ARTICLE
Mucocutaneous surfaces are constantly exposed to an array of exogenous
antigens including environmental proteins, peptides and low molecular
weight and microbial pathogens. These tissues are covered by an epithelium
which exerts both the role of a barrier, limiting the penetration of microbes
and of hydrophylic antigenic moieties, but at the same time ensures that
antigens which penetrate through the epithelium are rapidly captured and
transported to draining lymph nodes for initiation of a specific immune
response. Epithelial dendritic cells represent the immunocompetent cells
responsible for the dynamic uptake and presentation of antigen entering
peripheral tissues, and are unique in their efficiency in triggering the
immune system and in initiating a primary immune response.
The skin: a site of immune surveillance
The skin epithelium, the epidermis, is a stratified squamous keratinized
epithelium formed by multilayers of keratinocytes (KC), covered by a corneal
layer preventing absorption of macromolecules and pathogens. KC are resident
cells of non hematopoietic origin, tightened together at the lateral side
by desmosomes and at the basal side by hemidesmosomes. The basal KC layer,
in contact with the underlying dermis, contains slow dividing stem cells,
which differentiate into more superficial and rapidly dividing KC. The
skin contains professional antigen-presenting dendritic cells, including
Langerhans cells (LC) in the epidermis and dermal dendritic cells (DC)
in the dermis. LC, morphologically characterized by their long and slender
cytoplasmic processes, from a network of cells protruding their dendrites
between KC. Following antigen entry through the skin, LC are rapidly recruited
from dermal DC precursors to the epidermis, where they capture and internalize
the antigen by endocytosis and start processing native antigen into peptides
which are retained in class II vesicles of the endocytic pathway as well
as in the endoplasmic reticulum. Antigen capture by LC induces morphological
changes and mobilization of the LC, which emigrate through afferent lymphatics
(as veiled cells) to the T cell area of draining lymph nodes, where they
can be identified as interdigitating DC, forming close contacts with lymph
node T cells. At this stage they have acquired expression of costimulatory
molecules of the B7 family, upregulated surface expression of MHC class
I and class II molecules bound to peptides, and the ability to secrete
high levels of proinflammatory cytokines such as IL-12, IL-1. Thus, interdigitating
DC are highly specialized in the activation of naive T cells into antigen-specific
T cells.
DC, in contrast to macrophages, are unique in their capacity to prime
naive T cells against soluble antigens administered in the absence of
an adjuvant. Indeed, although monocytes/macrophages are professional APC
capable of the processing and presentation of live or non replicating
antigens to activated and memory T cells, they are unable to prime naive
T cells efficiently. This is due to the fact that macrophages do not traffic
from peripheral tissues to secondary lymphoid organs but are instead attracted
at the site of the antigen inflammation, where upon the release of proinflammatory
cytokines and chemokines they induce activation of pre-existing antigen
specific T and B (memory) cells, recruited from the circulation. Thus,
macrophages contribute to antigen presentation during secondary immune
responses, but are generally inefficient at priming naive T cells because
they do not transport antigen from peripheral sites to T cell areas of
secondary lymphoid organs. The migratory property of LC from epithelial
tissues to draining lymph nodes together with their unique dynamic changes
in antigen processing and presentation functions allow the immune system
to recognize foreign antigens encountered in peripheral tissues [reviewed
in 1, 2]. In addition, DC were shown to internalize macromolecules by
an endocytic pathway which intersects the endogenous cytosolic pathway,
allowing class I presentation of exogenous non replicating antigens. Thus,
exogenous antigens not synthetized within the DC could elicit cytotoxic
T cell responses [3].
During recent years, it has become increasingly clear that manipulation
of the immune system for therapeutic (i.e. anti-tumoral) or vaccination
(anti-infectious) purposes, requires immunization routes allowing efficient
antigen uptake by DC. The feasibility of antigen delivery through the
skin for the induction of protective immunity against infections has emerged
from the field of vaccinology and more recently illustrated by the delivery
of sub-unit vaccines. The concept of sub-unit vaccines is based on the
idea that harmful complications and potential reversion to virulence,
which may occur after vaccination with live attenuated virus or bacteria,
thereby limiting their use in immunocompromised individuals, may be overcome
by vaccination with antigenic components of the pathogens in the form
of recombinant proteins, synthetic peptides, or DNA encoding for the antigen
of interest.
