Measuring the Talent Management Function – Human Resources Assignment


Assignment: Measuring the Talent Management Function

Many organizations that use metrics to measure the success of their talent management function continuously outperform their competition (Pace, 2010). Nevertheless, many organizations still do not have metrics in place for measuring the initiatives’ success (Pace, 2010). Metrics give HR professionals and stakeholders the confidence in their talent management function by depicting the overall performance and enabling them to see when changes are necessary.

Suppose you have been asked by an organization what the best metrics are for determining the success of a talent management function.

With this in mind, respond to the following in a paper:

  • Identify five metrics that you would include in measuring the success of an organization’s talent management function.
  • Analyze each metric to determine its application and relevance for success.
  • Would you assign an equal weight to the metrics you chose, or would you assign different weights? Why? If you would assign different weights, how would you weight them and why?
  • APA Format.


– Galagan, P. (2007). Trend watch.

T+D, 61

(2), 1. Retrieved from the Walden library databases.

This article discusses perspectives on the growing trend among business enterprises to outsource many administrative human resources functions; such as, payroll and benefits administration. Organizations are shying away from outsourcing strategic functions that affect strategy-talent management, employee engagement, and training.

– Gordon, E. E. (2010). Talent challenge: Renewing the vision.

T+D, 64

(6), 42-47. Retrieved from the Walden library databases.

This article explores the roles professionals might fill in workplace learning and development today and over the next decade, to mitigate the current jobs skill crisis scenario. It also discusses the economic conditions that have shouldered much of the blame for the job market’s woes and the major problems with the quality of the U.S. and global workforces.

– Oshima, M., Kao, T., & Tower, J. (2005).

Achieving post-outsourcing success. Human Resource Planning, 28

(2), 7-11.

The article focuses on the new HR functions in the post-outsourcing world. It also identifies the need to understand that outsourcing success means having the courage to transform the makeup, organizational design, and strategic priorities of HR.

Editor’s Note
Trend Watch
For years this magazine has stepped
lightly over most HR topics in favor of
subjects deemed closer to the heart of
readers’ work: training, development,
talent management, and other efforts
to increase people’s ability to work
But in our interview for this issue with Mike Mussallem,
CEO of Edwards Lifesciences, we encountered one more instance of a trend we see brewing: companies are outsourcing
many of their administrative HR functions such as payroll and
benefits administration, but keeping the work that makes a difference to strategy—talent management, employee engagement, and training and development, for example. And in Mussallem’s case,
he specifically sought out an HR executive who would focus on growing talent. As he put it, “I want
someone with a development mindset as the head of HR.”
I wouldn’t take one CEO’s preference as evidence of a trend if it were not for several recent studies
of HR outsourcing among lots of big companies. They all say pretty much the same thing—the focus
of HR departments is shifting toward work that supports strategies and builds leadership, and the rest
is being outsourced. For the hard facts, see the Intelligence article on page 10. And for more on Mike
Mussallem’s philosophy of putting top talent in critical jobs, see pages 30 and 38.
These facts about shifting priorities also support another trend we have our eyes on: the steady
upward creep of responsibility for employee development. Companies such as The Gap and WalMart have created executive-level positions for managing talent. They see a skills gap coming as
older workers retire and the numbers of capable younger workers dwindle, and they understand that
growing talent drives performance. Their response has been to move the work of finding, attracting,
developing, and motivating employees to the level of strategy.
These are welcome developments for training professionals who have looked longingly at the seats
of power in organizations and failed to attain them. Could this signal the end of a reason to grouse
about not having a seat at the table?
Pat Galagan
Executive Editor, ASTD
| T+D | February 2007
Douglas T. Fearon
James M. Carr
Aino Telaranta
Marlene J. Carrasco
James E. D. Thaventhiran
The rationale for the IL-2independent generation of the
self-renewing central memory
CD8+ T cells
Authors’ address
Summary: Clones of CD8+ T cells that have been selected in the primary
Douglas T. Fearon, James M. Carr, Aino Telaranta,
Marlene J. Carrasco, James E. D. Thaventhiran
response must have a mechanism by which they can continuously or
intermittently generate new effector cells. Several years ago, this mechanism was proposed to involve a self-renewing, stem cell-like subset that
could avoid the differentiating effects of interleukin-2 (IL-2). The model
considered the stem cell subset to be contained within the central
memory population of CD8+ T cells (TCM). This proposal was inconsistent with subsequent findings suggesting that all antigen-activated
CD8+ T cells differentiated to effector cells (TEFF) during the primary
response and that TCM developed during the memory phase by dedifferentiating from effector memory cells (TEM). However, findings
have since been reported that support the stem cell model. First, studies
indicate that TEM do not serve as the precursors of TCM. Second, transcriptional repressors of IL-2 signaling do enhance the memory response.
Third, memory cells lacking effector functions and with a capacity to
replicate in a secondary response develop in the absence of signaling
through the IL-2/IL-15 receptor. Taken together, these findings suggest that antigen-activated CD8+ T cells with a stem cell-like capability
for maintaining proliferative potential develop by an unknown IL-2independent process. The challenge is now to identify this unknown
pathway of clonal expansion.
Department of Medicine, School of Clinical
Medicine, University of Cambridge,
Cambridge, UK.
Correspondence to:
Douglas T. Fearon
Department of Medicine
School of Clinical Medicine
University of Cambridge
Hills Road
Cambridge CB2 2QH
Tel.: +44 1223 330528
Fax: +44 1223 336815
These studies were supported by grants from the
Wellcome Trust and the National Institutes of Health.
Immunological Reviews 2006
Vol. 211: 104–118
Printed in Singapore. All rights reserved
ß 2006 The Authors
Journal compilation ß 2006 Blackwell Munksgaard
Immunological Reviews
The biological purpose of the antigen-dependent phase of
T-cell development is to generate numbers of cells with
differentiated effector functions that are sufficient to resolve
or control an infection. This function is especially challenging.
Within the pool of naı̈ve T cells, the frequency of clones that
are specific for individual microbial antigens is low, and, at
least for CD8+ T cells, the number of TEFF that are needed to
control infections is high, so that each clone must generate
perhaps as many as 105 TEFF in 7–10 days. Moreover, if the
primary response controls but does not eliminate the
infectious agent, continuous, uninterrupted production of
TEFF may be required. Even if the host eliminates the
Fearon et al Generation of the self-renewing CD8+ TCM
infectious agent during the primary response, it may persist
in the environment and possibly cause a secondary infection,
requiring the production of new TEFF. Therefore, CD8+ T-cell
clones that have been selected in the primary response must
have the potential for generating TEFF continuously or
intermittently for the individual’s lifetime. To do so, antigen-selected CD8+ T cells should have a self-renewing,
immortal, stem cell-like stage of development. A stem cell
stage during antigen-triggered CD8+ T-cell development was
proposed in 2001 (1), and here we summarize experimental
findings bearing on this proposal that have since been
The most appropriate experimental system in which to
study the possible role of self-renewal in the continuous
generation of CD8+ TEFF would be a persistent viral infection.
Although there have been studies of the CD8+ T-cell response
in persistent viral infections in the mouse, these have either
emphasized the ability of the infection to overwhelm the
CD8+ T-cell response, as occurs with clone 13 of lymphocytic
choriomeningitis virus (LCMV) (2) that impairs the antigenpresenting functions of dendritic cells (3), or have shown
continued expansion of epitope-specific CD8+ T cells, as in
murine cytomegalovirus (mCMV) infection (4), but have not
yet evaluated the developmental basis of the continuing CD8+
T-cell response. Instead, the question of how antigen-stimulated CD8+ T cells maintain their proliferative potential has
been studied mainly in relation to the development of
memory CD8+ T cells that mediate the replicative component
of a secondary response. Therefore, review of work that
pertains to a self-renewing stage of antigen-stimulated CD8+
T-cell differentiation must be done in the context of intermittent rather than a continuous stimuli for clonal expansion,
and it must be assumed that the memory CD8+ T-cell that
generates the burst of new effector cells in a secondary
response is closely related to the proposed stem cell that
maintains the production of new TEFF during persistent viral
infections. This emphasis on the proliferation of memory
CD8+ T cells also accounts for the absence in this review of
a discussion of the role of memory CD8+ T cells that have
immediate effector function to provide early defense against
secondary infections.
Memory CD8+ T cells are comprised of central memory
and effector memory subsets
In 1999, Lanzavecchia and colleagues (5) discovered that
assessing human memory CD4+ and CD8+ T cells for their
expression of the chemokine receptor CCR7 revealed the
presence of two functionally distinct subsets: CCR7+ memory
T cells that tended to resemble naı̈ve T cells in being CCR7+
and in lacking effector functions, and CCR7– memory T cells
that were capable of immediate effector activities. They
named the CCR7+ memory T cells TCM, to acknowledge
their potential for homing to secondary lymphoid organs,
and termed the CCR7– memory T cells TEM, because of their
relatively differentiated status. TCM were considered to be
responsible for the generation of new TEFF during secondary
responses, and TEM to mediate rapid host defense while new
TEFF were being produced. The correspondence between a
memory cell’s potential for homing to lymphoid or nonlymphoid tissues, the absence or presence of potential effector
function, and a cellular division of labor with respect to the
two tasks of a secondary response, proliferation and immediate effector function, was so intuitively appropriate that
immunologists quickly adopted this nomenclature.
The occurrence of the TCM and TEM subsets of memory cells
now underpins most analyses of the function and development of antigen-stimulated T cells. TCM not only express CCR7
but also high levels of CD62L, enabling them to recirculate
between the blood and secondary lymphoid organs, while
TEM, by definition being CCR7– and CD62Llow, tend to be
found primarily in peripheral tissues and the non-lymphoid
zones of the spleen (6–8). Lymphoid tissue is adapted
for supporting antigen-dependent, secondary proliferative
responses of memory T cells leading to the generation of
new TEFF, while peripheral, non-lymphoid tissue does not
usually have this capability and represents potential areas of
infection, where it would be appropriate for memory T cells
to exercise immediate effector function. Although mice deficient in lymphotoxin-a and lacking organized secondary lymphoid tissue have a primary response to pulmonary influenza
infection (9), this does not necessarily indicate that a lymphoid architecture is not required for antigen-induced CD8+
T-cell expansion, as the response is both delayed and diminished in these mice, and it may reflect the function of bronchoalveolar lymphoid tissue (10). The proposed differing
capabilities for replication in the TCM and TEM CD8+ T-cell
subsets also acknowledges the possibility that development of
effector functions might be associated with loss of replicative
function, as usually occurs with effector differentiation in
other cellular lineages. If so, then the possibility that TEM
and TEFF might be depleted in the early phases in a secondary
infection provides the rationale for the occurrence of the TCM
subset, which retain the proliferative capability of the naı̈ve
CD8+ T cells. Therefore, the proposal that memory T
cells could be divided into these two subsets was insightful
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
and constructive, in that it defined new experimental questions for understanding the function and development of
memory CD8+ T cells. These questions included the following. Is it correct that TCM accounts for the replicative component of the secondary response? What are the characteristics
of TCM that allow this subset to retain replicative function?
What are the developmental relationships between naı̈ve
CD8+ T cells, TCM, TEM, and TEFF?
Models for the development of central and effector
memory CD8+ T cells
Three general models have been proposed for the developmental pathways leading from naı̈ve CD8+ T cells to TCM, TEM,
and TEFF (Fig. 1). These models are presented in this section,
and the experimental evidence for them is in the next
section. Lanzavecchia and Sallusto (11) proposed in 2000
Fig. 1. Three models for the development of central memory and
effector memory CD8+ T cells during the primary immune response.
Shown are the progressive-differentiation model (11), the stem cellassociated differentiation model (1), and the linear differentiation model
(13). In the stem cell-associated model of differentiation, X refers to the
unknown agonist that causes IL-2/IL-15-independent proliferation
leading to the development of the precursor central memory cell.
Immunological Reviews 211/2006
the ‘progressive-differentiation model’, which puts forward
the concept that naı̈ve T cells progress through ‘hierarchical
thresholds’ for proliferation and differentiation as the strength
and duration of interaction with dendritic cells and cytokines
is increased. T cells receiving the weakest signals are ‘unfit’
and do not survive. The next threshold of signaling generates
the precursors of TCM that are fit and survive but have not
acquired effector functions. Even more signaling leads to
acquisition of effector functions, but not loss of essential
long-term survival characteristics, such as expression of
CD127 [interleukin-7 receptor-a (IL-7Ra)]. The highest
level of signaling causes development of terminally differentiated TEFF that cannot survive into the memory phase. These
different levels of signaling are a consequence of stochastic interactions between T cells and dendritic cells, which may reflect
not only the level of activation of the antigen-presenting
dendritic cells but also competition among proliferating T
cells for interaction with dendritic cells. After clearance of
antigen, the precursors of TCM, being the least differentiated
of the antigen-experienced memory cells and maintaining the
proliferative potential and homing receptors of naı̈ve cells,
become resting TCM and populate secondary lymphoid tissues,
where they are poised to mount secondary proliferative
responses. TEM, having progressed to an effector stage of
development and having changed their chemokine receptors
because they had received stronger signals during the primary
response than did TCM, populate peripheral non-lymphoid
tissues. The most stimulated TEFF have lost their capacity to
respond to survival and homeostatic cytokines and do not
persist into the memory phase.
The critical elements of this model are that there is a
hierarchy of irreversible differentiation, proceeding from the
naı̈ve T cell ! TCM ! TEM ! TEFF. The TCM stage, being the
least differentiated of the antigen-stimulated T cells, retains
essentially all of the developmental options of a naı̈ve cell,
including its capacity for marked clonal expansion. Also, the
developmental process is governed by the relative availability
of a common set of signals and not by specific signals for each
stage, and for this reason, it is appropriately characterized by
its proponents as a stochastic developmental process.
In 2001, a second proposal was considered that might be
termed the ‘stem cell model’ (1). It resembles the progressive
differentiation model in believing the differentiation process
to be irreversible, but it differs in that the TCM subset is
considered to be a self-renewing, stem cell-like stage of development that enables it to serve as an inexhaustible source of
TEFF for both chronic and intermittent infections. In the sense
that this subset is proposed also to provide the precursors of
Fearon et al Generation of the self-renewing CD8+ TCM
effector cells in primary and persistent infections, not only for
secondary responses, the TCM designation is not quite appropriate. In persistent infections, these replicating T cells (preTCM?) do not require one of the cardinal properties of a
memory cell, which is long-term survival in the absence of
antigen. However, the known ability of stem cells in other
organ systems to adjust the rate at which they replicate according to the relative need for cells with differentiated functions
encompasses both continuous and intermittent generation of
effector cells by TCM. A self-renewal function is implied in the
progressive differentiation model with ‘weak’ signals inducing
replication without differentiation. However, in the stem cell
model of antigen-dependent CD8+ T-cell differentiation, the
TCM stage is the result of either unique proliferation signals
that do not cause differentiation or an active process that
prevents effector differentiation of the antigen-stimulated
cells. This stem cell model believes that the high number of
cell cycles that TCM go high through to account for their clonal
expansion during a primary response high indicates that they
have received ample signaling, and not the weak signal proposed
by the progressive differentiation model. The biological
problem that the existence of the TCM subset presents is how
such extensive cellular replication can be uncoupled from
effector differentiation. The only known means for CD8+ Tcell replication, stimulation by IL-2 or IL-15, if repetitively
applied, causes effector differentiation (12). The stem cell
hypothesis proposed that a mechanism analogous to that of
BCL-6, a transcriptional repressor of the germinal center B cell
that inhibits IL-2-induced differentiation to a plasma cell, has
a role in arresting the differentiation of the developing TCM.
The third model was proposed in 2003, termed the ‘linear
differentiation model’ (13). This model differs from the two
previous models in putting forward the concept that TEFF
develop directly from naı̈ve CD8+ T cells, TEM from TEFF after
the primary infection is resolved, and then TCM only during the
memory phase by a gradual process of de-differentiation from
the TEM subset. The only signal that has been considered to be
required for the conversion of TEM to TCM is IL-7 (14), but
this cytokine probably functions only as a survival factor (15)
and not to regulate differentiation. One of the most interesting
implications of this model is that it does not provide a
mechanism for the ability of CD8+ T cells to cope with
persistent infections. If TCM is the subset that has secondary
proliferative function and if the generation of new TEFF during
a persistent infection is functionally analogous to this process
in a secondary response, the proposal that a ‘rest period’ is
required for the development of the TCM subset means that
CD8+ T cells would not be able to generate new TEFF during
persistent viral infections. Despite this aspect of the model not
being in accord with many examples of the CD8+ T-cell
response coping with persistent viral infections, with perhaps
the most striking example being the continuous production
for years of CD8+ TEFF in patients infected with human
immunodeficiency virus until virally induced severe depletion
of CD4+ T cells occurs, the experimental findings in the study
presenting the linear differentiation model were compelling.
This model was widely accepted as the mechanism for the
development of memory CD8+ T cells (16).
Experimental analyses of development models
In the original description of the CD8+ TCM and TEM subsets,
the CCR7+CD8+ TCM subset had no effector functions, including an absence of interferon-g (IFN-g) production (5). Most
studies of murine CD8+ T-cell differentiation have used high
expression of CD62L as the phenotypic marker of the TCM
subset, and they have occasionally found cells with effector
function in the CD62Lhigh memory population. Although this
finding would apparently argue against a clear association
between distinct tissue homing functions and immediate
effector capabilities, it is more likely that CD62Lhigh memory
cells with effector functions indicate an asynchronous differentiation process, making homing receptor expression an
imperfect marker of differentiation. For the sake of consistency with the original description of these cells, when TCM
are referred to in the review, the definition remains that of
Lanzavecchia and colleagues (5), this being a central homing
memory cell without immediate effector function.
The primary study forming the basis of the linear differentiation model set the agenda for subsequent investigators by
showing that TCM, defined as CD62Lhigh memory CD8+ T cells
taken at least one month after the resolution of a primary
infection, mediated a more effective secondary response when
adoptively transferred to recipient mice than did CD62Llow
TEM taken at the same time (13). The efficacy of the transferred cells was measured by viral clearance from several
different sites of challenge and by expansion and generation
of new effector cells. The development of CD62Lhigh memory
cells appeared to be from CD62Llow cells, because very few of
the former were found during or immediately after the primary response, and during the memory phase the proportion
of memory cells that were CD62Lhigh increased. Furthermore,
25 days after adoptive transfer of an almost homogenous
population of CD62Llow memory cells, 45% of the transferred
cells were CD62Lhigh. These findings were interpreted as
indicating that secondary replicative responses were mediated
Immunological Reviews 211/2006

