Peter Parham talks with
ScienceWatch.com and answers a few questions about
this month's Fast Moving Front in the field of
Article: MHC class I molecules and KIRs in human
history, health and survival
Journal: NAT REV IMMUNOL, 5 (3): 201-214 MAR 2005
Addresses: Stanford Univ, Dept Biol Struct, Stanford, CA
Stanford Univ, Dept Biol Struct, Stanford, CA 94305
Stanford Univ, Dept Microbiol & Immunol, Stanford, CA
Why do you think your paper is highly
There are likely to be several factors at work. The paper provided a unique
synthesis of a new and growing immunological field with direct clinical
relevance, but also strong implications for the evolution of the immune
system and human populations. The main points made in this review go well
beyond the contents of the papers to which it makes reference. It also
highlighted an aspect of the human immune system that is significantly
different from its counterpart in the laboratory mouse, the main
immunological model, and all other mammalian species. Such species-specific
differences, which have major impact on the clinical translation of basic
research, used to be casually brushed aside as minor inconveniences or the
consequences of methodological differences, but are now beginning to be
acknowledged as being of practical importance and scientific interest. I
would like to think that the synthesis provided by my paper, as well as the
spectacular failures of some clinical trials based upon animal models,
contributed to this enlightenment.
The editors of Nature Reviews Immunology, said they liked the
review (possibly because it needed little editing) and they promoted the
paper by having the cover of the journal show artist Neil Smith’s
representation of the bigger evolutionary issues touched on by the paper:
in it maps of the world set the scene; a city of skyscrapers where trees
are but a memory looms at center stage; a parade of exotic animals marches
to extinction in the limelight and Darwin looks grimly on from the wings.
human population there are many
different forms of NK cell
receptors. The effect of this
heterogeneity is that the
combinations of receptors expressed
by the NK cells of unrelated
individuals are different. This
imbues the human population as a
whole with a greater resistance to
the range of virus infections than
any one individual."
Taking a more mundane approach to the extent of citation, reviews are in
general more highly cited than primary research papers, hence the recent
trend for almost all journals to publish reviews as a tactic to boost their
impact factors. Moreover, in writing the introductions to their papers,
many authors use reviews as a convenient way to acknowledge a research
group’s contribution without having to sift through the details of
the primary literature. This raises the unanswerable question as to what
proportion of the citations were made by authors who had actually read the
paper rather than just peering at the Pub-med abstract?
Does it describe a new discovery, methodology, or
synthesis of knowledge?
This paper is a synthesis of the first eight years of research in an area
of NK-cell immunogenetics that was opened up by an investigation from my
group published in two papers in the journal Immunity in December
1997. The background context for that work was the biology of the natural
killer cell (NK), a type of lymphocyte that functions in innate immunity
and which, for many years had been considered by immunologists to be rather
boring and something of a Cinderella to the ugly sisters: the B cells and T
cells. In the 1990s that view began to change with discovery of a variety
of differentially expressed NK-cell receptors, which in their combination
define numerous NK-cell subpopulations.
Capturing most attention were receptors that bind to major
histocompatibility complex (MHC) class I molecules. These glycoproteins
were already well known for presenting peptide antigens to T-cell
receptors, for being the most polymorphic genes in the human genome, and
for causing the rejection of transplanted organs. We had indirect evidence
that the NK-cell receptors, called killer cell immunoglobulin-like
receptors (KIR), that bind to polymorphic MHC class I were also
polymorphic, a property that could account for unexplained donor-specific
effects described either in the literature or the folk-lore of the field.
Our 1997 papers addressed the question directly and demonstrated a degree
of genetic and functional diversity in the KIR gene family that was high
and unprecedented for a receptor of innate immunity. This variation derived
from two factors: haplotypic variation in the numbers of genes encoding
activating and inhibitory receptors, also called gene-content variation,
and allelic polymorphism at the individual genes.
At that time, the immunogenetics of the human MHC (the HLA region) had been
intensively studied for more than 40 years, and numerous HLA factors were
correlated with susceptibility and resistance to a wide range of
infectious, autoimmune, and allergic diseases as well as the outcome of
transplantation. In many of these diseases HLA is the strongest genetic
association, ankylosing spondylitis, a type of arthritis, for example.
Our discovery that the KIR were also highly polymorphic opened up the
possibilities that KIR factors would also be correlated with disease and
that combinations of alleles at the KIR and HLA class I genes, which
segregate independently in the human population, would be more strongly
correlated with disease and provide new insights into disease mechanisms
that had proved notoriously difficult to define.
First reported in 2001 by investigators (Cornelia Weyand, Jorg Goronzy, and
colleagues) at Mayo Clinic was an association of KIR with an autoimmune
disease, rheumatoid arthritis, and this was followed in 2002 by the
demonstration from investigators at the National Cancer Institute (Mary
Carrington, Stephen O’Brien, and colleagues) that a compound genotype
of KIR and HLA class I slowed the progression of human immunodeficiency
virus (HIV) infections to acquired immunodeficiency (AIDS). This trickle
soon became a torrent of KIR associations, with or without HLA, with an
impressive variety of diseases.
