Inhibitors of DREAM Complex Assembly and/or Function For Use In Repairing DNA Damage

Abstract

The present invention relates to the pharmaceutically-induced enhancement of DNA damage repair in cells and potential therapeutic applications thereof. In particular, the present invention is directed to an inhibitor of DREAM complex assembly and/or function for use in repairing DNA damage in a cell of a subject. The inhibitor of DREAM complex assembly may be an inhibitory nucleic acid inhibiting the expression of a component of the DREAM complex or an antibody to a component of the DREAM complex. Further disclosed is an inhibitor of DREAM complex assembly and/or function for use in treating and/or preventing a condition associated with DNA damage in a cell in a subject, such as age-related diseases and symptoms of ageing, a progeroid syndrome, acute radiation syndrome, chronic radiation syndrome, Xeroderma pigmentosum, Nijmegen breakage syndrome, Fanconi anemia and ataxia. The invention further discloses a method for obtaining at least one cell with reduced DNA damage, wherein the method comprises steps of providing at least one cell, and treating said at least one cell with an inhibitor of DREAM complex assembly and/or function to repair DNA damage in said cell. Finally, the invention provides pharmaceutical compositions comprising the at least one cell obtainable by said method.

Claims

1. A method for repairing DNA damage in a cell of a subject, comprising administering to the subject an inhibitor of dimerization partner, Retinoblastoma (RB)-like, E2F and multi-vulval class B (DREAM) complex assembly and/or function in an amount effective to repair the DNA damage in the cell of the subject.

2. The method of claim 1, wherein the inhibitor of DREAM complex assembly and/or function is selected from the group consisting of an inhibitory nucleic acid, an antibody, and a small molecule inhibitor.

3. The method of claim 1, wherein the inhibitor of DREAM complex assembly is an inhibitory nucleic acid inhibiting the expression of a component of the DREAM complex or an antibody to a component of the DREAM complex.

4. A method for in treating and/or preventing a condition associated with DNA damage in a cell in a subject comprising administering to the subject an inhibitor of dimerization partner, Retinoblastoma (RB)-like, E2F and multi-vulval class B (DREAM) complex assembly and/or function in an amount effective to repair the DNA damage in the cells of the subject, wherein optionally the condition is selected from the group consisting of age-related diseases and symptoms of ageing, a progeroid syndrome, acute radiation syndrome, chronic radiation syndrome, Xeroderma pigmentosum, Nijmegen breakage syndrome, Fanconi anemia, and ataxia.

5. A method for obtaining at least one cell with reduced DNA damage, wherein the method comprises steps of: a) providing at least one cell, optionally, from a sample from a subject, and b) treating said at least one cell with an inhibitor of dimerization partner, Retinoblastoma (RB)-like, E2F and multi-vulval class B (DREAM) DREAM complex assembly and/or function to repair DNA damage in said cell, wherein the method is an in vitro or ex vivo method.

6. The method of claim 4, wherein the inhibitor of DREAM complex assembly and/or function is selected from the group consisting of an inhibitory nucleic acid, an antibody, and a small molecule inhibitor.

7. The method of claim 6, wherein the inhibitor of inhibitory nucleic acid is an inhibitory RNA that inhibits the expression of a component of the DREAM complex.

8. The method of claim 6, wherein the inhibitor of DREAM complex assembly is an antibody to a component of the DREAM complex.

9. The method of claim 6, wherein the inhibitor of DREAM complex assembly is a DYRK1A inhibitor selected from the group consisting of an antibody, an inhibitory RNA, and a small molecule inhibitor selected from the group comprising epigallocatechin-3-gallate (EGCG), harmine, (1Z)-1-(3-ethyl-5-hydroxy-2(3H)benzothiazoylidene)-2-propanone (INDY), 3,5-di(polyhydroxyaryl)-7-azaindole (DANDY), (3-(4-fluorophenyl)-5-(3,4-dihydroxyphenyl)-1H-pyrrolo[2,3-b]pyridine (F-DANDY), folding intermediate-selective inhibitor of DYRK1A (FINDY), GNF4877, Roscovitine, Purvalanol A, 4,5,6,7-tetrabromo-1H-benzotriazole (TBB), CR8, quinazolinone, quinalizarin, Apigenin, Flavokavain A, Emodin, a meriolin, Ageladine A, and Leucettine L41.

10. The method of claim 4, wherein the DNA damage is selected from the group consisting of cyclobutane pyrimidine dimers (CPDs), pyrimidine (6-4) pyrimidone photoproducts (6-4PPs), DNA alkylations, base oxidations, base deaminations, interstrand crosslinks (ICLs), DNA mismatches, base loss, DNA single-strand breaks (SSBs), and DNA double-strand breaks (DSBs).

11. The method of claim 4, wherein the cell is a quiescent cell selected from the group consisting of a tissue-specific stem cell, a mature hepatocyte, or a fully differentiated cell selected from the group consisting of a neuronal cell and a kidney cell.

12. The method of claim 4, wherein the inhibitor of DREAM complex assembly is targeted to the cell of the subject using a cell-specific vector.

13. The method of claim 4, wherein the progeroid syndrome is selected from the group consisting of Werner syndrome, Bloom syndrome, Rothmund-Thomson syndrome, Cockayne syndrome, trichothiodystrophy, combined xeroderma pigmentosum-Cockayne syndrome, Wiedemann-Rautenstrauch syndrome, and Hutchinson-Gilford progeria syndrome.

14. The method of claim 4, wherein the age-related disease is cancer, neurodegeneration, or stem cell exhaustion.

15. A cell produced by the method of claim 5, wherein, optionally, the cell is a stem cell.

16. A pharmaceutical composition comprising the cell of claim 15 and at least one pharmaceutically acceptable excipient.

Description

FIGURE LEGENDS

[0108] FIGS. 1A to 1J. Mutations of components of the DREAM complex confer resistance to UV-induced DNA damage during development and adulthood.

[0109] (FIG. 1A) Sequence logos of the motifs found upon analysis of the promoters of the 211 DDR genes using HOMER. A sequence logo is a graphical representation of the sequence conservation of nucleotides. A sequence logo is created from a collection of aligned sequences and depicts the consensus sequence and diversity of the sequences. The relative sizes of the letters indicate their frequency in the sequences. The total height of the letters depicts the information content of the position, in bits. HOMER called the DPL-1, EFL-1 and LIN-15B binding motifs as follows: DPL-1 (E2F): TAGCGCGC, EFL-1 (E2F): TGCAARYGCGCTCYA (SEQ ID NO: 38) and LIN15B (Zf): CARTGGAGCGCRYTTGCATT (SEQ ID NO: 39) (FIG. 1B) Result of the motif search of the CDE-CHR DREAM complex motif (BSSSSSNNNTTYRAA, SEQ ID NO: 40) in the promoters of the DDR genes using HOMER. (FIG. 1C) L1 larvae of WT, lin-52(n771), lin-35(n745), dpl-1(n2994) and efl-1(se1) mutants were irradiated with UV-B or mock-treated and grown at 20° C. Larval stages were determined 48 h post-irradiation. Representative graph from one out of at least three independent experiments. Mean of each larval stage out of three biological replicates per experiment; standard deviation (SD) between replicates is shown. Two-tailed t-test between the fraction of each larval stage of a mutant compared to WT in the same treatment condition showed p<0.01 upon UV treatment of each mutant to WT. n>35 worms per replicate and condition. (FIG. 1D) L1 larvae of WT, lin-52(n771), dpl-1(n2994) mutants and the combined mutation of lin-52(n771) and dpl-1(n2994) were irradiated with UV-B or mock-treated and let to grow at 20° C. Larval stages were determined after 48 h. Representative graph from one out of three independent experiments. Mean of each larval stage out of three biological replicates per experiment; standard deviation (SD) between replicates is shown. Two-tailed t-test between the fraction of each larval stage of a mutant compared to WT in the same treatment condition showed p<0.01 upon UV treatment of each mutant to WT but no difference between double mutant and single mutants. n>37 worms per replicate and condition. (FIG. 1E-1J) Day one adult worms were irradiated or mock treated with UV-B and incubated at 20° C. Survival of the worms was analyzed every other day by discarding dead worms (immobile and unresponsive). log-rang (Mantel-Cox) test was performed to compare the lifespan of the DREAM mutants and WT in the same conditions. n>128 individuals per condition and genotype. Bar-graphs in (I) and (J) show the % decreased mean lifespan of each strain irradiated with UV-B compared to the mock-treated worms of the same strain. * P<0.05, ** P<0.01, *** P<0.001, **** P<0.0001.

[0110] FIGS. 2A to 2C. DREAM complex mutants show enhanced repair of UV-induced DNA lesions and alleviates the UV sensitivity of csb-1 and xpc-1 mutants.

[0111] (FIGS. 2A-2B) Representative slot blot out of three independent experiments labelled with antibodies against CPDs and SYBR™ Gold for DNA staining. DNA samples were collected from irradiated worms (WT, xpa-1(ok698), lin-52(n771) and dpl-1(n2994)) right after or 24 h after UV-B irradiation. Graphs in (FIG. 2B) represent the mean of the improved or decreased repair of the mutants tested compared to WT worms (mean with SD). Two-tailed t-test was used to compare the repair with WT worms. * P<0.05, ** P<0.01. (FIG. 2C) Quantification of CPDs nuclei signal intensity normalized with DAPI, in the heads of adult worms irradiated and collected immediately after or 60 h after irradiation. The number of nuclei quantified was n>1300, from 5-7 heads per strain and condition. **** P<0.0001.

[0112] FIGS. 3A to 3F. The DREAM complex directly represses multiple DNA damage response genes involved in the main DNA repair pathways.

[0113] (FIG. 3A) Volcano plot representing the RNAseq data of lin-52 vs WT. On the far left and right sides, all the genes with differential expression with a p-adj<0.05 and FC<−1.5 and FC>1.5 respectively. Highlighted in black are the genes belonging to the G0 term “Cellular response to DNA damage stimulus” that were upregulated above 1.5 FC. (FIG. 3B) qPCR data showing that different mutations for the DREAM complex (lin-52(n771), dpl-1(n2994) and efl-1(se1)) present an upregulation compared to WT of DDR genes that were previously found upregulated in the RNA-seq of lin-52 vs WT. mRNA levels were normalized to the expression of the house keeping genes eif3-c, Y45F10D.4 and vha-6. Graphs show the mean of three biological replicates with SD. (FIG. 3C) Area-proportional overlap between the genes upregulated in lin-52(n771) and the genes involved in the main DNA repair pathways (data from the G0 consortium). The overlap between repair pathways is not represented; nucleotide excision repair (NER), interstrand crosslink repair (ICL), base excision repair (BER), homologous recombination repair (HRR), mismatch repair (MMR), non-homologous end joining (NHEJ). Pathway genes data from G0 database released on 2019-10-08. (FIG. 3D) Area-proportional Venn diagram showing the overlap between the DDR genes found up-regulated in lin-52 compared to WT worms and two published transcriptome datasets on lin-35(n745). (FIG. 3E) Venn diagram showing the overlap between the DDR genes found up-regulated in lin-52 compared to the genes that were found to be bound by DREAM complex in the promoter area (41 in promoter area, 43 in total). Re-analysis of reference (Goetsch, Garrigues and Strome, 2017). (FIG. 3F) Overlap found between the genes bound by DREAM in the promoter area and all the DDR genes (76 in promoter area, 80 in total). Overlap p-value calculated by performing Fisher's Exact test.

[0114] FIGS. 4A to 4D. Mutations in the DREAM complex confer DNA damage resistance against multiple damage types.

