John Welch.jpg

John S. Welch, MD, PhD

Assistant Professor
Department of Medicine
Oncology Division
Stem Cell Biology

Research Interests

  • Acute myeloid leukemia
  • Myelodysplastic syndrome
  • Genomics


  • 314-362-2626 (tel)
  • 314-362-9333 (fax)
  • Southwest Tower, Room 626 (office)
  • Southwest Tower, Room 626 (lab)
  • Division of Oncology
    Section of Stem Cell Biology
    Campus Box 8007
    Washington University Medical School
    660 South Euclid Avenue
    St. Louis, MO 63110


The broad goal of our lab is to improve patient care through translational and basic science; we apply clinical insight to inform scientific hypotheses, use molecular technologies to illuminate disease pathogenesis, and then seek to integrate this knowledge back into clinical diagnostics and therapeutics.

Our lab is currently focused on understanding hematopoietic self-renewal and differentiation, and how these processes are altered by leukemia-associated mutations. We are currently focused on two sets of proteins:


Cohesins are highly conserved proteins that form ring structures that encircle DNA. This functions to either hold sister chromatids together during anaphase, or to facilitate enhancer/promoter interactions. We recently found that four of the cohesins are recurrently mutated in patients with acute myeloid leukemia. These mutations tend to be loss-of-function, mutually exclusive of one another, and are nearly always associated with a normal karyotype. It is unclear how these mutations lead to leukemia. We are currently using mouse models and tissue culture approaches to better define how cohesin mutations contribute to leukemia.

Ligand-dependent nuclear receptors

Nuclear receptors are ligand-dependent transcription factors. The retinoid receptors represent an important class of nuclear receptors. These transcription factors can be turned on and off by the presence or absence of a small hormone molecule (a vitamin A metabolite called a retinoid). This makes them ideal targets for drugs. The retinoid receptors play important roles maintaining hematopoietic self-renewal and inducing myeloid maturation. One retinoid receptor, RARA, is associated with recurrent fusion translocations in AML. There are multiple drugs that have been developed against the retinoid receptors, and against other nuclear receptors, but there has been limited clinical impact of these drugs. We believe this is partly because the distribution and regulation of natural retinoids has not been well defined, so it is difficult to predict when the addition of an agonist or an antagonist might be beneficial.

We built an in vivo reporter mouse that can detect the presence of natural retinoids. This system is highly sensitive and specific. We are using this to define which bone marrow cells are exposed to retinoid receptor ligands, when this changes, whether bone marrow cells respond to hematopoietic stress by modulating growth and differentiation through the retinoid receptors, and whether we can rationally identify new conditions that would be amenable to treatment by using a synthetic retinoid drug.

There are a total of 48 known human nuclear receptors. Most of these are modulated by the presence and absence of a ligand. Many of these are expressed during blood cell development. On-going projects will apply this reporter system to discover which other nuclear receptors are being activated by a ligand, when, and whether we could use this information to predict new applications for existing small molecules.


receptor construct

Figure: Specific response of UAS-GFP reporter mice to Gal4-fusion proteins. Bone marrow cells were transduced with MSCV-Gal4-RARA-IRES-mCherry, or MSCV-Gal4-RARG-IRES-mCherry, and then treated with receptor-specific ligands (RARA: BMS753; RARG: BMS961). Note, that the GFP+ cells only occur when treated with the receptor-appropriate ligands, and that the GFP+ cells are also mCherry+.