Pervasive duplication of tumor suppressors in Afrotherians during the evolution of large bodies and reduced cancer risk
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
The risk of developing cancer is correlated with body size and lifespan within species. Between species, however, there is no correlation between cancer and either body size or lifespan, indicating that large, long-lived species have evolved enhanced cancer protection mechanisms. Elephants and their relatives (Proboscideans) are a particularly interesting lineage for the exploration of mechanisms underlying the evolution of augmented cancer resistance because they evolved large bodies recently within a clade of smaller-bodied species (Afrotherians). Here, we explore the contribution of gene duplication to body size and cancer risk in Afrotherians. Unexpectedly, we found that tumor suppressor duplication was pervasive in Afrotherian genomes, rather than restricted to Proboscideans. Proboscideans, however, have duplicates in unique pathways that may underlie some aspects of their remarkable anti-cancer cell biology. These data suggest that duplication of tumor suppressor genes facilitated the evolution of increased body size by compensating for decreasing intrinsic cancer risk.
Article activity feed
-
-
Reviewer #3:
One example of this problem is in the estimation of cancer risk. The risk is estimated on the basis of body size and lifespan. However, that lifespan is itself phylogenetically estimated from body size at least for the non-extant species. It is not clear to me from the manuscript whether all lifespans are so estimated, or whether observations are used for the lifespan of the extant species. If the latter, caution is indicated, because lifespan data are highly uneven and often given as observed maximal lifespans, which can be misleading if taken from, for instance, zoo specimens. In either case, the manuscript needs to more clearly emphasize that these are statistically-predicted risks, not measured risks.
At a larger scale, the authors have done their best with a dataset that suffers from a couple of problems. First, all of …
Reviewer #3:
One example of this problem is in the estimation of cancer risk. The risk is estimated on the basis of body size and lifespan. However, that lifespan is itself phylogenetically estimated from body size at least for the non-extant species. It is not clear to me from the manuscript whether all lifespans are so estimated, or whether observations are used for the lifespan of the extant species. If the latter, caution is indicated, because lifespan data are highly uneven and often given as observed maximal lifespans, which can be misleading if taken from, for instance, zoo specimens. In either case, the manuscript needs to more clearly emphasize that these are statistically-predicted risks, not measured risks.
At a larger scale, the authors have done their best with a dataset that suffers from a couple of problems. First, all of the extant very large-bodied animals form a single clade, with the hyrax as the sole small-bodied member of that clade. And since the titanohyrax is extinct, among the extant organisms (an available large-bodies species with genomes) there is then a true large-bodied clade of the sirenia and elephants and relatives. I understand that other evolutionary data make it clear that these represent two (three including titanohyrax) independent transitions to large-body sizes. But with only the modern or nearly modern genomes to work with, I am not sure that the duplication inference procedures and their coupling to the body size analysis statistically represents more than a single observation (e.g., a default of a single transition to large size along the tethytheria branch).
Similarly, the authors observe what appears to be a number of independent duplications of tumor suppressors in African and Asian elephants: duplications that are lacking in many of the ancient genomes considered. I know that the authors used rigorous statistical methods to correct for the fragmented nature of these ancient genomes, but it is very hard not to wonder if some of the data in Figure 4 is really not an artifact of using ancient genomes, where detecting recent gene duplications may be very difficult (several of the Asian and African elephant duplications in Figure 4 appear to be of the same genes). If these events are truly independent and not genome assembly/annotation artifacts, there is then an alternative hypothesis to propose. Thus, are the authors suggesting that there is a rapid turnover in the duplication of tumor suppressors, such that all elephants have such duplicates, but the particular duplications have short life spans and differ from species to species?
Finally, it would be nice to see a few more comments on the manatee genome and why it does (or doesn't) show the expected patterns for the genome evolution in the face of the evolution of larger body sizes.
I would also note that Figure 3 and 4 would benefit from greatly expanded captions: I do not fully understand what is being illustrated in, for instance, Figure 3B-why are certain dots connected with lines? Intersections between what in the y-axis label?
