University of Groningen
On the molecular mechanisms of hematopoietic stem cell aging Lazare, Seka Simone
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Lazare, S. S. (2018). On the molecular mechanisms of hematopoietic stem cell aging. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
CHAPTER 6
Summary and Discussion
Seka Lazare
1, Leonid Bystrykh
*1, Gerald de Haan*
1,21 European Research Institute for the Biology of Ageing,
2 Mouse Clinic for Cancer and Aging, University Medical Center Groningen, University of Groningen Antonius Deusinglaan 1, 9713 AV, Groningen, the Netherlands,
110 CHAPTER 6
In this thesis, we have investigated hematopoietic stem cell aging using varied methodologies in an effort to further the understanding of its development and underlying mechanisms. We focused on what we term the commonalities of aging, seeking factors that are most frequently associated with HSC aging.
In chapter 1, we highlighted postulated mechanisms of HSC aging like senescence,
polarity and epigenetic reprogramming and the debate of cell intrinsic vs extrinsic causes of aging. Phenotypic changes in HSCs as they age have been well-documented. For example, we described how the HSC pool expands upon aging as if to compensate for the reduction in function observed. Yet, while phenotypic changes are well accepted, the underlying molecular changes remain elusive. We mentioned here the idea of heterogeneity of HSCs – the fact that while the aging pool is often analyzed and investigated as a whole, it contains subsets of HSCs with varied functional and molecular characteristics. It is plausible that there are subsets of ‘old’ and ‘young’ HSCs within a single HSC pool. We also touched upon the investigation of epigenetic sources of HSC aging and how disparities in reported candidate genes may suggest that rather than the consequence of single ‘aging genes’, the underlying mechanism of HSC aging may likely be attributed to changes in random combinations of genes.
In chapter 2, we investigated an experimental model based on one of these postulated
mechanisms of HSC aging – DNA damage and show that HSCs are particularly susceptible to reduced function as a result of this compared to other tissues. The presence of reactive oxygen species and the resulting DNA damage have been one of the theories attributed to aging. We used a mouse model deficient in Rev-1, a core translesion synthesis polymerase, either alone or in combination with Xpc deficiency (rendering mice deficient in global-genome nucleotide excision repair). This was a highly collaborative project lead by the Niels de Wind Lab in LUMC and involving several labs in a Marie Curie Aging ITN Consortium. In this network, various labs involved in this publication used their expertise to resolve the Rev1 phenotype in multiple tissues and cell types using diverse techniques. Our focus involved the hematopoietic system, HSCs in particular. We used in vitro and in vivo assays to quantify the functional defect in Rev-1 HSC compared to wild-type littermates and the extent to which this could be rescued by ROS scavengers. We showed that the hematopoietic stem cells are especially susceptible to Rev-1 deficiency compared to other tissues, such as stromal cells. Rev-1 HSCs showed compromised repopulation even at fetal stage (fetal liver HSCs) and double knock-out of Rev-1 and Xpc (Rev1/ Xpc) resulted in severe HSC exhaustion. Remarkably, HSC function in Rev1/Xpc -/- mice could be rescued by bone marrow transplant in unconditioned recipients, supporting the fact that the HSC defect was cell intrinsic.
In chapter 3, we utilized the only intervention currently known to delay aging in
multiple organisms, caloric restriction, to investigate if it may similarly delay aging at the tissue level in murine HSCs and whether a high fat diet may conversely accelerate HSC aging. We first highlighted phenotypic changes of the hematopoietic system upon
while caloric restriction was able to delay aspects of the hematopoietic system (such as bone marrow cellularity) and a high-fat diet accelerate this, neither were able to affect HSC aging parameters such as increased frequency, nor were they able to rescue HSC function in vivo.
In chapter 4, we analyzed gene expression changes that occur in HSC aging. We
highlighted increased variation with age for HSC and HSC-related parameters and along these lines make a case for analyzing age-related changes, including gene expression, taking into account individual heterogeneity. We showed increasing transcriptional activity of HSCs through increased Polymerase II and RNA levels in aged HSCs. We revealed, surprisingly, that previous studies showed great discrepancy not only in the number of genes deregulated with age, but also with respect to the identity of the genes themselves. Across 7 comparable studies (including our own), not a single transcript was found to be transcriptionally affected in all studies. Accounting for individual (mouse to mouse) variation of differentially expressed genes with age, and consistency with previous studies, we compiled a list of the most consistent differentially expressed genes with HSC aging. We showed that a large majority of these genes consist of cell surface receptors, DNA packaging genes, histones and epigenetic modifiers, many of which have not been reported in hematopoiesis or HSC functioning before.