Fearon et al Generation of the self-renewing CD8+ TCM
by TCM and that TCM were not present during the primary
response but slowly developed from TEM during the memory
phase. However, additional findings suggested that selective
expansion of the CD62Lhigh cells also contributed to the shift
in the phenotype of the memory cells. CD62Lhigh memory
cells were shown to replicate more rapidly than CD62Llow
cells, whether measured in non-lymphopenic or lymphopenic
environments by dilution of carboxyfluorescein diacetate succinimidyl ester (CFSE) or by uptake of bromodeoxyuridine
(BrdU). Thus, the TCM mediating effective secondary
responses may have been present in small numbers early in
the memory phase as CD62Lhigh cells, but they required time
for homeostatic expansion to become the dominant memory
population. Interestingly, this report also showed that TCM, in
addition to expressing a naı̈ve cell’s level of CD62L, had
another characteristic of a less differentiated cell, which was
the ability to produce IL-2. Therefore, the ‘linear differentiation pathway’ is truly a de-differentiation pathway. Although
de-differentiation has been reported to occur in plasma cells in
which BCL-6 has been introduced (17), the phenomenon is
sufficiently rare in biological systems to merit publication of
its experimental demonstration. Even though an inability to
account for the capacity of CD8+ T cells to cope with persistent viral infections and the need to posit a process of dedifferentiation weaken the case for the linear progression
model, it is important to recognize that the study on which
this model is based is the first to demonstrate that the TCM
subset is responsible for secondary proliferation.
Five studies using adoptive transfer of memory cells have
now tested aspects of this model (Table 1). Only one has not
found that replicative memory responses are mediated by the
TCM subset. Roberts and Woodland (18) demonstrated that
CD62Llow memory CD8+ T cells elicited by primary infection
with Sendai virus were more effective than CD62Lhigh cells in
generating new effector cells after adoptive transfer to recipient mice and secondary challenge with intranasal Sendai virus.
However, a second report by this group one year later (19)
modified this conclusion by showing that the relative capabilities of the TEM and TCM subsets were dependent on the time at
which they were assayed for this function. When taken one
month post-primary infection, CD62Llow memory cells were
more effective, whereas at 13 months post-primary infection,
the CD62Lhigh TCM subset was more effective in a secondary
infection with Sendai virus. Although this observation could
reflect a more rapid acquisition of replicative function than of
high CD62L expression by the memory cells destined to
become TCM, the first report by this group may be the only
instance in which cells identified as TEM by their relative
expression of CD62L were more effective in generating new
TEFF in a secondary response than TCM. The authors suggested
that mucosal infections, such as that caused by Sendai virus,
may alter the memory response as compared to infections at
other sites, a possibility that needs to be investigated further,
given the sometimes different strategies adopted by the
immune system to cope with mucosal infections.
In contrast to the analysis by Roberts and Woodland
(18), that by Bachmann and colleagues (20), which involved
infection of mice with LCMV, confirms many of the findings
of Wherry et al. (13), the most notable being that the
CD127+CD62Lhigh TCM subset of memory cells was most
Table 1. Central memory versus effector memory CD8+ T cells and secondary replicative responses
Conversion TEM
Immunological challenge Memory subset transferred 2° proliferation to TCM
Wherry et al. (13)
Roberts and Woodland (18) Sendai
Bachmann et al. (20)*
Marzo et al. (23)†
Male bone
marrow cells
Roberts et al. (19)
Bouneaud et al. (21)
1–2 mon. CD62Lhigh
1–2 mon. CD62Llow
>1.5 mon. CD62L+
>1.5 mon. CD62L–
12 mon CD62Lhigh
12 mon CD62Llow
Less effective
More effective
More effective
Less effective
2° proliferation of
‘converted’ cells
Not assessed
Not assessed
Not assessed
Not assessed
2/7 mice responded with
47% of cells converted
Not applicable
Not assessed
Not assessed
*In the Bachmann study, adoptively transferred gp33-specific TCR transgenic CD8+ T cells were taken 10 days after LCMV infection, and subsets were
re-transferred to naı̈ve recipients that were then challenged with LCMV or vaccinia expressing gp33. Thus, these cells were not memory cells but were
acutely responding cells, demonstrating the replicative potential of cells with a TCM phenotype at the peak of a primary response.
†In the Marzo study, conversion of TEM to TCM was observed only when TEM were derived from a large number of naı̈ve cells.
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
effective in secondary replicative responses and that this subset
had a greater capacity for the production of IL-2 than that of
TEM or TEFF (20). These investigators also found that after
adoptive transfer, CD127–CD62Llow cells gradually reverted
to a CD127+CD62Lhigh phenotype, and CD127+CD62Llow
memory cells became CD127+CD62Lhigh. However, the
apparent conversions between subsets may have represented
expansions of contaminating CD62Lhigh memory cells of
greater homeostatic proliferative function, because it was
noted that whereas recovery of CD127+CD62Lhigh cells was
‘almost quantitative’, that of the CD127–CD62Llow subset was
poor. This study had an additional finding that may relate to
the time at which TCM develop. Adoptive transfer of LCMVspecific, CD127+CD62L+CD8+ T cells obtained on day 10 or
15 (in the text it is written day 10, and in the legend for the
figure describing this experiment it is written day 15) of a
primary response provided naı̈ve recipients with a capacity for
a replicative secondary response to viral infection initiated one
day after transfer. Therefore, in contrast to the linear differentiation model stating that TCM develop slowly over a period
of weeks during the memory phase, this finding suggests that
they are present during the primary response (day 10) or very
early in the memory phase (day 15). Even if it is argued that
the cells that transferred the capability for a secondary
response cannot be considered as memory cells because they
were taken during or at the end of the primary response, they
must be considered at least as ‘pre-TCM’, which indicates that
the commitment for TCM development can occur earlier than
the memory phase.
The possibility that an early development of TCM may occur
is also suggested by a study of the development of TCM in vivo
by using a non-viral system of administering male bone
marrow cells to female mice to which had been transferred
T-cell receptor (TCR) transgenic naı̈ve CD8+ T cells specific
for the H-Y antigen Smcy3 (21). A number of findings were
made in this complex study, but two of them are most
relevant in the context of this review. First, the CD62Lhigh
TCM subset, not the CD62Llow TEM subset, transferred replicative secondary responses, and, second, those memory cells
that had converted from CD62Llow to CD62Lhigh during the
memory phase were impaired in their ability to replicate
when secondarily challenged. That is, the cells apparently
belonging to the TCM subset that developed during the memory phase, which the linear differentiation model presents as
the subset mediating a secondary proliferative response, did
not share with the TCM developing earlier in the immune
response the capacity for vigorous secondary proliferation.
This finding did not precisely contradict the observations of
Wherry et al. (13), which had assessed the secondary replicative function of all CD62Lhigh TCM cells taken during the
memory phase and not only the cells that had converted
to the CD62Lhigh state during this period. Since a CD62Lhigh
phenotype mediates homing to secondary lymphoid tissues,
this observation also suggests that although the lymphoid
environment may be necessary for a full proliferative recall
response of a memory cell, it is not sufficient. Thus, the
correlation noted between homing of in vitro activated, adoptively transferred CD8+ T cells to secondary lymphoid tissue
and non-lymphoid sites and the presence or absence, respectively, of antigen-stimulated proliferative function (22) does
not explain fully the basis for proliferative function of the TCM
subset. The TCM apparently must have two attributes that
enable it to have a replicative secondary response, lymphoid
homing capability and intrinsic proliferative potential.
Therefore, these reports (20, 21), while validating the importance of the TCM for secondary replicative responses, began to
question whether the TCM subset develops from the TEM subset.
The most recent report relevant to this analysis of the
development of the memory CD8+ T-cell subsets presented
evidence that this conversion is not physiological but is caused
by the adoptive transfer of numbers of TCR transgenic naı̈ve T
cells that exceed the expected precursor frequency of naturally
occurring naı̈ve cells (23). CD62Llow memory CD8+ T cells
derived from a larger than normal number of adoptively
transferred naı̈ve cells converted to a CD62Lhigh phenotype
during the memory phase of the response, whereas conversion was not observed with CD62Llow memory cells that had
been generated from a low number of naive cells. It was
proposed that the presence of large numbers of naı̈ve cells
leads to competition for limiting activation signals and the
development of a ‘transitional TEM’ subset that, because of the
limiting activation signals, had not completed differentiation
to a TEM stage. The transitional TEM is therefore ‘unstable’ and
reverts to the CD62Lhigh phenotype, and it may be related to a
recently described CCR7+CD62Llow phenotype of responding,
adoptively transferred, TCR transgenic CD8+ T cells (24).
Overall, this report argues against the linear differentiation
model, which is based on experiments using adoptively transferred TCR transgenic T cells, and supports the progressive
differentiation model that predicts that limiting activation signals are associated with the development of transitional cells
that are intermediate between two stages of development.
In summary, studies of the behavior of adoptively transferred memory CD8+ T cells have almost always found that
the CD62Lhigh TCM subset is principally responsible for the
proliferative component of the secondary response. The
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
frequent finding of CD62Llow memory cells converting to a
CD62Lhigh state during the memory phase has been recently
correlated with a high precursor frequency of antigen-specific
naı̈ve cells, perhaps indicating that the TEM to TCM conversion
is an artifact of the adoptive transfer model. However, secondary responses, which are initiated with a high precursor
frequency of antigen responsive TCM, might conceivably generate the ‘transitional’ TEM subset. Even if there are circumstances in which the TEM to TCM conversion might normally
occur, the CD8+ T cells that become CD62Lhigh during the
memory phase apparently do not have the proliferative function of the TCM subset in a secondary response. This observation has another potential meaning, which is the likelihood of
heterogeneity within the CD62Lhigh population of memory
CD8+ T cells, with some cells having this phenotype not
being able to mediate a secondary replicative response.
These findings strongly suggest that a model in which TCM
develop from TEM during the memory phase does not apply to
most circumstances of antigen-dependent CD8+ T-cell development. TCM must therefore develop during the primary
response, and the report by Bachmann et al. (20) provides
direct evidence in favor of this conclusion. In the absence of
the conversion of TEM to TCM during the memory phase, as
shown by the study of Marzo et al. (23), the improvement
with time of replicative memory must reflect changes within
the TCM subset itself. This may be explained by the greater
homeostatic expansion of the TCM subset than within the TEM
subset, leading to an absolute increase in the number of TCM.
These adoptive transfer studies identifying TCM as the subset
that generates TEFF in a secondary response also imply but do
not clearly show that TCM precede TEM in development and are
therefore less differentiated than T EM . Although Bouneaud
et al. (21) and Marzo et al. (23) suggest that all three products
of a primary response, TCM, TEM, and TEFF, develop independently, the ability of the TCM subset to give rise also to TCM, TEM,
and TEFF (24) during a secondary response suggests that preTCM could also serve as the source of TEM and TEFF during a
primary response. This function for pre-TCM, which remains to
be demonstrated, would be most consistent with the stem cell
model of development.
With regard to this question of whether CD8+ TCM are less
differentiated than TEM, in the original description of TCM and
TEM, CD4+ TCM could produce IL-2 but not IFN-g or IL-4, the
effector cytokines marking T-helper 1 (Th1) and Th2 development, whereas CD4+ TEM did produce these effector cytokines (5). This finding was taken as evidence that CD4+ TCM
were less differentiated than CD4+ TEM. It is curious, then, that
the relatively greater capacity of CD8+ TCM than TEM to
Immunological Reviews 211/2006
produce IL-2 is not necessarily considered as indicating that
CD8+ TCM are less differentiated than the CD8+ TEM that can
make IFN- g. Even though IL-2 has no antiviral effects and it is
clear that the purpose of differentiation of CD8+ T cells is to
create an antiviral effector cell, the finding of cells at the end
of the primary response that can rapidly produce IL-2 has
been ‘considered to reflect maximum functional differentiation’ (25). Thus, one of the difficulties in assigning relative
positions to TCM and TEM in the developmental pathway of
antigen-stimulated CD8+ T cells is that there is no widely
accepted stepwise pattern of acquisition of cytokine-producing
potential during development of CD8+ T cells as there is for
CD4+ T cells. When this difficulty is combined with the additional problem of CD62Lhigh CD8+ TCM being capable of
producing only IL-2 in some studies (26) and both IL-2 and
IFN-g in other studies (20), one is left with few phenotypic
markers that would allow characterization of the relative differentiation of the CD8+ TCM and TEM subsets. Perhaps the
definition of the TCM subset based on homing receptor expression is not sufficiently stringent, or differentiation within the
CD8+ T-cell lineage is not as defined or synchronously regulated as in the CD4+ T-cell lineage.
The adoptive transfer experiments, however, have provided
one means by which the TCM subset can be understood as
being less differentiated than the TEM subset, with the ability
of TCM but not TEM to mediate a secondary proliferative
response. Replication is the essential property of a naı̈ve
lymphocyte in a clonal immune system. The development of
a subset of antigen-activated CD8+ T cells that retains this
capability must occur before development of subsets that
have effector activity because the former can produce the
latter but the latter cannot generate the former. Therefore, to
understand how the TCM develops, one must know the signals
that cause the effector differentiation of CD8+ T cells. These
are the signals that TCM must modify or avoid during their
development and replication. When one accepts that the TCM
subset is less differentiated than the TEM subset, it follows that
the simplest model for TCM development is a pathway that
allows for clonal expansion without effector differentiation.
Evidence for the role of transcriptional repressors
of IL-2 signaling in the development of memory CD8+
T cells
IL-2 and IL-15, which stimulate T cells through receptors
that share b and common g chain (gc) subunits, cause CD8+
T cells to proliferate and differentiate to TEFF with cytolytic
activity and a capacity to produce effector cytokines. Whether
Fearon et al Generation of the self-renewing CD8+ TCM
IL-15 always causes differentiation or not is not clear, but
it has been shown to do so when CD8+ T cells were restimulated in the presence of this cytokine (12). Therefore,
for CD8+ T cells to regulate their differentiation, they would
need at least to control signals induced by IL-2 and possibly
IL-15. A potential means for cell autonomous regulation of IL-2
responses was initially suggested by studies (27) of a transcriptional repressor, BCL-6, expressed by the germinal center B
cell. Ectopic expression of BCL-6 in the murine B lymphoma
BCL-1 prevented IL-2 from causing the differentiation of these
cells to immunoglobulin M (IgM)-secreting plasma cells. This
effect of BCL-6 was mediated by repression of signal transducer
and activator of transcription 3 (Stat3)-dependent expression
of B-lymphocyte-induced maturation protein-1 (Blimp-1), the
master regulator of plasma cell differentiation. These and similar
findings by Staudt and his colleagues (28) indicated that by
arresting terminal differentiation, BCL-6 enables the germinal
center B cell to undergo iterative cycles of expansion, somatic
hypermutation of Ig genes, and selection for high affinity
variants. At a more general level, they also demonstrated that
the immune system has a mechanism for controlling cytokinedriven effector differentiation. Although this finding related to
B cells, the frequent parallels between the B and T-cell pathways of activation and differentiation at least made plausible
the suggestion that CD8+ T cells might also control IL-2induced differentiation by a similar mechanism. Thus, it was
suggested that in the developmental pathway triggered by
antigenic stimulation of naı̈ve CD8+ T cells, there is a stage
of self-renewal in which cells were restrained from differentiating into effector cells by a BCL-6-like transcriptional repressor (1). This pool of self-renewing cells could then serve as an
inexhaustible source of TEFF during persistent infections and as
the precursor for the TCM subset that mediates memory
responses for intermittent infections.
This hypothesis for active repression of effector differentiation in the development of memory CD8+ T cells has been
tested by two groups, one examining a potential role for BCL6 itself (29, 30) and the other evaluating the participation of a
paralogue, BCL6b (originally termed BAZF) (31). BCL-6 has
been shown to be expressed as protein in germinal center B
cells and, curiously, in cortical thymocytes (32), although no
abnormality in thymocyte development has been reported in
BCL-6-deficient mice (33–35). Mice with targeted interruption of the BCL6 gene do not develop germinal centers,
indicating an essential role in the germinal center B-cell.
They also develop a spontaneous inflammatory disease that
is characterized by infiltration of the heart, spleen, gut, liver,
and skin with monocytes, eosinophils, and CD4+ Th2 cells.
The cause of this process is thought to be dysregulated secretion
of chemokines by macrophages (36). Even though an underlying abnormality of lymphocytes is not thought to be basis for
the inflammation, it presents a problem when interpreting the
responses of memory CD8+ T cells in BCL-6–/– mice.
Both loss-of-function BCL-6–/– mice and gain-of-function
BCL-6 transgenic mice have been evaluated (29, 30). In
studies using vaccinia/ovalbumin (OVA), there was either
no significant difference or only a 50% decrease in the percentage of CD8+ T cells that were OVA-specific 10 weeks
post-infection in BCL-6–/– mice relative to the response of
normal controls. The function of the BCL-6–/– memory cells
was apparently unimpaired in that the fold increase in epitope-specific CD8+ T cells after boosting with OVA peptide
did not differ between wildtype and BCL-6–/– mice. In contrast, in mice with the BCL-6 transgene, which was regulated
by the lck proximal promoter that is usually employed to limit
expression to thymocytes, there was enhanced homeostatic
expansion of CD8+ T cells in lymphopenic mice and a twofold
increase in OVA-specific CD8+ T cells 10 weeks after infection
with vaccinia/OVA. The secondary response of the BCL-6
transgenic mice was also higher than in the wildtype mice.
These results indicate that the forced expression of BCL-6 in
CD8+ T cells promotes their primary and secondary responses,
suggesting that suppressing IL-2-induced differentiation may
indeed have a role. However, the loss-of-function studies with
BCL-6–/– mice did not show a marked reduction in the
expansion of CD8+ T cells, and these findings are ambiguous
because of the immunological abnormality contributing to
spontaneous inflammatory disease in these mice. Thus, a
transcriptional repressor of IL-2-induced differentiation may
promote the responses of memory CD8+ T cells, but the
evidence that BCL-6 is the relevant transcription factor is not
The second approach assessed the possible role of the BCL-6
paralogue BCL6b (31). The original report identifying the
BCL6b gene by its cross-hybridization with BCL-6 cDNA
showed that it was expressed in several non-immune organs,
such as the heart and lungs, and in splenocytes activated with
phorbol ester and ionomycin (37). The participation of
BCL6b in the function of memory CD8+ T cells was first
suggested by the finding that human and murine TCM and
TEM had 10- to 20-fold more BCL6b mRNA than did naı̈ve
CD8+ T cells. This increase reflected expression of BCL6b in a
small subpopulation of memory cells, but it has not been
possible to isolate these cells for further characterization.
Mice in which the BCL6b gene had been interrupted had
normal numbers of CD4+ and CD8+ T cells in the spleen
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
and lymph nodes, although there was a small decrease in the
number of CD4 and CD8 single positive thymocytes. Despite
expression of BCL6b in the heart and lungs of wildtype mice,
the BCL6b–/– mice exhibited no cardiac or pulmonary
abnormalities and did not exhibit spontaneous autoimmune
or inflammatory problems. The humoral response of
BCL6b–/– mice to a T-dependent antigen was normal with
respect to specific IgG1 and IgG2a antibody titers and the
generation of germinal center B cells. Therefore, there was
normal development in the BCL6b-deficient mice of CD4+ T
cells that provide help to B cells, which is dependent on the
expression of CD40L, a means by which CD4+ T cells also
facilitate development of memory CD8+ T cells.
Primary CD8+ T-cell responses to infection with vaccinia or
influenza in BCL6b–/– mice were normal, as was the maintenance of memory CD8+ T cells up to 10 weeks post-infection.
However, the secondary response to intranasal influenza
infection was diminished in the BCL6b–/– mice. Following a
normal initial increase in influenza-specific CD8+ T cells in the
BCL6b–/– mice, the response plateaued at a level in the mediastinal lymph nodes which was only one-third that of the peak
response in BCL6b+/+ mice. There was a corresponding
decrease in the number of influenza-specific CD8+ T cells
accumulating in the lungs of the BCL6b-deficient mice, as
would be expected since replicating cells in the lymph nodes
are the precursors of the TEFF in infected peripheral tissue.
BCL6b-deficient memory cells that were adoptively transferred
into wildtype mice also exhibited an impaired response to
influenza infection. Therefore, BCL6b is required for the
heightened magnitude rather than the accelerated kinetics of
the secondary response. This finding suggests that it is not
