Science 13Mar2020

are no longer expressed, showed no difference
in mRNA degradation between WT and DKO
(fig. S15), suggesting that BTG1/2 contribute to
lowering mRNA amounts through mRNA turn-
over in naïve but not activated T cells.
To examine if poly(A) length is indeed
elongated in DKO cells, we performed mTAIL-
seq (fig. S16A) ( 24 ). In line with the stabiliza-
tion of mRNA, the poly(A) tail length of DKO
T cells was longer than that of WT (Fig. 4, E
and F, and table S4). Again, mtRNA poly(A)
tail lengths were nearly unchanged (Fig. 4F),
suggesting that BTG1/2-mediated deadenyla-
tion was responsible for degrading the bulk of
cytoplasmic mRNA in naïve T cells. Notably,
the difference in poly(A) tail length in naïve
T cells was diminished in activated T cells
(fig.S16,BandC,andtableS5).
In conclusion, we propose a model to ex-
plainhowthequiescentstateinTcellsis
maintained (Fig. 4G). BTG1 and BTG2 are
highly and specifically expressed in quiescent
T cells and promote mRNA deadenylation and
degradation. Unlike BTG1/2, deadenylases ap-
pear to be ubiquitously expressed at a low level,
implying a specialized function of BTG1/2 in
immune cells (fig. S17). Because BTG1/2 inter-
actwithPABPandtheCNOTcomplex,which
bind nonspecifically to mRNAs ( 25 ), BTG1/2
can direct the down-regulation of mRNAs at
a global level. This mechanism is seemingly
inefficient in terms of cost, but the availability
of presynthesized mRNA provides a benefit to
quiescent T cells of a rapid response to acti-
vation signals. Given that naïve T cells differ-
entiate into multiple lineages, having such a
primed state on a hair trigger would be ulti-
mately beneficial. Upon activation, BTG1/2
quickly disappear, which results in an accumu-
lation of mRNAs and exit from the quiescent
state (Figs. 1D and 4G).

Here, we show the functional consequences

of BTG1/2-mediated deadenylation in vivo.

B cells may also use a similar mechanism

to maintain quiescence through BTG1/2, as

BTG1/2 are commonly dysregulated in leuke-

mia and lymphoma ( 15 , 26 – 28 ). Likewise,

other cofactors, in addition to BTG1/2, may

recruit and enhance mRNA deadenylation and

degradation in certain cell types. Thus, the

mechanism that we propose may be broadly

applicable to other cells and tissues, serving as

a general system to secure the quiescent state.

REFERENCES AND NOTES

1. T. H. Cheung, T. A. Rando,Nat. Rev. Mol. Cell Biol. 14 , 329– 340
   (2013).


2. G. J. van der Windt, E. L. Pearce,Immunol. Rev. 249 ,27–42 (2012).


3. N. J. MacIver, R. D. Michalek, J. C. Rathmell,Annu. Rev.
   Immunol. 31 , 259–283 (2013).


4. J. Zhu, H. Yamane, W. E. Paul,Annu. Rev. Immunol. 28 ,
   445 – 489 (2010).


5. J. Sprent, C. D. Surh,Nat. Immunol. 12 ,478–484 (2011).


6. W. Ouyang, O. Beckett, R. A. Flavell, M. O. Li,Immunity 30 ,
   358 – 371 (2009).


7. X. Fenget al.,Nat. Immunol. 12 , 544–550 (2011).


8. Y. M. Kerdileset al.,Nat. Immunol. 10 , 176–184 (2009).


9. W.Ouyang,M.O.Li,Trends Immunol. 32 ,26–33 (2011).


10. D. Tzachaniset al.,Nat. Immunol. 2 , 1174–1182 (2001).


11. K. Yang, G. Neale, D. R. Green, W. He, H. Chi,Nat. Immunol. 12 ,
    888 – 897 (2011).


12. D. Hniszet al.,Cell 155 , 934–947 (2013).


13. F. Mauxion, C. Y. Chen, B. Séraphin, A. B. Shyu,Trends
    Biochem. Sci. 34 , 640–647 (2009).


14. G. S. Winkler,J. Cell. Physiol. 222 ,66–72 (2010).


15. L. Yuniati, B. Scheijen, L. T. van der Meer, F. N. van Leeuwen,
    J. Cell. Physiol. 234 , 5379–5389 (2019).
    16.J.Henao-Mejiaet al.,Cold Spring Harb. Protoc. 2016 ,
    prot090704 (2016).


