Stability and flexibility in chromatin structure and transcription underlies memory CD8 T-cell differentiation

Abstract

The process by which naïve CD8 T cells become activated, accumulate, and terminally differentiate as well as develop into memory cytotoxic T lymphocytes (CTLs) is central to the development of potent and durable immunity to intracellular infections and tumors. In this review, we discuss recent studies that have elucidated ancestries of short-lived and memory CTLs during infection, others that have shed light on gene expression programs manifest in individual responding cells and chromatin remodeling events, remodeling factors, and conventional DNA-binding transcription factors that stabilize the differentiated states after activation of naïve CD8 T cells. Several models have been proposed to conceptualize how naïve cells become memory CD8 T cells. A parsimonious solution is that initial naïve cell activation induces metastable gene expression in nascent CTLs, which act as progenitor cells that stochastically diverge along pathways that are self-reinforcing and result in shorter- versus longer-lived CTL progeny. Deciphering how regulatory factors establish and reinforce these pathways in CD8 T cells could potentially guide their use in immunotherapeutic contexts.

Full text


Open Peer Review
F1000FacultyReviewsarewrittenbymembersof
theprestigious .TheyareF1000Faculty
commissionedandarepeerreviewedbefore
publicationtoensurethatthefinal,publishedversion
iscomprehensiveandaccessible.Thereviewers
whoapprovedthefinalversionarelistedwiththeir
namesandaffiliations.
Anycommentsonthearticlecanbefoundatthe
endofthearticle.
REVIEW
Stability and flexibility in chromatin structure and transcription
underlies memory CD8 T-cell differentiation [version 1; peer
review: 2 approved]
HuitianDiao,MatthewPipkin
DepartmentofImmunologyandMicrobiology,TheScrippsResearchInstitute,Jupiter,FL,USA
Abstract
TheprocessbywhichnaïveCD8Tcellsbecomeactivated,accumulate,
andterminallydifferentiateaswellasdevelopintomemorycytotoxicT
lymphocytes(CTLs)iscentraltothedevelopmentofpotentanddurable
immunitytointracellularinfectionsandtumors.Inthisreview,wediscuss
recentstudiesthathaveelucidatedancestriesofshort-livedandmemory
CTLsduringinfection,othersthathaveshedlightongeneexpression
programsmanifestinindividualrespondingcellsandchromatinremodeling
events,remodelingfactors,andconventionalDNA-bindingtranscription
factorsthatstabilizethedifferentiatedstatesafteractivationofnaïveCD8T
cells.Severalmodelshavebeenproposedtoconceptualizehownaïvecells
becomememoryCD8Tcells.Aparsimonioussolutionisthatinitialnaïve
cellactivationinducesmetastablegeneexpressioninnascentCTLs,which
actasprogenitorcellsthatstochasticallydivergealongpathwaysthatare
self-reinforcingandresultinshorter-versuslonger-livedCTLprogeny.
Decipheringhowregulatoryfactorsestablishandreinforcethesepathways
inCD8Tcellscouldpotentiallyguidetheiruseinimmunotherapeutic
contexts.
Keywords
MemoryCD8Tcells,chromatinstructure,transcriptionalcontrol
 
Reviewer Status
 InvitedReviewers

version 1
published
31Jul2019

1 2
,TheUniversityofMelbourne,Axel Kallies
Parkville,Australia
1
,InstituteofBiomedicalPeter N Cockerill
Research,UniversityofBirmingham,
Birmingham,UK
2
31Jul2019, (F1000FacultyRev):1278(First published: 8
)https://doi.org/10.12688/f1000research.18211.1
31Jul2019, (F1000FacultyRev):1278(Latest published: 8
)https://doi.org/10.12688/f1000research.18211.1
v1
Page 1 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019


MatthewPipkin( )Corresponding author: mpipkin@scripps.edu
 :Writing–Review&Editing; :FundingAcquisition,Writing–OriginalDraftPreparation,Writing–Review&EditingAuthor roles: Diao H Pipkin M
Nocompetinginterestsweredisclosed.Competing interests:
ThisworkwassupportedbyNationalInstitutesofHealthgrantsR01AI095634andU19AI109976andUSDepartmentofGrant information:
DefensegrantW81XWH-16-1-0006(toMEP)andtheFrenchman’sCreekWomenforCancerResearch.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
©2019DiaoHandPipkinM.Thisisanopenaccessarticledistributedunderthetermsofthe ,Copyright: CreativeCommonsAttributionLicence
whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
DiaoHandPipkinM.How to cite this article: Stability and flexibility in chromatin structure and transcription underlies memory CD8
F1000Research2019, (F1000FacultyRev):1278(T-cell differentiation [version 1; peer review: 2 approved] 8
)https://doi.org/10.12688/f1000research.18211.1
31Jul2019, (F1000FacultyRev):1278( )First published: 8 https://doi.org/10.12688/f1000research.18211.1
Page 2 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

Introduction
During a prototypical acute intracellular infection that will be
cleared, naïve antigen-specific CD8 T cells become activated
and their progeny accumulates dramatically, a period generally
referred to as the “effector” phase. Near the point of maximal
accumulation, cells in the responding population manifest sub-
stantial phenotypic and functional heterogeneity. As the infection
clears, most effector cells die and the population “contracts”.
Cells that survive this period ultimately give rise to an array
of memory CD8 T-cell subsets1.
Many excellent recent reviews have comprehensively outlined
the tapestry and importance of distinct memory CD8 T-cell
subsets that arise after infection2–4. An illustration of the main
effector and memory CD8 T-cell subsets in mice depicts their
general inter-relationships (Figure 1) (Table 1). Memory T
cells are classically categorized into central memory T (Tcm)
cells, which localize in secondary lymphoid organs (SLOs),
and effector memory T (Tem) cells, which recirculate between
peripheral tissues and SLOs4. However, at early memory time
points, a substantial fraction of the classically defined Tem
cells are more effector-like and have been termed effector-like
memory cells or long-lived effector (LLE) cells5,6. Moreover,
another subset of classic Tem cells, called peripheral memory
T cells, has been delineated as those that recirculate through
peripheral tissues via SLOs and has been distinguished from
Tem cells that do not recirculate7. In addition, memory T cells
that enter and stably reside within tissues have been defined as
tissue resident memory (Trm) cells8. Further emphasizing the
diversity of memory T-cell subsets is that analysis of human
Figure 1. Patterns and inter-relationships of effector and memory CD8 T-cell subsets induced by acute intracellular infection.
(A) Antigen presentation, co-stimulation, and additional inflammatory signals induce multiple individual naïve CD8 T cells to undergo a
prototypical pattern of geometric expansion. (B) Individual cells within the nascent CTL population of early effector (EE) cells differentiate
along any one of multiple trajectories. (C) Multiple phenotypic subsets with distinct memory CD8 T-cell potentials are detectable at the
peak response, near the time when most pathogen has been eliminated. Cells that are KLRG1hi CD127lo have the shortest half-lives after
the infection resolves and are referred to as short-lived effector cells (SLECs) or simply terminal effector (TE) CD8 T cells (red). Conversely,
KLRG1lo CD127hi cells are termed memory precursor (MP) effector CD8 T cells (light blue) because they most efficiently generate memory
CD8 T cells. However, some double-positive (DP) effector cells that are KLRG1hi CD127hi9 (purple) downregulate KLRG1 and give rise to
memory CD8 T cells. Trm precursors (light green) derived from KLRG1lo intermediates in the spleen begin populating non-lymphoid tissues
(NLTs) near the peak response. (D) Most TE cells persist poorly into the memory phase. At early memory time points, some KLRG1hi cells
persist and have been termed effector-like memory cells or long-lived effector (LLE) cells but they wane over time. Tem cells preferentially
localize in the vasculature (light red background), some of which convert into Tcm cells (dark blue) that reside in secondary lymphoid organs
(light blue background) later during the memory phase. Arrows indicate the general ancestry of the different cell populations and are colored
according to the main classes of effector and memory CTL subsets.
Page 3 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

CD8 T cells using cytometry by time of flight has demon-
strated that substantial heterogeneity exists between individual
cells defined classically as Tcm and Tem cells10. The extent to
which all of these phenotypically distinguishable populations of
effector and memory T cells comprise stable cell “lineages” is
an open set of questions3.
Although a generally agreed upon concept is that memory
CD8 T cells derive from effector cells, this general explana-
tion is somewhat unsatisfying because of the semantics in
defining what an “effector” cell is11–13. Phenotypically dis-
tinct populations of cells that arise in the effector phase differ
in their propensity to form specific types of memory cytotoxic
T lymphocytes (CTLs). The phenotypes of cells representing
some of these populations are relatively stable and do not read-
ily interconvert whereas others do so more easily7,9,14,15, which
likely reflects a spectrum of differentiated states that, on the
one end, are terminally differentiated and have relatively
short-term roles and, on the other, are stem cell–like and
participate in populating and re-populating multiple memory
cell niches during iterative infections over time. It is still
unclear exactly how all of these differentiated states are initially
established and how they are maintained.
Here, we discuss recent studies that have helped to define
how activated CD8 T cells terminally differentiate or become
memory CD8 T cells, and we focus specifically on the regulation
of gene expression and chromatin structure in distinct effector
CD8 T-cell populations. Our conclusion is a model that incor-
porates many of these observations and that might help to
clarify how memory CD8 T cells develop from activated cells
in the effector phase.
The descent of memory T cells: individual naïve
CD8 T cells initiate memory CD8 T-cell programming
rapidly and stochastically undergo terminal
differentiation
A brief encounter of T-cell receptors (TCRs) on naïve CD8
T cells with their cognate peptide–major histocompatibility
complex together with co-stimulation is sufficient to induce
Table 1. Key definitions.
Effector phase: Time period between the initial infection and when the accumulation of effector cells has peaked.
Contraction phase: Time period between the peak accumulation of effector cells and when the decreasing effector population numbers
have stabilized.
Memory phase: Time period after pathogen clearance and when the effector cell population has contracted and the antigen-specific cell
numbers have stabilized.
Effector cells: The antigen-activated cells that expand during infection and then die during contraction of the response as pathogen is
cleared.
Memory cells: The stable populations of antigen-specific cells that persist after the effector cell population undergoes contraction.
Early effector (EE) cells: KLRG1lo CD127lo cells defined around the time of peak cellular accumulation in response to infection. EE cells
retain potential to give rise to terminal effector (TE), double-positive (DP), and memory precursor (MP) cells and ultimately memory T cells.
Terminal effector (TE) or short-lived effector cells: KLRG1hi CD127lo cells identified around the time of peak cellular accumulation in
response to infection. TE cells are prone to apoptosis during contraction and manifest very weak persistence into the memory phase and
weak secondary proliferative capacity upon re-stimulation.
Memory precursor (MP) effector cells: KLRG1lo CD127hi cells identified around the time of peak cellular accumulation in response to
infection. MP cells efficiently give rise to effector and central memory T cells (Tcm) and manifest strong capacity for persistence and
secondary proliferation upon re-stimulation.
Double-positive (DP) effector cells: KLRG1hi CD127hi cells defined around the time of peak cellular accumulation in response to
infection. Intermediate capacity to contribute to effector memory and Tcm.
Tcm cells: CD62Lhi CCR7hi CD44hi (also CD127hi and KLRG1lo and CD27hi and CX3CR1lo) cells defined after expanded T-cell numbers
following infection have contracted and stabilized. Mainly reside in secondary lymphoid organs, exhibit lower constitutive expression of
effector molecules, and manifest strongest proliferation upon re-stimulation.
Effector memory T (Tem) cells: CD62Llo CCR7lo CD44hi (also CD127hi and KLRG1lo/hi and CD27lo and CX3CR1hi) cells defined after
expanded T-cell numbers following infection have contracted and stabilized. Mainly reside in vasculature and intravascular spaces,
exhibit higher constitutive expression of effector molecules, and manifest less strong proliferation upon re-stimulation compared with Tcm
cells.
Peripheral memory T (Tpm) cells: CX3CR1int cells defined after expanded T-cell numbers following infection have contracted and
stabilized. Tpm cells are located in both intravascular spaces and recirculating through secondary lymphoid organs and exhibit strong
homeostatic renewal.
Tissue resident memory (Trm) cells: Operationally defined antigen-specific cells that enter non-lymphoid tissues during the effector
phase, that are non-vascular-associated, and that do not recirculate. Trm cells have variable phenotypes depending on their host tissues
but are frequently CD69+ and CD103+.
Long-lived effector (LLE) or effector-like memory (ELM) cells: LLE cells are KLRG1hi CD127hi/lo (and CD62Llo) and are mainly CD27lo
and CD43lo (defined as ELM with these markers), probably correspond to most CD27lo CX3CR1hi cells, and are most frequent at early
times of the memory phase. LLE/ELM cells have strong protective capacity and expression of effector molecules but weak capacity for
proliferation upon secondary antigen stimulation.
Page 4 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

a complete program of memory cell differentiation16,17. Indi-
vidual naïve T cells have the potential to differentiate into all
phenotypic effector cell subsets and ultimately memory CD8
T cells9,18,19. Aspects of this decision could be programmed
during the first naïve cell division, as antigen-presenting cell
contact establishes molecular asymmetry in nascent daughter
T cells which is associated with their ultimate fate20,21, and cells
that have undergone their first cell division exhibit distinct
single-cell mRNA expression profiles that can be correlated
with either gene expression signatures from mature KLRG1hi
IL-7Rα (CD127)lo terminal effector (TE) CD8 T cells at the
peak response, or from memory CD8 T cells22,23. However, the
gene expression profiles in single cells 4 days later are
neither strongly distinct between each other nor analogous to
the profiles observed after the initial cell division. The expres-
sion profiles in single cells on day 4 are also distinct from those
in mature TE and memory CD8 T cells23. However, the day
4 cells could be classified as putative pre-terminal and pre-
memory cells on the basis of their expression of “fate-
classifier” genes associated with mature memory or TE CD8
T cells23. Therefore, distinctly fated cells could be present at
early times. However, it is unclear whether the distinct gene
expression patterns in cells after the initial division derived
from the same or different naïve parents and whether the fate-
associated gene expression regimes in the single cells are reinforced
in their descendants or whether they convert.
The ancestry of CD8 T cells at the single-cell level indicates
that the overall pattern of TE and memory precursor (MP) CD8
T-cell differentiation is an average resulting from stochastic
behavior of cells recruited into the response (Figure 2). Stud-
ies applying DNA barcodes to follow CD8 T-cell families from
Figure 2. The descent of individual naïve CD8 T cells into effector and memory CD8 T-cell progeny on the basis of lineage tracing
and single-cell transfer studies. (A) Individual naïve CD8 T cells are recruited into the response and undergo geometric accumulation
resulting in distinct CD8 T-cell families (numbers) derived from individual naïve cells. (B) Each naïve cell has the potential to differentiate into
progeny that exhibit central memory T (Tcm) (blue), effector memory T (Tem) (purple), or terminal effector (TE) (red) CD8 T-cell phenotypes.
(C) Central memory precursors (light blue) are composed of diverse families that divide slowly, (D) some of which give rise to faster-dividing
Tem precursors (purple). (E) TE CD8 T cells comprise relatively few CD8 T-cell families that have accumulated dramatically and most die.
Page 5 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

individual naïve cells using next-generation DNA sequencing24,
or the transfer of individual congenically marked cells9,18,
concur that the differentiated fates of single cells are highly
variable19. The overall response comprises relatively few
clones that grow into very large CD8 T-cell families whose
individual members manifest a phenotype that is indicative of
shorter-lived TE CD8 T cells (Figure 2A–E), together with
many smaller CD8 T-cell families derived from a larger number
of initial clones that manifest an MP CD8 T-cell phenotype that
develop into most long-lived memory cells (Figure 2B–D). These
data are best fit into a model in which naïve cells differenti-
ate linearly into MP cells that proliferate slowly and serve as
precursors of more rapidly dividing Tem cells that finally give
rise to shorter-lived TE CD8 T cells18,19.
Activated naïve CD8 T cells acquire effector cell
attributes before diverging into subsets with distinct
potential to form memory cells
Very soon after naïve CD8 T cells become activated, they dif-
ferentiate into a population of nascent CTLs that express genes
which are indicative of multiple effector cell functions11,25, even
though only some of these cells terminally differentiate while
others give rise to memory CTLs14. Moreover, although cells
from early times after infection that express higher amounts
of KLRG1 produce fewer memory cells, both KLRG1hi and
KLRG1int subsets generate substantial memory cell numbers25. In
addition, gene expression in KLRG1hi cells at day 5 after
lymphocytic choriomeningitis virus (LCMV) infection is
substantially different than in canonical TE CD8 T cells on
day 8 after infection26,27, and gene expression profiles in single
activated CD8 T cells 4 days after Listeria infection are
distinct from those in single cells on day 1 after infection
as well as those in single cells at the peak response on day 7
and in the memory phase23. These results imply that, at early
times, gene expression in the nascent CTL population is not
fixed, despite having established the capacity for multiple effec-
tor functions, and that this gene program diverges as cells
become TE and MP subsets as defined by KLRG1 and CD127
expression near the peak response.
The flexibility in gene expression of nascent CTLs is consistent
with the stochastic nature of whether activated CD8 T cells will
terminally differentiate or become memory T cells and is also
born out of recent genetic experiments. An engineered reporter
mouse in which Cre-recombinase is expressed from the endog-
enous Klrg1 locus to activate constitutive expression of fluo-
rescent proteins and indelibly mark cells which have expressed
Klrg1 in their history demonstrates that a substantial fraction of
KLRG1lo cells are marked with the reporter prior to the abso-
lute peak effector response, indicating that they had previously
expressed Klrg1 and subsequently downregulated it28. These
“exKLRG1” cells also frequently derived from KLRG1hi CD127hi
double-positive (DP) effector cells at the peak response and
are found in all memory CD8 T-cell populations at later
times (Figure 1).
The strong memory potential of exKLRG1 cells is an indi-
cation that many, if not all, memory cells are the progeny of
nascent CTLs that manifest promiscuous gene expression regimes
before acquiring a more stably differentiated phenotype. This sug-
gests that unstable gene expression in nascent CTLs facilitates
differentiation along both memory and terminal differentiation
paths, which are reinforced in only some progeny stochastically,
a process that might be similar to multi-lineage gene expres-
sion in hematopoietic precursors which precedes and primes
lineage commitment of myeloid and monocyte subsets29.
TCR stimulation rapidly induces chromatin
remodeling in naïve cells which persists in
differentiated effector and memory T cells
Initial TCR stimulation induces widespread alterations in chro-
matin accessibility of cis-regulatory regions prior to the ini-
tial cell division of naïve CD4 and CD8 T cells27,30. Analysis of
enriched DNA motifs encoded within differentially accessible
regions has provided insight into the potential transcription
factors (TFs) that control the early programming of effector
and memory T cells. Sequences in regions that become acces-
sible during initial TCR stimulation in naïve cells, and that
are also accessible in mature memory CD8 T cells, most
frequently encode enriched motifs recognized by TFs in
the RUNX, ETS, bZIP, T-BOX, IRF, RHD, PRDM1, and
ZF-KLF families, which implies that these TFs might induce
transcriptional reprogramming of naïve CD8 T cells, and also
stabilize the differentiation of memory CD8 T cells27,31–33. Many
TFs that can bind these motifs have established functions for
driving the differentiation of both effector and memory CD8
T-cell subsets and have been reviewed in detail fairly recently,
but still many others have yet to be explored34–36.
The mechanism that reprograms the chromatin structure of
cis-regulatory regions and promotes effector and memory
CD4 and CD8 T-cell differentiation involves transient activa-
tion of TFs that are activated by TCR signals (that is, NFAT and
AP-1), which facilitates binding of constitutively expressed or
lineage-specific TFs, such as the ETS and RUNX family TFs,
and presumably others30,37,38. TCR stimulation drives transient
chromatin accessibility at “inducible” regions of accessibil-
ity in conjunction with adjacent “primed” regions that remain
accessible persistently after cessation of TCR signals in the
differentiated progeny30. Sequences within inducible regions
are strongly enriched with binding sites for NFAT (RHD fam-
ily) and AP-1 (bZIP family) TFs, whereas sequences within
primed regions are enriched with ETS and RUNX binding sites30.
This process results in ETS and RUNX family TFs and presum-
ably many others, gaining stable access to cis-acting regions
in immune activation–relevant genes27,38.
Chromatin remodeling of distal cis-regulatory
regions correlates with the stability of gene
expression in naïve and distinct effector and memory
CD8 T-cell subsets
Analysis of chromatin accessibility in purified naïve, effec-
tor, exhausted, reinvigorated, and memory CD8 T-cell subsets
indicates that an extensive accessible cis-regulatory landscape
develops during the differentiation of both TE and MP cells,
most of which is preserved in memory CD8 T cells31,32,39–42.
Page 6 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

Even though TE and MP CD8 T-cell populations have distinct
proclivities to form memory CD8 T cells, there is consider-
able similarity in the chromatin accessibility profiles between
both cell subsets. Consistent with the notion of a common early
path of differentiation, accessibility to many of the regions
from both effector cell subsets is established within the first 24
hours of TCR stimulation of naïve cells27,32. Moreover, some
regions that are accessible in memory CD8 T cells but not TE
cells are established by initial TCR stimulation, which indi-
cates that specific aspects of memory CTLs are induced prior to
extensive effector cell differentiation.
TE and MP CD8 T cells both manifest more accessible regions
than memory CD8 T cells, when one considers regions that
are also different than in naïve T cells, and most are located dis-
tal to gene transcription start sites (TSSs)31,32,42. However, con-
sistent with MP cells being more efficient precursors of memory
CD8 T cells than TE cells, their accessibility profile is biased
toward that found in memory CD8 T cells32. Nevertheless, the
differences in the numbers of accessible regions between MP
cells and memory CTLs indicate that both chromatin condensa-
tion and chromatin opening likely occur as effector cells convert
into mature memory CD8 T cells. Consistent with this, other
changes to chromatin structure, such as DNA methylation, are
acquired during the effector phase but are erased as MP CD8
T cells convert into memory CTLs15.
The most well-defined alterations to chromatin structure that dif-
fer between effector and memory CD8 T-cell subsets appear to
occur in distal intragenic regions. Distinct histone modification
profiles occur at TSSs compared with transcriptional enhanc-
ers and have been used to define cis-regulatory function and
transcriptional activity in ex vivo CD8 T-cell subsets. Chromatin
immunoprecipitation and sequencing (ChIP-seq) analyses of mul-
tiple histone modifications (H3K4me3, H3K4me1, H3K27me3,
and H3K27Ac) combined with algorithms trained to predict
enhancer regions based on these modifications have identi-
fied many distal intergenic regions that potentially comprise
enhancers in specific CD8 T-cell subsets42–50. The apparent dif-
ferential activity of these putative enhancers based on histone
modifications42,44–46 and three-dimensional interactions with their
target gene promoters44 positively correlates with gene expres-
sion signatures of naïve, TE, and memory CD8 T cells. Thus,
cis-regulatory regions, mainly in distal intergenic regions,
undergo dynamic alterations as naïve CD8 T cells become
activated and differentiate into distinct populations of effector and
ultimately memory CD8 T cells.
Promoter proximal regulation is also likely to be important for
the gene activity that defines the distinct differentiated states of
CD8 T-cell subsets. Although neither differential histone modi-
fications near TSSs44 nor the accessibility of promoter-proximal
regions in TE and memory CD8 T cells correlates with the
differential gene expression patterns between these subsets32,44,
a complete assessment of chromatin modifications that influ-
ence promoter activities has not been performed in CD8
T cells51, and additional analyses could reveal important differ-
ences. In line with this idea, the occupancy of RNA polymerase
II (Pol II) at the promoters of multiple effector genes differs in
naïve, effector, and memory CD8 T cells52, which suggests that
recruitment and activity of Pol II at target gene promoters are
associated with effector and memory CD8 T-cell differentia-
tion. In addition, both subunits of P-TEFb (positive transcription
elongation factor b) are essential for TE cell differentiation53.
P-TEFb is recruited to paused Pol II molecules at TSSs
and is necessary for inducing transcriptional elongation54.
Therefore, overcoming Pol II pausing might be a key step that
drives terminal differentiation, whereas ensuring Pol II paus-
ing could be a mechanism that ensures that MP CD8 T cells
form and perhaps the transcriptional “capacitance” of effec-
tor genes in memory CD8 T cells. Such promoter-proximal
regulation is likely conferred by the differential activity and
long-range interactions observed at distal cis-regulatory regions in
distinct CD8 T-cell subsets.
Chromatin structure and transcriptional regulation
that initializes effector CTL differentiation and
preserves memory CTL potential
Memory CTL differentiation involves de-activating gene expres-
sion programs of naïve cells and concurrently establishing gene
expression that accounts for effector functions, tissue relocali-
zation, persistence, and re-expansion after a secondary antigen
encounter. The enrichment of Runx motifs in accessible regions
that are induced during TCR stimulation and that persist in
memory CTLs suggests that they might contribute to this proc-
ess. Indeed, insufficiency in either Runx3 or Cbfb (the partner of
all three Runx TFs that is obligatory for DNA binding) impairs
the acquisition of key effector functions of CTLs27,55,56, and the
activated cells do not differentiate into genuine MP CTLs or
circulating memory CTLs and instead preferentially develop
a TE-like phenotype27. Moreover, Runx3-deficient cells do
not repress Tcf7 and Bcl6, which results in aberrant acquisi-
tion of a follicular T helper cell phenotype and trafficking into
B-cell follicles56. In addition, Runx3 deficiency impairs the
differentiation of Trm cells and their homeostasis in non-
lymphoid tissue (NLT)39; the transcriptional control of Trm
cell differentiation was recently reviewed in detail36. Runx2-
deficient T cells also exhibit defects in memory CTL generation
and long-term persistence57, which confirms an earlier compu-
tational prediction that Runx2 is critical for memory CD8 T-cell
development58. Thus, Runx family TFs drive programming of
effector attributes of nascent effector CTLs and also ensure
that these cells develop into memory CTLs.
Runx3 is required during TCR stimulation to establish chro-
matin accessibility of cis-regulatory regions that form stably
in effector and memory CD8 T cells27 and most likely depends
on cooperativity with many additional TFs. The cis-regulatory
regions that do become accessible in CD8 T cells lacking
Runx3 encode many fewer motifs for RUNX, IRF, bZIP, RHD,
PRDM1, and T-BOX motifs, suggesting that TFs which normally
bind these sites in Runx3-sufficient cells could be collaborat-
ing factors. Runx and T-box motifs frequently co-occur within
stably remodeled cis-regulatory regions of memory CD8
T cells, and binding regions for the two T-box TFs—T-bet and
Eomesodermin—each extensively overlap those of the obligate
Page 7 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

Runx TF partner Cbfb33. Together, these observations indicate
that cooperativity between Runx and T-box proteins is a core
regulatory mechanism that establishes the identity of effector
and memory CD8 T cells27,33,55,59, perhaps by outcompeting
nucleosomes that otherwise would form at these sites60.
Furthermore, the overlapping binding of Runx3, IRF4, and mul-
tiple bZIP family TFs suggests that potential cooperativity with
these TFs is also important27. Thus, complex cooperative inter-
actions between multiple TFs are likely to establish a chromatin
accessibility landscape during initial naïve CD8 T-cell stimulation
that induces effector CD8 T-cell subsets and is stabilized in
memory cells.
In addition, the cis-regulatory regions that are operational in Tcm
cells relative to TE cells encode multiple TF motifs that pre-
dict potential TFs that promote memory CTL differentiation42,44.
Binding motifs for Tcf1, Lef1, Foxo1, Foxp1, Eomes, Stat5,
Gabpa, Gfi1, and Nr3c1 (as well as others) are enriched in these
regions, suggesting that these TFs promote cis-regulatory activ-
ity that establishes and maintains memory CTL differentiation.
Most of these TFs have established roles for activating gene
expression that promotes T-cell quiescence, lymphoid homing,
homeostasis, and the potential for self-renewal61–65. At the
same time, memory CTL differentiation appears to involve
actively repressing the activity of some genes to prevent
terminal differentiation. RNA interference (RNAi)-mediated
suppression of the glucocorticoid receptor (Nr3c1) and its
canonical co-repressor encoded by the chromatin regulator
Ncor1 both impaired the differentiation of MP cells and memory
CTLs and increased terminal differentiation42.
A somewhat paradoxical feature of memory CTL cell differen-
tiation is that genes that promote memory CTL formation and
homeostasis are initially downregulated during activation of naïve
CD8 T cells, only to be re-expressed in some effector cells that
become memory CTLs. The entire population of effector cells
near the peak response to infection exhibits increased CpG DNA
methylation genome-wide, including at representative genes
such as Il7r, Sell (CD62L), and Tcf7, which correlates with their
reduced expression in most effector CD8 T cells at the peak
response15. A large fraction of CD62Llo MP CD8 T cells
upregulate Sell and undergo demethylation of its locus prior
to their initial homeostatic cell division, indicating that
CpG methylation is actively removed as MP cells from the
effector phase convert into memory CTLs. This process does not
occur at an appreciable rate in TE CD8 T cells, which remain
CD62Llo. CD8 T cells from mice in which the de novo DNA
methyltransferase (Dnmt3a) was deleted early during the effec-
tor response undergo more rapid re-expression of Il7r, Sell,
and Tcf7 genes near the peak response and during the contrac-
tion phase, which indicates that initiation of DNA methyla-
tion at early times correlates with gene silencing that enforces
terminal differentiation of some cells but that, in others, it
can be erased at later times15,66. CD8 T cells lacking the
maintenance DNA methyltransferase Dnmt1 also appear to
have reduced differentiation of effector cells, and although
memory CTLs appear to form, they exhibit defective recall
function67. Thus, DNA methylation appears to be important
for proper memory CTL differentiation, although the mecha-
nisms that account for why some cells are able to undergo
demethylation of key loci that promote memory CTL
development whereas others do not and progress toward terminal
differentiation are not yet clear. However, multiple chromatin
reader proteins that bind methylated DNA and recruit addi-
tional chromatin-modifying factors or enzymes that chemically
convert methylated cytosine residues appear to be involved.
CD8 T cells from mice lacking Mbd2, one of four genes
encoding methyl CpG-binding DNA proteins, exhibit skewing
toward terminal differentiation and defective differentia-
tion of memory CTLs that are not protective68. In addition,
CD8 T cells deficient in methylcytosine dioxygenase ten-
eleven translocation 2 (Tet2) exhibit DNA hypermethylation in
multiple transcriptional regulators and enhanced memory
CD8 T-cell differentiation69.
Molecular regulation that imposes terminal
differentiation on effector CD8 T cells
Terminal differentiation of activated CD8 T cells posi-
tively correlates with extensive proliferative history19. Even
though the outcome is probabilistic at the single-cell level, the
pattern of terminal differentiation of the population of cells
seems to be programmed by signals received very early during
activation18,24. Stimulation of T cells via their antigen recep-
tors and co-stimulatory molecules, together with inflamma-
tory cytokines (for example, interleukin-12 [IL-12] and IL-2),
integrates to form a calculus that determines the amount of cell
division in the resulting progeny70. Cells accumulating larger
sums of the integrated signals during priming extends their
proliferative capacity and likely predisposes them to ter-
minal differentiation14,71–74. The same signals that induce
extensive proliferation in the responding cell population also
prolong their responsiveness to these stimuli, which sustains
or increases the expression of TFs (such as T-bet, Zeb2, and
Blimp-1) that jointly promote terminal differentiation14,72,74–76.
A critical feedforward transcriptional circuit involving the TFs
T-bet and Zeb2 positively regulates terminal differentiation77,78.
T-bet binds to the Zeb2 locus and induces Zeb2 expression,
and both TFs appear to be necessary for optimal T-bet bind-
ing to cis-regulatory regions it controls; although (owing to the
lack of a reliable antibody) Zeb2 occupancy was not analyzed,
its putative binding motif was highly enriched within T-bet
occupied regions, and T-bet binding was compromised in
Zeb2-deficient CD8 T cells77. In addition, both factors are
expressed in LLE cells from the memory phase but are more
highly expressed in TE cells from the peak response, which sug-
gests that each TF has roles in both terminally differentiated
and memory CTLs5.
Consistent with this, memory CD8 T cells remain differenti-
ated from TE CD8 T cells in part by preventing high expres-
sion of T-bet and Zeb214,77. An antagonistic relationship between
the TFs Zeb1 and Zeb2 and the action of mir-200 family
microRNAs79 form an important regulatory circuit that deter-
mines the memory potential of effector T cells. Zeb1 is
necessary for memory CD8 T-cell differentiation and is induced
Page 8 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

