PMID:35333350

[manulera]

PMID:35333350

Rec8 Cohesin-mediated Axis-loop chromatin architecture is required for meiotic recombination.

Misc

• Meiotic prophase, synaptonemal complex.
• Uses Hi-C:
  • Crosslink DNA
  • Sequence to see which fragments interact with each other and build maps.
• In human somatic cells cohesin complex is associated with kleisin Rad21 -> Link sister chromatids
• In human miotic cells, Rec8 is the kleisin -> Link sister chromatids as well
• Cohesin complex in somatic cells has been shown to make loops that extrude part of the genome, and bring together distant regions.
• Formation of the LinEs (syneptonemal complex-like structures) requires the phosphorilated Rec11
• Cohesin links can form in the absence of Rec8, but cannot form lateral elements (see cartoon)

Experiment conditions

• They synchronise cells:
  • Background: diploid pat1-as2 (analog sensitive kinase)
  • Meiosis induced by addition of ATP analogue.
  • G1 arrest nitrogen starvation

Findings

Fig. 1 and Fig. S1

• In vegetative growth, they can detect clustering of centromeres in the hotspots where the centromeres match. (Rab1 orientation)
• In meiosis, the telomers match instead, indicating formation of the bouquet (Fig. 1 A-B).
• Substraction maps (red and blue) substract the contact map of Meiosis 0hr to see the condition-specific contacts.
• Diagonal lines that go from telomer to telomer in Meiosis 3h represent interactions within the same chrosome due to bouquet formation. The line is strongest close to the telomer becauset that's the tethering point I guess?
• In the bottom representation (triangle) they represent the contact of each chrosome with itself. By using this triangular coordinates and not a matrix the vertical line in the center corresponds to the point of contact of a chromosome and its opposite part. The coordinates of those points are (x,L-x) where L is the length of the chromosome and x is the position from the begining of the chromosome. Then they apply some threshold along the vertical axis and see how far the strong part of the vertical line reaches. This would indicate how closely the chromosomes are paired through their telomere. Worth mentioning that this is independent of centromere position in the chromosome.

Fig. 2

NOTE: rec10D does not form LinEs, while rec8 forms aberrant LinEs. Why rec12?

• rec10 and rec12 are not needed to form the loop.
• rec8 needed for the bouquet, and it was known that it was reuquired for normal horsetail.

Row in curs table	Genotype	Phenotype	Figure
rec8D	Decreased alignment index	Fig. 2A, B

Fig. 3 and S3

• By magnifying you can see diamonds in the matrix, which are small loops. These loops overlap with known regions of binding of Rec8 (Ref. 85).
• Fig. 3E shows that long range interactions, TADs disappear in meiosis, at the same time that the Rec8-dependent small loops appear.
• This compacts the chromosome linearly (see the cartoons of synaptonemal complexes above).

Row in curs table	Genotype	Phenotype	Figure
rec8D	Decreased loops	Fig. 3C
rec10D	A bit decreased loops?	Fig. 3C

Fig. 4

• It was shown that Wap1 (Pombe wpl1) affects the TAD big loops, does it also affect the small loops formed by Rec8?
• wpl1D has lower chromosome pairing during meiosis (they image distance between ade8 locus -lac sequence + lac binding protein tagged with GFP- and a telomeric protein, they call the system ade8-GFP). This distance is high if there is no pairing. I guess this distance is reduced if chromosomes cannot be stretched. So if longer loops are formed, they are more loose rather than closely linked?
• wpl1D also reduced torsional turning during horsetail.
• The meiosis loops of wlp1D were bigger than wt (Fig. 4C).
• deleting wpl1D and:
  • rec10D: no LinEs but still thick structures
  • bunch of recombination genes: still thick structures.
  • In summary, formation of these structures does not require LinEs or recombination.

