03 Feb 2018

Researchers have long noted an inverse correlation between the odds of getting cancer and of getting a neurodegenerative disease. Now, a new genomic study of aging puts meat on the bones of that idea. In the January 30 Nature Communications, researchers led by Christoph Kaleta at Christian-Albrechts University in Kiel, Germany, reported that the transcriptional changes seen in aging people, and in older vertebrates, are similar to those that occur in neurodegeneration, but different from changes in cancer. While aging and neurodegeneration were marked by induction of inflammatory genes and suppression of cell-cycle genes, the opposite happened in cancer. The group dichotomy even held true at the level of particular genes, because variants that raised the risk for cancer lowered the risk for neurodegenerative disease, and vice versa. Thus, aging-related changes may represent a trade-off between cancer and neurodegeneration risk, the authors suggested.

Others said the genetic data make a valuable contribution. “The existence of this balance makes sense, and is elegantly shown in this paper,” Jose Bras at University College London wrote to Alzforum (see comment below). Julie Williams at Cardiff University in Wales was particularly intrigued by the immune genes. “The idea that immune processes play an opposite role in cancer and neurodegenerative disease has been around for some time, but there had been very little compelling evidence to support it,” she said.

Epidemiological studies first uncovered the antagonistic relationship between the two types of disease, finding that people with neurodegeneration, including Alzheimer’s disease, were at lower risk for cancers, even though they still occur (Dec 2009 news). The idea that these disorders are opposites appealed to many, because in one case cells become immortalized, while in the other they die too soon (Oct 2010 conference news). 

Cancer vs. Neurodegeneration. Mortality data indicates that as people age, neurodegenerative disease deaths rise as cancer deaths fall. [Courtesy of Aramillo Irizar et al., Nature Communications.]

Because the incidence of both diseases is higher in older people, Kaleta and colleagues wondered what role aging plays. The authors isolated RNA from the brain, liver, and skin of zebrafish, a model organism for aging, as well as killifish, the shortest-lived vertebrate, whose lifespan is a mere three months. These samples were compared to RNA that the authors took from mouse brain, liver, skin, and blood, as well as human RNA from blood and skin samples. Human brain was not included in this study. For each species, the authors sampled four or five time points across the lifespan. Each age group comprised about five individuals, except for the human data, which had 15 people per age group.

First author Peer Aramillo Irizar saw consistent differences during aging across species and tissues. Expression of cell cycle and developmental genes waned; immune genes spiked. This pattern matched changes seen in published disease data sets for cardiovascular disease, diabetes, cognitive decline, Alzheimer’s, and Parkinson’s disease, but belied those seen in cancer. Deaths from cancer peak around age 60 and then fall off, while those from other conditions continue to rise, the authors noted (see image above). “These data tell us there might be a much stronger connection between all diseases of aging than has been thought,” Kaleta told Alzforum. “Although aging looks different in different tissues, it may be based on the same underlying molecular processes.”

Focusing in on specific genes, Kaleta and colleagues found that those whose expression changed across the lifespan frequently contained binding sites for the transcription factors E2F1, NFAT, CEBPB, AP1, suggesting these as master regulators of aging-related processes. All of these factors have been implicated previously in both cancer and neurodegeneration, and affect both immune and cell-cycle processes (e.g. Palomer et al., 2011; Vukic et al., 2009; Zahnow et al., 2009). 

Likewise, when the authors compared known genetic risk variants for cancer and neurodegenerative disease, they found 40 shared single nucleotide loci. At 36 of these, one polymorphism raised cancer risk while protecting against degeneration, while another SNP did the opposite. The genetic locus with the largest number of shared risk SNPs contains a noncoding RNA, dubbed ANRIL, which controls expression of several nearby cell-cycle regulator genes (Congrains et al., 2013). Another highly involved locus, SH2B3, controls inflammatory signaling.