By 2005 it was still possible to cite all the papers in the field and stay
within the limits set by Nature Reviews Immunology, today that is
no longer the situation. My goals in writing the review were threefold:
first to extract general and simplifying principles from the clinical
associations, second to place NK-cell receptor variation in an evolutionary
context, both ancient and modern, and third to provide some indication of
how the field might move forward.
Would you summarize the significance of your paper in
NK cells are white blood cells that provide an early defense against viral
infection. For many years NK cells were believed to be homogeneous, but the
research reviewed in this paper changed that view. NK cells are actually
very diverse; they have different combinations of receptors on their
outside surfaces that allow NK cells to interact with other cells of the
body and determine whether they are healthy or infected by a virus. By
selectively killing the infected cells, the NK cells interrupt the
production of virus and prevent further infection of healthy cells. This
heterogeneity allows a person’s population of NK cells to detect many
more types of viral infection than is possible by NK cells with a single
Within the human population there are many different forms of NK cell
receptors. The effect of this heterogeneity is that the combinations of
receptors expressed by the NK cells of unrelated individuals are different.
This imbues the human population as a whole with a greater resistance to
the range of virus infections than any one individual.
Comparison of different mammalian species showed that each species has a
unique set of NK cell receptors. Thus the human receptors are quite
different from those of the mouse and even show significant differences
from the chimpanzee, our closest relative in the living animal kingdom.
These comparisons demonstrated that these receptors are evolving rapidly.
This is consistent with their job in fighting viral infections. Viruses
such as those causing influenza are always changing in ways that escape
human immune systems. By using new forms of NK cell receptor the human
population can keep up with changes in the virus. In conclusion, the NK
cell is now seen to be a highly diverse and dynamic white blood cell that
is constantly changing to adapt to new and emerging infections.
How did you become involved in this research and were
any particular problems encountered along the way?
Since I was a graduate student in the 1970s my research had focused on the
genetics and diversity of human MHC class I. These molecules present
peptide antigens to cytotoxic T lymphocytes and up until the 1980s that was
considered their only function. At around that time, various lines of
evidence were pointing to an MHC class I involvement in the NK cell
response and in 1991, I was invited to attend the Seventh International
Workshop on Natural Killer Cells in Stockholm to give a didactic lecture on
MHC class I. As I found out at the meeting, the relevance of MHC class I
was hotly disputed, but from the evidence presented there it seemed to me
to be the most parsimonious interpretation by far.
That workshop planted several seeds, including a conversation with Lewis
Lanier about possible collaboration. Not long afterwards, Jenny Gumperz, a
rotating graduate student, declared that she wanted to do her doctoral
research in my group on the topic of NK cell receptors for MHC class I.
Remembering our Stockholm conversation, I suggested Jenny ask Lewis, who at
that time was working just down the road from Stanford at the DNAX Research
Institute, to become a member of her Ph.D. thesis committee and to
collaborate with us on the project. That was how I got involved in this
research, which is now my major interest.
The biggest problem along the way was a conceptual one. In cytotoxic T-cell
responses, engagement of the MHC class I receptor generate positive signals
that cause the target cell to be killed. Whereas for NK cells, engagement
of the MHC class I receptor generates negative signals that prevent killing
of the target cell. Adapting to this seemingly simple difference took time,
because in the design of experiments and interpretation of data it required
thinking in a kind of reciprocal space of double negatives.
Where do you see your research leading in the
As was discussed in the paper, NK cells and their variable receptors have
major influences on success in defeating infection and success in
reproduction. Both these activities are absolutely essential for the
survival of a species. A recurring characteristic in the immunogenetics of
human MHC class I and KIR is an apparent balance between two distinctive an
opposing forms. To give just one example, the KIR gene family exists in two
forms, A haplotypes and B haplotypes, that are distinguished by their
relative content of genes for activating and inhibitory receptors.
This could arise because there is advantage to being a heterozygote for
these two forms, or alternatively these two forms might have advantages
that are beneficial at different times and in different places. Thus one
form might favor success in reproduction and be less good for immune
defense, while the other is better at immune defense and not so fit for
reproduction. In the future I hope to understand how the competing
pressures for reproduction and immune defense produce the patterns of KIR
and MHC class I variation observed in the human population.
Do you foresee any social or political implications for
There are medical implications for our work in bone marrow transplantation
and hematopoietic cell transplantation for leukemia and other malignant
diseases. At present, in choosing donors to match recipient patient only
the MHC variability is taken into account. By also considering KIR
variability in the choice of donors it should be possible in some
situations to improve graft survival and the patient’s quality of
life. Currently there is also interest in the potential use of NK cell
transfusions from healthy donors as short-term therapies, in a manner
analogous to the well-established blood transfusion. Here again knowledge
of both KIR and MHC type could inform the choice of donor.
Both the KIR and MHC systems of diversity exhibit considerable differences
between human populations of different ethnicity and history. Understanding
these differences and their implications for human health and survival will
only come from the study and comparison of all populations. For over 40
years, the field of human MHC genetics has been both highly "globalized"
and highly collaborative. Through a series of international workshops it
set the gold standard for international co-operation in science and for the
establishment and maintenance of a useful and consistent nomenclature. This
experience and template is now being used in the study of KIR.
Peter Parham, Ph.D.
Professor of Structural Biology and Microbiology & Immunology
Stanford University School of Medicine
Department of Structural Biology
Stanford, CA, USA