This web page was produced as an assignment for an
     undergraduate course at Davidson College.

Terminal Deoxynucleotidyl Transferase (TdT)

  1. Introduction
  2. Overview of Development of Lymphocyte Diversity
  3. Normal Function of TdT
  4. Effect of Missing TdT
  5. Drug Interactions with TdT Activity
  6. Laboratory Applications of TdT
  7. References



The transfer of genetic information from one generation to the next is dependent on accurate replication of parental DNA.  To accomplish this goal, DNA polymerases (along with many other enzymes) have evolved which are able to copy DNA from a template strand very effectively.  However, not all DNA polymerization that takes place in an organism is based on a template.  Terminal deoxynucleotidyl transferase (TdT) is a template-independent DNA polymerase which is capable of catalyzing the elongation of a DNA strand by the addition of nucleotides from the surrounding solution.  Such nucleotide addition is required in a few specific situations, the most studied of which is lymphocyte development.  This web page will explore the role of TdT in the development of lymphocyte diversity.

Return to the Top of the Page

Overview of Development of Lymphocyte Diversity

Adults have an incredibly diverse repertoire of B and T cell antigen receptors.  It has been estimated that the numbers approach 1014 different specific immunoglobins (Ig's)  and 1018 distinct T cell receptors (TCR's).  This incredible diversity greatly exceeds the number of different genes in the entire human genome, making it impossible for each specific antigen receptor to be coded by an individual gene.  Instead, mechanisms have evolved which derive a large amount of diversity from a relatively small number of genes.  This web page will focus on the development of diversity in Ig's, but it should be noted that similar mechanisms are at work in the development of diversity in TCR's.  In humans, there are several mechanisms which work together to generate Ig diversity.  Ig's are composed of two identical light chains combined with two identical heavy chains, all four of which are composed of a variable (V) region and a constant (C) region.  Since each individual can form many different light and heavy chains, association of different light and heavy chains leads to significant diversity.  Further diversity is obtained in the formation of individual light and heavy chain V-regions by splicing together different regions of the gene segment to form an active exon.  These regions, called V and J in the light chain and V, D, and J in the heavy chain, are present in multiple copies in each gene segment.  Therefore the choice of specific V(D)J joining adds considerable diversity to the immunoglobin.  The amount of diversity is increased yet again by the fact that the splicing of V, D, and J regions described above is not precise.  Nucleotides are often added to or deleted from the joint.  Nucleotides can be added to the joint by two methods.  The first method occurs through the formation and cleavage of hairpins during DNA splicing and results in the insertion of pallindromic segments of DNA (P-nucleotides).  The second method is the random insertion of non template (N) nucleotides, which is catalyzed by TdT.  Diversity is increased yet again by a process known as somatic hypermutation much later in the process of immunoglobin development (Janeway et al., 1999).  This page will focus on the role of TdT in the insertion of N-nucleotides.  For a more detailed look at the development of lymphocyte diversity, check out Dr. Campbell's somatic recombination animation (Campbell, 2000).

Return to the Top of the Page

Normal Function of TdT


Early research in to the joining of D and J segments led to the discovery that some nucleotides are present in the joint that were not part of either segment before they were joined (Alt et al., 1982).  This study was among the first to implicate TdT as a possible mechanism for the insertion of these nucleotides. This conclusion was based on TdT's high concentration in the thymus and bone marrow (where such recombination events occur) and the fact that many of the inserted nucleotides were guanine (G) residues, which TdT had been shown to prefer over other nucleotides (Alt et al., 1982).  Later research has shown that in addition to its G utilization preference, TdT is affected by steric interactions of nucleotide stacking which favor the insertion long strings of purines or pyrmidines, making the process somewhat less random than previously thought (Gauss et al., 1996).  The rate of nontemplated DNA addition by TdT has been found to be at least 76-fold slower than that of templated synthesis (Clark, 1988).

Regulation of Expression

Because TdT is capable of adding non template nucleotides to any exposed 3' end of DNA, it is very important that the body limits its expression to times and places where its function is beneficial.  Studies in mice have shown that TdT is not detected in neonatal thymi, but begins to appear about four days after birth (Bogue et al., 1993).  There is then a one to two day lag period before N-nucleotide addition is observed (Bogue et al., 1992).  It has been suggested that the lack of neonatal TdT is related to the smaller amount of self-reactivity observed in immunoglobins produced without the addition of N-nucleotides (Bogue et al., 1992).  If one considers that a fetus is still under the immune protection of the mother, it makes sense that the added immunoglobin diversity that N-nucleotide addition provides is not selected for strongly enough to outweigh the problem of creating many self-reactive antibodies.

Expression of TdT is tightly controlled in adults as well.  TdT expression in the thymus and bone marrow closely follow that of RAG-1 and RAG-2, which combine to form a complex which cuts one strand of the DNA prior to V(D)J joining.  The only difference in expression is that TdT is expressed only during stages while heavy chain rearrangement is occurring while RAG-1 and RAG-2 are expressed during both heavy and light chain rearrangement (Li et al., 1993).  This tight control of expression leads to the observation that N-nucleotide addition is common in heavy chain rearrangement and very rare in light chain rearrangement.  The stage in B cell development in which the heavy chain is rearranged is known as the pro-B cell stage.  Therefore only cells in this stage exhibit TdT.  Staining for TdT can reveal the location of these cells as shown in Figure 1 below.

