Scientific Highlights

In Amsterdam, I made mouse IgG₁ monoclonal antibodies to peroxidase to link peroxidase to cell surface antigens via mouse IgG₁ monoclonal antibodies against cell surface antigens using polyclonal anti-mouse IgG₁ (1,2). Based on the bivalency of IgG antibodies and the presence of two identical heavy chains in each of such antibody molecules, I speculated that monoclonal antibodies specific for selected epitopes present on the heavy chain of mouse IgG₁ molecules might be able to cross-link two mouse IgG₁ monoclonal antibodies into a stable tetrameric antibody complex. To test this hypothesis, I made (and found!) suitable (rat) monoclonal antibodies specific for mouse IgG₁ (3). See Figure 1.

Figure 1. Tetrameric complexes of monoclonal rat and mouse antibodies (from (3). A. Staining for peroxidase and alkaline phosphatase following agar gel electrophoresis of mouse IgG₁ anti-enzyme antibodies before and after addition of selected monoclonal rat anti-mouse IgG₁. Lane 1: mouse IgG₁ anti-peroxidase (a). Lane 2: mouse IgG₁ anti- alkaline phosphatase (b). Lane 3: mouse IgG₁ anti peroxidase plus an equimolar amount of rat anti-mouse IgG₁. Note: no free mouse antibody. Lane 4: mouse IgG₁ anti- alkaline phosphatase plus rat anti-mouse IgG₁. Lane 5: equimolar amounts of mouse IgG₁ anti- peroxidase and mouse IgG₁ anti- alkaline phosphatase mixed before addition of an equimolar amount of rat anti-mouse IgG₁. Note middle band with bispecific monoclonal tetrameric antibody complexes! Lane 6:  A mixture of the complexes shown in lane 3 and 4. Note the absence of bispecific complexes, indicating high stability of tetrameric antibody complexes. Lane 7. As lane 5, using a limiting amount of rat anti-mouse IgG₁ which shows free mouse IgG1 antibodies. B. Time course of tetrameric antibody complex formation. Mouse IgG₁ anti-peroxidase (lane 1) was mixed with an equimolar amount of monoclonal rat anti-mouse IgG₁. At the indicated time points the mixture was subjected to agarose gel electrophoresis and anti-peroxidase was visualized after blotting as in A. Note that antibody complexes are formed within seconds and not affected by electrophoresis. C. Scanning electron microscopy of purified tetramolecular antibody complexes. Scale bar 10 nm.

The ability to easily crosslink two different monoclonal antibodies into stable tetrameric antibody complexes (US patent No. 4,868,109) has found numerous applications in biotechnology. For example, we described applications in cell separation (4,5) and fluorescent labeling of cell surface antigens (6). Tetrameric antibody complexes are key to cell separation and many other technologies commercialized by STEMCELL Technologies in Vancouver, the largest biotechnology company in Canada.  

References: 

1. Lansdorp, P.M., Astaldi, G.C., Oosterhof, F., Janssen, M.C. and Zeijlemaker, W.P. (1980) Immunoperoxidase procedures to detect monoclonal antibodies against cell surface antigens. Quantitation of binding and staining of individual cells. J Immunol Methods, 39, 393-405.
2. Lansdorp, P.M., van der Kwast, T.H., de Boer, M. and Zeijlemaker, W.P. (1984) Stepwise amplified immunoperoxidase (PAP) staining. I. Cellular morphology in relation to membrane markers. J Histochem Cytochem, 32, 172-178.
3. Lansdorp, P.M., Aalberse, R.C., Bos, R., Schutter, W.G. and Van Bruggen, E.F. (1986) Cyclic tetramolecular complexes of monoclonal antibodies: a new type of cross-linking reagent. Eur J Immunol, 16, 679-683.
4. Thomas, T.E., Abraham, S.J., Otter, A.J., Blackmore, E.W. and Lansdorp, P.M. (1992) High gradient magnetic separation of cells on the basis of expression levels of cell surface antigens. J Immunol Methods, 154, 245-252.
5. Thomas, T.E., Sutherland, H.J. and Lansdorp, P.M. (1989) Specific binding and release of cells from beads using cleavable tetrameric antibody complexes. J Immunol Methods, 120, 221-231.
6. Wognum, A.W., Thomas, T.E. and Lansdorp, P.M. (1987) Use of tetrameric antibody complexes to stain cells for flow cytometry. Cytometry, 8, 366-371.

