Preface
Our laboratory is interested in mutations, not only their roles in human cancers
but also how these processes can be harnessed to destroy pathogens.
While mutation is often thought of in a negative context, mutations at the
DNA level can also be protective and beneficial. In the past three years,
our research has been instrumental in the discovery of two fundamental mutation-generating
mechanisms. In each, the purposeful use of mutation governs a potent pathogen
defense. One mechanism protects our cells from retroviruses such as
HIV-1, and the other helps strengthen our antibodies such that an effective
immune response occurs when we are infected. The common link between these
important immune defenses systems is that they are underpinned by related
human proteins that use identical strategies --- the chemical conversion of
the DNA base cytosine into uracil.
Research Overview
The human immune system uses two major levels of defense.
On the front line are the built-in, ‘innate’ defense mechanisms,
which are often capable of immediately neutralizing invading molecules such
as viruses. In reserve, however, are powerful, ‘adaptive’ defense
systems, which are capable of invader recognition and ultimately contribute
to specific and long-term defenses. Both the innate and the adaptive immune
processes are critical for the provision and maintenance of our good health.
Remarkably, mutation by DNA cytosine deamination is an essential part of both defenses. Cytosine is one of the four Watson-Crick DNA bases. In the DNA double-helix, cytosine normally ‘base-pairs’ with guanine. However, removal of an amine group (NH2) from cytosine converts it to uracil, a base normally found only in RNA. Uracil, like its DNA-based analogue thymine, base-pairs with adenosine. Thus, the conversion of DNA-based cytosine to uracil, followed by the fixation of the uracil lesion by other cellular processes, can result in a specific type of base substitution mutation, a C/G to T/A transition mutation.
Our cells have at least two enzymes, APOBEC3F
and APOBEC3G, which block retroviral infection by deaminating cytosines in
retroviral DNA. Retroviruses like HIV-1 are dependent on host-cell functions
for their replication. Retroviruses use a multi-step process to copy their
RNA genome into a DNA double-helical form that can insert in our DNA and thereby
immortalize the virus. This strategy is a major barrier in ‘curing’
AIDS because it enables the virus to hide in our DNA, effectively evading
immune defenses. The first step requires the virus to synthesize a DNA version
of its RNA genome. Then the RNA is degraded and a second DNA strand, complementary
to the first, is made. The finished product, a double-helical viral DNA, can
insert randomly in our genome. APOBEC3F and –3G exploit this strategy
by deaminating cytosines in the first viral DNA strand, such that when the
second strand is copied the resulting uracils cause the insertion of adenines
(instead of guanines) and so many mutations result that the virus is effectively
destroyed (Figure 1). This mechanism is clearly important in HIV infected
patients, as the mutational ‘signatures’ of APOBEC3F and –3G
are visible in retroviral DNA isolated from these individuals.
Figure 1: APOBEC3G
and Vif are key determinants of retroviral infectivity. This figure
presents a model that depicts how the APOBEC proteins and Vif (virion infectivity
factor) can influence the retroviral life cycle. APOBEC3G (red) that is expressed
in the producer cell is incorporated into the budding virion together with
other components of the virus, including its genomic RNA. HIV-1 Vif (green)
can reduce or eliminate APOBEC3G incorporation into budding virions by targeting
it for proteasomal degradation. However, should APOBEC3G escape Vif, gain
access to the virion and subsequently reach a target cell, it can deaminate
cytosine residues in the first retroviral DNA strand (blue). The resulting
uracil residues function as a template for the incorporation of adenine, which,
in turn, can result in strand-specific C/G to T/A transition mutations that
affect virus viability. Uracil residues have also been hypothesized to trigger
degradation of the retroviral DNA before it can integrate into the host-cell
genome, although this hypothesis awaits rigorous experimentation. LTR, long
terminal repeat; RT, reverse transcriptase. Figure reproduced from Harris
and Liddament, 2004; Nature Reviews Immunology 4, 868-877.
Of course, these APOBEC proteins are not 100% effective at inhibiting infection because the disease caused by HIV-1 infection, AIDS, remains a prominent global threat. This can be attributed in part to the fact that HIV-1 has an effective counter-defense mediated by one of its own proteins, Vif, which usually mediates the destruction of the APOBEC proteins before they can attack retroviral DNA (Figure 1). Thus, we hypothesize that the outcome of an infection can be determined by the winner of the APOBEC-Vif battle. We envisage that specific chemical inhibitors of Vif, or protectors of APOBEC3F and –3G, can be developed and used to treat infected individuals. However, a pre-requisite to such a treatment, is a fundamental understanding of the basic activities of these APOBEC proteins. This is a major goal of our laboratory.
In adaptive immunity, some lymphoid B cells
are capable of producing antibodies, proteins which both neutralize the pathogen
directly and trigger cell-based defenses. In order for the antibody response
to be highly effective, antibodies are matured in such a way that an initial
low-strength recognition of the pathogen becomes up to a thousand times stronger.
This remarkable increase in strength or ‘affinity’ at the protein
level is programmed at the DNA level by mutations in the antibody genes themselves,
a process called somatic hypermutation. A protein called AID (activation-induced
deaminase) underpins this antibody strengthening process by deaminating cytosines
in antibody gene DNA. The majority of evidence accumulated thus far indicates
that AID triggers antibody affinity maturation by DNA cytosine deamination.
AID also uses this activity to trigger antibody diversification by gene conversion
in some vertebrates (e.g., birds) and antibody class switching (Figure 2).
We are interested in the post-translational regulation of AID as this may
very well provide the key to how its activity can be directed specifically
to the antibody genes and how its activity can result in such remarkable distinct
events.
Figure 2: Programmed
immunoglobulin-gene diversification by AID. Activationinduced deaminase
(AID)-dependent cytosine deamination of functional immunoglobulin genes triggers
gene conversion, somatic hypermutation and class-switch recombination. Uracil
recognition and processing by the cellular uracil DNA glycosylase UNG2 is
a pivotal step in these processes. V, D, J, S and C denote the immunoglobulin-gene
variable, diversity, joining, switch and constant regions, respectively. Figure
reproduced from Harris and Liddament, 2004; Nature Reviews Immunology 4, 868-877.
Both the innate and the adaptive mutational responses appear remarkably specific, APOBEC3F and –3G to retroviral cDNA and AID to antibody gene DNA, occurring at frequencies up to a million-fold above levels at other genomic loci. One of our primary research goals is to understand how these proteins are so exquisitely targeted to their DNA substrates, thereby leaving other possible targets in our genome unscathed. The molecular answers to these mysteries will undoubtedly advance our understanding of the mechanisms behind these important immune processes and they will also help us understand how these potent mutator proteins might also contribute to carcinogenesis when misregulated.