In chapter 5, we investigated one of the two of the most consistently upregulated
genes in HSC aging according to our analysis in chapter 4. We showed that expression of this gene, Neogenin, which had not previously been implicated in HSC or HSC aging, was enhanced in LT-HSC and its expression in HSC and progenitor cells correlated with self-renewal potential. Knock-down of Neogenin reduced both young and old HSC proliferation in vitro, but did not affect young HSC in vivo. Old HSCs, however were unable to engraft upon Neogenin knockdown. We also showed that, similar as in mouse, Neogenin is upregulated in human HSC aging and that a subset of human umbilical cord blood HSC express Neogenin to a high degree. These Neohi cells showed reduced in vitro stem cell potential. Here, we not only highlighted Neogenin as a novel receptor able to affect HSC function and aging, but we also lend support to our methodology of determining putative HSC aging candidate genes based on their consistency in individual aged mice and previous studies.
Future Perspectives
Our mechanistic investigations into hematopoietic stem cell aging set apart HSCs as unique in their response. While we saw clear indicators that diet could affect aging parameters in C57BL/6 mice such as preventing increased weight gain, and bone marrow cellularity, HSCs were largely immune to the effect of diet. Conversely, although many parameters were affected by REV1 deletion, HSCs were particularly affected to a large degree, even to the point of exhaustion, and the effect seemed to be at least in some part cell-intrinsic. From these two studies, it seems that HSCs have
112 CHAPTER 6
selective susceptibility to various aging-related pathways that differ even from more differentiated hematopoietic cells. This is a very interesting observation as it brings in the fundamental question: what are the qualities that set HSCs apart from other somatic stem cells and from other cells of the hematopoietic system and how do these qualities contribute to the susceptibilities of HSCs to various aging effectors? It is clear that while some interventions are able to increase longevity in general, they may not do so equally for all tissues and tissue stem cells. With aging, there is increased incidence of blood malignancies and infection, many of which are linked to phenotypes observed in HSC aging (dampened immune system, myeloid bias and enhanced self-renewal). Whilst it is important to combat aging as a whole, special attention should be given to understanding of aging tissues like the hematopoietic system and their malfunctions, which are likely to further increase as life expectancy does.
Of note, all studies in mice shown here were done using wild type C57BL/6 mice, or mice on a C57BL/6 background. While this is the most widely used mouse strain, making our experiments comparable to many others, there are strain-to-strain variations in the mouse for many aspects of aging, including lifespan 1–3. It remains to be seen whether affecting these aging-related pathways would have the same effect in other strains, for example the shorter-lived DBA/2 mouse. This has already been shown for caloric restriction in general 4, but not at the tissue level analyzing long-term hematopoietic stem cells both phenotypically and functionally as we have done here. This remains to be seen not just for caloric restriction and diet, but other aspects like genomic instability (Rev-1 as an example), telomere attrition or epigenetic alterations. The strain to strain differences already observed suggest the contribution of genetic loci that vary between these strains. Their identification would add to our understanding of HSC aging.
Our focus on consistent gene expression changes with age led to surprising results. Many of the most consistently differentially expressed genes were not known to play a role in hematopoiesis. The genes reported in individual studies varied greatly in number and identities. We stressed many times in this thesis that with increasing age also comes increasing mouse to mouse, and likely also cell to cell, variability. This manifests in the fact that in terms of phenotypic hematopoietic parameters old mice deviate from young, but are also more likely to be different from other old mice. This was also true for some differentially expressed genes. As we characterized gene expression in individual mice, we were able to look back at differentially expressed genes and see that for some genes, for example Clusterin and Stat3, not all old mice showed differential expression from young. The fact that (old) HSCs vary in their gene expression changes may offer some explanation as to the lack of consensus between studies using pooled mice, and highlight an importance in experimental design of sampling when investigating aging parameters. If pooling, sampling of large numbers of mice may be necessary to ensure consistency. This is also true when analyzing gene expression at the individual level. It is possible that were our sample sizes larger we may have detected larger number of genes or stronger fold-changes. However, modest
fundamentally still HSCs. Gene expression analysis between LT-HSC and ST-HSC has shown about a 3% change in gene expression 5 and so massive changes in gene expression within the LT-HSC compartment upon aging may be unlikely.