During my graduate studies in Amsterdam my research interest shifted from immunology to human stem cell biology and the purification of blood-forming stem cells using monoclonal antibodies became a research priority for me. At the time, it was believed that purified stem cells, together with defined culture conditions, would allow unlimited expansion (self-renewal) of stem cells for clinical purposes, such as gene therapy and transplantation. During a meeting in London I met with Allen Eaves, the director of the Terry Fox Laboratory at the time. Allen convinced me to visit Vancouver and to do postdoctoral work in the lab of his wife, Connie Eaves, enabled by a Terry Fox fellowship. We made several monoclonal antibodies with remaining relevance for stem cell research including antibodies to CD34 and CD90. Our antibodies proved to be superior reagents to detect e.g. CD34 cells in peripheral blood (1) and identify and isolate stem cells in combination with antibodies to CD38 (2). 

References:

1. Siena, S., Bregni, M., Brando, B., Belli, N., Ravagnani, F., Gandola, L., Stern, A.C., Lansdorp, P.M., Bonadonna, G. and Gianni, A.M. (1991) Flow cytometry for clinical estimation of circulating hematopoietic progenitors for autologous transplantation in cancer patients. Blood, 77, 400-409.
2. Terstappen, L.W., Huang, S., Safford, M., Lansdorp, P.M. and Loken, M.R. (1991) Sequential generations of hematopoietic colonies derived from single nonlineage-committed CD34+CD38- progenitor cells. Blood, 77, 1218-1227.

Before the use of xeno-transplants, the activity of “candidate” hematopoietic stem cells was primarily measured by the ability of cells to produce colony-forming cells in long-term “Dexter” cultures (1). Together with Connie Eaves and Heather Sutherland we established a long-term culture assay for human cells (2) using the limiting dilution principles I used for producing monoclonal antibodies. This assay was used to purify “candidate” stem cells and for development of a serum-free culture medium for human “candidate” stem cells (3). This serum-free culture medium is currently available as “StemSPAN” from STEMCell technologies.

References:

1. Dexter, T.M., Allen, T.D., Lajtha, L.G., Schofield, R. and Lord, B.I. (1973) Stimulation of differentiation and proliferation of haemopoietic cells in vitro. J Cell Physiol, 82, 461-473.
2. Sutherland, H.J., Eaves, C.J., Eaves, A.C., Dragowska, W. and Lansdorp, P.M. (1989) Characterization and partial purification of human marrow cells capable of initiating long-term hematopoiesis in vitro. Blood, 74, 1563-1570.
3. Mayani, H., Dragowska, W. and Lansdorp, P.M. (1993) Characterization of functionally distinct subpopulations of CD34+ cord blood cells in serum-free long-term cultures supplemented with hematopoietic cytokines. Blood, 82, 2664-2672.

Despite our novel and highly specific monoclonal antibodies, innovative cell purification techniques, improved cell culture media and assays for stem cells, our attempts to increase the number of adult blood-forming stem cells in tissue culture were a dismal failure. It was possible to maintain CD34+ cells in culture for many weeks. However, we could not markedly increase the number of CD34+ cells when cultures were initiated with purified “candidate” stem cells from adult bone marrow (1). A breakthrough came when we finally tested “candidate” stem cells from fetal liver and umbilical cord blood in our culture systems. The number of CD34+ cells increased several thousand-fold in cultures of fetal liver cells and several hundred-fold in cord blood cultures. We concluded that the self-renewal of stem cells is developmentally controlled (2). Similar findings were obtained with purified murine stem cell “candidates” (3). Developmental changes in stem cell function were confirmed many years later (4,5). 

References: 

1. Mayani, H., Dragowska, W. and Lansdorp, P.M. (1993) Lineage commitment in human hemopoiesis involves asymmetric cell division of multipotent progenitors and does not appear to be influenced by cytokines. J Cell Physiol, 157, 579-586.
2. Lansdorp, P.M., Dragowska, W. and Mayani, H. (1993) Ontogeny-related changes in proliferative potential of human hematopoietic cells. J Exp Med, 178, 787-791.
3. Rebel, V.I., Miller, C.L., Thornbury, G.R., Dragowska, W.H., Eaves, C.J. and Lansdorp, P.M. (1996) A comparison of long-term repopulating hematopoietic stem cells in fetal liver and adult bone marrow from the mouse. Exp Hematol, 24, 638-648.
4. Kim, I., Saunders, T.L. and Morrison, S.J. (2007) Sox17 dependence distinguishes the transcriptional regulation of fetal from adult hematopoietic stem cells. Cell, 130, 470-483.
5. Bowie, M.B., Kent, D.G., Dykstra, B., McKnight, K.D., McCaffrey, L., Hoodless, P.A. and Eaves, C.J. (2007) Identification of a new intrinsically timed developmental checkpoint that reprograms key hematopoietic stem cell properties. Proc Natl Acad Sci U S A, 104, 5878-5882.