Transcutaneous DNA vaccination
Basic principles of DNA vaccination
DNA vaccination is based on in vivo transfection of host cells
with a bacterial DNA plasmid encoding the antigen of interest which can
be expressed in host cells, in a way that is similar to that occuring
after natural infection.
Plasmid DNA can be delivered in three different forms:
"naked DNA", consisting of plasmid DNA in solution, is used for
immunizations by cutaneous or intramuscular injections;
plasmid DNA coated onto gold particles, is used for transepidermal
immunization by skin bombardment under helium pressure using a gene-gun;
plasmid DNA encapsulated into inert vectors such as biodegradable
microspheres, are suitable for cutaneous injection while live (viral or
bacterial) vectors, which can infect skin cells are appropriate for epicutaneous
immunization.
The plasmid
The bacterial plasmids employed for DNA immunizations generally contain
a procaryotic origin of replication and an antibiotic resistance gene
suitable for propagation in E. Coli. The gene of interest is under
the control of a viral promotor and is followed by a mRNA termination/polyadenylation
sequence allowing strong expression in mammalian cells (Fig.
1). The plasmid DNA is not infectious and is incapable of replication
in eucaryotic cells, is unable to integrate into the host genome and remains
episomal in the nucleus of transfected host cells. The encoded antigen
is biosynthetized in host cells with native post-translational modifications
and protein conformation. Expression of the antigen may persist for prolonged
periods of time in cells that are slowly dividing.
Mode of DNA delivery through the skin
The skin is an ideal target for DNA immunizations, because it contains
numerous and readily accessible bone-marrow-derived LC and dermal DC,
specialized in the initiation of the immune response. DNA vaccination
through the skin has been performed by coupling recombinant plasmid DNA
with < 1 µm gold particles and administration through the skin
by bombardement under helium pressure using a gene-gun. Transdermal injection
of naked DNA in PBS has also been reported, but is less efficient for
priming immune responses, presumably due to the limited uptake by phagocytic
cells, compared to DNA coated onto gold beads. Bacterial DNA introduced
into viral vectors which can infect a variety of cell types including
skin cells, can be administered epicutaneously (see chapter about topical
DNA vaccination).
In vivo priming of a protective immune response
An increasing number of studies have demonstrated that immunizations
with plasmid DNA promote effective immune responses against many bacteria,
virus and parasites in rodents [reviewed in 4]. DNA immunization through
the skin appears as the most efficient way to prime anti-viral protective
CTL responses [5]. Several routes of inoculations of plasmid DNA expressing
influenza virus hemagglutinin (subtype H1) have been compared for their
ability to induce CTL and antibody responses, and protection against challenge
with a mouse adapted influenza virus, expressing the same H1 subtype.
These include: (i) intramuscular injection of naked DNA, a route that
permits efficient transfection of resident cells, (ii) subcutaneous and
intraperitoneal injections, routes that result in less efficient transfection
but are frequently used for antigen administration to a test animal, (iii)
the epidermis and the upper respiratory tract (nares and trachea) which
result in less efficient transfection but deliver DNA to tissues with
high levels of local immune surveillance. Mice were immunized twice at
4 week intervals and were challenged on day 10 after the last immunization,
by inhalation of the virus into the lung. Apart from the intraperitoneal
route, each of these routes gave rise to at least some protection. Intramuscular
or intravenous inoculations of naked DNA in saline gave excellent protection
but required large quantities (around 100 µg) of DNA per immunization.