Figure 1:  Part of a transverse section of rat femur showing fluorecence labeled TdT cells (green).  These cells are most highly concentrated toward the outside of the bone (upper right corner) and less numerous toward the middle of the bone (lower left corner) because more developed cells that no longer express TdT are moving toward the middle of the bone.  (Photograph courtesy of D. Opstelten and M. Hermans.  From Janeway et al.,Immunobiology, 4th ed., Garland Publishing,1999, p. 198)

 Recent studies have shown that regulation of TdT expression is controlled on the DNA level by the selective methylation of promoter regions of genes which code for the protein (Nourrit et al., 1999).  It has also been shown that the RAG-1/RAG-2 complex plays a vital role in the regulation of N-nucleotide addition by persisting on the DNA after single strand cleavage, blocking TdT from adding nucleotides until both strands are cleaved (Grawunder et al., 1997).

Although the average number of N-nucleotides added to each junction correlates with the amount of TdT expression, it is thus far impossible to predict the number of N-nucleotides which will be added to a given junction.  It has been shown that the level of diversity is not coded in the stem cell and is probably dependent on environmental factors (Bogue et al., 1993).  Early work predicted that N-nucleotides would be added until a homology of one nucleotide occurred, at which point templated DNA polymerases would finish the joint (Alt et al., 1982).  However, later work has shown that such homologies are not necessary for N-nucleotide addition to stop, nor does their presence dictate a stop in N-nucleotide addition.  Junction formation has been observed in joints in which zero, one, two, or more homologies exist (Komori et al., 1993).

Return to the Top of the Page

Effect of Missing TdT

Studies on mice have shown that animals which lack TdT expression have en extremely small number of N-nucleotides (Gilfillan et al., 1993).  This lack of N-nucleotides limits the number of different antigen receptors seen in these mice due to fewer possible recombinations.  However, the observed decrease in diversity was much greater than could be explained by this logic alone.  Apparently, recombination in mice is much more strongly governed by homology than it is in humans, leading to the preferential formation of certain junction combinations at the expense of others.  TdT driven addition of N-nucleotides eliminates this bias for certain pairings and therefore plays a more important role than expected in the generation of diversity in mice (Gilfillan et al., 1993).

Return to the Top of the Page

Drug Interactions with TdT Activity

TdT adds nontemplate nucleotides to DNA sequences with an exposed 3' end.  This addition depends on the presence of a hydroxyl group at the 3' position on the terminal nucleotide.  Therefore, nucleotide addition can be limited to one nucleotide if a modified nucleotide lacking a 3' hydroxyl group is present in the solution.  One such nucleotide, [alpha]cordycepin-5'-triphosphate, has been shown to accomplish this limitation (Promega, 2000).  This interaction may prove applicable to the treatment of refractory TdT-positive leukemia, a type of leukemia in which the cancerous cells are TdT+.  One such treatment involving the nucleotide described above is currently in phase I of study (Division of Hematology / Oncology at New England Medical Center, 1999).

Return to the Top of the Page

Laboratory Applications of TdT

Because of its ability to add nucleotides to any DNA strand with a free 3' end, TdT has been adopted as a useful tool in immunology research.  When cells undergo apoptosis, their DNA is cleaved into many small segments which diplay free 3' ends.  These apoptotic cells can be labeled with biotin-coupled uridine by exposing them to a solution containing TdT and these uridine residues.  These biotin labels can then be detected with enzyme bound to streptavidin, which is known to bind to biotin.  If the subtrate to this ezyme, a colorless solution, is added to the culture, a colored precipitate will be produced in cells which have undergone apoptosis (Janeway et al., 1999).  A similar method which uses fluorescence techniques instead of biotin labeling has also been developed to locate apoptotic cells (Mitra, 1996).

Return to the Top of the Page


Alt, F.W. and D. Baltimore.  1982Procedings of the  National Academy of Science, USA.  79, 4118-4122.

Bogue, M., S. Gilfillan, C. Benoist, and D. Mathis. 1992. Procedings of the National Academy of Science, USA. 89,11011-11015.

Bogue, M., H. Mossmann, U. Stauffer, C. Benoist, and D. Mathis.  1993European Journal of Immunology. 23, 1185-1188.

Campbell, A.M.  2000.  Flash Animation of Somatic Recombination of Antibody Light Chain.  <>  Accessed 2000 March 1.

Clark, J.M.  1988Nucleic Acids Research16, 9677-9686.

Division of Hematology / Oncology at New England Medical Center.  1999.  Cancer Clinical Trials Office:  Cordycepin.  <>  Accessed 2000 Feb 28.

Gauss, G.H. and M.R. Lieber.  1995. Molecular and Cellular Biology16, 258-269.

Gilfillan, S., A. Dierich, M. Lemeur, C. Benoist, and D. Mathis.  1993Science261, 1175-1178.

Grawunder, U. and M.R. Lieber.  1997. Nucleic Acids Research25, 1375-1382.

Janeway, C.A., P. Travers, M. Walport, and J.D. Capra. 1999. Immunobiology, 4th ed.  Garland Publishing, New York.

Komori, T., A. Okada, V. Stewart, and F.W. Alt. 1993. Science261, 1171-1175.

Lewis, S.M.  1994Advances in Immunology. 56, 27-150.

Li, Y.S., K. Hayakawa, and R.R. Hardy.  1993. Journal of Exp. Medicine.  178, 195-960.

Mitra, G.  1996.  Detection of Apoptotic Cells using the Apoptosis Detection System, Fluorescein:  Promega Notes Magazine (57).  <>  Accessed 2000 Feb 28.

Nourrit, F., I. Coquilleau, M.F. D'Andon, F. Rougeon, and N. Doyen.  1999Journal of Molecular Biology. 292, 217-227.

Promega Corporation.  Feb 2000.  Abstract For Terminal Deoxynucleotidyl Transferase.  <>  Accessed 2000 Feb 28.

Return to the Top of the Page

Return to Bob Magnussen's Immunology Home Page

Return to Immunology Main Page

Send comments, questions, and suggestions to: