S C

IE N

T IF

IC H

IG H

LI G

H T

S S

T R

U C

T U

R A

L B

IO LO

G Y

5 9 I H I G H L I G H T S 2 0 2 1

stalled polymerases using Integrator s nuclease activity (INTS11); and 2) by recruiting protein phosphatases, PP2A-A and PP2A-C, which remove phosphorylation marks from the RNPII CTD domain and reset its transcriptional status [4]. Both mechanisms contribute to regulation of gene expression.

Although more than 16 subunits were identified to be components of the Integrator complex, its architecture remained unknown until recently. To address this problem, a series of stable human cell lines expressing tagged variants of Integrator s subunits were generated and tandem affinity purification was performed, followed by quantitative mass spectrometry (Figure 46). This experiment identified stable modules within the Integrator complex, which were then validated by recombinant expression in insect cells. Cryo-EM analysis of the Integrator Cleavage Module, which contains the nuclease catalytic subunit, INTS11, was performed at beamline CM01.

Cryo-EM reconstruction at a resolution of 3.5 Å provided first structural insights into the catalytic core of the integrator complex. This core is composed of two homologous proteins, the true nuclease subunit, INTS11, and its

catalytically impaired homologue, INTS9 (Figure 47). Both proteins belong to the MBL/b-CASP family. The structure revealed that they form a tightly intertwined heterodimer involving three major interfaces (Figure 47).

Structure-based biochemical assays indicate that dimerisation of the C-terminal domain 2 (CTD2) is necessary to tether both proteins to each other and only then can the secondary dimerisation domain (CTD1) be formed. This allows recruitment of a HEAT-repeat containing subunit INTS4, which wedges between the two nuclease domains.

A combination of structural data with biochemical assays made it possible to propose a model wherein a progressive formation of the dimeric interface is a prerequisite for the recruitment of additional Integrator factors, including INTS4. Analysis of the surface electrostatic potential revealed a highly positively charged interface, which suggests a potential RNA-binding path within the complex. Finally, the structure, together with the recently reported structure of the histone processing complex, uncovers common design principles shared between both 3 -end processing machineries.

PRINCIPAL PUBLICATION AND AUTHORS

Structure of the catalytic core of the Integrator complex, M.M. Pfleiderer (a), W.P. Galej (a), Mol. Cell 81(6), 1246-1259.e8 (2021); https:/doi.org/10.1016/j.molcel.2021.01.005 (a) EMBL Grenoble (France)

REFERENCES

[1] D. Baillat et al., Cell 123, 265-276 (2005). [2] N.D. Elrod et al., Mol. Cell. 76, 738-752.e7 (2019). [3] S. Lykke-Andersen et al., Mol. Cell 81, 514-529.e6 (2021). [4] H. Zheng et al., Science 370, eabb5872 (2020).

Fig. 47: a) Overall structure of the INTS4/9/11 complex, which constitutes Integrator s catalytic core. b) A close-up view at the nuclease active site of INTS11. c) The active site of CPSF73, a related endonuclease from the MBL/b-CASP family, showing conservation of the active site between the two enzymes. d) Degenerated active site of INTS9, which evolved to become a pseudoenzyme.