Alternatively, mice receiving naked DNA subcutaneously or intradermally
had only marginal protection (65-75% survival) and more severe signs of
influenza. Transepidermal gene-gun delivery of DNA coated onto gold beads
was by far the most efficient method, since protection was achieved with
200-2,000 times less DNA than direct inoculation of DNA in saline. Although
expression was transient, and lost in 2-3 days due to normal sloughing
of the epidermis, as little as 0.4 µg of DNA was sufficient to achieve
95% survival from lethal influenza challenge. These survivors developed
limited-to-no signs of post challenge influenza, in contrast to good survival
but more severe influenza in mice receiving naked DNA intranasally.
Similar to intramuscular naked DNA injection, bolistic immunization
with DNA coated onto gold beads induces high levels of antibodies to the
encoded antigen. However, in different viral systems, the gene-gun induced
a predominantly IgG1 (Th2) response with IL-4-producing cells, while intramuscular
or intradermal injection of naked DNA induced an IgG2a (Th1) response
[6, 7] with expansion of IFN-gamma-producing CD4+ T cells and
CD8+ CTL. The basis for these divergent responses may reflect
the different gene transfer methodologies. Indeed, at least 100 µg
of DNA in saline are required for naked DNA injections compared to nanogram
quantities of DNA coated onto gold beads for bolistic immunizations. Therefore,
injected DNA would provide more bacterial plasmid sequences containing
immunostimulatory CpG motifs (ISS), which are known to promote production
of proinflammatory cytokines (IL-12 and IFN-alpha) inducing a Th1-oriented
immune response.
That gene-gun vaccination provides protective anti-tumoral CTL responses
has been also demonstrated [8]. A single immunization by gene-gun delivery
to mice abdominal skin of 1 µg DNA-encoding OVA coated to gold particles,
generated OVA-specific CTL activity mediated by class I-restricted CD8+
T cells, as detected in indirect CTL assay, after 7 day restimulation
of splenocytes with a syngeneic OVA-transfected lymphoma line (EG7-OVA).
Two gene-gun immunizations with a total of 2 µg of DNA-encoding OVA,
conferred protection against lethal intradermal injection of OVA-transfected
B16 melanoma, while tumors were lethal in all control mice immunized with
DNA encoding the irrelevant antigen ß-galactosidase. Since the B16
melanoma is non immunogenic in C57BL/6 mice and still remains non immunogenic
after OVA transfection, and since OVA endogenously synthetized by OVA-transfected
B16 generates the epitope SIINFEKL associated with the class I molecule
Kb molecule, these data indicate that DNA immunization induces
antigen-specific, CTL-dependent protective tumour immunity.
Mechanisms of priming of
the immune response
Transfection efficiency versus vaccination efficiency
One of the most striking results of DNA vaccine trials in animal models
is that the efficiency of transfection does not necessarily determine
the vaccination efficiency. For example, the high ability of rodent muscle
to take up and express DNA did not result in better antibody production
and protection than intravenous or intranasal inoculations [5]. Thus,
it is more likely that transfection of even a limited number of professional
APC, such as DC, is preferred over transfection of a large number of resident
cells for inducing protective immunity. That DNA inoculation can induce
immunological memory is illustrated by the fact that serum titers of specific
IgG antibodies are quite low on day 10 after boost, but rapidly increase
after challenge, indicating mobilization of memory cells.
In vivo transfection of DC
Characterization of the nature of the in vivo transfected cells
after gene-gun immunization with DNA has been performed using DNA encoded
with reporter genes such as the green fluorescence protein (GFP) or ß-galactosidase,
whose expression in the transfected cells can be easily detected [8].
Gene-gun immunization with DNA results in in vivo transfection
of both resident KC but also of DC which are highly enriched in the skin.