[0115] (FIG. 4A) HRR in lin-52(n771) was evaluated by ionizing radiation (IR) treatment of early embryos of WT, lin-52(n771), the HRR deficient worms brc-1(tm1145); brd-1(dw1), and lin-52(n771); brc-1(tm1145); brd-1(dw1). The survival of the embryos was evaluated by the % of eggs hatched 24 h after irradiation. Representative graph of one out of three independent experiments, each with three biological replicates (mean with SD of the replicates). Two-tailed t-tests between the fraction of surviving embryos within the treatment are represented. n>55 per replicate and condition. (FIG. 4B) NHEJ repair in lin-52(n771) was evaluated by irradiating L1 worms of WT, lin-52(n771), the NHEJ deficient worms cku-70(tm1524) and the double mutant lin-52(n771); cku-70(tm1524), and let to grow at 20° C. Larval stages were determined after 48 h. Representative graph of one out of three independent experiments, each with three biological replicates (mean with SD of replicates). Two-tailed t-tests between the fraction of the larval stages of lin-52 compared to WT, and lin-52, cku-70 compared to cku-70 (non-significant), are represented (showed statistics indicated with upper bracket). n>31 per replicate and condition. (FIG. 4C) The response to alkylation damage in lin-52(n771) compared to WT was evaluated by treating L1 worms of WT, lin-52(n771), the alkylation damage sensitive animals polh-1(lf31) and double mutants lin-52(n771); polh-1(lf31) to different doses of MMS, and let to grow after washing at 20° C. Larval stages were determined after 48 h. Representative graph of one out of three independent experiments, each with three biological replicates (mean with SD of replicates). Two-tailed t-tests between the fraction of the larval stages of lin-52 compared to WT, and lin-52, polh-1 compared to polh-1 are represented (showed statistics indicated with upper bracket). n>20 per replicate and condition (average n=55). (FIG. 4D) The response of lin-52(n771) to cisplatin compared to WT was evaluated by treating WT and lin-52 L1 larvae with different concentrations of cisplatin (diluted in DMF), and let to grow after washing at 20° C. Worms were also given the maximum dose of DMF that was given for the cisplatin treatments as control. Larval stages were determined after 48 h. Representative graph of one out of three independent experiments, each with three biological replicates (mean with SD of replicates). Two-tailed t-tests between the fraction of the larval stages of lin-52 compared to WT are represented. n>39 per replicate and condition. * P<0.05, ** P<0.01, *** P<0.001, **** P<0.0001.

[0116] FIGS. 5A to 5F. Inhibition of DREAM using DYRK1A inhibitors confer DNA damage resistance in human cells.

[0117] (FIG. 5A) Overlap between the 530 genes with G0 term “DNA repair” in humans (Database released 2021-01-01) and the 328 genes found to be bound by DREAM in humans upon re-analysis of Litovchick et al., 2007. Overlap p-value calculated using Fisher's exact test. (FIG. 5B) Derepression of DREAM targets by DYRK1A inhibitors. Plot of the 270 of the 328 DREAM target genes that present significant differences upon either harmine or INDY (FDR<0.01 for at least one of the two datasets) upon 24 h treatment. DNA repair genes are marked in black, all remaining DREAM target genes are marked in white. The distribution of either all 270 genes (grey), only the DNA repair gene subset (black), or all DREAM target genes that are not included in the DNA repair gene annotation (white) are shown on top and on the right side of the plot. The treatments show a strong overlap and derepression with 161 out 270 genes upregulated in both treatments (2.39×over-enrichment, binomial q-value=9.56e-33). (FIGS. 5C-5D) Percentage of apoptotic U2OS cells (annexin V positive+annexin V positive/7-AAD positive) upon harmine (FIG. 5C) or INDY (FIG. 5D) treatment and UV exposure (2 mJ.Math.cm-2). Graphs show the mean with SD of a biological triplicate. Two-tailed t-tests between the populations under same irradiation conditions were performed. (FIG. 5E-5F) Percentage of apoptotic U2OS cells (annexin V positive+annexin V positive/7-AAD positive) upon harmine (FIG. 5E) or INDY (FIG. 5F) treatment and MMS treatment (2 mM for 2 hours). Graphs show the mean with SD of a biological triplicate. Two-tailed t-tests between the populations under same MMS conditions were performed. *** P<0.001, **** P<0.0001.

[0118] FIG. 6. Treatment with the DREAM kinase DYRK1A inhibitor harmine prevents photoreceptor cell loss in Ercc1 knockout (ko) model of premature aging.

[0119] Male and female Ercc1−/− mice on the third day after birth (postnatal day P3) were injected intraperitoneally 3 times/week with 10 mg/kg/bw of harmine hydrochloride (SMB00461, Sigma) diluted in 0.9% sodium chloride. Mice were sacrificed at postnatal day P15 for retina tissue isolation. Tissues were embedded in optimal cutting temperature (OCT) compound, cryosectioned and stained using in situ cell detection kit (TUNEL staining) (11684817910, Roche).

[0120] FIGS. 7A and 7B. Treatment with the DREAM kinase DYRK1A inhibitor harmine prevents loss of kidney function in Ercc1 knockout (ko) model of premature aging.

[0121] Podocyte-specific Ercc1 knockout mice at 3 weeks of age were mock treated (n=3) or treated with 10 mg/kg/day harmine (n=3) for 4 weeks. Proteinuria as endpoint for kidney dysfunction was assessed by Coomassie brilliant blue staining in urine 8 days (FIG. 7A) and 11 days (FIG. 7B) afterwards.

TABLE-US-00001 TABLE 1 SEQ ID NOs. SEQ ID NO: Description 1 Y45F10D.4 Forward primer 2 Y45F10D.4 Reverse primer 3 eif-3.C Forward primer 4 eif-3.C Reverse primer 5 vha-6 Forward primer 6 vha-6 Reverse primer 7 parp-1 Forward primer 8 parp-1 Reverse primer 9 polh-1 Forward primer 10 polh-1 Reverse primer 11 polk-1 Forward primer 12 polk-1 Reverse primer 13 mus-101 Forward primer 14 mus-101 Reverse primer 15 exo-3 Forward primer 16 exo-3 Reverse primer 17 lig-1 Forward primer 18 lig-1 Reverse primer 19 atm-1 Forward primer 20 atm-1 Reverse primer 21 HPRT Forward primer 22 HPRT Reverse primer 23 GPDH Forward primer 24 GPDH Reverse primer 25 CCNB1 Forward primer 26 CCNB1 Reverse primer 27 FEN1 Forward primer 28 FEN1 Reverse primer 29 PLK1 Forward primer 30 PLK1 Reverse primer 31 POLQ Forward primer 32 POLQ Reverse primer 33 RAD21 Forward primer 34 RAD21 Reverse primer 35 RAD51 Forward primer 36 RAD51 Reverse primer 38 EFL-1 binding motif 39 LIN-15B binding motif 40 C. elegans CDE-CHR motif 42 H. sapiens DREAM binding motif 43 H. sapiens DREAM binding motif

EXAMPLES

Example 1

[0122] In the following example, it is demonstrated in the nematode model organism C. elegans and in cultivated quiescent human cells how inhibition of DREAM complex assembly augments repair of various types of DNA lesions and confers resistance to genotoxic stress.

Material and Methods

[0123] Promoter Analysis—C. elegans

[0124] The set of 211 DDR genes was used as input for the findMotifs function of HOMER-4.11-2 (Heinz et al., 2010) with the parameters -len 8,10 -start -1000 -end 0. We first converted the Wormbase IDs to the sequence name with WormBase's SimpleMine (Harris et al., 2020) that were searched in the ‘worm.description’ file of HOMER to gain the corresponding Ref-Seq IDs. The p-values were calculated with the hypergeometric tests function in scipy(1.5.1). HOMER's seq2profile function (Heinz et al., 2010) was used to convert the previously reported CDE+CHR DREAM complex motif (Goetsch, Garrigues and Strome, 2017) with one mismatch and three random base pairs in between to a motif file usable by HOMER with the following parameter: seq2profile.pl BSSSSSNNNTTYRAA 1 (SEQ ID NO: 40). The constructed motif was searched with the findMo-tifs.pl -find function for the 211 DDR genes with the parameters -start -1000 -end 0. The back-ground enrichment of the motif was calculated for all 20174 protein-coding genes with a RefSeq ID included in the worm.description file of HOMER. The p-values were calculated with the hypergeometric tests function in scipy (1.5.1) (Virtanen et al., 2020).

Promoter Analysis—Human

[0125] Homer's seq2profile function was used to convert previously reported DREAM complex motifs (Litovchick et al., 2007) with no allowed mismatch with the following parameter:

TABLE-US-00002 seq2profile.pl TTTSSCGS 0 (SEQ ID NO: 42) seq2profile.pl VVCGGAAGNB 0  (SEQ ID NO: 43) seq2profile.pl BNBVNTGACGY 0  seq2profile.pl CWCGYG 0

[0126] The motifs were searched with the findMotifs.pl -find function with the parameters -start -1000 -end 0 for the up-, respective down-regulated genes, after harmine, respective INDY treatment with an FDR-cutoff of 0.01. The background enrichment of the motif was calculated for all 20102 protein-coding genes included in the homer.description file of HOMER. The p-values were calculated with the hypergeometric test function in scipy(1.5.1) (Virtanen et al., 2020) and Python's statsmodels(0.11.1) (Seabold and Perktold, 2010) was used to calculate the Benjamini-Hochberg FDR.

UV-Irradiation for Somatic Development

[0127] The effects of UV-B in worm development were analyzed as previously described (Rieckher et al., 2017). Upon standard alkaline hypochlorite solution treatment to adult worms, embryos were left in a rotating mixer at room temperature over-night in M9 buffer. Hatched L1 worms were filtered using an 11 μm hydrophilic filter (Millipore, NY1104700), counted, and plated in NGM plates. Worms were mock-treated or irradiated with a 310 nm PL-L 36W/UV-B UV6 bulb. OP50 E. coli was added to the plates and worms were incubated at 20° C. for 48 h.

IR Sensitivity Assay Dependent on NHEJ Repair

[0128] As previously described (Clejan, Boerckel and Ahmed, 2006; Johnson, Lemmens and Tijsterman, 2013), L1 worms repair DSBs mainly via NHEJ repair. Synchronized L1 worms were irradiated with different doses of IR using an IR-inducing cesium-137 source, and left 48 h at 20° C. to allow development.

IR Sensitivity Assay Dependent on HR Repair

[0129] Early embryos highly rely on HRR to repair DSBs (Clejan, Boerckel and Ahmed, 2006; Johnson, Lemmens and Tijsterman, 2013). Day-1 adults were left laying eggs on seeded NGM plates for no longer than 1.5 h. Upon removal of the adult worms, early eggs were irradiated using an IR-inducing cesium-137 source.

Alkylation Damage Induction by Using Methyl Methanesulfonate Treatment (MMS)

[0130] Synchronized L1 worms were incubated with different concentrations of MMS (Sigma, 129925) diluted in M9 buffer, moving for 1 hour at 20° C. Worms were washed three times with M9 buffer, and plated in seeded NGM plates.

Interstrand Crosslink Induction by Using Cisplatin

[0131] Synchronized L1 worms were exposed to different concentrations of cisplatin in dimethyl-formamide (DMF) diluted in M9 buffer or mock-treated with DMF (Sigma, 227056) diluted M9 buffer for 2 hours at 20° C. Afterwards, worms were washed three times with M9, allowed to grow for 48 h in NGM plates.

Lifespan Assay

[0132] Synchronized day-1 adult worms were irradiated or mock-irradiated with a 310 nm PL-L 36W/UV-B UV6 bulb (Waldmann, 451436623-00005077), then placed on fresh OP50-seeded NGM plates and incubated at 20° C. The worms were transferred to new plates every other day to avoid progeny overgrowth. Worms presenting internal hatching or protruding or ruptured vulvas were censored, and death worms were scored when no movement or pumping was observed even upon physical stimulus.

DNA Repair Capacity Assay in L1 Worms (Slot Blot)

[0133] The quantification of DNA repair via immunostaining of CPDs of DNA samples in a slot blot was performed as previously described with slight changes (Rieckher et al., 2017). Briefly, bleach-synchronized L1 worms (at least 30.000 per plate) were irradiated with UV-B light and split in two groups, one to be immediately quick-frozen in liquid nitrogen, the control of unrepaired damaged, and the other one to be left in seeded plates for 24 h at 20° C. to allow DNA repair to occur. After this, to avoid bacterial DNA contamination, worms were washed 5 times, incubated 2 hours to permit the removal of intestinal bacteria, washed another 5 times and finally quick-frozen.

[0134] DNA extraction was performed using the Gentra® Puregene® Tissue Kit (Qiagen, 158667) and the protocol for DNA Purification from Tissue. 500 μl Cell Lysis Solution, 2.5 μl Puregene Proteinase K, 2.5 μl RNase A Solution, 170 μl Protein Precipitation Solution, 500 μl isopropanol and 500 μl 70% ethanol were used. Cell Lysis Solution was directly added to the thawed sample and additional step with Proteinase K was performed, incubating at 55° C. overnight. DNA concentration was measured using the Qubit® dsDNA HS Assay Kit (Invitrogen, Q32851). Serial dilutions of the DNA were denatured at 95° C. for 5 minutes and transferred onto a Hybond nylon membrane (Amersham, RPN119B) using a Convertible Filtration Manifold System (Life Technologies, 11055). DNA crosslinking to the membrane was performed by incubating at 80° C. for 2 hours. The membrane was blocked for 30 minutes in 3% milk/PBS-T (0.1%) at RT. The membrane was incubated overnight at 4° C. with anti-CPDs (Clone TDM-2, 1:10.000, Cosmo Bio, CAC-NM-DND-001), then washed three times with PBS-T (5 minutes at RT), and blocked for 30 minutes with 3% milk/PBS-T. The secondary antibody used was a goat anti-mouse AffiniPure peroxidase-conjugated secondary antibody (1:10.000, Jackson Immuno Research, 115-0.5-174), followed by three washes in PBS-T and incubation with ECL Prime (Amersham, RPN2232). The DNA lesions were visualized by using a Hyperfilm ECL (Amersham, 28906836).