These findings are purely correlative, and reveal no causal connections, Kaleta emphasized. Still, he would like to know what drives age-related changes in gene expression. He believes the root cause might be accumulating genomic damage, although there are other possibilities as well, such as oxidative damage or microbiome changes. In future studies, Kaleta will explore potential molecular mechanisms in more detail. He thinks there may be a vicious cycle, where genomic damage kicks off inflammation, which could in turn drive further damage, escalating age-related deterioration.

Commenters suggested replicating the findings in larger sample sets, as well as examining specific neurodegenerative diseases and brain regions. If the data hold up, some of the genes altered in aging might make good therapeutic targets for treating degenerative diseases, Bras said. Williams pointed out that it will be important to make sure that targeting those genes does not raise cancer risk. Kaleta agreed this is a concern, noting that anti-inflammatory statins, for example, have been found to boost the odds of cancer.—Madolyn Bowman Rogers

Comments

1. • Jose Bras
     Van Andel Research Institute
   • Posted: 03 Feb 2018

   This is a very interesting paper that provides evidence for a concept that is intuitive, but that I don’t think has been thoroughly explored: cancer—a disease of dysregulated cellular proliferation—has an opposite (transcriptomic) profile to late-onset neurodegenerative diseases—diseases characterized by cell death. Of course, humans are complex systems and so the existence of these balances makes complete sense. This is elegantly shown in this paper. I would note, however, that the data sets are not particularly large (531 samples in five age points), and so it is difficult to generalize the findings and place them in the context of human biology. It would have been interesting to break down the group of degenerative aging-associated diseases and test for which of them these results hold true. For people studying neurodegeneration, it is interesting to see transcriptomic profiles that are negatively associated with late-onset neurodegenerative disease; these may provide novel therapeutic targets in the future.

   View all comments by Jose Bras
2. • Christoph Kaleta
     Kiel University, Institute for Experimental Medicine
   • Posted: 05 Feb 2018

   Thank you for your comment concerning our work. Please note that this data set covers four species and four tissues (thereof 86 samples from humans). Thus, the data set might appear small compared to other studies, but we find this antagonism across species and tissues which strongly supports its generality (at least in vertebrates).

   Moreover, we have tested the disease antagonism in gene-expression data from several previously published human-aging studies. Among them, one study in brain covered 55 samples and another one in blood (marked as "cross-cohort" in Figure 3 of our manuscript) comprised approximately 15,000 participants. For the gene-expression analysis we report the associations between individual diseases and the aging data sets in Figure 3. For the genetic analyses, the breakdown into individual diseases can be found in Supplementary Data 1 (sheet "GWAS Cancer vs. DAD").

   Indeed it would also be interesting to look into specific sub-phenotypes of diseases such as late-onset AD, if corresponding expression data is available.

   View all comments by Christoph Kaleta

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Research Brief: Epidemiological Study Links Cancer, AD 27 Dec 2009
2. Two Faces of Evil: Cancer and Neurodegeneration 13 Oct 2010

Paper Citations

1. Palomer X, Álvarez-Guardia D, Davidson MM, Chan TO, Feldman AM, Vázquez-Carrera M. The interplay between NF-kappaB and E2F1 coordinately regulates inflammation and metabolism in human cardiac cells. PLoS One. 2011;6(5):e19724. Epub 2011 May 23 PubMed.
2. Vukic V, Callaghan D, Walker D, Lue LF, Liu QY, Couraud PO, Romero IA, Weksler B, Stanimirovic DB, Zhang W. Expression of inflammatory genes induced by beta-amyloid peptides in human brain endothelial cells and in Alzheimer's brain is mediated by the JNK-AP1 signaling pathway. Neurobiol Dis. 2009 Apr;34(1):95-106. PubMed.
3. Zahnow CA. CCAAT/enhancer-binding protein beta: its role in breast cancer and associations with receptor tyrosine kinases. Expert Rev Mol Med. 2009 Apr 8;11:e12. PubMed.
4. Congrains A, Kamide K, Ohishi M, Rakugi H. ANRIL: molecular mechanisms and implications in human health. Int J Mol Sci. 2013 Jan 10;14(1):1278-92. PubMed.

Further Reading

News

1. Neurons and Cancer Cells Share Survival Tactics 18 Jul 2014