Direct evidence for the uptake of DNA by epidermal DC has, however, not
been provided, most likely due to their limited number and their rapid
emigration from the skin to draining lymph nodes. However, electron dense
1 µm gold particles have been detected in the cytoplasm of interdigitating
lymph node DC, 24 hrs after bolistic immunization [8]. Gene expression
by lymph node DC was demonstrated by the green fluorescence of cells with
dendritic morphology within T cell areas of the lymph nodes resulting
from expression on the green fluorescence protein (GFP) encoded in the
DNA used for immunization. Evidence that these in vivo transfected
DC originated from the skin was further provided by experiments in which
skin was painted with the fluorescent hapten rhodamine, immediately before
bolistic immunization. Twenty-four hours later, clusters of skin-derived
cells (red fluorescence) were evident in the lymph node in the region
of afferent lymph flow, and several of these cells were double positive
cells (expressing GFP and rhodamine). One cannot formally exclude the
possibility that DNA-beads could have travelled through lymphatics to
draining lymph nodes and be captured by LN DC or to hepatic lymph and
spleen, where they could have been phagocytosed by immature DC, also present
at these sites. In this respect, DNA-beads injected subcutaneously can
gain access to the cytosol of phagocytic cells. However, because DNA coated
onto gold is rapidly solubilized (> 95% within 3 min) in aqueous media,
it is unlikely that DNA could survive trafficking to LN uptake by resident
DC and endosomal transport. Although solubilized DNA may traffic to lymph
nodes and may be captured by DC independently of beads, this could not
account for colocalization of beads and DNA expression in the same DC.
Relative role of keratinocytes (KC) and dendritic
cells (DC) to the induction of specific immunity and memory
When DNA vaccines are administered by gene-gun bombardment of the skin,
the majority of the plasmid is taken up by KC [9]. It has been suggested
that non migratory cells do not contribute to the development of immunity
since primary antibody and CTL responses can be generated despite immediate
removal of the site of vaccination [10, 11]. It has been shown that bone-marrow-derived
APC contribute to this process [12]. Consistent with this hypothesis,
DC in the skin were recently shown to take up DNA-coated beads and migrate
within 24 hrs to the draining lymph nodes where primary immunity develops
[8]. Recent studies have addressed the role of cells at the site of skin
bombardment to the magnitude of the primary response and the generation
of immunological memory. To this end, transfected skin was periodically
removed and transplanted onto naive recipients. Immediate removal of the
skin site prevented the outcome of a primary immune response in the vaccinated
mice, while both primary immunity and memory could be induced in naive
recipients engrafted with skin transferred 0-24 hrs post-vaccination.
Thus, skin DC are responsible for both the priming and maintenance of
immunological memory. However the magnitude of the primary immune response
increased the longer the vaccination site was left in place (for up to
2 weeks). Since transfected KC continue to produce the antigen for up
to 2 weeks, this indicates that KC influence the magnitude of the response
[13]. Further studies indicate that directly transfected DC exert a predominant
role in antigen presentation to CD8+ T cells after gene-gun
immunization. Although 24 hrs after immunization the number of lymph node
cells directly expressing the transfected DNA are rare (50-100 per individual
lymph node), there is a two fold increase in the number of CD11+
DC in the lymph nodes (20-30,000 DC/lymph node). This augmentation is
due to gold bombardement and is independent of the presence of plasmid
DNA. Using a mutant influenza NP gene which needs the costimulatory B7.2
gene for expression, it was also shown that directly transfected cells
were involved in CD8 T cell priming. Only mice immunized with beads co-coated
with both plasmids developed NP specific CD8+ T cells, while
no priming occured in mice immunized with each plasmid separately. This
indicated that antigen presentation was mediated by directly transfected
DC (Fig. 2) rather than
by cross priming due to phagocytosis of transfected cells by DC. Thus,
epidermal DNA immunization involves the small number of directly transfected
DC migrating to lymph nodes rather than a much larger number of migrating
DC that could potientially present the antigen expressed in epidermal
cells via a cross-priming mechanism [14].