[0135] Finally, in order to stain and quantify the total amount of DNA per sample, the membrane was incubated overnight at 4° C. in PBS with 1:10.000 SYBR™ Gold Nucleic Acid Stain (Invitrogen, S11494), then washed 5 times 10 minutes in PBS at RT, and imaged using a BIO-RAD Gel Dox XR+ Gel Documentation System (BIO-RAD, 1708195).

Adult Somatic DNA Repair Assay (Immunofluorescence)

[0136] Synchronized day-1 adult worms were irradiated or mock-irradiated with a 310 nm UV-B light Philips UV6 bulb in a Waldmann UV236B irradiation device. Half of the worms were left in seeded NGM plates for 60 h at 20° C. to allow DNA repair, whereas the others were collected directly after the irradiation. After irradiation or the incubation time respectively, worms were picked and placed in a drop of M9 buffer on top of a HistoBond+ Adhesion Microscope Slides (Marienfeld, 0810461). Using two standard disposable hypodermic needles we cut the worms close to the head, and then placed a coverslip over the slide and placed it at −80° C. for at least 30 minutes. The coverslip was removed quickly to perform freeze-cracking (Duerr, 2013), that besides the cutting of the worm, would improve the penetrance of the antibodies. Worms were fixated by placing the slides at liquid methanol at −20° C. for 10 minutes, then washed 5 minutes in PBS. 70 μl of 2M HCl was added for 30 minutes at RT to denature the DNA. Slides were washed three times with PBS, and blocked with 70 μl of 20% Fetal Bovine Serum (FBS) in PBS for 30 minutes at 37° C. After washing the blocking solution, the slides were incubated with 70 μl of 1:10.000 anti-CPDs (Clone TDM-2, 1:3.000, Cosmo Bio, CAC-NM-DND-001) in PBS containing 5% FBS, at 4° C., overnight in a humid chamber. After washing three times in PBS for 5 minutes each, 70 μl of secondary anti-mouse Alexa Fluor 488 (1:300, Invitrogen, A21202) in 5% FBS PBS was added for 30 minutes at 37° C. The slides were washed three times for 5 minutes and mounted using 5 μl of Fluoromount-GTM with DAPI (Invitrogen, 00495952).

Image Quantification of Heads

[0137] Image stacks of the heads of adult worms were analyzed using the microscopy image analysis software, Imaris (Oxford Instruments). Nuclei in the area anterior to the pharyngeal-intestinal valve were determined by using the DAPI staining and setting a threshold of size and intensity. False positive nuclei (due to bacteria in the pharynx) were manually discarded. CPDs signal was quantified by using the maximum spherical volume fitting inside each of the nuclei. The 2-way ANOVA to analyze the interaction effect between the strain and the time components was calculated on log 10 transformed values with Python's pingouin v0.3.6.

Cell Culture and Treatments

[0138] U2OS were cultured in DMEM, high glucose GlutaMAX Supplement, Pyruvate (Thermo Fisher Scientific, 31966047) with 10% fetal bovine serum (FBS; Biochrom GmbH, S0615) and 1% Penicillin-Streptomycin (Thermo Fisher Scientific, 15140112). Cells were kept at 37° C. in a 5% CO.sub.2 incubator (Binder). Cell dissociation from the plates was performed with Accutase (Sigma, A6964). Cells were cultivated in FBS-free medium for 48 hours before genotoxic treatment were applied. 24 hours after FBS-free medium, cells were mock treated or received harmine hydrochloride (diluted in water) or INDY (diluted in DMSO) (Sigma, SMB00461 and SML1011) at 10 or 25 μM respectively. Before the genotoxic treatment, cells were washed with FBS-free medium. For the UV treatment, medium was removed from the plates and cells were irradiated using 254 nm UV-C light Phillips UV6 bulbs. The MMS treatment was performed by adding MMS at 2 mM for 2 hours, followed by 3 washes with FBS-free medium. Quantification via FACS of cell death and apoptosis was performed 24 hours after genotoxic treatment.

FACS Analysis

[0139] Collected cells were incubated in Annexin V Binding buffer (BioLegend, 422201) with Pacific Blue Annexin V (BioLegend, 640917) and 7-AAD (Thermo Fisher Scientific, 00699350) at 4° C. for 15 minutes. Cells were measured using a MACSQuant VYB (Miltenyi Biotec) and analyzed using FlowJo (BD).

RNA Extraction for RNA-Seq and qPCR Experiments

[0140] For the qPCRs and RNA-seq on L1 worms, around 10.000 (qPCR) or 40.000 (RNA-seq) bleach-synchronized L1 worms in triplicates (qPCR) or quadruplicates (RNA-seq) per strain and condition were placed in seeded NGM plates and left at 20° C. for 3 hours. Posteriorly, they were irradiated with a 310 nm UV-B light Philips UV6 bulb in a Waldmann UV236B irradiation device or left unirradiated, and left for 6 more hours to allow the DNA damage related transcriptional changes to be taking place. After this, worms were collected, washed three times with M9 buffer, and the pellet was placed in a tube containing 1 ml of TRIzol™ (Invitrogen, 15596018) and 1 mm zirconia/silica beads (Biospec Products, 11079110z). To extract the RNA, worms were first disrupted with a Precellys24 (Bertin Instruments, P000669-PR240-A), and then the RNA isolation was performed by using the RNeasy Mini Kit (QIAGEN, 74106) following the manufacturer's specifications, except for the use of 1-bromo-3-chloropropane (Sigma, B9673) instead of chloroform. The RNA was quantified using NanoDrop™ 8000 (ThermoFisher, ND-8000-GL). RNA extraction from U2OS cells was performed after 24 hours of harmine or INDY treatment of cells starved for a total of 48 hours, by using the RNeasy Mini Kit following the manufacturer's specifications. Cells were disrupted with RLT buffer and homogenized with QIAshredder spin columns (QIAGEN, 79656).

qPCRs

[0141] Reverse transcription to form cDNA was performed using Superscript III (Invitrogen, 18080044). The obtained cDNA was used to perform qPCR by using SYBR Green I (Sigma, S9460) and Platinum Taq polymerase (Invitrogen, 10966034) in a BIO-RAD CFX96 real-time PCR machine (BIO-RAD, 1855196). The analysis of the results was performed by using the comparative C.sub.T method(Schmittgen and Livak, 2008).

[0142] All C. elegans qPCR experiments were done in biological triplicates and the data was normalized to three housekeeping genes.

TABLE-US-00003 TABLE 2 C. elegans qPCR PRIMERS Forward primer Reverse primer House- keeping genes: Y45F10D.4 SEQ ID NO 1: SEQ ID NO 2: CGAGAACCCGC CGGTTGCCAGG GAAATGTCGGA GAAGATGAGGC eif-3.C SEQ ID NO 3: SEQ ID NO 4: ACACTTGACGA TGCCGCTCGT GCCCACCGAC TCCTTCCTGG vha-6 SEQ ID NO 5: SEQ ID NO 6: CTGCTATGT CGGTTACAAA CAATCTCGG TTTCAACTCC Genes of interest: parp-1 SEQ ID NO 7: SEQ ID NO 8: AGCGAATGAAG ACTAGGCGTTC AAACAATCCGA GATTACTTGTG polh-1 SEQ ID NO 9: SEQ ID NO 10: AGAAATATCG GTAGGTAATA CGACGCTAGC GCAGCCTGCA polk-1 SEQ ID NO 11: SEQ ID NO 12: GAGATACTGATG AGTAGTTGGA GAGAATCTTGAG TGTGCTCAGC mus-101 SEQ ID NO 13: SEQ ID NO 14: TCGAAAGCCATA ACAAGAACGGGA TACGATGAACC GTACTAGAGAC exo-3 SEQ ID NO 15: SEQ ID NO 16: GGAGGAGACGTT TAGATCACTGG TAAGAACTACAC CTTCTTCTCGT lig-1 SEQ ID NO 17: SEQ ID NO 18: TGATCAAGGCT AGCCTCAATTC GTTGCTAAAGC CTTGACATGC atm-1 SEQ ID NO 19: SEQ ID NO 20: GCGAAGTTCTT AGTTCGACACA ACACCTCGAC TTCTTCAGCA

[0143] U2OS qPCR experiments were done using six biological replicates and the data was normalized to HPRT and GPDH.

TABLE-US-00004 TABLE 3 U2OS qPCR PRIMERS Forward primer Reverse primer House- keeping genes: HPRT SEQ ID NO 21: SEQ ID NO 22: GACCAGTCAA CCTGACCAAG CAGGGGACAT GAAAGCAAAG GPDH SEQ ID NO 23: SEQ ID NO 24: GACCAGTCAA TTAAAAGCAG CAGGGGACAT CCCTGGTGAC Genes of interest: CCNB1 SEQ ID NO 25: SEQ ID NO 26: GATTGGAGAGG AGTCATGTGCT TTGATGTCGAG TTGTAAGTCCT FEN1 SEQ ID NO 27: SEQ ID NO 28: AAAGGCCAGT TTGCCATCAAA CATCCCTCC GACATACACGG PLK1 SEQ ID NO 29: SEQ ID NO 30: TATTCCCAAG TAGCCAGAAGT CACATCAACC AAAGAACTCGT POLQ SEQ ID NO 31: SEQ ID NO 32: GCAACTTCTACT ATACTCTCGC CTTTCTTCTGG CTACTGTGTC RAD21 SEQ ID NO 33: SEQ ID NO 34: ACAGACTACTGA GGGCTCATCG AGCTCTTTACAC ATAACATCAC RAD51 SEQ ID NO 35: SEQ ID NO 36: GTTCAACACAG CTACACCAAAC ACCACCAGAC TCATCAGCGA

RNA-seq

[0144] A triplicate of RNA samples from lin-52 mutant and WT L1 worms were rRNA depleted using Ribo-Zero Plus rRNA Depletion Kit (Illumina, 20037135) and sequenced using a Hiseq4000 (Illumina) with PE75 read length. RNA quality control showed RIN≥9.4 for all samples. The RNA-seq data were processed through the QuickNGS pipeline (Wagle, Nikolić and Frommolt, 2015), Ensembl version 85. Reads were mapped to the C. elegans genome using Tophat (Kim et al., 2013) (version 2.0.10) and abundance estimation was done using with Cufflinks (Trapnell et al., 2010) (Version 2.1.1). DESeq2 (Love, Huber and Anders, 2014) was used for differential gene expression analysis.

[0145] The human RNA-seq data were processed with Salmon-1.1 (Patro et al., 2017) against a decoy-aware transcriptome (gencode.v37 transcripts and the GRCh38.primary_assembly genome) with the following parameter: --validateMappings -gcBias -seqBias. The output was imported and summarized to the gene-level with tximport (1.14.2) (Soneson, Love and Robinson, 2016) and differential gene analysis was done with edgeR (3.28.1) (Robinson, McCarthy and Smyth, 2009).

Datasets

[0146] The list of 211 genes belonging to the G0 term “Cellular response to DNA damage stimulus” was obtained by using data from the Gene Ontology Consortium(Ashburner et al., 2000; Carbon et al., 2019) (database released on 2019-10-08). Gene IDs from previously published datasets were updated to current databases, and duplicated or dead IDs were eliminated accordingly. Obtention of updated gene IDs and transformation of names to Wormbase IDs or Ensembl IDs was done with WormBase's SimpleMine(Harris et al., 2020) and Ensembl's BIOMART (Yates et al., 2020) respectively. Overlap analysis were done by using Fisher's exact test in R.

Data Access

[0147] The C. elegans RNA-seq data used in this study are available from Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.goz/geo) with the accession number GSE152235 and secure token yvqvewuetzkjrsl. The human RNA-seq data is available with the accession number GSE168401 and secure token wjehcacaflkxrip.