in response to transforming growth factor-beta (TGF-β)
signals. Together with mir-200 family microRNAs, it represses
Zeb2 expression. Runx3 also prevents high expression of T-bet
and Zeb2 that normally occurs in TE cells27.
In addition, the bZIP family TF, Bach2, deactivates terminal
CTL differentiation by preventing TCR-induced AP-1–driven
signals by competing with Jun proteins for DNA occupancy
within cis-regulatory regions58,80 and is necessary for exKLRG1
cells to develop into memory CTLs28. Runx3 appears to
contribute to this process because Runx3-deficient CD8 T cells fail
to induce chromatin accessibility of cis-regulatory regions encod-
ing Bach2-binding motifs27. Therefore, negative feedback is
provided by TFs that initially drive acquisition of effector cell
attributes during CD8 T-cell activation which prevents terminal
differentiation.
Terminal CTL differentiation involves stable repression of
genes encoding stem cell–like qualities that normally pro-
mote the long-lived nature of Tcm cells64,81,82. Both T-bet and
Zeb2 repress features of memory CD8 T cells (for example, by
binding directly to the Il2 and Il7r genes and repressing their
expression). In addition, high Blimp-1 expression causes
repression of Id3, which retards the ability of effector cells
to contribute to the memory CTL compartment83,84. Also,
chromatin-level repression of genes that promote lymphoid hom-
ing and quiescence and other features of “stemness” that can be
considered “pro-memory” promotes commitment to terminal
differentiation15,26,48,85. Methylation of histone H3K9 and
H3K27 is a well-studied mechanism that promotes chroma-
tin condensation and gene silencing during cell development86.
Upon activation, naïve CD8 T cells rapidly accumulate islands
of histone H3K9 trimethylation (H3K9me3), especially at
genes such as Il7r and Sell85. H3K9me3 is deposited by multi-
ple methyltransferases, including the suppressor of variegation
3-9 homolog 1 (Suv39h1), and is a histone modification that
recruits multiple proteins in the chromobox (Cbx) family to
bind adjacent nucleosomes together, a process that reinforces
recruitment of additional Suv39h1 and promotes spreading
of H3K9me3 deposition87,88. In addition, Suv39h1 interacts
with Mbd family proteins, which suggests that DNA meth-
ylation could instigate or enhance Suv39h1 recruitment and
H3K9me3 deposition89. Suv39h1-deficient CD8 T cells fail
to repress naïve and stem cell–associated genes and exhibit a
loss in the inverse correlation between H3K9me3 density and
stem cell gene expression85. These cells accumulate poorly and
develop a normal TE CD8 T-cell phenotype inefficiently, and the
resulting memory cells are not protective85.
Similarly, repression of MP cell signature genes by enhanced
deposition of H3K27me3 in cis-regulatory regions of TE CD8
T cells also promotes terminal differentiation23,48. H3K27me3
deposition is catalyzed by the methyltransferase Ezh2, which
is upregulated upon stimulation of naïve CD8 T cells48, and
its mRNA is more highly expressed in a subset of responding
CD8 T cells classified as pre-terminal effector cells23. Disrup-
tion of Ezh2 impairs CD8 T-cell accumulation and effector cell
differentiation23,48. This phenotype correlates with reduced
H3K27me3 and enhanced expression of Eomes, Tcf7, and Klf2
genes, which encode TFs that promote competitive fitness of
Tcm cells, their maintenance, and lymphoid retention23,61,65,90.
Thus, coordinated targeting of histone methyltransferase activ-
ity in effector cells leads to methylation of H3K9 and H3K27
residues in nucleosomes of genes that are essential for memory
cell homeostasis, which represses their expression and may
ensure terminal differentiation.
Finally, the stability of phenotypes in CTL subsets, as in many
other developmental systems, is enforced by TFs that drive particu-
lar cell states by continuously directing the activity of chromatin
regulators to their appropriate gene targets91. In the earliest part
of the memory phase, LLEs that are KLRG1hi retain proper-
ties that endow them with additional effector capacities and per-
sistence at early times during the memory phase5 (Figure 1).
The phenotype of these cells depends on continued expression
of the proteins Id2 and Zeb26,78. Conditional disruption of Id2
in KLRG1hi cells after differentiation of LLE results in loss of
KLRG1 expression and in conversion of their transcriptional pro-
file into one reminiscent of that found in Tcm cells6. These results
demonstrate that the persistent activity of certain TFs is essen-
tial for maintaining the differentiated state of memory CTLs
after they have been generated. Thus, while these differentiation
programs depend on chromatin remodeling, they are maintained
by the continuous activity of specific TFs.
Toward a unified model of memory CD8 T-cell
differentiation
Several models have been proposed to conceptualize how naïve
CD8 T cells differentiate into memory CD8 T cells19,35,92. An
amenable solution that bears similarity to the decreasing poten-
tial and progressive differentiation models but that includes
insight from single-cell tracing studies and population analy-
ses of chromatin structure suggests that naïve CD8 T cells
rapidly acquire critical features of both effector and memory
CD8 T cells upon TCR activation and thus comprise effector
and memory precursor progenitor (EMPpro) cells (Figure 3).
Cells in the EMPpro population manifest metastable transcrip-
tional states characterized by promiscuous gene expression
among individual cells and stochastic proclivity for acceleration
or diversion into effector memory–like cells and further com-
mitment to extensive proliferation and terminal differentiation,
or reversion to a slowly dividing EMPpro state, relaxation into
MP cells, and ultimately differentiation into memory CD8
T cells. Between these extremes, some cells depart the spleen
and seed peripheral NLTs to form precursors of Trm CD8
T cells39. The probability that cells opt to proliferate extensively
and differentiate into TE CTLs is influenced by the integration
of signals that individual naïve cells experience during initial
activation. In addition, signals in the local microenviron-
ment as the nascent EMPpro families accumulate may sustain
or antagonize these signals in some cells93, and influence the
binding activity of specific TFs and alterations to chromatin
structure that drive the gene expression programs specific to
TE and MP cells, thus progressively reinforcing (or revers-
ing) the fates of individual cells that tend to diverge along these
pathways to memory. Therefore, even though individual T cells
Page 9 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