Adjuvanticity of plasmid DNA: immunostimulatory
DNA sequences (ISS)
DNA vaccines administered in saline are effective in preclinical animal
models without the need for adjuvants or delivery systems. Part of this
effectiveness may be due to the immunostimulatory effect of the bacterial
DNA itself. Specific nucleotide sequences, including unmethylated CpG
motifs are immunostimulatory for lymphocytes and are present in naked
DNA used for vaccination. For intradermal immunization, incorporation
of such stimulatory motif into a plasmid increased both humoral and cellular
responses for a weakly antigenic protein ß-galactosidase, encoded
by the same plasmid or a co-injected plasmid [15]. Furthermore these ISS
are responsible for the Th1 response generated after naked DNA injection
in the dermis, inasmuch as they suppress IgE production but promote IgG
and IFN-gamma production. They further initiate the production of IFN-alpha,
IFNß, IL-12 and IL-18, all of which foster Th1 responses and enhance
cell-mediated immunity [16]. KC and dermal DC transfected with the ISS-containing
DNA plasmid could produce IL-12 and IFN-alpha, involved in the induction
of a Th1 response against the encoded protein. In this respect in vitro
studies have shown that CpG-containing oligonucleotides induce activation
and maturation of a LC-like fetal skin-derived DC and stimulate production
of large amounts of IL-12; injection of CpG nucleotides into the dermis
also led to enhanced expression of MHC class II and CD86 (B7.2) molecules
by LC in the overlying epidermis with accumulation of
IL-12 in a subset of activated LC [17].
It should be noted that the nucleotides flanking CpG motifs also play
a role by influencing the nature of the cytokine produced. In addition,
the presence within the plasmid of ISS inhibitory sequences may abrogate
the immunostimulatory effect of CpG motifs [18].
Thus, the plasmid used for DNA immunizations, itself functions as an
adjuvant or an immunomodulator, and altering the nucleotide sequence of
the vector may affect the immunogenicity of DNA vaccines. Although ISS
are necessary for gene vaccination, they are unnecessary and may be harmful
for gene replacement therapy, because they down regulate gene expression
as a consequence of induction of IFN-alpha production by transfected cells.
CD4 dependency of DNA vaccination
Intramuscular DNA immunization can prime CD4+ T cells and
help antibody responses by class II presentation of exogenous antigens
released by in vivo transfected cells. Studies have shown that
CTL induced by DC pulsed with class I peptides require presentation of
class II epitopes and CD4+ T cell activation [19]. In contrast,
other studies reported that CTL induced by bolistic immunization using
gold beads coated with DNA encoding only a class I-restricted epitope
[20] or multiple CTL epitopes appeared to be generated independently of
CD4 help [21]. It is possible that antigen presented by directly transfected
DC are far more efficient at generating CTL precursors compared to DC
pulsed with exogenous class I peptide, not only because of the presence
of higher numbers of MHC class I/peptide complexes but also because of
continuous antigen synthesis providing repetitive stimulations of naive
T cells.
Increased efficiency of
transcutaneous DNA vaccines
Co-inoculation of plasmid expressing cytokines
The efficiency of co-inoculation with plasmids expressing cytokines
on the modulation of the immune response to a viral protein encoded in
a separate plasmid has been tested by intramuscular route. It was shown
that a plasmid vector encoding GM-CSF enhanced the antibody response to
a vector encoding rabies G protein, only if both plasmids were injected
simultaneously but not if the plasmids were injected separately, even
several hours apart. This indicated that either co-transfection of individual
cells with both plasmids or close proximity of the APC to GM-CSF secreting
cells was crucial, presumably reflecting the localized activity of the
cytokine. Co-administration of plasmids encoding GM-CSF resulted in enhanced
production of
IL-2 and IL-3 but not IL-4, suggesting that Th1 cell response was enhanced.
Moreover the presence of a plasmid encoding GMCSF improved the survival
of vaccinated mice to intramuscular challenge with rabies virus, from
40% to 80-100% protection [22]. Neutralizing antibodies protected the
mice from peripheral challenge by preventing the virus from reaching the
central nervous system. Conversely co-immunization with a plasmid encoding
mouse IFN-gamma, failed to enhance rabies specific T cell response in
vitro and even decreased specific antibody titers. This most likely
reflects the inability of transfected myoblasts to act as APC for stimulation
of the immune response, even in circumstances which would favor induction
of MHC class II molecules [22]. Epidermal gene-gun immunization with plasmids
encoding IFN-alpha and IL-12 along with antigen-encoding plasmids are
able to enhance CTL activity and shift the normally observed Th2 bias
of the immune response towards a Th1 phenotype (LD Falo, manuscript in
preparation).