16. N. Gadir, L. Haim-Vilmovsky, J. Kraut-Cohen, J. E. Gerst,
    RNA 17 , 1551–1565 (2011).


17. H. Yiet al.,Mol. Cell 70 , 1081–1088.e5 (2018).


18. J. P. Rouaultet al.,J. Biol. Chem. 273 , 22563–22569 (1998).


19. D. Prévôtet al.,J. Biol. Chem. 276 , 9640–9648 (2001).


20. F. Mauxion, C. Faux, B. Séraphin,EMBO J. 27 , 1039–1048 (2008).


21. R. Doidge, S. Mittal, A. Aslam, G. S. Winkler,PLOS ONE 7 ,
    e51331 (2012).


22. B. Stupfler, C. Birck, B. Séraphin, F. Mauxion,Nat. Commun. 7 ,
    10811 (2016).
    24. J. Lim, M. Lee, A. Son, H. Chang, V. N. Kim,Genes Dev. 30 ,
    1671 – 1682 (2016).
    25. I. A. Eliseeva, D. N. Lyabin, L. P. Ovchinnikov,Biochemistry
    (Mosc.) 78 , 1377–1391 (2013).
    26. R. P. Kuiperet al.,Leukemia 21 , 1258–1266 (2007).
    27. R. D. Morinet al.,Nature 476 , 298–303 (2011).
    28. E. Waanderset al.,PLOS Genet. 8 , e1002533 (2012).

ACKNOWLEDGMENTS

We are grateful to R. Medzhitov for support and discussions. We

thank J. Y. Hwang and W. J. Chae for helpful comments.

We also thank all members of the Flavell laboratory for

technical and administrative support. We thank M. Ha for the

gift of RNA spike-in and J. Choi for illustrating the graphical

abstract.Funding:This work was supported by the Howard

Hughes Medical Institute (R.A.F.). The work of S.S.H. was

supported by a Leslie H. Warner Fellowship from Yale Cancer

Center (YCC). The work of J.L. was supported by postdoctoral

fellowships from the HFSPO (LT000037/2018-L) and Jane

Coffin Childs Memorial Fund. G.R.L was supported by National

Research Foundation of Korea (NRF-2017R1A2B3008621).

H.B.L. was supported by the National Natural Science Foundation

of China (91753141) and the start-up fund from the Shanghai

Jiao Tong University School of Medicine.Author contributions:

S.S.H. conceived and initiated the project. S.S.H. and J.L. designed

the experiments, analyzed and discussed the results, and wrote

the manuscript. S.S.H. performed most of the experiments in

this study. J.L. performed total RNA-seq and mTAIL-seq, analyzed

the data, and illustrated the figures. Z.Y. and P.K. helped to

perform the FACS analysis and in vitro experiments. E.S. and

H.X. helped to carry out the colitis experiment. C.C.D.H. helped

to analyze our deep RNA-seq data. L.K.K., G.R.L., and H.-B.L.

contributed to key discussions. R.A.F. supervised the project,

participated in interpreting the results, and edited the manuscript.

Competing interests:R.A.F is a consultant for GSK and Zai Lab

Ltd. All other authors declare no competing financial interests.

Data and materials availability:Sequenced reads have been

deposited in the NCBI Gene Expression Omnibus (GEO) database

(accession no. GSE125890).

SUPPLEMENTARY MATERIALS

science.sciencemag.org/content/367/6483/1255/suppl/DC1

Materials and Methods

Figs. S1 to 17

Tables S1 to S6

References ( 29 – 33 )

View/request a protocol for this paper fromBio-protocol.

14 February 2019; resubmitted 22 October 2019

Accepted 19 February 2020

10.1126/science.aax0194

Hwanget al.,Science 367 , 1255–1260 (2020) 13 March 2020 5of5

RESEARCH | REPORT