required for the development or maintenance of memory
CD8+ T cells that mediate replication but is required for their
continued secondary expansion in the lymphoid tissue. BCL6b
may enhance expansion of memory cells by suppressing IL-2
signaling to inhibit effector differentiation, which would allow
cells to continue to replicate and accumulate in the draining
lymph node. This function would be analogous to that of BCL6 in the germinal center B cell.
The clear demonstrations of a defect in the memory
response of CD8+ T cells in BCL6b–/– mice and of enhancement of the response in BCL-6 transgenic mice indicate that a
transcriptional repressor of IL-2 signaling has a positive role.
The distinct abnormalities in both in vitro and in vivo secondary
responses in the BCL6b–/– mice and the less clear-cut findings
in the BCL-6–/– mice suggest that BCL6b is the relevant
transcriptional repressor. Thus, as was initially proposed, a
BCL-6-like transcriptional repressor does contribute to the
Immunological Reviews 211/2006
memory response of CD8+ T cells. The contribution is less
than originally envisioned, since only the full potential of
memory cells for proliferation was affected by BCL6b, not
their development, so that active repression of IL-2-induced
differentiation by a BCL-6-like transcription factor was not
absolutely necessary for the generation of the self-renewing
population of antigen-stimulated CD8+ T cells. However, an in
vitro analysis of whether BCL6b could separate the growth and
differentiation effects of IL-2 on CD8+ T cells provided
another clue that supported the stem cell model of development, which was that IL-2 may not have a non-redundant role
in the clonal expansion of antigen-stimulated CD8+ T cells.
The thought that BCL6b could be selective in suppressing
only the differentiation-inducing effects of IL-2 was prompted
by a consideration of how it recognizes its target genes. DNA
recognition by BCL-6 and BCL6b is mediated by six and five,
respectively, Krüppel-type, C-terminal zinc fingers that are
94% identical between the two proteins. The consensus
DNA element recognized by these transcriptional repressors
is similar to the core TTC(T/C)N(G/A)GAA IFN-g-activated
sequence motif to which the Stat family of transcription
factors binds. Consistent with this observation, BCL-6
repressed transcription from reporter constructs containing
Stat3 (27, 38) or Stat6 (33, 39) elements. Since it is likely
that the zinc finger domains of BCL-6 and BCL6b interact with
Stat elements in their target genes in a manner distinct from
that of the Stat proteins, it seemed possible that these transcriptional repressors could selectively inhibit the IL-2- and
Stat-activated genes involved in differentiation but not the
genes involved in cellular replication. This uncoupling of
these two processes would permit self-renewal to occur in
IL-2-stimulated T cells, as is thought to occur in the germinal
center B cell. This possibility was also consistent with the
selective inhibition of B-cell differentiation and not of proliferation, when BCL-6 was ectopically expressed in primary
murine B cells stimulated with CD40 and various cytokines
acting through receptors containing the gc chain (27).
However, when BCL6b was assessed for its effects on IL-2
responses by primary murine CD8+ T cells in vitro, it was
found to suppress proliferation induced by the cytokine
(31). Therefore, BCL6b is not capable of distinguishing
between IL-2 target genes involved in proliferation and differentiation. The original hypothesis of the relatively undifferentiated state of the TCM subset being dependent on selective
repression of some effects of IL-2 signaling was not correct.
However, these results did present an apparent paradox in that
BCL6b suppressed IL-2-induced proliferation in vitro, but promoted the expansion of memory CD8+ T cells in vivo. The
Fearon et al Generation of the self-renewing CD8+ TCM
simplest explanation for resolving this paradox would be that
antigen-induced proliferation of CD8+ T cells is largely IL-2independent.
A response of CD8+ T cells in the absence of IL-2/
IL-15 signaling?
The question of whether CD8+ T cells can have significant
clonal expansion without signaling through the IL-2R has not
been clearly resolved, but the preponderance of evidence
suggests that this may occur and that CD8+ T-cell growth in
the absence of signaling through the IL-2/IL-15 receptor
actually may account for most of their expansion in a primary
and perhaps a secondary response (Table 2). The first analysis
of the primary antiviral CD8+ T-cell response in the absence
of IL-2 was more than 10 years ago, and it concluded
that infection with vaccinia or LCMV of mice in which the
IL-2 gene had been functionally inactivated led to the development of normal numbers of virus-specific cytotoxic T
lymphocytes (CTLs), as measured by ex vivo lysis of peptideloaded target cells (40). The possibility of redundant cytokine
functions was considered, and the study was repeated with
mice that were both IL-2- and IL-4-deficient. CTLs were still
generated in response to an infection with LCMV but not
vaccinia in double-deficient mice (41). The authors recognized that a potential role of IL-15 had not been excluded.
Another group at approximately this time also evaluated the
response of IL-2–/– mice to infection with LCMV, and they
found diminished production of CTLs and of IFN-g (42).
Although no reasons were offered to account for the disparate
outcomes of these two sets of experiments, both groups of
investigators acknowledged the potentially confounding effect
of lymphoproliferative disease in the IL-2-deficient mice,
which we now know to be caused by the absence of the
Foxp3-expressing regulatory T-cell lineage. These early studies were also hindered by being dependent on antigen-specific effector functions of CD8+ T cells to detect their
presence, which could mean that TCM were missed. Both
problems were avoided in a study of the effects of IL-2
deficiency in CD8+ T cells expressing a transgenic TCR specific for the class I-restricted influenza nucleoprotein peptide
(43). Although the main aim of the study was to evaluate
thymic development in the absence of IL-2, which was normal, it also examined the response of the TCR-transgenic IL2–/– CD8+ T cells to peptide immunization and found normal
expansion but absence of CTL effector activity; the development of CTL effector function required IL-2. This is the first
report suggesting that effector differentiation, not replication,
is the essential, non-redundant role for IL-2 in the developmental pathway of antigen-stimulated CD8+ T cells. However,
the in vivo peptide stimulation of the CD8+ T cells was not
physiological, lacking the inflammatory components of a normal viral infection, and the role of IL-2 and/or IL-15 in the
primary response of CD8+ T cells to viral infection remained
In 2002, this question was examined again by measuring
the response of CD25-deficient or IL-2-deficient adoptively
transferred OT-I CD8+ T cells specific for the SIINFEKL peptide (44). Infection of recipient mice with vesicular stomatitis
virus expressing this peptide induced equivalent expansion of
wild type, CD25-deficient, and IL-2-deficient OT-I cells in
secondary lymphoid organs. Interestingly, expansion of the
CD8+ T cells, especially in the lamina propria but also in other
non-lymphoid organs, was diminished in the absence of IL-2
Table 2. Studies of the role of IL-2/IL-15 signaling in the primary and secondary responses of CD8+ T cells
Genetic deficiency
Immunological challenge
Kundig et al. (40)
Kramer et al. (43)
Bachmann et al. (41)
Cousens et al. (42)
D’Souza et al. (44)
LCMV, vaccinia
Yu et al. (48)
Clonal expansion*
Effector differentiation
Secondary expansion in vivo
Slight reduction
Not assessed
Not assessed
Not assessed
*In the early studies, class I tetramers were not available. In experiments not involving TCR transgenic mice, clonal expansion could not be determined
independently of effector function.
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
signaling, suggesting that IL-2 may promote growth mainly in
CD8+ T cells that have left secondary lymphoid tissues, possibly linking IL-2 stimulation to a commitment to effector
differentiation. The converse, that a response to IL-2 is not
required for expansion of the relatively less differentiated
CD8+ T cells in secondary lymphoid tissues, also holds. A
subsequent study using similar methods concluded that IL-2
is not required for the early primary proliferative response of
CD8+ T cells (45). If, as seems likely, there is a temporal
progression from an initial phase of expansion of antigenspecific naı̈ve CD8+ T cells in the secondary lymphoid tissues
to a later phase of expansion of responding cells in the
periphery, this second study reinforces the possibility that
IL-2 promotes only the growth of CD8+ T cells that have
committed to effector differentiation. Thus, relatively undifferentiated CD8+ T cells, perhaps the pre-TCM, may proliferate
in an IL-2-independent manner, and more differentiated cells,
TEM and TEFF, may require IL-2 for proliferation. Other inferences regarding the role of IL-2 in the development of TCM,
TEM, and TEFF cannot be made, because the study did not
assess the effector functions of CD8+ T cells that expanded
with or without IL-2 signaling, their expression of homing
receptors or ligands, or their ability to mediate secondary
replicative responses. The report also could not exclude the
participation of IL-15, and some studies (12, 46) but not
others (47) have presented findings suggesting that IL-15
promotes the primary response of naı̈ve CD8+ T cells.
However, these studies with CD25- and IL-2-deficient CD8+
T cells do overcome three problems in early analyses of the role
of IL-2 in the response of CD8+ T cells: by the use of TCR
transgenic T cells and their adoptive transfer into normal recipients, the spontaneous lymphoproliferative disease associated
with absence of IL-2 signaling was no longer a confounding
factor; by enumerating responding CD8+ T cells by their physical presence instead of functional response, the measurement
of the response did not require that the responding cells have
acquired effector function; and the use of viral vectors to
express peptide epitopes presented the CD8+ T cells with a
physiological stimulus. One could conclude unambiguously
from the results of these studies that IL-2 signaling does not
had a non-redundant function in the clonal expansion of naı̈ve
CD8+ T cells within the secondary lymphoid tissues.
The three questions of whether IL-15 signaling can compensate for the absence of IL-2 in the expansion of naı̈ve CD8+
T cells, whether the differentiation of responding CD8+ T cells
to effector cells is dependent on IL-2 or IL-15 or not, and
whether the development of memory cells requires stimulation by these cytokines or not were addressed by Malek and
Immunological Reviews 211/2006
his colleagues in 2003 (48). To exclude effects of IL-2 and
IL-15, mice were developed in which CD122, the shared
b-subunit of the IL-2/IL-15 receptor, was deleted by targeted
interruption of the CD122 gene. To compensate for the
requirement for IL-2 signaling in the development of Foxp3expressing regulatory T cells, the mice also had a transgene
expressing CD122 under control of the lck proximal promoter,
which restricted expression of CD122 to developing thymocytes. Consistent with the success of this strategy, the mice did
not have lymphoproliferative disease, and their peripheral
mature T cells did not respond to IL-2 in vitro, demonstrating
that the CD122 transgene was expressed only in the thymocytes. Remarkably, these mice demonstrated normal expansion of antigen-specific CD8+ T cells to primary infection with
a recombinant vaccinia virus expressing the immunodominant