Analysis of DNA Repair Kinetics in Human U2OS Cells Upon Addition of an Inhibitor of DREAM Complex Assembly

[0148] Cells are cultured as described above. Next, cells are mock treated or receive harmine hydrochloride (diluted in water) or INDY (diluted in DMSO) (Sigma, SMB00461 and SML1011) at 10 or 25 μM respectively. Before the genotoxic treatment, cells are washed with FBS-free medium. For the UV treatment, medium is removed from the plates and cells are irradiated using 254 nm UV-C light Phillips UV6 bulbs. The MMS treatment is performed, e.g., by adding MMS at 2 mM for 2 hours, followed by 3 washes with FBS-free medium. Cells are fixed in a solution comprising PFA on HistoBond+ Adhesion Microscope Slides (Marienfeld, 0810461). To estimate the total amount of DNA damage per cell, the fixed cells are incubated overnight at 4° C. in PBS comprising anti-γH2AX antibody (Mouse anti-phospho-Histone H2A.X (Ser139), clone JBW301, Millipore, Cat #: 05-636), then washed 5 times 10 minutes in a suitable buffer at RT, prior to incubation for at least two hours with a secondary antibody. Finally, the slides are washed another 5 times 10 minutes in PBS-T and dried overnight. The next morning, slides are sealed with a cover slip.

Results

DNA Damage Response Gene Promoters Carry the CDE-CHR DREAM Binding Motif

[0149] We assessed whether specific transcription factor binding sites might be overrepresented in promoters of DDR genes. We performed an unbiased DNA motif enrichment analysis of the 211 DDR genes in C. elegans, based on the Gene Ontology Consortium classifications (Ashburner et al., 2000; Carbon et al., 2019). This analysis revealed a significant enrichment of the DPL-1, EFL-1 and LIN-15B binding motifs in the DDR gene promoters (FIG. 1A). DPL-1 and EFL-1 form the E2F-DP heterodimer that directly contacts promoters by binding to the cycle-dependent element (CDE). E2F-DP is linked via the pocket protein LIN-35 to the MuvB subcomplex, which binds the cell cycle genes homology region (CHR) in promoters to then, together, form the DREAM transcriptional repressor complex (Müller and Engeland, 2010; Müller et al., 2012; Goetsch, Garrigues and Strome, 2017). To address whether the DREAM complex target sites were present in the DDR genes, we queried whether the CDE-CHR motif was overrepresented. Strikingly, a total of 125 of the 211 DDR genes contained the CDE-CHR motif in the promoter area (FIG. 1B, one mismatch allowed, 0 to −1000 bp before transcription start site; p-value=5.16e-07), which suggested that the DREAM complex might be a regulator of DDR genes.

Loss-of-Function Mutations of DREAM Complex Components Confer Resistance to UV-Induced DNA Damage

[0150] We next determined whether mutations in DREAM components might influence DNA damage sensitivity. UV-B irradiation induces cyclobutane pyrimidine dimers (CPDs) and pyrimidine (6-4) pyrimidone photoproducts (6-4PPs) that are repaired by NER. We exposed synchronized L1 larvae carrying mutations in the DREAM complex to different doses of UV-B. 48 h post UV, the developmental growth was scored by counting the stages from L1 to the consecutive L2, L3, L4 larval and adult stages. The DREAM mutants lin-52, dpl-1, efl-1, lin-53, and lin-35 showed a significant improvement in proceeding through somatic development compared to wild-type (WT) worms following UV exposure (FIG. 1C). lin-52; dpl-1 double mutants that carry a defect in each of the two subcomplexes that form the DREAM showed a similarly improved developmental growth as the respective single mutants, indicating that their role as DREAM complex subunits conferred the UV resistance (FIG. 1D). Thus, DREAM mutants confer resistance to UV-induced DNA damage during development, indicating that the specific function of lin-52, dpl-1, efl-1, lin-53 and lin-35 as subunits of the DREAM complex determine the animals' ability to overcome DNA damage-induced developmental impairments.

[0151] To evaluate whether mutations in the DREAM complex could affect DNA damage-driven organismal aging, we UV-treated the DREAM complex mutant worms on day 1 of adulthood and assessed their lifespan (FIG. 1E-J). While in humans, mutations in DNA repair genes are sufficient to accelerate aging and lead to premature death (da Silva and Schumacher, 2019), C. elegans cultured under laboratory conditions requires exogenous DNA damage to shorten lifespan (Lans et al., 2013; Mueller et al., 2014; Bianco and Schumacher, 2018). UV irradiation led to a pronounced decrease in lifespan, which was significantly milder in all the DREAM complex mutants tested. lin52, dpl-1, efl-1 and lin-35 mutants significantly outlived wild-type worms upon DNA damage, despite the fact that some were short-lived without irradiation.

[0152] In addition to survival, we also assessed the motility of the worms as an important parameter of the maintenance of organismal health (Keith et al., 2014). Upon UV treatment, lin-52 mutants retained motility to a greater extent than WT animals, indicative of improved organismal health upon DNA damage. Overall, due to the phenotypic specificity with, comparatively, mild adverse effects under unperturbed conditions and the strong UV-resistance phenotype, we decided to focus mainly on lin-52 mutants to further investigate the role of the DREAM complex in regulating genome stability from here onwards.

DREAM Complex Mutants Improve Repair of UV-Induced DNA Lesions

[0153] We next tested whether the UV resistance of the DREAM complex mutants might be a consequence of enhanced DNA repair activity. We measured the worms' capacity to repair cyclobutane pyrimidine dimers (CPDs), which is the main UV-induced DNA lesion type that can be detected by an established and highly specific anti-CPD antibody (FIG. 2). We irradiated L1 larvae and quantified the removal of CPDs after 24 h using an anti-CPD antibody in a DNA slot blot. The DREAM complex mutants tested showed a significantly improved CPD repair compared to WT, whereas, as expected, a mutation in the NER core component xpa-1, showed a diminished CPD removal (FIG. 2A-B). To exclude that the enhanced removal of CPDs might be a consequence of CPD dilution due to overall DNA replication, we performed an EdU incorporation assay in L1 worms upon the same dose of UV irradiation. lin-52 mutants and WT worms showed comparable DNA replication events regardless of UV irradiation. Thus the decreased amount of CPDs was a consequence of improved repair capacity.

[0154] We next assessed whether the DREAM complex also regulates the DNA repair capacity in adult animals. Day 1 adult worms were UV-irradiated and the CPDs were quantified by anti-CPD antibody staining in situ (FIG. 2C). We focused on the head of the worm because of the high density of somatic nuclei in this region. Similarly, as in the case of L1 larvae, DREAM complex mutants showed an improved removal of CPD lesions compared to WT worms indicating an augmented repair capacity.

[0155] NER is initiated either by the TC-NER protein CSB-1 or by the GG-NER protein XPC-1 (Lans et al., 2010). We next evaluated whether mutations in the DREAM complex required one of these branches to confer the improved resistance to UV-induced DNA damage. The lin-52 mutation alleviated the UV sensitivity of csb-1, csa-1, and xpc-1 mutants. Consistent with the requirement for NER for removing UV lesions, the mutant lin-52 had no effect on completely NER deficient xpa-1 or csb-1; xpc-1 double mutants. Therefore, a mutation in the DREAM complex improved both GG-NER and TC-NER and could thus partially compensate for defects in either one of the NER initiating systems, while the enhanced UV resistance depends on the presence of the NER machinery.

The DREAM Complex Represses Multiple DNA Repair Genes and Pathways

[0156] We hypothesized that the DREAM complex could be directly regulating genes involved in the DNA repair process, thus curbing the repair capacity of somatic cells. We performed RNA-sequencing of lin-52(n771) and WT L1 larvae (FIG. 3). The majority of differentially expressed genes were upregulated in lin-52 mutants and these genes included a range of genes involved in DNA repair mechanisms (FIG. 3A and Tab. 4). Gene Ontology (G0) analysis of the significantly upregulated genes (p-adj<0.05) in lin-52 mutants compared to WT showed that the terms “Cellular response to DNA Damage stimulus”, “DNA repair” and “DNA metabolic process” were highly enriched (FDR corrected p-values of 1.23e-27, 1.10e-25 and 8.42e-31, respectively). We confirmed the upregulation of DNA repair genes by qPCR analysis, and also determined that DDR genes were upregulated in other DREAM complex mutants (FIG. 3B). Thus, the DREAM complex represses the expression of genes involved in DNA repair mechanisms. A total of 53 genes involved in cellular responses to DNA damage were significantly upregulated (p-adj<0.05) in lin-52 mutants compared to WT worms (Tab. 4). Among these 53 genes, there are genes involved in NER, inter-strand crosslink (ICL) repair, base excision repair (BER), HRR, mismatch repair (MMR), and NHEJ, suggesting that the DREAM complex represses components of all the main DNA repair pathways (FIG. 3C).

[0157] We next asked whether the DNA repair genes we found de-repressed in lin-52 mutants, and confirmed in efl-1 and dpl-1 mutants (FIG. 3B), would also be present in other DREAM complex transcriptomic data. The induced DNA repair genes in lin-52 mutants showed a remarkably consistent induction in two transcriptome datasets on lin-35(n745) mutants (Latorre et al., 2015; Sierra et al., 2015). We find that 52 DDR genes were upregulated in lin-35 L1s and 62 in lin-35 L3 worms. Furthermore, we observe a striking overlap between these up-regulated DDR genes and those found in lin-52. Out of the 53 DDR genes up-regulated in lin-52, 35 were common with lin-35 L1s, and 44 with lin-35 L3 worms (FIG. 3D and Tab. 4). The high amount and overlap of the DDR genes regulated by lin-52 and by lin-35 mutants in two independent studies together with the qPCR results in mutants for dpl-1 and efl-1 (FIG. 3B) indicate that the DREAM complex regulates an extensive amount of DDR genes.

[0158] To address whether the induced DDR genes might be directly repressed by the DREAM complex, we analyzed a published ChIP-seq dataset on the DREAM complex from late embryos (Goetsch, Garrigues and Strome, 2017). 43 out of the 53 DDR genes upregulated in lin-52 mutants are bound by the DREAM complex (41 in the promoter area, 2 intergenic or intronic), i.e. by at least 6 out of 7 DREAM components tested (Goetsch, Garrigues and Strome, 2017) (Fisher's Exact p-value<2.2e-16) (FIGS. 3E-F and Tab. 4). We confirmed the direct binding of DREAM to DDR genes by analyzing a ModENCODE ChIP seq dataset for DREAM in L3-stage worms (Latorre et al., 2015). 34 of the 53 DDR related genes that were upregulated in lin-52 mutants were bound by at least one member of the DREAM complex and 17 by all of them (one not found), a highly significant enrichment (Fisher's Exact p-value<2.2e-16 and 8.348e-14, respectively).

[0159] We analyzed whether among all the genes classified as cellular responses to DNA damage, there are more directly bound by the DREAM complex beyond our findings with lin-52. Using the embryonic ChIP-seq data (Goetsch, Garrigues and Strome, 2017), we found that 80 out of 211 DDR genes were bound by the DREAM complex (76 in the promoter, 4 intergenic/intronic, FIG. 3E-F). This overlap was highly significant (Fisher's Exact p-value<2.2e16). A similar result was obtained when analyzing the ModENCODE L3-stage data (Latorre et al., 2015): 200 out of 211 genes had data available, of which 61 were bound by at least one component of the DREAM complex (Fisher's Exact p-value<2.2e-16). These analyses reveal that the DREAM complex directly binds and represses multiple genes involved in the DDR. The consistency of the induction of DDR genes in DREAM mutants and the direct binding of DREAM components to DDR gene promoters across studies using embryos and different larval stages indicate that DREAM constitutively represses DDR genes in somatic cells. In conclusion, our analysis of RNA-seq and ChIP-seq data indicates that the DREAM complex constitutively represses DDR genes operating in all major DNA repair mechanisms in somatic cells.