arrive at their fates randomly, the patterns of memory CTL
differentiation are influenced deterministically.
Abbreviations
CTL, cytotoxic T lymphocyte; EMPpro, progenitor effector
and memory precursor; H3K9me3, histone H3 lysine 9 tri-
methylation; H3K27me3, histone H3 lysine 27 tri-methylation;
IL, interleukin; LLE, long-lived effector; MP, memory precursor;
NLT, non-lymphoid tissue; Pol II, polymerase II; P-TEFb,
positive transcription elongation factor b; SLO, secondary lym-
phoid organ; Tcm, central memory; TCR, T-cell receptor; TE,
terminal effector; Tem, effector memory; TF, transcription factor;
Trm, tissue-resident memory; TSS, transcription start site.
Grant information
This work was supported by National Institutes of Health
grants R01 AI095634 and U19 AI109976 and US Depart-
ment of Defense grant W81XWH-16-1-0006 (to MEP) and the
Frenchman’s Creek Women for Cancer Research.
The funders had no role in study design, data collection and
analysis, decision to publish, or preparation of the manuscript.
Figure 3. An integrated model of memory CD8 T-cell differentiation. (A) Individual naïve CD8 T cells undergo memory CD8 T-cell
programming wherein they acquire fundamental traits of fully developed memory CD8 T cells prior to the first cell division. (B) The growing
nascent CD8 T-cell population comprises a transitional population of effector and memory precursor progenitor (EMPpro) cells that are
metastable at the chromatin and transcriptional level and can give rise to all subsets of differentiated CD8 T-cell progeny. (C) A small number
of EMPpro cells randomly undergo massive proliferation coupled to higher expression of multiple transcription factors (TFs) in response
to inflammatory signals that drive transcription underlying the phenotypic and functional profiles of terminally differentiated CD8 T cells.
(D) Multiple chromatin regulatory factors that methylate DNA and histones persistently repress genes that otherwise favor quiescence,
lymphoid retention, and overall “stemness” and thus enforce terminal differentiation. Factors generally associated with terminal CD8 T-cell
differentiation (red) and memory (blue) are highlighted by color but are not intended to imply exclusive correlations.
References
F1000 recommended
1. Kaech SM, Wherry EJ, Ahmed R: Effector and memory T-cell differentiation:
implications for vaccine development. Nat Rev Immunol. 2002; 2(4): 251–62.
PubMed Abstract
|
Publisher Full Text
2. Masopust D, Soerens AG: Tissue-Resident T Cells and Other Resident
Leukocytes. Annu Rev Immunol. 2019; 37: 521–46.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
3. Jameson SC, Masopust D: Understanding Subset Diversity in T Cell
Memory. Immunity. 2018; 48(2): 214–26.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
4. Mueller SN, Gebhardt T, Carbone FR, et al.: Memory T cell subsets, migration
patterns, and tissue residence. Annu Rev Immunol. 2013; 31: 137–61.
PubMed Abstract
|
Publisher Full Text
Page 10 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

5. Olson JA, McDonald-Hyman C, Jameson SC, et al.: Effector-like CD8+ T cells in
the memory population mediate potent protective immunity. Immunity. 2013;
38(6): 1250–60.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
6. Omilusik KD, Nadjsombati MS, Shaw LA, et al.: Sustained Id2 regulation of E
proteins is required for terminal differentiation of effector CD8+ T cells. J Exp
Med. 2018; 215(3): 773–83.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
7. Gerlach C, Moseman EA, Loughhead SM, et al.: The Chemokine Receptor
CX3CR1 Defines Three Antigen-Experienced CD8 T Cell Subsets with Distinct
Roles in Immune Surveillance and Homeostasis. Immunity. 2016; 45(6): 1270–84.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
8. Schenkel JM, Masopust D: Tissue-resident memory T cells. Immunity. 2014;
41(6): 886–97.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
9. Plumlee CR, Sheridan BS, Cicek BB, et al.: Environmental cues dictate the
fate of individual CD8+ T cells responding to infection. Immunity. 2013; 39(2):
347–56.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
10. Newell EW, Sigal N, Bendall SC, et al.: Cytometry by time-of-flight shows
combinatorial cytokine expression and virus-specific cell niches within a
continuum of CD8+ T cell phenotypes. Immunity. 2012; 36(1): 142–52.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
11. Bannard O, Kraman M, Fearon DT: Secondary replicative function of CD8+
T cells that had developed an effector phenotype. Science. 2009; 323(5913):
505–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
12. Harrington LE, Janowski KM, Oliver JR, et al.: Memory CD4 T cells emerge from
effector T-cell progenitors. Nature. 2008; 452(7185): 356–60.
PubMed Abstract
|
Publisher Full Text
13. Opferman JT, Ober BT, Ashton-Rickardt PG: Linear differentiation of cytotoxic
effectors into memory T lymphocytes. Science. 1999; 283(5408): 1745–8.
PubMed Abstract
|
Publisher Full Text
14. Joshi NS, Cui W, Chandele A, et al.: Inflammation directs memory precursor
and short-lived effector CD8+ T cell fates via the graded expression of T-bet
transcription factor. Immunity. 2007; 27(2): 281–95.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
15. Youngblood B, Hale JS, Kissick HT, et al.: Effector CD8 T cells dedifferentiate
into long-lived memory cells. Nature. 2017; 552(7685): 404–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
16. Kaech SM, Ahmed R: Memory CD8+ T cell differentiation: initial antigen
encounter triggers a developmental program in naïve cells. Nat Immunol. 2001;
2(5): 415–22.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
17. van Stipdonk MJ, Lemmens EE, Schoenberger SP: Naïve CTLs require a single
brief period of antigenic stimulation for clonal expansion and differentiation.
Nat Immunol. 2001; 2(5): 423–9.
PubMed Abstract
|
Publisher Full Text
18. Buchholz VR, Flossdorf M, Hensel I, et al.: Disparate individual fates compose
robust CD8+ T cell immunity. Science. 2013; 340(6132): 630–5.
PubMed Abstract
|
Publisher Full Text
19. Buchholz VR, Schumacher TN, Busch DH: T Cell Fate at the Single-Cell Level.
Annu Rev Immunol. 2016; 34: 65–92.
PubMed Abstract
|
Publisher Full Text
20. Chang JT, Ciocca ML, Kinjyo I, et al.: Asymmetric proteasome segregation
as a mechanism for unequal partitioning of the transcription factor T-bet
during T lymphocyte division. Immunity. 2011; 34(4): 492–504.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
21. Chang JT, Palanivel VR, Kinjyo I, et al.: Asymmetric T lymphocyte division
in the initiation of adaptive immune responses. Science. 2007; 315(5819):
1687–91.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
22. Arsenio J, Kakaradov B, Metz PJ, et al.: Early specification of CD8+ T lymphocyte
fates during adaptive immunity revealed by single-cell gene-expression
analyses. Nat Immunol. 2014; 15(4): 365–372.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
23. Kakaradov B, Arsenio J, Widjaja CE, et al.: Early transcriptional and
epigenetic regulation of CD8+ T cell differentiation revealed by single-cell RNA
sequencing. Nat Immunol. 2017; 18(4): 422–432.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
24. Gerlach C, Rohr JC, Perié L, et al.: Heterogeneous differentiation patterns of
individual CD8+ T cells. Science. 2013; 340(6132): 635–9.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
25. Sarkar S, Kalia V, Haining WN, et al.: Functional and genomic profiling
of effector CD8 T cell subsets with distinct memory fates. J Exp Med. 2008;
205(3): 625–40.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
26. Best JA, Blair DA, Knell J, et al.: Transcriptional insights into the CD8+ T cell
response to infection and memory T cell formation. Nat Immunol. 2013; 14(4):
404–12.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
27. Wang D, Diao H, Getzler AJ, et al.: The Transcription Factor Runx3 Establishes
Chromatin Accessibility of cis-Regulatory Landscapes that Drive Memory
Cytotoxic T Lymphocyte Formation. Immunity. 2018; 48(4): 659–74.e6.
PubMed Abstract
|
Publisher Full Text
28. Herndler-Brandstetter D, Ishigame H, Shinnakasu R, et al.: KLRG1+ Effector
CD8+ T Cells Lose KLRG1, Differentiate into All Memory T Cell Lineages, and
Convey Enhanced Protective Immunity. Immunity. 2018; 48(4): 716–29.e8.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
29. Laslo P, Spooner CJ, Warmflash A, et al.: Multilineage transcriptional
priming and determination of alternate hematopoietic cell fates. Cell. 2006;
126(4): 755–66.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
30. Bevington SL, Cauchy P, Piper J, et al.: Inducible chromatin priming is
associated with the establishment of immunological memory in T cells. EMBO
J. 2016; 35(5): 515–35.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
31. Scharer CD, Bally AP, Gandham B, et al.: Cutting Edge: Chromatin
Accessibility Programs CD8 T Cell Memory. J Immunol. 2017; 198(6): 2238–2243.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
32. Scott-Browne JP, Lopez-Moyadó IF, Trifari S, et al.: Dynamic Changes in
Chromatin Accessibility Occur in CD8+ T Cells Responding to Viral Infection.
Immunity. 2016; 45(6): 1327–40.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
33. van der Veeken J, Zhong Y, Sharma R, et al.: Natural Genetic Variation
Reveals Key Features of Epigenetic and Transcriptional Memory in Virus-
Specific CD8 T Cells. Immunity. 2019; 50(5): 1202–1217.e7.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
34. Chang JT, Wherry EJ, Goldrath AW: Molecular regulation of effector and
memory T cell differentiation. Nat Immunol. 2014; 15(12): 1104–15.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
35. Kaech SM, Cui W: Transcriptional control of effector and memory CD8+ T cell
differentiation. Nat Rev Immunol. 2012; 12(11): 749–61.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
36. Milner JJ, Goldrath AW: Transcriptional programming of tissue-resident
memory CD8+ T cells. Curr Opin Immunol. 2018; 51: 162–169.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
37. Bevington SL, Cauchy P, Cockerill PN: Chromatin priming elements
establish immunological memory in T cells without activating transcription:
T cell memory is maintained by DNA elements which stably prime inducible
genes without activating steady state transcription. BioEssays. 2017; 39(2).
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
38. Bevington SL, Cauchy P, Withers DR, et al.: T Cell Receptor and Cytokine
Signaling Can Function at Different Stages to Establish and Maintain
Transcriptional Memory and Enable T Helper Cell Differentiation. Front
Immunol. 2017; 8: 204.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
39. Milner JJ, Toma C, Yu B, et al.: Runx3 programs CD8+ T cell residency in non-
lymphoid tissues and tumours. Nature. 2017; 552(7684): 253–257.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
40. Pauken KE, Sammons MA, Odorizzi PM, et al.: Epigenetic stability of
exhausted T cells limits durability of reinvigoration by PD-1 blockade. Science.
2016; 354(6316): 1160–5.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
41. Sen DR, Kaminski J, Barnitz RA, et al.: The epigenetic landscape of T cell
exhaustion. Science. 2016; 354(6316): 1165–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
42. Yu B, Zhang K, Milner JJ, et al.: Epigenetic landscapes reveal transcription
factors that regulate CD8+ T cell differentiation. Nat Immunol. 2017; 18(5):
573–82.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
43. Calo E, Wysocka J: Modification of enhancer chromatin: what, how, and why?
Mol Cell. 2013; 49(5): 825–37.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
44. He B, Xing S, Chen C, et al.: CD8+ T Cells Utilize Highly Dynamic Enhancer
Repertoires and Regulatory Circuitry in Response to Infections. Immunity.
2016; 45(6): 1341–54.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
45. Russ BE, Olshanksy M, Smallwood HS, et al.: Distinct Epigenetic Signatures
Delineate Transcriptional Programs during Virus-Specific CD8+ T Cell
Differentiation. Immunity. 2014; 41(5): 853–65.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
46. Russ BE, Olshansky M, Li J, et al.: Regulation of H3K4me3 at Transcriptional
Enhancers Characterizes Acquisition of Virus-Specific CD8+ T Cell-Lineage-
Specific Function. Cell Rep. 2017; 21(12): 3624–36.
PubMed Abstract
|
Publisher Full Text
47. Shen Y, Yue F, McCleary DF, et al.: A map of the cis-regulatory sequences
in the mouse genome. Nature. 2012; 488(7409): 116–20.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
Page 11 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

