Inflammation gone awry: aging or disease

Comment on Ferrucci et al, page [2294][1]

In a cohort analysis of elderly Italians, Ferrucci and colleagues found that the dysregulation of proinflammatory mediators, considered by many to be at the root of aging, may more aptly be considered at the root of cardiovascular disease.

Now and again,

N ow and again, unifying concepts emerge with relevance to diverse clinical domains. This is clearly true for the robust expansion of appreciation for inflammation gone awry, which has now been implicated in the pathogenesis of disease processes as diverse as gingivitis, diabetes, atherosclerosis, sarcopenia, osteoporosis, and dementia. For gerontologists interested in the biology of aging, the proinflammatory cytokine interleukin-6 (IL-6) has been particularly interesting, inasmuch as animal models and clinical series have time and time again demonstrated an age-associated rise in serum or tissue levels, with the most dramatic change observed in those who develop the phenotype of frailty. For example, in an epidemiologic study of more than 4000 community-dwelling elderly, IL-6 levels were positively associated with functional impairment (mobility and activities of daily living), depression, and mortality. 1 Other large cohorts, including the Health Aging and Body Composition Study (Health ABC) 2 and the Cardiovascular Health Study, 3 have demonstrated similar changes in IL-6 and other proinflammatory molecules or acute phase proteins, such as Creactive protein (CRP) and fibrinogen. Invariably, to the extent that these factors are present, negative clinical consequences are observed.
Yet it remains to be determined whether the presence of these mediators of inflammation is the consequence of aging (ie, a decline in those factors that regulate proinflammatory cytokines) or a reflection of an underlying disease, such as atherosclerosis. If the former were true, one would expect increased cytokine levels in at least some individuals in the absence of demonstrable inflammatory disease. In this issue of Blood, Ferrucci and colleagues report data from a large cohort of thoroughly studied older adults. From their analysis, there appears to be a tight correlation of inflammatory signals with cardiovascular disease, as most (but not all) of the measured markers were not demonstrably correlated with age once the data were adjusted for cardiovascular risk factors or morbidity. As is typical from well-constructed epidemiologic studies, these findings will raise as many questions as they answer. And the big one still remains: which comes first-the cytokines or the disease? Do those factors that render an individual at risk for cardiovascular disease do so by altering proinflammatory cytokine regulation? Is there a genetic predisposition to cytokine dysregulation increasing susceptibility to atherosclerosis?
The study by Ferrucci and colleagues goes as far as a cross-sectional analysis can. At this point, what is sorely needed is a longterm longitudinal analysis of healthy individuals as they traverse from middle age to old age, with interval checks of these important markers and a cataloguing of clinical conditions and incipient disease processes. From there, rational experimental interventions in laboratory animals and humans can be derived with the ultimate goal of retarding either, or both, aging and age-related diseases. ■
T he pathophysiology of lung injury after stem cell transplantation (SCT) is poorly understood. Idiopathic pneumonia syndrome (IPS) has been attributed to many causes, including preparative regimen toxicity, culturenegative infection, immune-mediated injury, and graft-versus-host disease (GVHD). In this issue of Blood, Hildebrandt and colleagues report that donor leukocyte-derived RANTES (a chemokine ligand of the CC chemokine family of proteins that promotes migration of T cells, eosinophils, basophils, and macrophages to sites of inflammation) is significantly elevated in recipients of an allograft after irradiation, compared with syngeneic controls. Elevated mRNA and RANTES protein levels were associated with increased mRNA expression of CCR5 and CCR1 and increased inflammatory cell infiltration into the lung in this mouse model. The importance of donor RANTES was confirmed by the use of RANTES-deficient T cells, where lung injury was significantly reduced but not eliminated.
These data support the idea that IPS is a complex disorder due to a cascade of events. Preparative regimen toxicity (especially from irradiation) generates a proinflammatory environment that damages host tissues, augments the allostimulatory capacity of host dendritic cells, and alters the chemokine environment. Thus, donor T cells may act as effectors and facilitators of lung injury after allogeneic SCT, For personal use only. on July 22, 2018. by guest www.bloodjournal.org From initiating and/or enabling a cascade, which perpetuates the lung injury. This model suggests that GVHD and IPS share common effector pathways. As importantly, it suggests that targeting T-cell recruitment may be effective in preventing (and to a lesser degree treating) IPS.
Does this mean that IPS is pulmonary GVHD? No. The authors found that GVHD pathology was not significantly altered in the bowel and liver in RANTESdeficient animals when pulmonary damage was significantly decreased. Is IPS an example of a disorder in which an initiating event such as preparative regimen toxicity creates a milieu where bystander allogeneic T cells become activated, further harming damaged tissues and perpetuating the process? What are the relationships among the mechanisms involved in IPS and bronchiolitis obliterans and bronchiolitis obliteransorganizing pneumonia? Further work is clearly needed to better define the pathophysiology and the relationships among these frequently fatal disorders. A better understanding of the mechanisms should lead to effective therapy for these disorders. ■