The failure to significantly expand “candidate” hematopoietic stem cells from adult bone marrow in tissue culture was a set-back for therapeutic strategies counting on significant production of stem cells in tissue culture. The situation did not improve when we found that hematopoietic stem cells lost telomeric DNA with each cell divisions (1), similar to human fibroblasts (2).

Reference:

1. Vaziri, H., Dragowska, W., Allsopp, R.C., Thomas, T.E., Harley, C.B. and Lansdorp, P.M. (1994) Evidence for a mitotic clock in human hematopoietic stem cells: loss of telomeric DNA with age. Proc Natl Acad Sci U S A,91, 9857-9860.
2. Allsopp, R.C., Vaziri, H., Patterson, C., Goldstein, S., Younglai, E.V., Futcher, A.B., Greider, C.W. and Harley, C.B. (1992) Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A,89, 10114-10118.

With PNA and Q-FISH tools, we contributed to a number of important studies, the most cited one being our contribution to the study of the telomerase knock-out (KO) KO mouse (1) (> 2000 citations). This paper was already under review at Cell when we got involved. Our findings significantly changed the message, essentially from “telomerase is not as important as we believed: it can be knocked out without consequences” to “telomerase KO mice lose 5 kb of telomeric DNA with each generation and uncapped telomeres at late generations cause chromosome fusions and genome instability”. See Figure. 

Figure. Telomerase KO mice lose about 5 kb of telomere repeats per generation. Q-FISH analysis of embryonic fibroblasts from the indicated subsequent generations of telomerase KO mice. 1 TFU corresponds to ~1kb of telomere repeats. Note that in wild-type (WT) animals all chromosomes are “capped” with readily detectable telomere repeats, whereas at generation 6 (bottom panel and chromosome spread) many chromosome ends have no telomere repeats. Note the dicentric chromosomes without telomere repeats at the point of fusion point (e.g. at 9 pm). 

Reference: 

1. Blasco, M.A., Lee, H.W., Hande, M.P., Samper, E., Lansdorp, P.M., DePinho, R.A. and Greider, C.W. (1997) Telomere shortening and tumor formation by mouse cells lacking telomerase RNA. Cell, 91, 25-34.

We developed and perfected a method to measure the average length of telomeres in specific nucleated cell types in peripheral blood using PNA probes, fluorescence in situ hybridization (FISH) and flow cytometry, a technique we called flow FISH (1). With this method, we showed that there is a large variation in the average telomere length between individuals of the same age and that the most dramatic decline in telomere length occurs in the first few years of life (see Figure).  Interestingly, at birth and throughout life the average telomere length was found to be longer in females than in males (2) and Figure, left panel). Based on reported sex differences in gene expression between cells from human embryos, I recently proposed that such telomere length differences could be established before embryo implantation (3). In collaboration with hematologists and geneticists at NIH, we established that the telomere length in patients with genetic defects in telomere pathway genes, such as TERT, TERC, RTEL1 and others could readily be distinguished from patients with other causes for their bone marrow failure (reviewed in (4). Because our flow FISH measurements could also distinguish carriers from telomerase mutations from siblings without such genetic defect (Figure, right panel), we obtained many requests from clinicians to perform diagnostic flow FISH experiments for their patients in our research lab. This clearly was not ideal, and I decided to set up a company, Repeat Diagnostics Inc., to provide clinical telomere length measurements. This company employs 8 people and is processing 10-20 samples per day for clinicians around the world looking after patients suspected of inherited or acquired “telomere biology disorders”.  My take on the role of telomeres and telomerase in aging and cancer is summarized in two review papers (5,6). 

Encouraged by the results described in our Nature paper (1), we embarked on developing the single cell DNA template strand sequencing (Strand-seq) technique (2). At the time, we could not have predicted the wide range of applications that this technique would have outside our initial goal: to distinguish sister chromatids in order to test the “silent sister” hypothesis. Strand-seq results are illustrated in the Figure and in a recent review (3).  