TABLE-US-00005 TABLE 4 The DREAM complex binds and represses DNA damage response (DDR) gene expression. Induction of the expression of DDR gene in lin-52 mutants is highly consistent in lin-35 mutants at L1 (Kirienko and Fay, 2007) and L3 stage (Latorre et al., 2015) and promoters of these genes are bound by DREAM (re-analysis of embryonic ChlP-seq data (Goetsch et al., 2017)). DDR list based on the GO databased released on 2019 Oct. 8. p-adj = p-value with Benjamini-Hochberg (FDR) adjustment, FC = Fold Change, NA = not applicable due to the gene information not being available or in the dataset, in brackets = the FC was non-significant, ✓ = bound by at least 6 DREAM components in the promoter, ✓* = bound by at least 6 DREAM components in intronic or intergenic areas. FC lin- FC lin- FC lin- Bound 52 vs 35 vs 35 vs by Gene WT p-adj WT (L1) WT (L3) DREAM atm-1 1.53 1.91e−53 NA 1.52 baf-1 1.15 2.96e−03 NA 1.35 ✓ brc-1 1.34 2.39e−06 1.95 2.57 ✓ brd-1 1.23 9.11e−03 1.78 2.39 ✓ chk-1 1.18 1.39e−02 2.18 1.73 ✓ cku-80 1.32 1.63e−08 2.43 2.31 ✓ clsp-1 1.21 6.22e−05 2.27 2.59 ✓ crn-1 1.30 7.34e−11 1.57 1.96 csa-1 1.38 1.15e−04 4.37 4.87 ✓ ctf-4 1.16 2.23e−03 2.07 1.70 ✓ dog-1 1.27 1.74e−05 1.76 2.23 ✓ exo-1 1.27 4.24e−05 2.64 2.43 ✓ exo-3 2.08 4.70e−37 2.99 2.07 ✓ F10C2.4 1.35 3.99e−19 1.62 1.22 ✓ fan-1 1.32 5.34e−05 1.79 1.55 ✓* fcd-2 1.14 9.27e−03 NA [1.16] ✓ H21P03.2 1.34 2.97e−06 1.61 1.63 him-1 1.10 8.10e−03 NA [1.10] ✓ his-3 1.24 2.14e−23 NA NA ✓ hmg-12 1.74 1.50e−39 1.87 [1.16] ✓ hpr-17 1.31 2.18e−03 2.17 1.93 ✓ hsr-9 1.18 2.03e−09 NA [1.15] ✓ JC8.7 1.24 1.43e−03 NA [1.21] lig-1 1.65 2.37e−42 2.93 2.66 ✓ M03C11.8 1.37 5.61e−30 NA 1.26 ✓ mcm-3 1.19 3.23e−06 1.80 2.06 ✓ mcm-4 1.11 7.38e−04 1.63 1.63 ✓ mcm-6 1.12 1.06e−03 NA 1.40 ✓ mcm-7 1.20 7.23e−10 2.53 2.59 ✓ mre-11 1.21 1.63e−04 2.15 1.67 ✓ mrt-1 1.58 7.63e−10 2.32 2.48 msh-6 1.25 2.43e−10 2.62 2.12 ✓ mus-101 1.87 7.35e−72 4.13 2.51 ✓ parg-1 1.45 5.98e−32 1.66 1.89 ✓* parp-1 2.35 5.46e−85 3.36 3.27 pms-2 1.29 2.58e−03 NA 1.35 ✓ polh-1 2.14 6.85e−45 3.40 3.57 ✓ polk-1 1.63 4.18e−15 2.75 3.18 ✓ rad-50 1.51 1.05e−30 2.11 1.35 ✓ rad-51 1.27 3.44e−04 1.86 1.97 ✓ rad-54 1.48 8.45e−09 NA 1.60 ✓ rnf-113 1.23 3.15e−04 NA 1.29 rpa-1 1.08 2.43e−02 NA 1.25 ✓ ruvb-1 1.13 9.05e−03 NA [0.92] ✓ smc-3 1.20 4.95e−10 NA 1.38 ✓ smc-5 1.42 4.82e−18 2.20 1.79 ✓ smc-6 1.15 1.32e−03 1.56 [1.14] ✓ sws-1 1.35 2.06e−04 3.93 3.37 ✓ tdpt-1 1.30 2.85e−04 NA 1.86 ✓ tim-1 1.27 2.36E−13 1.97 2.04 tipn-1 1.70 2.31e−22 5.34 6.98 ✓ trr-1 1.12 1.94e−07 NA [1.02] ung-1 1.37 6.70e−05 NA 1.90

DREAM Mutations Confer Resistance to a Wide Variety of DNA Damage Types

[0160] Given that we found a range of DNA repair pathways induced in the soma of lin-52 mutants, we hypothesized that DREAM complex mutants would show a resistance phenotype to a wide variety of DNA damaging insults that require different repair machineries for their removal. We first evaluated how early embryos would be affected by the induction of double-strand breaks (DSBs) by ionizing radiation (IR). Somatic cells in the early embryo are highly replicative and only during this early stage employ HRR, which is initiated by the BRC-1/BRD-1 complex (Clejan, Boerckel and Ahmed, 2006; Johnson, Lemmens and Tijsterman, 2013; Janisiw et al., 2018). Upon egg-laying of day 1 adults, eggs were exposed to different IR doses. 24 h later the number of hatched larvae was evaluated for WT, lin-52(n771), the HRR deficient brc-1(tm1145); brd-1(dw1) double mutant and the lin-52, brc-1; brd-1 triple mutant (FIG. 4A). The proportion of egg hatching upon IR in lin-52(n771) was significantly higher than in WT worms, and as expected, brc-1; brd-1 deficient eggs were highly IR sensitive. Surprisingly, mutant lin-52 rescued the IR sensitivity of brc-1; brd-1 double mutants to levels even above that of WT animals, suggesting that in the absence of HRR, a mutation in lin-52 can induce alternative DSB repair pathways that are able to compensate for the lack of HRR. The IR resistance conferred by mutant lin-52 in early embryos, which are comprised of the most rapidly proliferating cells in C. elegans, indicates that the repression of DNA repair genes is a property of a somatic function of the DREAM complex and not just associated with cellular quiescence or terminal differentiation.

[0161] We wished to know whether a deficiency in the DREAM complex would also render the worms resistant to DSBs in a NHEJ repair-dependent fashion. WT, lin-52(n771), the NHEJ deficient mutant cku-70(tm1524) and the double mutant lin-52; cku-70 L1 larvae were exposed to IR and their developmental growth assessed 48 h later. IR-treated lin-52 mutants showed a significantly improved developmental growth compared to WT worms. However, this enhanced repair was dependent on NHEJ, as lin-52 could not alleviate the sensitivity to IR of cku-70 mutants (FIG. 4B). We also analyzed two other strains that are sensitive to IR (Roerink et al., 2012), xpa-1(ok698) and polh-1(lf31), and observed a strong sensitivity to IR during development, but when crossed with mutants for lin-52 this sensitivity was significantly rescued for both mutants. These results indicate that mutations in lin-52 enhance the NHEJ-dependent DSB repair, resulting in augmented IR resistance in IR-sensitive strains that have the canonical NHEJ pathway intact.

[0162] We evaluated how a mutation in the DREAM complex would respond to alkylation damage, a complex DNA insult known to be repaired by several mechanisms involving DNA methyltransferases, AlkB enzymes and BER (Soil, Sobol and Mosammaparast, 2017). L1 worms were exposed to an acute dose of methyl methanesulfonate (MMS) and their development was assessed 48 h later. We employed WT, lin-52(n771), and polh-1(lf31), and the double mutant lin-52; polh-1 (FIG. 4C). The mutation in the DREAM complex component lin-52 led to an improved development following MMS treatment and suppressed the MMS sensitivity of polh-1 mutant animals to levels comparable to the one of WT worms.

[0163] We assessed the response to cisplatin, a commonly used antitumor drug known to cause intra- and inter-strand crosslinks (Deans and West, 2011). L1 larvae were treated with increasing doses of cisplatin and the development of WT and lin-52(n771) was evaluated 48 h later (FIG. 4D). A mutation in the DREAM complex significantly alleviated the growth retardation upon cisplatin induced DNA-damage for all doses tested.

[0164] Taken together, these data show that lin-52 mutants are resistant to a wide array of DNA damage types and alleviate DNA damage sensitivity of various mutants in single repair systems.

Inhibition of the DREAM Complex Kinase DYRK1A Confers DNA Damage Resistance in Human Cells and Boosts DNA Repair Activity

[0165] Based on our C. elegans findings, we next wondered whether inhibition of the highly conserved DREAM complex could provide a pharmacological approach to augment DNA repair capacities in human cells. We queried whether similarly to nematodes, the human DREAM complex might also bind DDR gene promoters. We analyzed ChIP-seq data from quiescent human cells (Litovchick et al., 2007) and searched for bound DNA repair genes following a similar criteria than the analyzed dataset from C. elegans (Goetsch and Strome, 2019) (FIG. 3E-F) and our promoter analysis (FIG. 1A, B). We selected genes bound by at least LIN9, p130 and E2F4 simultaneously, where the binding occurred in 5′ between 0 and −1000 bp of the TSS. Among the 328 genes whose promoters were bound by DREAM, 67 genes are G0-classified as “DNA repair” (significantly overrepresented, Fisher's Exact p-value<2.2e-16, FIG. 5A). Thus, the DREAM transcription repressor complex directly binds to promoters of DNA repair genes indicating that the DREAM-mediated DNA repair gene regulation is highly conserved in C. elegans and humans.

[0166] In mammals, DREAM components not only form the DREAM repressor complex but can also associate in other complexes that, unlike the DREAM complex, induce transcription (Osterloh et al., 2007; Mannefeld, Klassen and Gaubatz, 2009; Sadasivam, Duan and DeCaprio, 2012). We, therefore, decided to use chemical inhibitors of the DYRK1A kinase, which phosphorylates LIN52, a modification required for the assembly of the DREAM complex thus allowing its specific abrogation (Litovchick et al., 2011). We employed two potent but distinct chemical inhibitors of the DYRK1A kinase: The beta-carboline alkaloid harmine that has been widely used as specific DYRK1A inhibitor (Göckler et al., 2009) but could also inhibit monoamine oxidase A (MAOA) and the benzothiazole derivative INDY that was established as a highly selective DYRK1A inhibitor (Ogawa et al., 2010). As the DREAM complex represses gene expression in G0 cells, we serum starved confluent U2OS cells to obtain quiescent cell populations.

[0167] To confirm that the DYRK1A inhibitors abrogated DREAM-mediated gene repression, we performed RNAseq analysis of quiescent cells treated with either harmine hydrochloride or INDY. Among the significantly upregulated genes (FDR adjusted p-value<0.01) all the known motifs bound by DREAM in human cells (Litovchick et al., 2007) were significantly overrepresented, thus substantiating that INDY and harmine resulted in gene upregulation by inhibiting the DREAM complex. To confirm that the DYRK1A inhibitors lead to a similar derepression of DREAM targets, we plotted all the genes bound by the DREAM complex (FIG. 5A and Table 2) that were significantly up- or down-regulated (270 out of 328 genes with FDR corrected p-value<0.01) upon harmine or INDY treatment (FIG. 5B). Most of the DREAM target genes were upregulated upon both treatments (corrected binomial test for the upper right quadrant p-value=9.56e-33). Of the 67 DNA repair genes bound by DREAM (FIG. 5A), 58 were upregulated upon harmine and 46 upon INDY treatment, 45 of which were upregulated in with both treatments (highlighted with black, FIG. 5B). These results indicate that the pharmacological inhibition of DYRK1A with harmine or INDY results in upregulation of DREAM target genes, including the majority of DREAM targets encoding DNA repair genes.

[0168] To directly assess whether DYRK1A inhibitor treatment could augment DNA damage resistance, we exposed harmine- or INDY-treated quiescent cells to UV irradiation. We measured the apoptotic response to the DNA damage using Annexin V and 7-AAD analysis by FACS (FIG. 5C-D). Strikingly, both DYRK1A inhibitor treatments resulted in a highly significant reduction in UV-induced apoptosis compared to mock treated cells. As the DREAM complex targets multiple DNA repair genes both in nematodes and humans (FIGS. 3 and 5A), we hypothesized that the harmine and INDY treated cells could be used to elevate resistance also to other DNA damage types. Consistently, DYRK1A inhibitor treatment boosted the resistance to the alkylating agent MMS (FIG. 5E-F). In conclusion, pharmacological inhibition of the DREAM complex kinase DYRK1A increased the expression of DREAM target genes involved in DNA repair and conferred resistance to distinct types of DNA damage, suggesting a highly conserved function of the DREAM complex in regulating DNA repair capacities.

[0169] To analyze whether DYRK1A inhibitor treatment augments DNA repair in quiescent cells, harmine- or INDY-treated quiescent cells are exposed to UV irradiation followed by antibody staining against the well-known DNA damage response marker γH2AX. In line with the results obtained from the CPD-staining performed in C. elegans, human quiescent cells that are treated with the DYRK1A inhibitors are expected to exhibit significantly lower amounts of γH2AX hours after UV-irradiation as compared to control cells that receive the UV treatment but are not exposed to either harmine or INDY, and, thus, that UV-induced lesions are repaired more readily in quiescent cells upon inhibition of DREAM assembly.