Mickie Bhatia ROBARTS RESEARCH INSTITUTE
Notch signaling is viewed as a potent regulator of mammalian hematopoietic stem cells (HSCs). Surprisingly, a study by Radtke's group demonstrates that donor or recipient mice with inactivated Notch1 or Jagged1 were capable of normal HSC reconstituting function. This observation challenges the essential nature of Notch signaling in homeostatic function of HSCs.
I n addition to myeloid and lymphoid development, accumulated studies have associated Notch signaling with mammalian hematopoietic stem cells (HSCs). These experimental approaches have demonstrated Notch-mediated expansion of HSC number during ex vivo culture, and have fostered the view that Notch must therefore be a crucial part of HSC physiology. In this issue of Blood, an elegant and concise study by Mancini and colleagues challenges the essentiality of Notch in homeostatic function of HSCs.
Mancini's group employs a sophisticated inducible Cre-loxP transgenic mouse system to specify and control the inactivation of Notch1 (N1), and/or its putative ligand, Jagged1 (J1). Mice with inactivated N1 or J1 had normal HSC function, and were capable of normal hematopoietic development and reconstitution of this function upon transplantation of HSCs into recipients. Importantly, this study not only addresses the autonomous role of J1-mediated Notch signaling among HSCs and progenitors, but also HSC interactions with the bone marrow microenvironment in recipients with inactivated J1 that retain a normal hematopoietic system after HSC transplantation.
The mechanisms that govern HSC biology remain elusive. However, independent categories that delineate signaling pathways and transcription factors that are essential versus those that augment existing HSC function are emerging. These categories may be related, but allow access to unique components of HSC behavior. Those that fall in the category of essential are likely to disrupt HSC function upon inactivation, and probably do not include J1-mediated Notch signaling or (as also demonstrated by Mancini's group) ␤-cateninbased signaling required for the Wnt pathway. Despite these observations, both Notch and Wnt signaling expand or augment baseline HSC function.
Caveats persist. In the case of such complex signaling pathways, inactivation studies capable of targeting specific signaling machinery are plagued by the lack of definitive evidence to exclude the pathway involvement due to potential compensatory mechanisms. For example, in the normal condition, the Notch pathway consists of more than 4 receptors and more than 5 ligands, whereas the Wnt pathway consists of more than 9 frizzled receptors and 12 ligands. To further complicate matters, these 2 pathways crosstalk, and since the interdependency is unknown, not only could other Notch-Notch ligand-mediated interactions compensate for inactivation of J1 or N1, it is also possible that Wnt signaling may rescue potential HSC deficiencies due to an absence of appropriate Notch signaling. It is likely that some level of certainty will be required to address these issues by evaluating other components of the targeted pathway and other crosstalking signaling pathways in HSCs after inactivation.
The study by Mancini et al embodies the power of such experimental approaches to test paradigms of HSC biology that evolve from the interpretation of other studies attempting to define a means to modulate HSCs. This study, and future studies like it, will assist in identifying additional questions to understand the elusive mechanisms that regulate HSCs. In the meantime, introduction of Notch and/or Wnt ligands during ex vivo culture of HSCs remains a possibility to augment HSC number, independent of evidence that supports an essential role of Notch or Wnt signaling in HSCs. ■