S510 epitope from the spike protein of mouse hepatitis virus.
Further, these antigen-specific cells differed from those that
developed in wild type mice in that they lacked two effector
functions, the ability to produce IFN-g when stimulated with
antigen in vitro and, in some assays, the ability to kill antigenexpressing target cells immediately ex vivo. Therefore, in the
absence of IL-2 and IL-15 signaling, CD8+ T cells clonally
expanded and replication was not accompanied by the acquisition of effector function. This study made another critical
observation. The memory CD8+ T cells that developed in the
CD122-deficient mice expanded normally when challenged
with a secondary infection with Sindbis virus expressing the
same S510 epitope. Since effector CD8+ T cells were not
generated in the primary response, the precursors of the
memory cells that mediated secondary replicative responses,
presumably the TCM subset, must have developed directly
from naı̈ve CD8+ T cells. Although the complex genetic strategy used to generate these mice may lead to concerns about a
possible ‘leakiness’ of the transgene expressing CD122, the
absence of effector function in the virus-specific CD122-deficient CD8+ T cells suggests that signaling via the IL-2/IL-15
receptor had not occurred and that expression of the CD122
transgene was appropriately restricted to developing
Implications of IL-2/IL-15-independent signaling for
models of memory CD8+ T-cell development
These studies on the role of IL-2 and IL-15 in the response of
CD8+ T cells to viral infections lead to four conclusions that
place constraints on the pathways that one can propose to
account for the development of TCM, TEM, and TEFF. First,
clonal expansion that produces the expanded pool of memory
Fearon et al Generation of the self-renewing CD8+ TCM
cells that is responsible for the secondary proliferative
response is not mediated by IL-2 or IL-15. This consideration
is based on the finding that memory CD8+ T cells in CD122deficient mice mediated a normal secondary replicative
response (48). Thus, an assumption derived from over
20 years of in vitro studies that IL-2 is the essential T-cell
growth factor must be revised (49), and the mechanism by
which CD8+ T cells expand in secondary lymphoid organs
needs to be defined, as this is the cellular response that
establishes replicative memory.
Second, clonal expansion and effector differentiation are
not coupled when expansion is driven by this unknown
mechanism that is independent of IL-2 and IL-15. This finding
suggests that the unknown means for CD8+ T-cell growth
permits self-renewal, at least in the sense that the progeny of
cells dividing in response to this unknown signal are not
obligated to become effector cells and maintain the replicative
potential and undifferentiated phenotype of the parental cell.
Presumably, this is the mechanism for clonal expansion that
avoids exhaustion of antigen-selected clones.
Third, expansion that is mediated by IL-2 is coupled to
effector differentiation. By analogy to myelomonocytic differentiation, IL-2 may be to the CD8+ T cell as colony-stimulating factors are to developing granulocytes and monocytes.
Each cell cycle of myelomonocytic cells growing in response
to colony stimulating factors, or of CD8+ T-cell responding to
IL-2, is associated with further differentiation to an effector
cell. The study by Malek and his colleagues (48) further
suggests that IL-2 and/or IL-15 is absolutely required for at
least some aspects of effector differentiation, although under
unusual circumstances and with in vitro studies, other cytokines signaling through other receptors using the gc chain,
such as IL-4, can drive effector differentiation of the CD8+ Tcell. Whether this occurs with most viral infections in vivo,
which tend to induce a Th1 response, may not be likely.
The cells that replicate and differentiate in response to IL-2
and/or IL-15 therefore must be the precursors of TEM, since
the memory cells generated from CD122-deficient CD8+ T
cells lacked the effector functions characteristic of this subset.
Unfortunately, analysis of the expression of homing receptors,
which could have supported this conclusion, was not conducted. The distinction between TEM and TEFF is, in essence,
operational with the former being defined as cells with effector function that persist into the memory phase. The molecular distinction between TEM and TEFF may be the expression by
the former of the IL-7Ra, since IL-7 is required for the CD8+
T cells responding in the primary response to persist into the
memory phase (14, 50). Thus, IL-2/IL-15 may broadly
regulate the development of cells with effector function,
while other, perhaps ‘inflammatory’, signals that regulate
the expression of the IL-7Ra may determine whether an IL2-stimulated CD8+ T cell becomes a TEM or a TEFF.
Fourth, if clonal expansion of CD8+ T cells via the IL-2/IL15-independent pathway is associated with absence of effector
differentiation and is responsible for the TCM subset and if
effector differentiation is mediated by the two transcription
factors T-bet (51) and eomesodermin (52), then the TCM
derived from the IL-2/IL-15-independent subset may not
have expressed T-bet or eomesodermin. If this assumption is
correct, then the memory cells that mediate a replicative
secondary response may not be dependent on IL-15 for their
maintenance. This possibility is based on the recent finding
that T-bet and eomesodermin are required for the upregulation
of CD122 during activation of CD8+ T cells (53).
Increased expression of CD122 had been established as being
associated with responsiveness to IL-15. Although IL-15 has
been shown to be necessary for the slow replication of
memory CD8+ T cells that mediates their long-term persistence
(47, 54, 55), two studies suggest that not all memory CD8+ T
cells require IL-15 for their maintenance. First, the subset of
gp33-specific memory CD8+ T cells that expands in response to
a secondary infection with LCMV was maintained after three
months in the absence of IL-15, despite a reduction of 90% in
total gp33-specific memory CD8+ T cells, relative to the
number in IL-15-sufficient mice (47). Second, among the
CD44+ memory CD8+ T cells that arise apparently in response
to environmental antigens, a subset of CD122low cells were
maintained in the absence of IL-15, even while the CD122high
subset was markedly depleted (54). This CD44highCD122low
IL-15-independent population is not as evident in an elicited
memory CD8+ T-cell pool, and it would be interesting to
determine whether the residual gp33-specific memory population in the IL-15-deficient mice that mediated a normal secondary response was CD122low. It may be that the cytokine
requirements for maintenance may differ among the TCM and
TEM subsets of memory cells, with perhaps only the TEM subset
being IL-15-dependent, because they have induced T-bet and
eomesodermin and upregulated CD122.
It is informative to examine how these four inferences
square with the different models for the development of
antigen-stimulated CD8+ T cells. First, the finding that TCM
can develop in the absence of TEM indicates that the linear
differentiation model, which places TEFF and TEM before TCM
in the developmental pathway, must be modified to include a
direct naı̈ve T-cell to TCM developmental step. This step is
consistent with the progressive differentiation model, the
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
observations of Bachmann et al. (20), Bouneaud et al. (21) and
Marzo et al. (23), and the stem cell model for antigen-stimulated CD8+ T-cell development (1). Second, the development
of TCM but not TEM in the absence of IL-2/IL-15 signaling is
also consistent, in part, with the prediction of the progressive
differentiation model that TEM and TEFF development require
‘stronger’ signals. However, it may be that ‘different signals’
would more closely reflect the respective requirements of
these subsets for development. Third, as was inherent in the
original description of the TCM and TEM subsets in which the
latter but not the former had potential effector functions, and
as IL-2 induces effector differentiation of CD8+ T cells, the
TCM subset can and perhaps must develop independently of
the differentiating effects of IL-2/IL-15. The mechanism for
this clonal expansion probably is the basis for the essential
stem cell characteristic of antigen-stimulated CD8+ T cells of
maintaining the production of differentiated cells in persistent
viral infections.
An essential next step: definition of the IL-2/IL15-independent mechanism for clonal expansion
of CD8+ T cells
A critical problem now is to discover the means by which
the CD8+ T cell grows in the absence of signaling through the
IL-2/IL-15 receptor. The phenotypic requirements for such
growth are threefold: it must be dependent on ligation of the
TCR and, in some way, on the activation of dendritic cells by
microbial products acting on Toll-like receptors and possibly
by ligation of CD40; it must not induce the production of IL2, so as not to induce differentiation to TEFF or TEM; and it
must maintain the expression of the receptors and ligands
required for residence in secondary lymphoid tissue and of
the IL-7Ra for survival. There is not an absolute requirement
for a BCL-6/BCL6b-like transcriptional repressor of IL-2induced differentiation to create this self-renewing stem cell,
because this as yet undefined IL-2/IL-15-independent pathway of CD8+ T-cell clonal expansion obviates the need to
uncouple cytokine-induced proliferation and differentiation.
However, it is possible that when the number of antigen-
specific CD4+ and CD8+ T cells reaches a critical density
within the T-cell zone of the responding secondary lymphoid
organ, paracrine IL-2 could begin to drive effector differentiation and restrict the expansion of the self-renewing pool of
CD8+ T cells. In this circumstance, the expression of BCL6b by
self-renewing CD8+ T cells would insulate them from IL-2
and allow their continued IL-2-independent growth and
retention in the lymphoid environment (Fig. 1). This process
would increase the pool of self-renewing cells and, since these
are the precursors of TEFF, correspondingly increase the rate at
which effector cells are generated.
In summary, the following pathway is proposed to account
for the observations that have been referred to in this review.
Naı̈ve CD8+ T cells are ‘programmed’ by interaction with
appropriately activated mature dendritic cells that present the
relevant antigen. The programming allows the CD8+ T cells
either to respond to an unidentified ligand that is present on
dendritic cells, which induces proliferation without differentiation, or to IL-2/IL-15, which induces proliferation that is
coupled to differentiation. The former constitute a self-renewing population from which cells can ‘spin-off’ and respond to
IL-2, either autocrine or paracrine. These IL-2-responding
cells then proliferate in a manner that is coupled to differentiation to TEFF and leave the lymphoid environment. At the
end of the primary response, the self-renewing population
gives rise to resting TCM, and the IL-2-responding subset
yields TEM. Both require IL-7 for survival, but only the TEM
require IL-15 to maintain their numbers. In a secondary
response, the high precursor frequency of both CD4+ and
CD8+ TCM causes a rapid build-up of replicating cells that
would produce relatively large amounts of IL-2, which, by
driving differentiation, could limit the expansion of the
self-renewing pool. BCL6b represses the effects of this IL-2,
blocking differentiation, allowing further expansion of the
self-renewing pool and enabling generation of a larger
number of TEFF than occurs in the primary response.