Targeting DNA to sites of immune induction
Because the availability of antigen in lymphoid organs is important
in generating an immune response, it has been postulated that antigen
availability may also be important in the response to DNA vaccines, because
immune responses are stronger when antigen is secreted from DNA-transfected
cells. In order to direct expression of the transgene to lymphoid organs,
vaccination was performed using DNA encoding antigen-ligand fusion proteins.
The model antigen human IgG was targeted to lymph nodes or APC by two
ligands, L-selectin or CTLA4-Ig, respectively binding to receptors that
are present on endothelial venule cells of lymph nodes and on antigen
presenting cells.
L-selectin, expressed on the surface of naive lymphocytes and by binding
to CD34 on high endothelial venule cells, initiates entry into lymph nodes,
CTLA4 on activated T cells binds to B7-expressing APC, which are required
to initiate the immune response. Intramuscular injection of DNA encoding
each of these fusion proteins dramatically enhanced the humoral response
to human IgG (by 104 fold within 2 to 8 weeks) as well as T
cell proliferative responses. Targeting of the DNA vaccine to endothelial
venules by L-selectin enhanced the Th1 response (IgG2a and IgG2b) that
was observed after intramuscular immunization with plasmid encoding Hu
IgG alone, while targetting to APC by CTLA4Ig induced a switch towards
a Th2 type response as shown by a 7,000 fold increase in IgG1 level [23].
Another approach to increase the efficiency of DNA immunization is targeting
DNA to APC. As several types of APC, including DC, phagocytose particulate
material in the micron size range, plasmid DNA has been encapsulated in
biodegradable polylactide-co-glycolide (PLGA) microspheres. Such microparticles
when engulfed by APC in vitro showed expression of the encoded
protein within 24 hrs and up to 3 days. In addition, microspheres containing
2 µg of plasmid DNA encoding for a single class I epitope of Vesicular
Stomatitis Virus, elicited a VSV-specific CTL in vivo, when injected
subcutaneously. It is interesting to note that the CTL response induced
by a single immunization with 2 µg of encapsulated DNA was higher
than that induced by two intramuscular injections with 200 µg naked
DNA or VSV peptide in complete Freund adjuvant. The data illustrate that
the efficiency of DNA encapsulation may be due both to the protective
nature of the polymer coating and to the increased uptake of DNA by phagocytic
APC [24].
Epicutaneous immunization with DNA or proteins
Topical DNA vaccination by utilization of viral
vector
Viral vectors such as adenovirus, which has a broad host range and can
infect non dividing cells, offer a tool for gene therapy of skin diseases.
Adenovirus carrying plasmid DNA encoding ß galactosidase, can be
delivered topically on to the skin of mice. This is performed by first
tape-stripping of the corneal layer, then application of the recombinant
virus onto 1 cm2 of skin by an occlusive technique. Transduction
of the antigen appears by 2-3 hrs and within 24-48 hrs the antigen is
expressed on the entire skin, including all keratinocytes and some follicular
cells, with a mild mononuclear cell and neutrophil infiltrate of the epidermis.
At day 2, the level of ß-galactosidase expression in the skin after
topical adenovirus vector is 14 fold higher than that observed after parallel
gene-gun immunization with the same plasmid. In addition, topical skin
immunization for 2 days with an adenovirus vector expressing TGF-alpha,
results 4 days later in hyperkeratosis and acanthosis demonstrating that
the transduced gene is biologically active [25]. This demonstrated that
replication defective viral vector can be used for efficient topical skin
immunization.
Epicutaneous immunization with proteins
The skin's great barrier properties limits the penetration of macromolecules
greater than 500 Da, thus preventing epicutaneous delivery of the high
molecular weight therapeutics as well as non invasive transcutaneous immunization.
However, it has been recently shown that transcutaneous immunization can
be achieved by simple application of protein antigens mixed with cholera
toxin (CT) onto the shaved mouse skin. CT is a member of ADP-ribosylating
bacterial exotoxins, widely used experimentally as an adjuvant to enhance
immune response to vaccine components, by the oral and the nasal routes.