Row in curs table	Genotype	Phenotype	Figure
9,10 how to tell the difference with strong recruitment?	wpl1D	Increased recruitment of Rec8-GFP to the structures during horsetail + more obvious elongated structure rather than fuzzy (suggesting that the axial structure becomes thicker)	Fig. 4A
8, but check specificity of term	wpl1D	Reduced chromosome pairing during horsetail	Fig. S4B
11	wpl1D	Longer loops	Fig. 4B
wpl1D	Reduced chromosome pairing in meiosis and VG (see how the centromere connexion is the same actually)	Fig. S4B
wpl1Drec10D	No LinEs (probably not worth curating since it's like rec10D)	Fig. S4F
wpl1Drec10D	Formation of thick elements (meaning that formation of LinEs is not required for this thick structures)	Fig. S4F

Fig. 5 and S5

• You can express Rad21 during meiosis and this compensates for deletion of Rec8 (in which way?). Is this ectopic forced with a promoter, or is it a compensation/leak? Ref. 57, 59. This does not seem to be curated if expression is induced.
• Their localization on wpl1D is different.
• They want to see if they can find a mutant that keeps the cohesin activity, but still forms the loops.
• For that they use a background of rad21D psc3D (mitotic cohesins) and transform with randomly mutagenised DNA of rec8. They know that colonies that live must have cohesin activity, otherwise they would die.

Row in curs table	Genotype	Phenotype	Figure
rec8-F204S + background	Normal cell growth	Fig. S5B
rec8-F204S + background	No formation of thick structures, but cohesin activity	Fig. 5B
3, but mei4 maybe should be added?	rec8-F204S + background only mei4 and locus tagged	No change in sister chromatid pairing. This is an assay for cohesin activity, no horsetail!	Fig. 5C
6	rec8-F204S no background	Phenotype in which the spb moves with horsetail but not the rest. Not entirely sure what the difference between deletion and mutant are	Fig. 5D
rec8-F204S no background	Stretched telomere - ade8 distance. Like deletion	Fig. 5E

• Physical interaction:
  • rec8-F204S: still binds to Psm1. Co-IP (Fig. S5F).

Fig. 6 and S6

Row in curs table	Genotype	Phenotype	Figure
rec8-F204S	Lower big loop formation Hi-C as deletion? looks like it	Fig. 6A
4	rec8-F204S	Lower small loop formation Hi-C lower severity than deletion	Fig. 6A
rec8-F204S	Lower recruitment to known Rec8-enriched sites of cohesin rec8 (ChIP)	Fig. S6B
rec8-F204S	Normal recruitment to other genome positions (ChIP)	Fig. S6B

Fig. 7 and S7

Row in curs table	Genotype	Phenotype	Figure
1	rec8-F204S	aberrant LinEs as seen with Rec10-mCherry	Fig. 7A
2,5?	rec8-F204S	Lower recombination (known consequence of aberratn LinEs)	Fig. 7B
rec8-F204S	Normal phosphorylation of Rec11 (not to curate I guess)	Fig. 7C

Discussion

• Rec8 is involved in the formation of axis-loop structure.
• These structures are required for formation of the LinEs.
• The LinEs are important for recombination (this was known).

Questions

• diploid genotypes
• Worth mentioning synchronisation in experiments?
  • For background
  • For the fact itself that this condition allows induction of meiosis.
• Something on the localization of ectopically expressed Rad21?
• Something on the localization of Rec8-GFP specifically to these structures.
• Is there a GO-term for these structures?
• Difference in horsetail between rec8D and mutant?
• Should the other point mutations be curated? (Fig. S5H)
• They also test recombination with a plasmid. Is this something worth anotating? (Fig. S7B).
• They also show phosphomimetic Rec11 and fusion Rec11-Rec10 do not prevent LinE formation or recombination defects. I guess referee asked whether Rec8 could be involved in this process. Worth curating? (Fig. S7A and C).
• Regarding the wlp1 phenotype. If rec8 goes to the bottom of the loops, why would the wpl1D mutant have a higher rec8 signal. It should have fewer longer loops?

curs table questions

• What is the difference between intragenic and intergenic? Is this the plasmid thing?
• Annotation 7 -> Should this be kept?
• Annotation 12. Where does it come from
• Annotation 13, 14. I had not included them originally.

TODO

• [x] Check annotations from previous papers:
  • Ref. 90: hop1D and mek1D prevent formation of LinEs
  • Ref. 57, 58: Partial rescue of rec8D by expression of Rad21 during meiosis.