TABLE-US-00006 TABLE 5 DREAM binds DDR genes in the promoters of quiescent cells. List including all the genes bound in the promoter (0 to −1000 bp) by at least three components of DREAM (re-analysis of reference (Litovchick et al., 2007)), among them, the 67 genes involved in DNA repair (“✓”, database released 2021 Jan. 1), of which 58 and 46 we identified as upregulated upon harmine or INDY treatments respectively (“✓”). Harmine INDY DNA DNA DNA Gene repair repair repair Gene stable ID name genes UP UP ENSG00000004700 RECQL ✓ ✓ ✓ ENSG00000013573 DDX11 ✓ ✓ ✓ ENSG00000051180 RAD51 ✓ ✓ ✓ ENSG00000051341 POLQ ✓ ✓ ✓ ENSG00000062822 POLD1 ✓ ✓ ✓ ENSG00000071539 TRIP13 ✓ ✓ ✓ ENSG00000072501 SMC1A ✓ ✓ ✓ ENSG00000073111 MCM2 ✓ ✓ ✓ ENSG00000077152 UBE2T ✓ ✓ ENSG00000080345 RIF1 ✓ ✓ ✓ ENSG00000083093 PALB2 ✓ ✓ ✓ ENSG00000085999 RAD54L ✓ ✓ ✓ ENSG00000092853 CLSPN ✓ ✓ ✓ ENSG00000097046 CDC7 ✓ ✓ ✓ ENSG00000100479 POLE2 ✓ ✓ ✓ ENSG00000104889 RNASEH2A ✓ ✓ ✓ ENSG00000106268 NUDT1 ✓ ✓ ✓ ENSG00000108384 RAD51C ✓ ✓ ✓ ENSG00000109674 NEIL3 ✓ ✓ ✓ ENSG00000111206 FOXM1 ✓ ✓ ✓ ENSG00000111602 TIMELESS ✓ ✓ ✓ ENSG00000117748 RPA2 ✓ ✓ ENSG00000121988 ZRANB3 ✓ ✓ ENSG00000123374 CDK2 ✓ ✓ ✓ ENSG00000125871 MGME1 ✓ ✓ ✓ ENSG00000125885 MCM8 ✓ ✓ ENSG00000132646 PCNA ✓ ✓ ✓ ENSG00000132781 MUTYH ✓ ✓ ✓ ENSG00000133119 RFC3 ✓ ✓ ✓ ENSG00000134574 DDB2 ✓ ✓ ENSG00000136492 BRIP1 ✓ ✓ ENSG00000138376 BARD1 ✓ ✓ ENSG00000139618 BRCA2 ✓ ✓ ✓ ENSG00000141499 WRAP53 ✓ ✓ ✓ ENSG00000146263 MMS22L ✓ ✓ ENSG00000146670 CDCA5 ✓ ✓ ✓ ENSG00000147536 GINS4 ✓ ✓ ✓ ENSG00000152422 XRCC4 ✓ ✓ ENSG00000154920 EME1 ✓ ✓ ✓ ENSG00000162607 USP1 ✓ ✓ ✓ ENSG00000163918 RFC4 ✓ ✓ ✓ ENSG00000164611 PTTG1 ✓ ✓ ✓ ENSG00000164754 RAD21 ✓ ✓ ✓ ENSG00000166801 FAM111A ✓ ✓ ENSG00000168496 FEN1 ✓ ✓ ENSG00000175279 CENPS ✓ ✓ ✓ ENSG00000178295 GEN1 ✓ ✓ ✓ ENSG00000178966 RMI1 ✓ ✓ ✓ ENSG00000182481 KPNA2 ✓ ✓ ✓ ENSG00000183763 TRAIP ✓ ✓ ✓ ENSG00000183765 CHEK2 ✓ ✓ ✓ ENSG00000185480 PARPBP ✓ ✓ ✓ ENSG00000186280 KDM4D ✓ ✓ ENSG00000196584 XRCC2 ✓ ✓ ✓ ENSG00000197275 RAD54B ✓ ✓ ✓ ENSG00000197299 BLM ✓ ✓ ✓ ENSG00000221829 FANCG ✓ ✓ ✓ ENSG00000227345 PARG ✓ ✓ ENSG00000020922 MRE11 ✓ ENSG00000127922 SEM1 ✓ ENSG00000136504 KAT7 ✓ ✓ ENSG00000140451 PIF1 ✓ ENSG00000168148 H3-4 ✓ ENSG00000181218 H2AW ✓ ENSG00000247746 USP51 ✓ ENSG00000258366 RTEL1 ✓ ENSG00000270882 H4C14 ✓

Discussion

[0170] Given the complexity of the repair mechanisms that respond to the distinct types of DNA lesions, it has remained elusive whether a mechanism exists that regulates the overall repair capacities of an organism. We uncovered that the C. elegans DREAM complex actively represses DNA repair gene expression in somatic tissues, thus curbing their repair capacity and consequently limiting developmental growth, organismal health and lifespan upon DNA damage. Our data indicate that pharmacological targeting of DREAM, e.g., of the assembly of DREAM using DYRK1A inhibitors can be applied for augmenting DNA damage resistance in human cells.

[0171] Our analysis revealed that DNA repair genes across all major DNA repair systems contain motifs in their promoters that are recognized by DREAM, resulting in their DREAM-mediated repression in the soma. Abrogating DREAM leads to de-repression of DNA repair gene expression, boosting DNA repair activity—as we demonstrated for the removal of UV-induced CPDs—and conferred significant resistance to every genotoxic agent that we tested. The recognition of DNA lesions is mediated by recognition factors that are constitutively present, such as CSB and XPC that indirectly or directly recognize helix-distorting lesions, glycosylases that recognize distinct oxidative base modifications, and DSB binding proteins such as the Ku70/80 heterodimer and the MRN complex initiating NHEJ and HRR, respectively. Transcriptional regulation of individual DNA repair genes has thus far been found as part of the DDR to further elevate levels of a specific DNA repair mechanism when responding to a specific genotoxic insult (Christmann and Kaina, 2013). In contrast, we found that abrogation of the DREAM complex leads to a constitutive induction of genes operating in all major DNA repair pathways.

[0172] DREAM is a highly conserved transcription repressor complex that regulates the induction and maintenance of cellular quiescence by repressing cell-cycle genes across multiple species (Sadasivam and DeCaprio, 2013; Uxa et al., 2019). Similar to the somatic cells in C. elegans, quiescent mammalian cells have limited DNA repair capacities (Iyama and Wilson III, 2013). For example, while some mammalian stem cell compartments such as intestinal stem cells utilize HRR, quiescent hematopoietic stem cells (HCSs) and hair follicle stem cells (HFSCs) employ NHEJ (Al zouabi and Bardin, 2020), an error-prone DSB repair mechanism, which was shown to be accountable for the increased mutagenesis during HSC aging (Mohrin et al., 2010). In contrast, cell cycle entry promotes DNA repair in HSCs (Beerman et al., 2014) as well as in postmitotic neurons (Schwartz et al., 2007). Similarly to the DREAM targets in C. elegans, we found a significant number of DNA repair genes bound by DREAM in human ChIP-seq data suggesting that the function of DREAM across multiple species includes the transcriptional repression of DNA repair genes. In human cells, the DREAM complex represses cell cycle genes that are induced in the G1/S-transition and include DNA repair genes (Fischer et al., 2016). Here, we established the applicability of pharmacological targeting of the DREAM complex to augment DNA damage resistance in quiescent human cells. Indeed, two independent and highly specific DYRK1A inhibitors, harmine and INDY, increased the expression of DREAM targets, including DNA repair genes, and resulted in a strikingly elevated resistance to two distinct DNA damage types (FIG. 5). It is thus likely that DREAM complex inhibition in mammals could prevent DNA-damage driven conditions, such as stem cell exhaustion (McNeely et al., 2019), neurodegeneration (Madabhushi, Pan and Tsai, 2014) and premature aging (Niedernhofer et al., 2018). Intriguingly, the DYRK1A kinase is overexpressed in Down-Syndrome and involved in neurodegeneration in these patients, as well as in Alzheimer's, Parkinson's and Pick's disease (Ferrer et al., 2005; Dowjat et al., 2007; Liu et al., 2008). Considering that the conserved DREAM complex is highly active in post-mitotic cells and that neurons are particularly susceptible to DNA repair defects such as TC-NER defects in Cockayne syndrome, DSB repair deficiencies in Ataxia telangiectasia, or impaired SSB repair in cerebellar ataxia (Madabhushi, Pan and Tsai, 2014), targeting of the DREAM complex provides new therapeutic avenues for a range of congenital DNA repair deficiencies in humans.

[0173] Of particular relevance to such monogenetic DNA repair deficiency syndromes, the overall elevated DNA repair gene expression revealed an interesting compensatory role between the distinct repair pathways. Abrogation of the DREAM complex enhances the removal of UV-induced DNA lesions and even suppresses defects in GG-NER and TC-NER but requires the presence of the NER machinery. These data suggest that in DREAM mutants each NER sub-pathway is elevated and their respective compensation is enhanced. Strikingly, DREAM mutants even suppress HRR deficiency, likely through elevated NHEJ. The suppression of polh-1 mutants' sensitivity to ICLs and alkylating damage suggest that the consequence of DREAM abrogation might not necessarily be more error-prone repair. Instead, we show that DREAM mutants lead to restoration of developmental growth upon DNA damage in these mutants. Likewise, DNA damage-driven aging—in C. elegans induced upon exogenous DNA damage—is also alleviated in the absence of DREAM.

[0174] Our data establish the DREAM complex as a master regulator of DNA repair gene expression in somatic tissues and quiescent cells. Abrogation of the DREAM-mediated DNA repair gene repression elevates somatic repair capacities and enhances DNA damage resistance. We show that the DREAM complex restricts somatic DNA repair and its alleviation confers germline-like DNA repair capacities to the soma. Given the central role of nuclear genome stability in the aging process (Schumacher et al., 2021), the DREAM complex provides a valuable target intervening at a root cause of age-related diseases. Moreover, the suppression of various DNA repair defects such as GG- and TC-NER and HRR shows that the DREAM complex provides therapeutic opportunity also in congenital DNA repair deficiency syndromes that cause developmental growth failure and premature aging.

Example 2

[0175] In example 2, it is demonstrated how inhibition of the DREAM kinase DYRK1A prevents photoreceptor cell loss as well as loss of kidney function in Ercc1 knockout mouse models of premature aging.

Material and Methods

[0176] Analysis of Photoreceptor Cell Loss in an Ercc1 Knockout Mouse Model of Premature Aging after Treatment with the DYRK1A Inhibitor Harmine

[0177] Male and female Ercc1−/− mice on the third day after birth (postnatal day P3) were injected intraperitoneally 3 times/week with 10 mg/kg/bw of harmine hydrochloride (SMB00461, Sigma) diluted in 0.9% sodium chloride. Mice were sacrificed at postnatal day P15 for retina tissue isolation. Tissues were embedded in optimal cutting temperature (OCT) compound, cryosectioned and stained using in situ cell detection kit (TUNEL staining) (11684817910, Roche).

Analysis of Kidney Function in an Ercc1 Knockout Mouse Model of Premature Aging after Treatment with the DYRK1A Inhibitor Harmine

[0178] Podocyte-specific Ercc1 knockout mice at 3 weeks of age were mock treated (n=3) or treated with 10 mg/kg/day harmine (n=3) for 4 weeks. Proteinuria as endpoint for kidney dysfunction was assessed in urine 8 days and 11 days afterwards by assessing overall protein levels in urine samples via Coomassie brilliant blue staining after electrophoretic separation in a polyacrylamide gel.

Results and Discussion

Inhibition of the DREAM Complex Kinase DYRK1A Prevents Photoreceptor Cell Loss and Loss of Kidney Function in Ercc1 Knockout Mouse Models of Premature Aging

[0179] Loss of photoreceptor cells in the outer nuclear layer (ONL) causes age-dependent macular degeneration in humans. Patients suffering from Cockayne syndrome and related nucleotide excision repair (NER) defects age prematurely and display retinal degeneration that is driven by apoptosis in photoreceptor cells (Gorgets et al., 2007). Therefore, the progeroid NER deficient Ercc1.sup.−/− mice provide a model for age-related macular degeneration. Here, we showed that treatment with the DYRK1A inhibitor harmine, which we established to result in effective inhibition of the DREAM complex, prevented apoptosis in photoreceptor cells (FIG. 6). This data provides evidence that DYRK1A and DREAM inhibition can be applied for treatment of age-related macular degeneration. Specifically dry AMD with its clear link to the loss of photoreceptor cells provides a clinical application for DYRK1A and/or DREAM inhibition.