Clearly, the critical element needed to support this proposed pathway is the mechanism of IL-2/IL-15-independent
clonal expansion. From this basic finding, the various other
predictions can then be tested.
1. Fearon DT, Manders P, Wagner SD.
Arrested differentiation, the selfrenewing memory lymphocyte,
and vaccination. Science 2001;
2. Wherry EJ, Blattman JN, Ahmed R. Low CD8
T-cell proliferative potential and high viral
load limit the effectiveness of therapeutic
vaccination. J Virol 2005;79:8960–8968.
Immunological Reviews 211/2006
3. Sevilla N, McGavern DB, Teng C, Kunz S,
Oldstone MB. Viral targeting of hematopoietic progenitors and inhibition of
DC maturation as a dual strategy for
immune subversion. J Clin Invest
Fearon et al Generation of the self-renewing CD8+ TCM
4. Karrer U, et al. Memory inflation: continuous
accumulation of antiviral CD8+ T cells over
time. J Immunol 2003;170:2022–2029.
5. Sallusto F, Lenig D, Forster R, Lipp M,
Lanzavecchia A. Two subsets of memory
T lymphocytes with distinct homing potentials and effector functions. Nature
6. Reinhardt RL, Khoruts A, Merica R, Zell T,
Jenkins MK. Visualizing the generation of
memory CD4 T cells in the whole body.
Nature 2001;410:101–105.
7. Masopust D, Vezys V, Marzo AL, Lefrancois L.
Preferential localization of effector memory
cells in nonlymphoid tissue. Science
8. Weninger W, Crowley MA, Manjunath N,
von Andrian UH. Migratory properties of
naive, effector, and memory CD8+ T cells.
J Exp Med 2001;194:953–966.
9. Lund FE, et al. Lymphotoxin-alpha-deficient
mice make delayed, but effective, T and B cell
responses to influenza. J Immunol
10. Moyron-Quiroz JE, et al. Role of inducible
bronchus associated lymphoid tissue (iBALT)
in respiratory immunity. Nat Med
11. Lanzavecchia A, Sallusto F. Dynamics of
T lymphocyte responses: intermediates,
effectors, and memory cells. Science
12. Manjunath N, et al. Effector differentiation is
not prerequisite for generation of memory
cytotoxic T lymphocytes. J Clin Invest
13. Wherry EJ, et al. Lineage relationship and
protective immunity of memory CD8 T cell
subsets. Nat Immunol 2003;4:225–234.
14. Kaech SM, Tan JT, Wherry EJ, Konieczny BT,
Surh CD, Ahmed R. Selective expression of the
interleukin 7 receptor identifies effector CD8
T cells that give rise to long-lived memory
cells. Nat Immunol 2003;4:1191–1198.
15. Pellegrini M, Bouillet P, Robati M, Belz GT,
Davey GM, Strasser A. Loss of Bim increases
T cell production and function in interleukin
7 receptor-deficient mice. J Exp Med
16. Gourley TS, Wherry EJ, Masopust D, Ahmed R.
Generation and maintenance of immunological
memory. Semin Immunol 2004;16:323–333.
17. Fujita N, et al. MTA3 and the Mi-2/NuRD
complex regulate cell fate during B lymphocyte differentiation. Cell 2004;119:75–86.
18. Roberts AD, Woodland DL. Cutting edge:
effector memory CD8+ T cells play a
prominent role in recall responses to
secondary viral infection in the lung.
J Immunol 2004;172:6533–6537.
19. Roberts AD, Ely KH, Woodland DL.
Differential contributions of central and
effector memory T cells to recall responses.
J Exp Med 2005;202:123–133.
20. Bachmann MF, Wolint P, Schwarz K, Jager P,
Oxenius A. Functional properties and lineage
relationship of CD8+ T cell subsets identified
by expression of IL-7 receptor alpha and
CD62L. J Immunol 2005;175:4686–4696.
21. Bouneaud C, Garcia Z, Kourilsky P, Pannetier C.
Lineage relationships, homeostasis, and recall
capacities of central- and effector-memory CD8
T cells in vivo. J Exp Med 2005;201:579–590.
22. Harris NL, Watt V, Ronchese F, Le Gros G.
Differential T cell function and fate in lymph
node and nonlymphoid tissues. J Exp Med
23. Marzo AL, Klonowski KD, Le Bon A, Borrow P,
Tough DF, Lefrancois L. Initial T cell frequency
dictates memory CD8+ T cell lineage commitment. Nat Immunol 2005;6:793–799.
24. Unsoeld H, Pircher H. Complex memory
T-cell phenotypes revealed by coexpression of
CD62L and CCR7. J Virol
25. La Gruta NL, Turner SJ, Doherty PC.
Hierarchies in cytokine expression profiles for
acute and resolving influenza virus-specific
CD8+ T cell responses: correlation of cytokine
profile and TCR avidity. J Immunol
26. Huster KM, et al. Selective expression of IL-7
receptor on memory T cells identifies early
CD40L-dependent generation of distinct
CD8+ memory T cell subsets. Proc Natl Acad
Sci USA 2004;101:5610–5615.
27. Reljic R, Wagner SD, Peakman LJ, Fearon DT.
Suppression of signal transducer and activator
of transcription 3-dependent B lymphocyte
terminal differentiation by BCL-6. J Exp Med
28. Shaffer AL, Yu X, He Y, Boldrick J, Chan EP,
Staudt LM. BCL-6 represses genes that
function in lymphocyte differentiation,
inflammation, and cell cycle control.
Immunity 2000;13:199–212.
29. Ichii H, et al. Role for Bcl-6 in the generation
and maintenance of memory CD8+ T cells.
Nat Immunol 2002;3:558–563.
30. Ichii H, Sakamoto A, Kuroda Y, Tokuhisa T.
Bcl6 acts as an amplifier for the generation and
proliferative capacity of central memory CD8+
T cells. J Immunol 2004;173:883–891.
31. Manders PM, et al. BCL6b mediates the
enhanced magnitude of the secondary
response of memory CD8+ T lymphocytes.
Proc Natl Acad Sci USA 2005;102:7418–7425.
32. Hyjek E, Chadburn A, Liu YF, Cesarman E,
Knowles DM. BCL-6 protein is expressed in
precursor T-cell lymphoblastic lymphoma
and in prenatal and postnatal thymus. Blood
33. Dent AL, Shaffer AL, Yu X, Allman D,
Staudt LM. Control of inflammation, cytokine
expression, and germinal center formation by
BCL-6. Science 1997;276:589–592.
34. Ye BH, et al. The BCL-6 proto-oncogene
controls germinal-centre formation and
Th2-type inflammation. Nat Genet
35. Fukuda T, et al. Disruption of the Bcl6 gene
results in an impaired germinal center formation. J Exp Med 1997;186:439–448.
36. Toney LM, et al. BCL-6 regulates chemokine
gene transcription in macrophages. Nat
Immunol 2000;1:214–220.
37. Okabe S, et al. BAZF, a novel Bcl6 homolog,
functions as a transcriptional repressor. Mol
Cell Biol 1998;18:4235–4244.
38. Yu RY, et al. BCL-6 negatively regulates
macrophage proliferation by suppressing
autocrine IL-6 production. Blood
39. Harris MB, et al. Transcriptional repression
of Stat6-dependent interleukin-4-induced
genes by BCL-6: specific regulation of
iepsilon transcription and immunoglobulin
E switching. Mol Cell Biol
40. Kundig TM, Schorle H, Bachmann MF,
Hengartner H, Zinkernagel RM, Horak I.
Immune responses in interleukin-2-deficient
mice. Science 1993;262:1059–1061.
41. Bachmann MF, et al. Antiviral immune
responses in mice deficient for both interleukin-2 and interleukin-4. J Virol
42. Cousens LP, Orange JS, Biron CA.
Endogenous IL-2 contributes to T cell
expansion and IFN-gamma production
during lymphocytic choriomeningitis virus
infection. J Immunol
43. Kramer S, Mamalaki C, Horak I, Schimpl A,
Kioussis D, Hung T. Thymic selection
and peptide-induced activation of T cell
receptor-transgenic CD8 T cells in
interleukin-2-deficient mice. Eur J Immunol
44. D’Souza WN, Schluns KS, Masopust D,
Lefrancois L. Essential role for IL-2 in the
regulation of antiviral extralymphoid
CD8 T cell responses. J Immunol
45. D’Souza WN, Lefrancois L. IL-2 is not
required for the initiation of CD8 T cell
cycling but sustains expansion. J Immunol
46. Schluns KS, Williams K, Ma A, Zheng XX,
Lefrancois L. Cutting edge: requirement for
IL-15 in the generation of primary and
memory antigen-specific CD8 T cells. J
Immunol 2002;168:4827–4831.
Immunological Reviews 211/2006
Fearon et al Generation of the self-renewing CD8+ TCM
47. Becker TC, et al. Interleukin 15 is required
for proliferative renewal of virus-specific
memory CD8 T cells. J Exp Med
48. Yu A, et al. Efficient induction of primary and
secondary T cell-dependent immune
responses in vivo in the absence of functional
IL-2 and IL-15 receptors. J Immunol
49. Gillis S, Gillis AE, Henney CS. Monoclonal
antibody directed against interleukin 2. I.
Inhibition of T lymphocyte mitogenesis and
the in vitro differentiation of alloreactive
cytolytic T cells. J Exp Med
50. Schluns KS, Kieper WC, Jameson SC,
Lefrancois L. Interleukin-7 mediates the
homeostasis of naive and memory CD8 T
cells in vivo. Nat Immunol
51. Glimcher LH, Townsend MJ, Sullivan BM,
Lord GM. Recent developments in the
transcriptional regulation of cytolytic
effector cells. Nat Rev Immunol
52. Pearce EL, et al. Control of effector CD8+
T cell function by the transcription factor
Eomesodermin. Science
Immunological Reviews 211/2006
53. Intlekofer AM, et al. Effector and memory
CD8+ T cell fate coupled by T-bet and
eomesodermin. Nat Immunol
54. Judge AD, Zhang X, Fujii H, Surh CD,
Sprent J. Interleukin 15 controls both
proliferation and survival of a subset of
memory-phenotype CD8+ T cells. J Exp Med
55. Goldrath AW, et al. Cytokine requirements
for acute and Basal homeostatic proliferation
of naive and memory CD8+ T cells. J Exp Med
CURRENT PRACTICES William G. Stopper, Editor
Achieving Post-Outsourcing
Oshima, HR Outsourcing Change Solution Leader, Tina Kao & Jennifer Tower,
Talent and Organization Consultants, Hewitt Associates
Many companies faii to realize the full
financial and strategic benefits of cmtsotircing hecause in their enthusiasm for “what
goes” they devote insufficient attention to
“what stays. ” HR leaders need to understand
that outsourcing success means havtng the
courage to transform the make up, organizational design, and strategic priorities of HR.
For the retained HR, this represents a oncein-a-lifetitne opportunity to enhance its value
to the hiisiness.
ccording to a rcccnr survey by
I lewitt Associates (Hewitt Research
Study, 2004), more than 60 percent
of companies outsource part or all of HR
administration. By contracting out technology maintenance, data management, HR
customer service, and other HR processes,
companies aim to reduce costs, improve service levels, and free up time for their HR
leaders to focus on critical business issues. As
the HR outsourcing industry matures and
brand name companies such as Prudential,
Sony Electronics, I’roctor and Gamble, and
What Is HRBPO?
HRBPO is the integrated delivery and
operation of five or more HR processes with
an emphasis on process, service levels, and
quality of information. In an HRBPO
arrangement, the third party becomes the single point of contact for all HR transactions
and HR customer service. Employees, managers, and HR professionals have one place to
go (web-enabled desktop) to execute transactions and retrieve data and one place to call
(the HR contact center) for questions about
HR policies and programs. Although the
HRBPO may be serving a client from a variety of locations, using different technologies,
and coordinating a network of vendors, the
infrastructure is invisible to tbe employees
and managers.
Sun Microsystems sign multiyear HR
business process outsourcing (HRBPO)
agreements, more companies are exploring
broad based HR outsourcing.
The focus now is on the new HR function in tbe post-outsourcing world. Many
companies fail to realize the full financial
and strategic benefits of outsourcing because
in their enthusiasm for “what goes” they
devote insufficient attention to “what stays.”
HR leaders need to understand that outsourcing success means having the courage to
transform the make up, organizational
design, and strategic priorities of HR. F{)r the
retained HR, this represents a once-in-a-lifetime opportunity to enhance its value to the
Why Companies Outsource HR
Outsourcing is increasingly part of business strategy, particularly at high-growth
firms. Cost savings is an important part of
the equation, but in a recent survey at least
half of executive respondents said that a
“better focus on core competency” and
“improved business processes” were among
tbetr top three reasons for outsourcing
(Hewitt Research Study, 2004). The most
common reasons for outsourcing, whether
it is benefits, payroll, employee data management, or all HR administration, include:
