltemes-Harris. (2011).  Turning Transcription ON or OFF with STAT5:when more is less.  Nat Immunol 12: 1139-1140.

  • J.H. Rowe, J.M. Ertelt, M.N. Aguilera, C.Y. Law, M.A. Farrar, S.S. Way.  (2011). Expanded Foxp3+ regulatory T cells sustain pregnancy, but impair host defense exploited by prenatal bacterial pathogens. Cell Host & Microbe, 10: 54-64.

  • L. Heltemes Harris*, M Willette*, L.B. Ramsey*, Y.H. Qui, E.S. Neeley, N. Zhang, D.A. Thomas, T. Koeuth, E.C. Baechler, S.M. Kornblau, M.A. Farrar. 2011  Ebf1 or Pax5 haploinsufficiency synergizes with STAT5 activation to initiate acute lymphoblastic leukemia. The Journal of Experimental Medicine.  208: 1135-1149.

  • K.B. Vang, J. Yang, A.J. Pagan, L. Li, J. Wang, J.M. Green, A.A. Beg and M.A. Farrar.  (2010).  Cutting edge:  CD28 and c-Rel-Dependent Pathways Initiate Regulatory T Cell Development. J. Immunol. 184, 4074 -4077.

  • L. Li, C.A. Goetz, C.D.S. Katerndahl, N. Sakaguchi and M.A. Farrar (2010). A Flt3 and Ras-dependent Pathway Primes B Cell Development by Inducing A State of IL7-responsiveness.  J Immunol, 184: 1728-36.

  • M.A. Burchill, J.J Moon, H.H. Chu, K.B. Vang, C-W.J. Lio, A.L. Vegoe, C-S. Hsieh, M.K. Jenkins, and M.A. Farrar.  (2008).  Linked T Cell Receptor  and Cytokine Signaling Govern the Development of the Regulatory T cell repertoire. Immunity.  28: 112-121.

  • M.A. Burchill, J. Yang, C. Vogtenhuber, B.R. Blazar, and M.A.  Farrar.  (2007). Interleukin-2-Receptor-dependent STAT5  Activation is Required for the Development of Foxp3+ Regulatory T Cells. J. Immunol. 178: 262-270.

  • C.A. Goetz, I.R. Harmon, J.J. O’Neil, M.A. Burchill, and M.A. Farrar. STAT5 Activation Underlies Interleukin-7-dependent B Cell but not T Cell Development.
    J Immunol. Apr 15;172(8):4770-8, 2004.