[0180] Podocyte-specific Ercc1 knockout animals develop progressive kidney failure as a result of premature loss of podocytes leading to glomerulosclerosis, interstitial fibrosis and tubular atrophy that is reminiscent of focal segmental glomerulosclerosis (FSGS) (Braun et al., 2020). Proteinuria is a clinical outcome of loss of kidney function and occurs within 2 months from birth in mice with a podocyte-specific knockout of Ercc1. Treatment of podocyte-specific Ercc1 knockout mice at 3 weeks of age with 10 mg/kg/day harmine (n=3) for 4 weeks resulted in the mitigation of proteinuria (FIG. 7A-B), indicating that DYRK1A and/or DREAM inhibition can mitigate age-dependent loss of renal function, specifically FSGS.

REFERENCES

[0181] Arnoult, N. et al. (2017) ‘Regulation of DNA repair pathway choice in S and G2 phases by the NHEJ inhibitor CYREN’, Nature. Nature Publishing Group, 549(7673), pp. 548-552. doi: 10.1038/nature24023. [0182] Ashburner, M. et al. (2000) ‘Gene Ontology: tool for the unification of biology’, Nature Genetics, 25(1), pp. 25-29. doi: 10.1038/75556. [0183] Beerman, I. et al. (2014) ‘Quiescent hematopoietic stem cells accumulate DNA damage during aging that is repaired upon entry into cell cycle’, Cell Stem Cell. Elsevier Inc., 15(1), pp. 37-50. doi: 10.1016/j.stem.2014.04.016. [0184] Bianco, J. N. and Schumacher, B. (2018) ‘MPK-1/ERK pathway regulates DNA damage response during development through DAF-16/FOXO’, Nucleic Acids Research, 46(12), pp. 6129-6139. doi: 10.1093/nar/gky404. [0185] Branzei, D. and Foiani, M. (2008) ‘Regulation of DNA repair throughout the cell cycle’, Nature Reviews Molecular Cell Biology, 9(4), pp. 297-308. doi: 10.1038/nrm2351. [0186] Braun, F. et al. (2020) ‘DNA damage activates mTORC1 signaling in podocytes leading to glomerulosclerosis’, bioRxiv 2020.09.13.295303; doi:10.1101/2020.09.13.295303. [0187] Brenner, S. (1974) ‘The genetics of Caenorhabditis elegans’, Genetics, 77(1), pp. 71-94. [0188] Carbon, S. et al. (2019) ‘The Gene Ontology Resource: 20 years and still GOing strong’, Nucleic Acids Research. Oxford University Press, 47(D1), pp. D330-D338. doi: 10.1093/nar/gky1055. [0189] Ceol, C. J. et al. (2006) ‘Identification and classification of genes that act antagonistically to let-60 ras signaling in Caenorhabditis elegans vulval development’, Genetics, 173(2), pp. 709-726. doi: 10.1534/genetics.106.056465. [0190] Chatterjee and Walker (2017) ‘Mechanisms of DNA damage, repair, and mutagenesis’. Environ Mol Mutagen 58(5), 235-263 [0191] Chen, B.-R. et al. (2021) LIN37-DREAM Prevents DNA End Resection and Homologous Recombination at DNA Double Strand Breaks in Quiescent Cells. bioRxiv. https://doi.org/10.1101/2021.04.21.440786 [0192] Christmann, M. and Kaina, B. (2013) ‘Transcriptional regulation of human DNA repair genes following genotoxic stress: Trigger mechanisms, inducible responses and genotoxic adaptation’, Nucleic Acids Research, 41(18), pp. 8403-8420. doi: 10.1093/nar/gkt635. [0193] Clejan, I., Boerckel, J. and Ahmed, S. (2006) ‘Developmental modulation of nonhomologous end joining in Caenorhabditis elegans’, Genetics, 173(3), pp. 1301-1317. doi: 10.1534/genetics.106.058628. [0194] Cui, M. et al. (2006) ‘SynMuv Genes Redundantly Inhibit lin-3/EGF Expression to Prevent Inappropriate Vulval Induction in C. elegans’, Developmental Cell, 10(5), pp. 667-672. doi: 10.1016/j.devcel.2006.04.001. [0195] Deans, A. J. and West, S. C. (2011) ‘DNA interstrand crosslink repair and cancer.’, Nat Rev Cancer, 11(7), pp. 467-480. [0196] Dowjat, W. K. et al. (2007) ‘Trisomy-driven overexpression of DYRK1A kinase in the brain of subjects with Down syndrome’, Neuroscience Letters, 413(1), pp. 77-81. doi: 10.1016/j.neulet.2006.11.026. [0197] Duerr, J. S. (2013) ‘Antibody staining in C. elegans using “freeze-cracking”’, Journal of Visualized Experiments, (80), pp. 1-9. doi: 10.3791/50664. [0198] Engeland, K. (2018), ‘Cell cycle arrest through indirect transcriptional repression by p53: I have a DREAM’, Cell Death & Differentiation, 25, pp. 114-132. [0199] Fausto N. (2004) ‘Liver regeneration and repair: Hepatocytes, progenitor cells, and stem cells.’ Hepatology 39(6), 1477-1487. [0200] Ferrer, I. et al. (2005) ‘Constitutive Dyrk1A is abnormally expressed in Alzheimer disease, Down syndrome, Pick disease, and related transgenic models’, Neurobiology of Disease, 20(2), pp. 392-400. doi: 10.1016/j.nbd.2005.03.020. [0201] Fischer, M. et al. (2016) ‘Integration of TP53, DREAM, MMB-FOXM1 and RB-E2F target gene analyses identifies cell cycle gene regulatory networks’, Nucleic Acids Research, 44(13), pp. 6070-6086. doi: 10.1093/nar/gkw523. [0202] Gartner, A. et al. (2000) ‘A conserved checkpoint pathway mediates DNA damage-induced apoptosis and cell cycle arrest in C. elegans.’, Molecular Cell, 5(3), pp. 435-443. [0203] Gockler, N. et al. (2009) ‘Harmine specifically inhibits protein kinase DYRK1A and interferes with neurite formation’, FEBS Journal, 276(21), pp. 6324-6337. doi: 10.1111/j.1742-4658.2009.07346.x. [0204] Goetsch, P. D., Garrigues, J. M. and Strome, S. (2017) ‘Loss of the Caenorhabditis elegans pocket protein LIN-35 reveals MuvB's innate function as the repressor of DREAM target genes’, PLoS Genetics, 13(11), pp. 1-25. doi: 10.1371/journal.pgen.1007088. [0205] Goetsch, P. D. and Strome, S. (2019) ‘DREAM Interrupted: Severing MuvB from DREAM's pocket protein impairs gene repression but not DREAM assembly on chromatin’, BioRxiv. [0206] Gonzalo and Kreienkamp (2015) ‘DNA repair defects and genome instability in Hutchinson-Gilford Progeria Syndrome.’ Curr Opin Cell Biol. 34, 75-83. [0207] Gorbunova, V. et al. (2007) ‘Changes in DNA repair during aging’, Nucleic Acids Research, 35(22), pp. 7466-7474. doi: 10.1093/nar/gkm756. [0208] Gorgels, T. G. M. F. et al. (2007) ‘Retinal degeneration and ionizing radiation hypersensitivity in a mouse model for Cockayne syndrome’, Mol. Cell Biol. 27, 1433-1441. [0209] Guard et al. (2019) ‘The nuclear interactome of DYRK1A reveals a functional role in DNA damage repair.’ Scientific Reports 9:6539 [0210] Harris, T. W. et al. (2020) ‘WormBase: a modern Model Organism Information Resource’, Nucleic acids research. Oxford University Press, 48(D1), pp. D762-D767. doi: 10.1093/nar/gkz920. [0211] Hecker et al. (2017) ‘Transcription factor decoy technology: A therapeutic update.’ Biochemical Pharmacology, 144, 29-34. [0212] Heinz, S. et al. (2010) ‘Simple Combinations of Lineage-Determining Transcription Factors Prime cis-Regulatory Elements Required for Macrophage and B Cell Identities’, Molecular Cell. Elsevier Inc., 38(4), pp. 576-589. doi: 10.1016/j.molcel.2010.05.004. [0213] Hou, J.-W. (2009) ‘Natural Course of Neonatal Progeroid Syndrome.’ Pediatr. Neonatol. 50(3): 102-109. https://en.wikipedia.org/wiki/Cell cycle [0214] Hustedt, N. and Durocher, D. (2016) ‘The control of DNA repair by the cell cycle.’, Nature cell biology, 19(1), pp. 1-9. [0215] Ionescu et al. (2012) ‘DYRK1A Kinase Inhibitors with Emphasis on Cancer.’ Mini-Reviews in Medicinal Chemistry 12, pp. 1315-1329. [0216] Iwamoto et al. (2010) ‘A general chemical method to regulate protein stability in the mammalian centralnervous system.’ Chem Biol. 17(9), 981-988. [0217] Iyama, T. and Wilson III, D. M. (2013) ‘DNA repair mechanisms in dividing and non-dividing cells’, DNA Repair, 12(8), pp. 620-636. [0218] Jackson, S. P. and Bartek, J. (2009) ‘The DNA-damage response in human biology and disease’, Nature, 461(7267), pp. 1071-1078. doi: 10.1038/nature08467. [0219] Janisiw, E. et al. (2018) ‘BRCA1-BARD1 associate with the synaptonemal complex and pro-crossover factors and influence RAD-51 dynamics during Caenorhabditis elegans meiosis’, PLoS Genetics, 14(11), pp. 1-30. doi: 10.1371/journal.pgen.1007653. [0220] Johnson, N. M., Lemmens, B. B. L. G. and Tijsterman, M. (2013) ‘A Role for the Malignant Brain Tumour (MBT) Domain Protein LIN-61 in DNA Double-Strand Break Repair by Homologous Recombination’, PLoS Genetics. Edited by N. Maizels, 9(3), p. e1003339. doi: 10.1371/journal.pgen.1003339. [0221] Keith, S. A. et al. (2014) ‘The C. elegans healthspan and stress-resistance assay toolkit’, Methods. Elsevier Inc., 68(3), pp. 476-486. doi: 10.1016/j.ymeth.2014.04.003. [0222] Kim, D. et al. (2013) ‘TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions’, Genome Biology, 14(4), p. R36. doi: 10.1186/gb-2013-14-4-r36. [0223] Kirienko, N. V. and Fay, D. S. (2007) ‘Transcriptome profiling of the C. elegans Rb ortholog reveals diverse developmental roles’, Developmental Biology, 305(2), pp. 674-684. doi: 10.1016/j.ydbio.2007.02.021. [0224] Lans, H. et al. (2010) ‘Involvement of Global Genome Repair, Transcription Coupled Repair, and Chromatin Remodeling in UV DNA Damage Response Changes during Development’, PLoS Genetics. Edited by N. Maizels, 6(5), p. e1000941. doi: 10.1371/journal.pgen.1000941. [0225] Lans, H. et al. (2013) ‘DNA damage leads to progressive replicative decline but extends the life span of long-lived mutant animals’, Cell Death & Differentiation, 20(12), pp. 1709-1718. doi: 10.1038/cdd.2013.126. [0226] Latorre, I. et al. (2015) ‘The DREAM complex promotes gene body H2A.Z for target repression’, Genes & Development, 29(5), pp. 495-500. doi: 10.1101/gad.255810.114. [0227] Lehner, B. et al. (2006) ‘Loss of LIN-35, the Caenorhabditis elegans ortholog of the tumor suppressor p105Rb, results in enhanced RNA interference.’, Genome biology, 7(1), p. R4. doi: 10.1186/gb-2006-7-1-r4. [0228] Lévy et al. (1999) ‘Analysis of a conditional degradation signal in yeast and mammalian cells.’ Eur J Biochem. 259(1-2), 244-252. [0229] Liang and Zhang (2013) ‘Genetic and epigenetic variations in iPSCs: potential causes and implications for application.’ Cell Stem Cell 13(2), 149-159. [0230] Litovchick, L. et al. (2007) ‘Evolutionarily Conserved Multisubunit RBL2/p130 and E2F4 Protein Complex Represses Human Cell Cycle-Dependent Genes in Quiescence’, Molecular Cell, 26(4), pp. 539-551. doi: 10.1016/j.molcel.2007.04.015. [0231] Litovchick, L. et al. (2011) ‘DYRK1A protein kinase promotes quiescence and senescence through DREAM complex assembly’, Genes Dev, 25(8), pp. 801-813. [0232] Liu, F. et al. (2008) ‘Overexpression of Dyrk1A contributes to neurofibrillary degeneration in Down syndrome’, The FASEB Journal, 22(9), pp. 3224-3233. doi: 10.1096/fj.07-104539. [0233] Love, M. I., Huber, W. and Anders, S. (2014) ‘Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2’, Genome Biology, 15(12), pp. 1-21. doi: 10.1186/s13059-014-0550-8. [0234] Lu, X. and Horvitz, H. R. (1998) ‘Lin-35 and lin-53, two genes that antagonize a C. elegans Ras pathway, encode proteins similar to Rb and its binding protein RbAp48’, Cell, 95(7), pp. 981-991. doi: 10.1016/S0092-8674(00)81722-5. [0235] Lynch, M. (2010) ‘Evolution of the mutation rate’, Trends Genet, 26(8), pp. 345-352. doi: 10.1016/j.tig.2010.05.003.Evolution. [0236] Madabhushi, R., Pan, L. and Tsai, L.-H. (2014) ‘DNA Damage and Its Links to Neurodegeneration’, Neuron, 83(2), pp. 266-282. doi: 10.1016/j.neuron.2014.06.034. [0237] Mannefeld, M., Klassen, E. and Gaubatz, S. (2009) ‘B-MYB is required for recovery from the DNA damage-induced G2 checkpoint in p53 mutant cells’, Cancer Research, 69(9), pp. 4073-4080. doi: 10.1158/0008-5472.CAN-08-4156. [0238] Maynard, S. et al. (2008) ‘Human Embryonic Stem Cells Have Enhanced Repair of Multiple Forms of DNA Damage’, Stem Cells, 26(9), pp. 2266-2274. doi: 10.1634/stemcells.2007-1041. [0239] McNeely, T. et al. (2019) ‘DNA damage in aging, the stem cell perspective’, Human Genetics. Springer Berlin Heidelberg, (0123456789). doi: 10.1007/500439-019-02047-z. [0240] Milholland, B. et al. (2017) ‘Differences between germline and somatic mutation rates in humans and mice’, Nature Communications. Nature Publishing Group, 8(1), p. 15183. doi: 10.1038/ncomms15183. [0241] Mohrin, M. et al. (2010) ‘Hematopoietic stem cell quiescence promotes error-prone DNA repair and mutagenesis’, Cell Stem Cell, 7(2), pp. 174-185. doi: 10.1016/j.stem.2010.06.014. [0242] Momcilovic, O. et al. (2010) ‘DNA damage responses in human induced pluripotent stem cells and embryonic stem cells’, PLoS ONE, 5(10). doi: 10.1371/journal.pone.0013410. [0243] Mueller, M. M. et al. (2014) ‘DAF-16/FOXO and EGL-27/GATA promote developmental growth in response to persistent somatic DNA damage’, Nature Cell Biology, 16(12), pp. 1168-1179. doi: 10.1038/ncb3071. [0244] Müller, G. A. et al. (2012) ‘The CHR promoter element controls cell cycle-dependent gene transcription and binds the DREAM and MMB complexes’, Nucleic Acids Research, 40(4), pp. 1561-1578. doi: 10.1093/nar/gkr793. [0245] Müller, G. A. and Engeland, K. (2010) ‘The central role of CDE/CHR promoter elements in the regulation of cell cycle-dependent gene transcription’, FEBS Journal, 277(4), pp. 877-893. doi: 10.1111/j.1742-4658.2009.07508.x. [0246] Niedernhofer, L. J. et al. (2018) ‘Nuclear Genomic Instability and Aging’, Annu Rev Biochem, 87(1), pp. 295-322. [0247] Nussbaum-Krammer, C. I. et al. (2015) ‘Investigating the Spreading and Toxicity of Prion-like Proteins Using the Metazoan Model Organism <em>C. elegans</em>’, Journal of Visualized Experiments, (95), p. 52321. doi: 10.3791/52321. [0248] Ogawa, Y. et al. (2010) ‘Development of a novel selective inhibitor of the Down syndrome-related kinase Dyrk1A’, Nature Communications. Nature Publishing Group, 1(7). doi: 10.1038/ncomms1090. [0249] Osterloh, L. et al. (2007) ‘The human synMuv-like protein LIN-9 is required for transcription of G2/M genes and for entry into mitosis’, EMBO Journal, 26(1), pp. 144-157. doi: 10.1038/sj.emboj.7601478. [0250] Ou, H.-L. et al. (2019) ‘Somatic Niche Cells Regulate the CEP-1/p53-Mediated DNA Damage Response in Primordial Germ Cells’, Developmental Cell, 50(2), pp. 167-183.e8. doi: 10.1016/j.devcel.2019.06.012. [0251] Patro, R. et al. (2017) ‘Salmon provides fast and bias-aware quantification of transcript expression’, Nature Methods. Nature Publishing Group, 14(4), pp. 417-419. doi: 10.1038/nmeth.4197. [0252] Reinke, V. et al. (2004) ‘Genome-wide germline-enriched and sex-biased expression profiles in Caenorhabditis elegans’, Development, 131(2), pp. 311-323. doi: 10.1242/dev.00914. [0253] Rieckher, M. et al. (2017) ‘A simple answer to complex questions: Caenorhabditis elegansas an experimental model for examining the DNA damage response and disease genes’, Journal of Cellular Physiology, 233(4), pp. 2781-2790. doi: 10.1002/jcp.25979. [0254] Robinson, M. D., McCarthy, D. J. and Smyth, G. K. (2009) ‘edgeR: A Bioconductor package for differential expression analysis of digital gene expression data’, Bioinformatics, 26(1), pp. 139-140. doi: 10.1093/bioinformatics/btp616. [0255] Roerink, S. F. et al. (2012) ‘A Broad Requirement for TLS Polymerases η and κ, and Interacting Sumoylation and Nuclear Pore Proteins, in Lesion Bypass during C. elegans Embryogenesis’, PLoS Genetics. Edited by S. Jinks-Robertson, 8(6), p. e1002800. doi: 10.1371/journal.pgen.1002800. [0256] Sadasivam, S. and DeCaprio, J. A. (2013) ‘The DREAM complex: master coordinator of cell cycle-dependent gene expression’, Nature Reviews Cancer, 13(8), pp. 585-595. doi: 10.1038/nrc3556. [0257] Sadasivam, S., Duan, S. and DeCaprio, J. a. (2012) ‘The MuvB complex sequentially recruits B-Myb and FoxM1 to promote mitotic gene expression’, Genes and Development, 26(5), pp. 474-489. doi: 10.1101/gad.181933.111. [0258] Saretzki, G. et al. (2004) ‘Stress Defense in Murine Embryonic Stem Cells Is Superior to That of Various Differentiated Murine Cells’, Stem Cells, 22(6), pp. 962-971. doi: 10.1634/stemcells.22-6-962. [0259] Schindelin, J. et al. (2012) ‘Fiji: An open-source platform for biological-image analysis’, Nature Methods, 9(7), pp. 676-682. doi: 10.1038/nmeth.2019. [0260] Schmittgen, T. D. and Livak, K. J. (2008) ‘Analyzing real-time PCR data by the comparative CT method’, Nature Protocols, 3(6), pp. 1101-1108. doi: 10.1038/nprot.2008.73. [0261] Schumacher, B. et al. (2021) ‘The central role of DNA damage in the ageing process’, Nature. Springer US, 592(April), pp. 695-703. doi: 10.1038/s41586-021-03307-7. [0262] Schwartz, E. I. et al. (2007) ‘Cell cycle activation in postmitotic neurons is essential for DNA repair’, Cell Cycle, 6(3), pp. 318-329. doi: 10.4161/cc.6.3.3752. [0263] Seabold, S. and Perktold, J. (2010) ‘Statsmodels: Econometric and Statistical Modeling with Python’, in Proceedings of the 9th Python in Science Conference, pp. 92-96. doi: 10.25080/Majora-92bf1922-011. [0264] Sierra, H. et al. (2015) ‘Confocal Imaging—Guided Laser Ablation of Basal Cell Carcinomas: An Ex Vivo Study’, Journal of Investigative Dermatology, 135(2), pp. 612-615. doi: 10.1038/jid.2014.371. [0265] da Silva, P. F. L. and Schumacher, B. (2019) ‘DNA damage responses in ageing’, Open Biology, 9(11), p. 190168. doi: 10.1098/rsob.190168. [0266] Soll, J. M., Sobol, R. W. and Mosammaparast, N. (2017) ‘Regulation of DNA Alkylation Damage Repair: Lessons and Therapeutic Opportunities’, Trends in Biochemical Sciences. Elsevier Ltd, 42(3), pp. 206-218. doi: 10.1016/j.tibs.2016.10.001. [0267] Soneson, C., Love, M. I. and Robinson, M. D. (2016) ‘Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences’, F1000Research, 4, p. 1521. doi: 10.12688/f1000research.7563.2. [0268] Stankunas et al. (2003) ‘Conditional protein alleles using knockin mice and a chemical inducer of dimerization.’ Mol Cell. 12(6), 1615-1624. [0269] Tabuchi, T. M. et al. (2011) ‘Chromosome-biased binding and gene regulation by the Caenorhabditis elegans DRM complex’, PLoS Genetics, 7(5). doi: 10.1371/journal.pgen.1002074. [0270] Tapia and Scholer (2016) ‘Molecular Obstacles to Clinical Translation of iPSCs.’ Cell Stem Cell 19, 298-309. [0271] Tichy, E. D. and Stambrook, P. J. (2008) ‘DNA repair in murine embryonic stem cells and differentiated cells’, Experimental Cell Research, 314(9), pp. 1929-1936. doi: 10.1016/j.yexcr.2008.02.007. [0272] Trapnell, C. et al. (2010) ‘Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation’, Nature Biotechnology. Nature Publishing Group, 28(5), pp. 511-515. doi: 10.1038/nbt.1621. [0273] Unhavaithaya, Y. et al. (2002) ‘MEP-1 and a Homolog of the NURD Complex Component Mi-2 Act Together to Maintain Germline-Soma Distinctions in C. elegans’, Cell, 111(7), pp. 991-1002. doi: 10.1016/S0092-8674(02)01202-3. [0274] Uxa, S. et al. (2019) ‘DREAM and RB cooperate to induce gene repression and cell-cycle arrest in response to p53 activation’, Nucleic acids research, 47(17), pp. 9087-9103. doi: 10.1093/nar/gkz635. [0275] Vermezovic, J. et al. (2012) ‘Differential regulation of DNA damage response activation between somatic and germline cells in Caenorhabditis elegans’, Cell Death & Differentiation, 19(11), pp. 1847-1855. doi: 10.1038/cdd.2012.69. [0276] Vitale et al. (2017) ‘ DNA Damage in Stem Cells.’ Molecular Cell 66, 306-319 [0277] Virtanen, P. et al. (2020) ‘SciPy 1.0: fundamental algorithms for scientific computing in Python’, Nature Methods, 17(3), pp. 261-272. doi: 10.1038/s41592-019-0686-2. [0278] Wagle, P., Nikon& M. and Frommolt, P. (2015) ‘QuickNGS elevates Next-Generation Sequencing data analysis to a new level of automation’, BMC Genomics, 16(1), pp. 1-8. doi: 10.1186/s12864-015-1695-x. [0279] Wang, D. et al. (2005) ‘Somatic misexpression of germline P granules and enhanced RNA interference in retinoblastoma pathway mutants’, Nature, 436(7050), pp. 593-597. doi: 10.1038/nature04010. [0280] Wilson, D. M. et al. (2017) ‘Systematic analysis of DNA crosslink repair pathways during development and aging in Caenorhabditis elegans’, Nucleic Acids Research, 45(16), pp. 9467-9480. doi: 10.1093/nar/gkx660. [0281] WO 2020/069418 A1 [0282] Yates, A. D. et al. (2020) ‘Ensembl 2020’, Nucleic acids research, 48(D1), pp. D682-D688. doi: 10.1093/nar/gkz966. [0283] Yoon et al. (2020) Small Molecule Inhibitors of DYRK1A Identified by Computational and Experimental Approaches Int. J. Mol. Sci. [0284] Al zouabi, L. and Bardin, A. J. (2020) ‘Stem Cell DNA Damage and Genome Mutation in the Context of Aging and Cancer Initiation’, Cold Spring Harbor Perspectives in Biology, p. a036210. doi: 10.1101/cshperspect.a036210.