-
Reviewer #2:
This manuscript addresses the question of whether duplication of tumor suppressors occurred coincidently with the enlarged body size and reduced cancer risk evolved independently in Afrotherians. Using the human genome as reference, the authors systematically searched for gene duplications in 13 publicly available Afrotherian genomes, including 9 extant and 4 extinct species. The authors also reconstructed the ancestral body sizes, cancer risks and gene duplication events across the Afrotherian phylogeny. These data showed that both increased body sizes and reduced cancer risks are gradually evolved. Reactome pathway enrichment analysis for gene duplicates showed unexpectedly that gene duplicates in both lineages with or without major increases in body size/lifespan/decreases in cancer risk are enriched in many cancer …
Reviewer #2:
This manuscript addresses the question of whether duplication of tumor suppressors occurred coincidently with the enlarged body size and reduced cancer risk evolved independently in Afrotherians. Using the human genome as reference, the authors systematically searched for gene duplications in 13 publicly available Afrotherian genomes, including 9 extant and 4 extinct species. The authors also reconstructed the ancestral body sizes, cancer risks and gene duplication events across the Afrotherian phylogeny. These data showed that both increased body sizes and reduced cancer risks are gradually evolved. Reactome pathway enrichment analysis for gene duplicates showed unexpectedly that gene duplicates in both lineages with or without major increases in body size/lifespan/decreases in cancer risk are enriched in many cancer related pathways. However, the authors found that 157 genes duplicated in Proboscidean stem-lineage, in which extremely large species evolved, were uniquely enriched in 12 cancer pathways. These genes might facilitate further body enlargement and cancer resistance evolution in Proboscidean. Most interestingly, the authors found that several genes both upstream and downstream of a famous tumor suppressor TP53 have also been duplicated, either before or after initial TP53 duplication. These genes are involved in transcriptional regulation of TP53 and may have facilitated re-functionalization of TP53 retroduplicates. Overall, this is an important and interesting study that can help us understand the evolution of body size, lifespan and cancer risk in mammals more deeply.
Major comments:
In general, the evolutionary fate of gene duplication includes: 1) Conservation of gene function; 2) Neofunctionalization; 3) Pseudogenization; 4) Subfunctionalization (doi:10.1016/S01695347(03)00033-8). To execute the function of tumor suppression, as this study focused on, gene duplicates were supposed to be functionally conserved or subfunctionalized. Gene duplicates that have been neofunctionalized or pseudogenized will not be helpful (also mentioned by authors in the Caveats section). Therefore, it might be more convincing to investigate the functional status of each gene duplicate, especially those in Fig 4C/D. In many cases, however, a related function, rather than an entirely new function, evolves by neofunctionalization after gene duplication, and also that to check new functions for a batch of genes is not realistic, the authors could simply check the coding sequences to ensure these genes duplicates are not pseudogenes and are functional. This is necessary because in Fig 4D, many genes have only 2 copies expressed. If one of them is a young pseudogene, it could be stochastically expressed and will encode a dysfunctional protein.
In Results section 3, the cancer pathway frequency data of many nodes seems not consistent with data shown in Table 2. For example, Line 293-296: "55.8% (29/52) of the pathways that were enriched in the Tethytherian stem-lineage..., 27.8% (20/72) of the pathways that were enriched in the Proboscidean stem-lineage...were related to tumor suppression", the cancer pathway percentages shown in Table 2 for these 2 nodes are 63.4% and 38.81%, respectively. While the frequency data in Table 2 are consistent with Supplementary Data File S3: "Atlantogenata_Reactome_ORA.xlsx". It is possible that the frequency data shown in the main text are specific to pathways of tumor suppression, rather than cancer related pathways. If this is the case, more detailed data should be shown somewhere else.
The titles of Results section 3 and section 4 are highly similar and actually the data in section 4 seems to be used to further solidify the conclusion of section 3. Therefore, is it possible to merge them into one single section?
-
Reviewer #1:
The strength of this paper is its coupling of careful phylogenetic work with genomics to demonstrate the take-home message: all afrotherians are equal, but some are more equal than others with respect to mechanisms that reduce cancer risk. This is a significant advance in our understanding of the evolution of cancer risk with body size, and in so doing it considerably lengthens the list of genes of interest. It also has interesting examples illustrating the logical criteria of consistency, necessity, and sufficiency that will make it quite useful in teaching critical thinking to students.
-