48. Gray SM, Amezquita RA, Guan T, et al.: Polycomb Repressive Complex
2-Mediated Chromatin Repression Guides Effector CD8+ T Cell Terminal
Differentiation and Loss of Multipotency. Immunity. 2017; 46(4): 596–608.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
49. Firpi HA, Ucar D, Tan K: Discover regulatory DNA elements using chromatin
signatures and artificial neural network. Bioinformatics. 2010; 26(13): 1579–86.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
50. Rajagopal N, Xie W, Li Y, et al.: RFECS: a random-forest based algorithm for
enhancer identification from chromatin state. PLoS Comput Biol. 2013; 9(3):
e1002968.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
51. Kouzarides T: Chromatin modifications and their function. Cell. 2007; 128(4):
693–705.
PubMed Abstract
|
Publisher Full Text
52. Zediak VP, Johnnidis JB, Wherry EJ, et al.: Cutting edge: persistently open
chromatin at effector gene loci in resting memory CD8+ T cells independent of
transcriptional status. J Immunol. 2011; 186(5): 2705–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
53. Chen R, Bélanger S, Frederick MA, et al.: In vivo RNA interference screens
identify regulators of antiviral CD4+ and CD8+ T cell differentiation. Immunity.
2014; 41(2): 325–38.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
54. Adelman K, Lis JT: Promoter-proximal pausing of RNA polymerase II: emerging
roles in metazoans. Nat Rev Genet. 2012; 13(10): 720–31.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
55. Cruz-Guilloty F, Pipkin ME, Djuretic IM, et al.: Runx3 and T-box proteins
cooperate to establish the transcriptional program of effector CTLs. J Exp
Med. 2009; 206(1): 51–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
56. Shan Q, Zeng Z, Xing S, et al.: The transcription factor Runx3 guards
cytotoxic CD8+ effector T cells against deviation towards follicular helper T cell
lineage. Nat Immunol. 2017; 18(8): 931–939.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
57. Olesin E, Nayar R, Saikumar-Lakshmi P, et al.: The Transcription Factor
Runx2 Is Required for Long-Term Persistence of Antiviral CD8+ Memory T
Cells. Immunohorizons. 2018; 2(7): 251–261.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
58. Hu G, Chen J: A genome-wide regulatory network identifies key transcription
factors for memory CD8+ T-cell development. Nat Commun. 2013; 4: 2830.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
59. Intlekofer AM, Takemoto N, Wherry EJ, et al.: Effector and memory CD8+ T cell
fate coupled by T-bet and eomesodermin. Nat Immunol. 2005; 6(12): 1236–44.
PubMed Abstract
|
Publisher Full Text
60. Li G, Levitus M, Bustamante C, et al.: Rapid spontaneous accessibility of
nucleosomal DNA. Nat Struct Mol Biol. 2005; 12(1): 46–53.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
61. Banerjee A, Gordon SM, Intlekofer AM, et al.: Cutting edge: The transcription
factor eomesodermin enables CD8+ T cells to compete for the memory cell
niche. J Immunol. 2010; 185(9): 4988–92.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
62. Chandele A, Joshi NS, Zhu J, et al.: Formation of IL-7Rαhigh and IL-7Rαlow CD8 T
Cells during Infection Is Regulated by the Opposing Functions of GABPα and
Gfi-1. J Immunol. 2008; 180(8): 5309–19.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
63. Kerdiles YM, Beisner DR, Tinoco R, et al.: Foxo1 links homing and survival
of naive T cells by regulating L-selectin, CCR7 and interleukin 7 receptor. Nat
Immunol. 2009; 10(2): 176–84.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
64. Lin WW, Nish SA, Yen B, et al.: CD8+ T Lymphocyte Self-Renewal during
Effector Cell Determination. Cell Rep. 2016; 17(7): 1773–82.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
65. Zhou X, Yu S, Zhao DM, et al.: Differentiation and Persistence of Memory
CD8+ T Cells Depend on T Cell Factor 1. Immunity. 2010; 33(2): 229–40.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
66. Ladle BH, Li KP, Phillips MJ, et al.: De novo DNA methylation by DNA
methyltransferase 3a controls early effector CD8+ T-cell fate decisions
following activation. Proc Natl Acad Sci U S A. 2016; 113(38): 10631–6.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
67. Chappell C, Beard C, Altman J, et al.: DNA Methylation by DNA
Methyltransferase 1 Is Critical for Effector CD8 T Cell Expansion. J Immunol.
2006; 176(8): 4562–72.
PubMed Abstract
|
Publisher Full Text
68. Kersh EN: Impaired memory CD8 T cell development in the absence of methyl-
CpG-binding domain protein 2. J Immunol. 2006; 177(6): 3821–6.
PubMed Abstract
|
Publisher Full Text
69. Carty SA, Gohil M, Banks LB, et al.: The Loss of TET2 Promotes CD8+ T Cell
Memory Differentiation J Immunol. 2018; 200(1): 82–91.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
70. Marchingo JM, Kan A, Sutherland RM, et al.: T cell signaling. Antigen affinity,
costimulation, and cytokine inputs sum linearly to amplify T cell expansion.
Science. 2014; 346(6213): 1123–7.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
71. Curtsinger JM, Valenzuela JO, Agarwal P, et al.: Type I IFNs provide a third signal
to CD8 T cells to stimulate clonal expansion and differentiation. J Immunol.
2005; 174(8): 4465–9.
PubMed Abstract
|
Publisher Full Text
72. Pipkin ME, Sacks JA, Cruz-Guilloty F, et al.: Interleukin-2 and inflammation
induce distinct transcriptional programs that promote the differentiation of
effector cytolytic T cells. Immunity. 2010; 32(1): 79–90.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
73. Starbeck-Miller GR, Xue HH, Harty JT: IL-12 and type I interferon prolong the
division of activated CD8 T cells by maintaining high-affinity IL-2 signaling
in vivo. J Exp Med. 2014; 211(1): 105–20.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
74. Kalia V, Sarkar S, Subramaniam S, et al.: Prolonged interleukin-
2Ralpha expression on virus-specific CD8+ T cells favors terminal-effector
differentiation in vivo. Immunity. 2010; 32(1): 91–103.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
75. Rutishauser RL, Martins GA, Kalachikov S, et al.: Transcriptional repressor
Blimp-1 promotes CD8+ T cell terminal differentiation and represses the
acquisition of central memory T cell properties. Immunity. 2009; 31(2): 296–308.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
76. Xin A, Masson F, Liao Y, et al.: A molecular threshold for effector CD8+ T cell
differentiation controlled by transcription factors Blimp-1 and T-bet. Nat
Immunol. 2016; 17(4): 422–32.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
77. Dominguez CX, Amezquita RA, Guan T, et al.: The transcription factors ZEB2
and T-bet cooperate to program cytotoxic T cell terminal differentiation in
response to LCMV viral infection. J Exp Med. 2015; 212(12): 2041–56.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
78. Omilusik KD, Best JA, Yu B, et al.: Transcriptional repressor ZEB2 promotes
terminal differentiation of CD8+ effector and memory T cell populations during
infection. J Exp Med. 2015; 212(12): 2027–39.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
79. Guan T, Dominguez CX, Amezquita RA, et al.: ZEB1, ZEB2, and the miR-200
family form a counterregulatory network to regulate CD8+ T cell fates. J Exp
Med. 2018; 215(4): 1153–1168.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
80. Roychoudhuri R, Clever D, Li P, et al.: BACH2 regulates CD8+ T cell
differentiation by controlling access of AP-1 factors to enhancers. Nat
Immunol. 2016; 17(7): 851–860.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
81. Gattinoni L, Zhong XS, Palmer DC, et al.: Wnt signaling arrests effector T cell
differentiation and generates CD8+ memory stem cells. Nat Med. 2009; 15(7):
808–13.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
82. Graef P, Buchholz VR, Stemberger C, et al.: Serial transfer of single-cell-
derived immunocompetence reveals stemness of CD8+ central memory T cells.
Immunity. 2014; 41(1): 116–26.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
83. Ji Y, Pos Z, Rao M, et al.: Repression of the DNA-binding inhibitor Id3 by Blimp-1
limits the formation of memory CD8+ T cells. Nat Immunol. 2011; 12(12): 1230–7.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
84. Yang CY, Best JA, Knell J, et al.: The transcriptional regulators Id2 and Id3
control the formation of distinct memory CD8+ T cell subsets. Nat Immunol.
2011; 12(12): 1221–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
|
F1000 Recommendation
85. Pace L, Goudot C, Zueva E, et al.: The epigenetic control of stemness in
CD8+ T cell fate commitment. Science. 2018; 359(6372): 177–86.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
86. Blackledge NP, Rose NR, Klose RJ: Targeting Polycomb systems to regulate
gene expression: modifications to a complex story. Nat Rev Mol Cell Biol. 2015;
16(11): 643–9.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
87. Bannister AJ, Zegerman P, Partridge JF, et al.: Selective recognition of
methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature. 2001;
410(6824): 120–4.
PubMed Abstract
|
Publisher Full Text
88. Lachner M, O'Carroll D, Rea S, et al.: Methylation of histone H3 lysine
9 creates a binding site for HP1 proteins. Nature. 2001; 410(6824): 116–20.
PubMed Abstract
|
Publisher Full Text
|
F1000 Recommendation
89. Rose NR, Klose RJ: Understanding the relationship between DNA methylation
and histone lysine methylation. Biochim Biophys Acta. 2014; 1839(12): 1362–72.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
90. Schober SL, Kuo CT, Schluns KS, et al.: Expression of the transcription factor
lung Krüppel-like factor is regulated by cytokines and correlates with survival
Page 12 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019