1. Foc«s on the core husiness. Companies
that outsource can free up staff from HR
administration to focus on more strategic
work: activities that are integral and
differentiating to the business.
2. Drive cost savings and efficiency. By outsourcing, companies can reduce operating
costs and minimize the need for additional capital expenditures such as technology purchases and upgrades. According to
a recent report by the Yankee Group
(Fersht, 2005), companies with more
than 10,000 employees can expect a savings of 20 percent from HR outsourcing.
.1. Improve service experience. Outsourcing
HR enables companies to provide new
capabilities and services for employees,
such as 24/7 access to their benefits via the
Internet, decision support tools, and
improved customer service. Also, in
broad-scope HR outsourcing, processes
can be seamlessly integrated across multiple HR areas (hiring, compensation., benefits, learning, etc.) to match how managers
and employees access HR information and
to eliminate the silos that are common in
many HR organizations.
4. Hiiild flexibility and responsiveness to
business needs. Outsourcing can help a
company be more flexible and responsive
to changing business needs by providing
additional capability, capacity, and unit
cost predictability. For example, highly
acquisitive companies have found outsourcing an invaluable way to scale their
organization while providing unit cost
predictability for HR costs tied to supporting the acquired populations. This is
critical in helping the business ensure that
integration costs are anticipated and
managed rigorously.
Sony Electronics Inc. is one company
that moved to outsourcing HR administration for a number of reasons, not solely to
reduce costs. An outdated technology infrastructure, inefficient HR processes, and
nonstandardized HR tools across businesses
and locations were all drivers behind the
decision. According to Patricia BoggiGibbons, VP of e-HR and Benefits, “managing HR operations detracted from our true
role as people and talent strategists, and
many other companies are coming to the
same conclusion. We became convinced tbat
the HR outsourcing model was the wave of
tbe future.”
What HRBPO Means for the HR
HRBPO gives companies the opportunity
to transfer all administrative, customer service, and technology infrastructure work to
an outside partner, so that they can devote
energy to strategy, design, and consulting to
the business. Based on Hewitt’s research (HR
Analyzer database), tbe typical pre-HRBPO
company devotes between 40 and 50 percent
of its HR hTEs to routine transaction processing, data entry, and responding to
employee and manager questions (Lawler,
2004). When that work shifts to the outsourcing provider, HR needs to make smart
decisions about overall HR headcount, the
role of the retained HR organization, and the
critical skills and competencies needed.
staffing/recruiting, compensation, training
administration, and emerging areas such as
HRMS and expatriate administration (HR
Analyzer database).
What Goes to HRBPO
In HRBPO, the vendor provides centralized HH administration and customer service,
maintains the employee portal and supporting
technology infrastructure, and coordinates
internal and external interfaces. Currently,
total or partial outsourcing is most common
in benefits and payroll, gaining ground in
staffing, training/development, compensation
and HRIT, and emerging as new service
offerings in areas such as performance management and communications planning.
Using four HRBPO pioneers (Lawler, et al.,
2004)—British Petroleum, Bank of America,
Prudential Financial, and International
Paper—as examples. Exhibit I shows that the
scope of services delivered via HRBPO
expand beyond the basics to include
What Stays in HR
The role of HR, and the opportunity to
contribute to business strategy, is enhanced by
HRBPO. The core responsibilities that stay in
HR are:
1. IIR strategy: Determine actions HR must
take to drive results for the business.
2. HK program design: Design of HR programs and practices tor benefits, compensation, talent programs, staffing, and
other talent management areas.
3. F-xecutive coaching and organizational consulting: Direct support for senior leadership
in improving their individual effectiveness
and the effectiveness of the organizations
with which they work.
4. Workforce
Interpretation and action planning based
on workforce data and external trends,
including staffing and talent strategies.
5. Problem solving and manager support:
Assist managers in HR and other related
issues including definition of staffing
needs, compensation decisions, and complex employee relations issues.
6. Change management: Support in strategy
and execution of firm-wide change initiatives, t>ften in the context of major
restructuring (e.g., merger, divestiture).
7. Budget management: Management of the
HR budget.
8. Vendor management: Management of
relationship with BPO provider; establishing and monitoring service levels.
One of the most common, and desired,
results of HRBPO is that HR business partners or HR generalists increase in visibility
and impact, as they enhance their role as
business advisors and HR content experts.
Scope of Services
HRBPO Delivery Models
Outsourcing Prevalence
HR Activity
Most Common
Gaining Ground
Outsource Outsource
Completely Partially
Bank of
Flexible spending administration
Health care benefits admin
Pension benefits admin
Payroll administration
Recruitment/staffing (non-exec)
Retirement/staffing (exec)
Training and development
Executive dev and coaching
Compensation administration
HR technology’
Employee communication plans
Expatriate administration
HR policy development
Strate^c business planning
Performance management
Source: HR activity data troin “Human Resource Outsourcing Survey Report” (SHRM, 201)4) and Hiinuin Resource Business Process (JutsoHrcmg, Liwler, ei al.
(Jossey Bass, 2004).
Freed up from a role o “jack of all tradcb,”
they are able to focus on creating solutions to
business problems—not on transactional support. Recent Hewitt (Exhibit 2) data found
that companies that implement HR shared
services (internal or outsourced) more than
double the reach of the HR generalist from a
231:1 ratio (employee to HR ^eneralist role)
to a 553:1 ratio. {Eilitor’s Note: In tbe Hewitt
data, the broader comparison of employee to
HR overall is 86:1, a number closer to the traditional 100:1 ratio.)
articulate outsourcing’s impact on the business, showing how HRBPO not only benefits
tbe HR function, but aiso bow it affects tbe
enterprise as a whole. This is critical to creating tbe right buy-in and establishing tbe
appropriate expectations from business leaders throughout the organization.
This tbeme has played out time and time
again in successful IIRHPO partnerships.
For example, at Britisb Petroleum, outsourcing HR aligned with a broader business initiative to e-enable all business
processes and support tbe company’s
explosive growth
electronics, aligning the internal and
was a critical aspect
of the outsourcing
business case. In
both cases tbe ability to reduce and
better manage costs
was a key factor in
making tbe outsourcmg decision.
Ratio of Company FTEs to HR
Business Partners/Generalists
Shared Services
(internal or
No Shared
Source; Hewitt HR Analyzer data, February 2005.
Four Steps for Achieving PostOutsourcing Success
Io rn.ixnni/e the value from HRBI’O, HR
needs a clear perspective on how tbis dramatic change fits into tbe overall HR strategic
vision. Smart decisions need to be made about
bow the retained HR organizati(tn will function in the new environment. At the outset of
an HRBPO transformation, HR leaders must:
I. Anchor HR oiirsourcing in tbe business
2. Designate a “cbampion” tor tbe outsourcing initiative.
^. Develop a vision and strategy for retained
4. Get tbe right people with tbe right skills and
tbe rigbt mindset.
Anchor HR Outsourcing in the
Business Strategy
HRBPO should fit witb tbe business
strategy, in its expected outcomes and
the road to getting tbere. HR leaders in
the organization need to understand and
although cost is universally one of the
500 600
main drivers for
outsourcing, most
organizations find that other factors such as
improved scalability and an enhanced
employee experience gain in importance after
the outsourcing relationship has heen initiated. This suggests an even greater ability of HR
to impact tbe business and deliver value on an
ongoing basis.
Value Delivered, Having a clearly articulated
vision of how HR outsourcing contributes to
the business will belp on two counts: (I) establisbing tbe right financial and performance
measures, and (2) monitoring progress toward
meeting tbose goals while helping ensure that
HR meets its cost savings targets within the
expected time frames.
Designate a “Champion” for the
Outsourcing Initiative
Ihc shift to HRBPO is a major transformation in the way HR services are delivered.
This requires buy-in and participation from
multiple stakebolders—leaders, managers,
employees, and HR professionals—to ensure
the shift is a success.
Companies that have the greatest success
in driving change designate a leader with credihility and passion to be the “champion,”
someone who will drive and own tbe transformation and ensure that the changes stick
and become “tbe way business gets done
around here.” Such a champion should focus
on the following activities:
1. F.ngd^e Inisiness leaders tu the process.
This includes sharing the vision on three
levels: (a) the business case and financial
benefits, (b) the transformation of the
customer experience, and (c) a status
report on progress during the implementation. For one uf Hewitt’s clients, tbe
champion spent time individually with
eacb key business leader and provided a
realistic picture of the direct impact of the
change on bis/her operation. Tbis allowed
the champion to build buy-in from top
leaders and set expectations about wbat
the implementation would look like.
2. Be il visible suf)portcr. This includes
publicly sharing tbe vision for how outsourcing aligns with the business and
enbances tbe employee experience.
3. Matiafic an ongoing change program.
Tbese changes are not one-rime events.
They require a sustained effort to change
tbe way people think about HR and its
role in the organization. The champion is
critical in ensuring tbat tbe changes for
how work gets done are embraced.
Value Delivered. Creating visihie leadership
support for outsourcing through a champion
accelerates tbe adoption of new delivery
models by employees and managers. It also
limits tbe prevalence of “shadow HR” organizations that can increase total costs and
undermine the effectiveness of tbe new serv ice
delivery solution t)ffered by tbe outsourcing
Develop o Vision and a Strategy for
Retoined HR
Outsourcing major components of HR is
an opportunity^ to reinvent tbe HR function.
Before outsourcing, one bas to understand
how HR, as a whole, will contribute to and
align with the organization’s business strategy.
Some of the questions that determine the strategy are:
1. Growth: What are tbe business’s expectations for growth? Wbere will that growth
be (geographic, business focus, employee
type, etc.)? What are the implications for
changes in tbe workforce and the ways in
which HR serves employees?
Sony Electronics: Meeting the
Challenge of Chonge
Percentage of Time Spent on Various
HR Roles
Sony Electronics Inc. (Sony) made its decision to
outsource HR administration in 2001, and went
“live” in the fall of 2002. Today, Sony’s outsourcing
includes the HR service center and employee portal,
HRMS, benefits, compensation administration,
staffing delivery for many locations, and elements of
workforce management. Three aspects of the HRBPO
made this a big change for Sony:
1 PostH Outsource
Pre- t)uts«urce