Figure. Strand-seq allows identification and analysis of sister chromatids. The two daughter cells (a and b) of a single parental CD34+ cell show perfect, complementary segregation of parental DNA template strands and only a single sister chromatid exchange event (arrow, bottom chr 1). Note that all chromosomes showing reads mapping to both strands of the reference genome (Chr 2, 5, 6, 7, 9, 11, 14, 16, 17 and 18 in this cell pair) can be used to map SNPs to parental genomes along entire chromosomes and assemble complete physical maps of parental haplotypes. c) Sister chromatid exchange events are very common in cells from patients with Bloom’s syndrome. In this cell 56 SCE events are detected (arrows). 

References:

1. Falconer, E., Chavez, E.A., Henderson, A., Poon, S.S., McKinney, S., Brown, L., Huntsman, D.G. and Lansdorp, P.M. (2010) Identification of sister chromatids by DNA template strand sequences. Nature, 463, 93-97.
2. Falconer, E., Hills, M., Naumann, U., Poon, S.S., Chavez, E.A., Sanders, A.D., Zhao, Y., Hirst, M. and Lansdorp, P.M. (2012) DNA template strand sequencing of single-cells maps genomic rearrangements at high resolution. Nat Methods, 9, 1107-1112.
3. Hanlon, V.C.T. and Lansdorp, P.M. (2026) Strand-seq and the future of personalized genomics. Nat Genet. 

My lab generated comprehensive maps of polymorphic inversions in the human genome. This work was spearheaded by three graduate students: Ashley Sanders, David Porubsky and Vincent Hanlon. The Figure shows examples of results. Following Ashley’s initial papers (1,2), David reported on inversions and genome instability (3) and Vincent developed software to visualize phased inversions in a genome browser (4). 

The diploid nature of the genome is often incompletely represented in “whole genome sequence” analysis, where a genome is typically represented as a set of unphased variants with respect to a reference genome. However, important biological phenomena such as compound heterozygosity and epistatic effects between enhancers and target genes can only be studied when haplotype-resolved genomes are available. Hence a method that can produce dense and accurate chromosome-length haplotypes is highly desirable. Using a well-studied trio from the HapMap consortium we showed that Strand-seq allows for accurate phasing along entire chromosomes (1). The principle of this method and phasing trio family members are shown in Figure 1.

Figure 1. a) Strand-seq provides a physical haplotype map of SNP’s along entire chromosomes. Multiple cells with Strand-seq sequence reads mapping to both strands of the reference genome for a given chromosome (orange and blue) are used to assemble haplotype backbones (1). Such chromosome-long haplotypes correspond accurately (>99%) with HapMap reference data for a child (b, yellow) but not her parents (father blue , mother red), where Strand-seq reveals the location of parental meiotic recombination events (resp. red and blue dots). The haplotype backbones generated by Strand-seq allow complete phasing of human genomes when combined with other type of sequence data (e.g. 10X or PacBio data (1, 2, 3).

When combined with other type of sequencing data, Strand-seq allows near complete phasing (>99% of SNP’s) of human genomes (Fig 2, from 2). 

Although Strand-seq data are increasingly in demand, the preparation of single cell Strand-seq libraries is technically challenging (1). Variable results and limited informative reads have hampered widespread application of the Strand-seq method despite its theoretical appeal. We recently reported a simplified “one pot” (OP) method to make Strand-seq libraries that overcomes many of the previous technical challenges (2). Specifically, we developed a nanoliter-volume protocol in which reagents are added cumulatively, DNA purification steps are avoided, and enzymes are inactivated with a thermolabile protease. OP-Strand-seq libraries routinely capture 10%–25% of the genome from a single cell with reduced costs and increased throughput, opening up a wide variety of studies that are being pursued in collaboration with investigators in and outside Vancouver and Canada.  A typical result of our current Strand-seq results is shown in the Figure and various issues related to Strand-seq library production are reviewed in (3). 

Genetic testing can inform a patient’s inherited risk for developing a disease. However, predicting which side of the family an autosomal variant comes from is often a limitation of current technology. Inability to determine a variant’s parent-of-origin (PofO) (i.e., if it is maternally or paternally inherited) can lead to uncertain patient management and ineffective genetic testing of family members. In collaboration with Intan Schrader, Steven Jones and their collaborators we have developed a protocol using Oxford Nanopore Technologies’ long read sequencing (ONT-LRS) in combination with chromosome-length haplotyping enabled by single cell Strand-seq to allow, for the first time, PofO haplotyping without any sequence information of the parents (1). The principle of this method is shown in the Figure.