SAMHD1 in cancer: curse or cure?

Abstract

Human sterile α motif and HD domain-containing protein 1 (SAMHD1), originally described as the major cellular deoxyribonucleoside triphosphate triphosphohydrolase (dNTPase) balancing the intracellular deoxynucleotide (dNTP) pool, has come recently into focus of cancer research. As outlined in this review, SAMHD1 has been reported to be mutated in a variety of cancer types and the expression of SAMHD1 is dysregulated in many cancers. Therefore, SAMHD1 is regarded as a tumor suppressor in certain tumors. Moreover, it has been proposed that SAMHD1 might fulfill the requirements of a driver gene in tumor development or might promote a so-called mutator phenotype. Besides its role as a dNTPase, several novel cellular functions of SAMHD1 have come to light only recently, including a role as negative regulator of innate immune responses and as facilitator of DNA end resection during DNA replication and repair. Therefore, SAMHD1 can be placed at the crossroads of various cellular processes. The present review summarizes the negative role of SAMHD1 in chemotherapy sensitivity, highlights reported SAMHD1 mutations found in various cancer types, and aims to discuss functional consequences as well as underlying mechanisms of SAMHD1 dysregulation potentially involved in cancer development.

Keywords: Cancer development; Cellular functions of SAMHD1; Mutations in SAMHD1; SAMHD1; dNTP regulation.

Fig. 1

SAMHD1, its functions, and implications for AGS and cancer Cellular functions of SAMHD1 and functional consequences for mutated/dysfunctional or downregulated SAMHD1 are depicted. Mutated SAMHD1 might lead to displacement of ssDNA into the cytoplasm, where it can be detected by intracellular DNA sensors like cGAS. cGAS then produces cyclic guanosine monophosphate-adenosine monophosphate (cGAMP) to activate STING which in turn activates interferon regulatory factor 3 (IRF3) and the NF-κB pathways through the kinases TANK-binding kinase 1 (TBK1) and IκB kinase (IKK), thus inducing an IFN response. Consequences of dysfunctional SAMHD1 on AGS and cancer are displayed in the lower part of the figure. Unclear relations and consequences are indicated by question marks. Image created with Servier Medical Art (https://smart.servier.com/)

Fig. 2

Donors affected by mutations in SAMHD1 per cancer type The distribution of all (a) and only coding somatic mutations (b) across the 20 most prevalent ICGC cancer studies is represented. The ICGC data portal offers clinical and analyzed data representing 81 cancer type datasets available from the ICGC Data Coordination Center for Release 28 (human genome hg19/GRCh37), processed as of March 27, 2019. We used open-access simple somatic mutations (SSM) calls. These include single and multiple base substitutions, and small (≤ 200 bp) insertions and deletions that appear in the tumor tissue, but not in the normal control tissues. The figure legends in a and b depict all surveyed cancers that are included in the pie charts along with the calculated percentage (%) of donors affected by each cancer type

Fig. 3

Expression of SAMHD1 in different cancer types Each point represents paired tumor/healthy samples and the relative difference of SAMHD1 expression between the two. The difference is represented as a z-score, which shows the number of standard deviations between the expression of SAMHD1 in the respective tumor sample and the mean expression of SAMHD1 in the healthy samples

Fig. 4

Graphical representation of the somatic mutation spectrum throughout the protein sequence of SAMHD1 In total, 177 coding mutations from ICGC cancer studies and other 53 mutations surveyed from the literature were visualized. The scale bar represents the length (amino acids) of the protein sequence. Each lollipop represents a somatic coding mutation. Lollipops are colored according to the consequence type: missense (red), frameshift (blue), stop-gain (purple), stop-lost (olive), deletion (yellow). The size of the lollipops represents the number of reported patients with the mutation. The lollipop diagram was created by using [106]. The domain structure is based on [100]. Supplemental Table 1 lists the 230 coding mutations inclusive cancer type and references