Strategic Business
Source: HR Business Process Outsourcing., by Ed Lawler, et al. tJosscy Bass, 2004).
2. Productivity: What steps is the company
taking to increase productivity? How can
HR support these initiatives? What is the
best way to cur hack on HR time for
3. Talent: How will the workforce change
over the next five years? What are the
implications for your HR programs?
What will managers need from HR to
better manage talent?
4. Expectations of HR: Do you want the
HR delivery model to be “high touch” or
“high tech”? What are the most critical
processes that you want to outsource
Exhibit 3 shows that among those companies that have gone through HRBI’O, the HR
function was able to reduce their administrative focus by more than half, and increase time
spent on strategic business partnering by
approximately 40 percent. Given these opportunities, HR needs to assess carefully “what
stays” and “what goes,” and then determine
the appropriate organizational structure that
will enable efficiency and performance. As
organizations think through the changes in
HR, some critical lessons learned are:
1. In forming reporting relationships, avoid
burying under one person a capability
that is needed by many.
2. Do not confuse shared responsibility witb
overlapping responsibility; jobs should
have clear responsibilities and outcomes.
3. Oeate line of sight between individuals
and the broader HR strategy; keep the
new HR model streamlined
enough so HR staff do not get
stuck in rigid silos.
4. Focus on competencies that communicate what it will take to be
successful in the new environment.,
for example: commumcatmg for
results, relationship management,
and a strategic orientation.
5. Clearly define what will not be
done, then enlist HR professionals in driving the change.
Value Delivered. Moving quickly to
detail the vision, structure, and composition of the new HR organization
is the only way to ensure that HR
moves in lock step with the new HR
direction. It also helps to avoid common traps:
— HR will not let go of transactional work.
— HR will not support behavioral
change for employees/managers.
Getting the Right People with
the Right Skills and the Right
The skills and competencies that
were so valuable in an insourced HR
delivery model—employee advocacy,
deep knowledge of HR programs and
HR technology—are not necessarily
the ones that are most critical in an
outsourced environment. In an
HRBPO environment, the successful
1. Going from paper-based to web-based HR processes;
2. Centralizing HR administration in an outsourced
service center;
3. Redefining the role of HR business partners and
Center of Expertise specialists to focus more closely
on the business.
For VP of e-HR and Benefits PattI Boggi-Gibbons,
it was important to introduce the new HR tnodel in a
way that built support for the change from both
employees and HR. Patti and other HR leaders made
the rounds of the business unit leaders and key line
managers to brief them on the reasons for outsourcing, how the processes would work, and the expected
benefits to Sony. “We made sure that our executives
saw how HR outsourcing fit into the bigger picture of
making SONY a more efficient, streamlined and webbased company. We also made the connection between
HR quality and Six Sigma.” HR relied on extensive
communications—posters, videos, flyers, emails,
personalized letters to the home—to inform employees
about the new HR portal and services. They also ran
training sessions to get people comfortable with the
new tools.
Boggi-Gibbons made a point of sharing information with her HR colleagues and making them a part
of the change: “HR could have seen this as a big
threat. We were upfront about how many jobs we
expected to eliminate. Almost everyone in HR supported the transformation in some form or another.
That’s a toolkit that they can take with them into a
number of different types of roles and companies.”
One thing Sony did to make sure that the peopie who
stayed in HR were up to the task was to offer training
on workforce analytics: “We want to give our HR
Business Partners and COE leaders all the tools they
need to help the business make smart decisions about
the workforce and HR.”
The work is not done for Boggi-Gibbons and her
team. Managing such a significant change has not
been easy; not everyone is convinced that the outsourced iTiodel is the best solution. The HR team is
still involved with some non-strategic areas—what
Boggi-Gibbons refers to as “program delivery support” for annual salary increases, annual incentive
plans, etc.—that need to be more streamlined. Over
titne, howevei; as processes and programs become
more aligned, as employees adjust to their new roles,
and the benefits to the business become apparent, the
change management piece gets easier.
British Petroleum: Enhancing HR Capability through Outsourcing
When British Petroleum (BP) outsourced HR to Exult in 1999, the recently merged company saw retooling and streamlining HR as consistent with broader business strategy. First,
there was a mandate…
Purchase answer to see full

We offer the bestcustom writing paper services. We have done this question before, we can also do it for you.

Why Choose Us

  • 100% non-plagiarized Papers
  • 24/7 /365 Service Available
  • Affordable Prices
  • Any Paper, Urgency, and Subject
  • Will complete your papers in 6 hours
  • On-time Delivery
  • Money-back and Privacy guarantees
  • Unlimited Amendments upon request
  • Satisfaction guarantee

How it Works

  • Click on the “Place Order” tab at the top menu or “Order Now” icon at the bottom and a new page will appear with an order form to be filled.
  • Fill in your paper’s requirements in the "PAPER DETAILS" section.
  • Fill in your paper’s academic level, deadline, and the required number of pages from the drop-down menus.
  • Click “CREATE ACCOUNT & SIGN IN” to enter your registration details and get an account with us for record-keeping and then, click on “PROCEED TO CHECKOUT” at the bottom of the page.
  • From there, the payment sections will show, follow the guided payment process and your order will be available for our writing team to work on it.