of memory T cells in vitro and in vivo. J Immunol. 1999; 163(7): 3662–7.
PubMed Abstract
91. Ptashne M: Epigenetics: core misconcept. Proc Natl Acad Sci U S A. 2013;
110(18): 7101–3.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
92. Kaech SM, Wherry EJ: Heterogeneity and cell-fate decisions in effector and memory
CD8+ T cell differentiation during viral infection. Immunity. 2007; 27(3): 393–405.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
93. Seo YJ, Jothikumar P, Suthar MS, et al.: Local Cellular and Cytokine Cues in the
Spleen Regulate In Situ T Cell Receptor Affinity, Function, and Fate of CD8+ T
Cells. Immunity. 2016; 45(5): 988–98.
PubMed Abstract
|
Publisher Full Text
|
Free Full Text
Page 13 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019


Open Peer Review
Current Peer Review Status:
Editorial Note on the Review Process
arewrittenbymembersoftheprestigious .TheyarecommissionedandF1000FacultyReviews F1000Faculty
arepeerreviewedbeforepublicationtoensurethatthefinal,publishedversioniscomprehensiveandaccessible.
Thereviewerswhoapprovedthefinalversionarelistedwiththeirnamesandaffiliations.
The reviewers who approved this article are:
Version 1
ThebenefitsofpublishingwithF1000Research:
Yourarticleispublishedwithindays,withnoeditorialbias
Youcanpublishtraditionalarticles,null/negativeresults,casereports,datanotesandmore
Thepeerreviewprocessistransparentandcollaborative
YourarticleisindexedinPubMedafterpassingpeerreview
Dedicatedcustomersupportateverystage
Forpre-submissionenquiries,contact research@f1000.com
Peter N Cockerill
InstituteofImmunologyandImmunotherapy,InstituteofBiomedicalResearch,UniversityofBirmingham,
Birmingham,UK
Nocompetinginterestsweredisclosed.Competing Interests:
1
Axel Kallies
DepartmentofMicrobiologyandImmunology,TheUniversityofMelbourne,Parkville,Australia
Nocompetinginterestsweredisclosed.Competing Interests:
2
Page 14 of 14
F1000Research 2019, 8(F1000 Faculty Rev):1278 Last updated: 31 JUL 2019