Cell surface markers allow us to isolate purified LT-HSCs but as yet we are unable to isolate a 100% pure population. The Lin-Sca-1+c-Kit+CD150+CD48- bone marrow population, as used in our analysis, has been shown to have a functional HSC frequency of 50% 6. This can be further improved by the addition of EPCR. As more markers are discovered to further purify LT-HSC, new gene expression analysis may further narrow the stem-cell specific changes that occur in HSCs with aging. Many of the top differentially expressed genes with age were cell surface receptors. We have functionally investigated one of these receptors, Neogenin and shown for the first time its implication in the regulation of HSC proliferation and function in both murine and human HSC aging. Not only did this serve as proof of concept of our methodology, but it also points to interesting experimental and therapeutic possibilities. As we have mentioned, isolation of subsets of HSCs based on these markers, as we have done with Neogenin, followed by functional and molecular characterization may be fundamental in understanding HSC aging. We have shown that aging mice may deviate in the degree and manifestation of the associated hematopoietic phenotypes. Along these lines, it is likely that the aging HSC pool also contain subsets of HSCs with different functions and contributions to aging phenotypes. We may be able to isolate ‘young’ and ‘old’ HSCs from a single mouse, for further testing and characterization. Due to the accessibility, and established HSC functional protocols, the hematopoietic system and HSCs are ideal for this kind of age-related analysis, allowing for analysis of contribution of different subtypes of stem cells to aging and the ability to test the idea of selective depleting of aged stem cells, leaving the young to remain in their place. The work presented in this thesis inevitably leads to the debate on two avenues of aging: cell intrinsic versus cell extrinsic causes of aging, and the idea of whether aging is pre-programmed or simply the consequence of random changes in the cell. Most of our data points towards cell intrinsic ‘causes’ of aging. For example, hematopoietic stem cell aging seemed to be impervious to the effects of diet, and Xpc bone marrow transplanted into Rev1Xpc bone marrow somewhat rescued the HSC dysfunction displayed by Rev1Xpc HSC. A significant number of differentially expressed genes in aging fell into the category of epigenetic modulators or DNA packaging genes, suggesting changes in epigenetic memory and DNA accessibility. However, the abundance of genes encoding cell surface markers in our gene lists suggested that there may be a difference in the way aged HSCs are able to interact with the bone marrow microenvironment. It is likely that both cell intrinsic and extrinsic factors play a role and may even have some interplay. For example, cell extrinsic signaling may lead to changes in intracellular gene expression that transforms cellular pathways, eventually becoming established and independent of signaling. Closer monitoring of changes at the single cell level of HSCs at a time-course during aging, and not just upon aging,
114 CHAPTER 6
would be useful to investigate this. Lastly, the deviation in HSC phenotypes, and gene expression changes in genes involved in global regulation such as gene expression and epigenetics suggest that aging may be the result of random dysfunction in processes that are well regulated in young. Our observation of increased transcriptional activity in aged HSCs supports our gene expression data of downregulation of DNA packaging genes such as histones, and upregulation of epigenetic modifiers. The loss of well-regulated epigenetic memory in aged cells may facilitate the clonal evolution of HSC subsets which may outcompete ‘young’ HSC in the bone marrow. However, although random, some genetic changes may lend an advantage to expansion of certain aged HSC subsets, as has been observed in leukemias, whose mutations often also involve epigenetic modifers 7,8. These are likely to occur more frequently in individuals manifesting hematopoietic aging phenotypes. For this reason, identification of the most prevalent molecular changes, e.g. the most frequent genes deregulated, in aged HSCs may be as important as the conceptual characterization of uniquely observed changes in aging studies.
References
1. Lang, D. H. et al. Quantitative trait loci (QTL) analysis of longevity in C57BL/6J by DBA/2J (BXD) recombinant inbred mice. Aging Clin Exp Res 22, 8–19 (2010).
2. De Haan, G. & Van Zant, G. Dynamic changes in mouse hematopoietic stem cell numbers during aging. Blood 93, 3294–301 (1999).
3. Kõks, S. et al. Mouse models of ageing and their relevance to disease. Mechanisms of Ageing and Development 160, 41–53 (2016). 4. Swindell, W. Dietary restriction in rats and mice: A meta-analysis and review of the evidence for genotype-dependent effects on lifespan. Ageing Research Reviews 11, 254–270 (2012).
5. Forsberg, E. C. et al. Differential expression of novel potential regulators in hematopoietic stem cells. PLoS Genet. 1, e28 (2005).
6. Kiel, M., Radice, G. & Morrison, S. Lack of Evidence that Hematopoietic Stem Cells Depend on N-Cadherin-Mediated Adhesion to Osteoblasts for Their Maintenance. Cell Stem Cell 1, 204–217 (2007).
7. Renneville et al. Cooperating gene mutations in acute myeloid leukemia: a review of the literature. Leukemia 22, 915–931 (2008). 8. Dingli, Traulsen, Lenaerts & Pacheco. Evolutionary Dynamics of Chronic Myeloid Leukemia. Genes & Cancer 1, 309–315 (2010).