CT binds to asialo-GM1 ganglioside expressed on a variety of epithelial
and hematopoietic cell types through its B subunit. It was observed that
two epicutaneous applications of 100 µg of CT alone, performed 3
or 8 weeks apart, induced specific serum IgG. Furthermore, CT could act
as a skin adjuvant for the common vaccine components diphteria toxoid
and tetanus toxoid, by promoting induction of specific antibodies to the
vaccines. No sign of redness or swelling of the skin at the site of application
were observed up to 72 hrs post- immunization and skin biopsies taken
at the site of exposure revealed no sign of inflammation. Thus, transcutaneous
immunization with large proteins without physical penetration of the skin
by needles can be achieved using CT. In addition, conversely to administration
by mucosal routes, CT is non toxic when applied epicutaneously, without
skin disruption [26].
Moreover, it was also shown that epicutaneous immunization with CT alone
induced anti-CT IgG and IgA antibodies in the serum, lung washes and stool
samples and that immunized mice were protected from an intranasal challenge
with a lethal dose of CT. Therefore epicutaneous CT administration induced
clinically relevant immunity against mucosal toxin challenge [27]. However,
although CT has been shown to induce a predominantly Th2-biased immune
response, it has been shown that IgE were also induced when CT was used
as adjuvant. More careful examination of IgE responses in these studies
has not been undertaken. This is, however, of major concern for skin immunization
with proteins, particularly because of the potential risk of inducing
immediate hypersensitivity reactions.
There is one exception to the theory that proteins cannot normally penetrate
through the skin: the skin of patients with atopic dermatitis which exhibits
deficient barrier functions [28]. Indeed, eczematous reactions can develop
after epicutaneous application of protein allergens on the non lesional
skin of atopic dermatitis patients, demonstrating that high mw proteins
could penetrate through this type of skin and induce an allergen-specific
DTH response [29]. These observations suggested that the skin, like the
gut and the lungs, could be a site of sensitization to environmental protein
allergens in these patients. In this respect, two independent studies
in normal inbred strains of mice have shown that epicutaneous exposure
to ovalbumin, in the absence of adjuvant, can sensitize the animals and
induce a dominant Th2-like response with high levels of specific IgE.
Moreover, repeated immunizations sustained elevated levels of IgE. The
protocole of epicutaneous immunization used was performed by applying
the immunizing solution on a patch left in place for 3 consecutive days
on the shaved back skin and immunization was renewed by application of
a freshly prepared patch for another 3 days [30]. Similar observations
were reported by another group who further showed that mice exposed to
3 one week 100 µg OVA skin patches, separated from each other by
2 week intervals, not only developed IgE responses, but also developed
dermatitis with skin infiltration with CD3+ T cells, eosinophils
and neutrophils. RNA for IL-4, IL-5 and IFN-gamma were detected in the
skin. More strikingly, a single epicutaneous exposure to OVA induced eosinophilia
in the bronchoalveolar lavage fluid, and airway hyperresponsiveness to
intravenous methacholine [31]. These data suggest that epicutaneous exposure
to protein antigens in atopic dermatitis may be involved in the development
of allergic asthma. Furthermore, this also raises questions about the
feasibility of epicutaneous protein antigen delivery for vaccination purposes.
The Th2-biased immune response following protein application contrasts
with the Th1-biased response generated by topical exposure to sensitizing
haptens. The explanation for these different types of immune response
are not clear. It is possible that the higher number of MHC/peptide complexes
generated by hapten immunization as compared to those generated by protein
processing, reflects the predominant Th1 response after contact sensitization
with haptens compared to the Th2 response generated with proteins. In
addition, contact sensitizers are all irritants and induce inflammation
while proteins are not inflammatory unless administered with adjuvants.
Finally, the nature of the protein itself and the presence of sequences
endowed with protease activity dictates the allergenic potential of proteins
[32], through as yet unknown mechanisms. The risk of developing a Th2-biased
response and allergy may thus also depend on the nature of the protein.
Prospective use of DNA
for the treatment of skin lesions
Insertion and expression of genes in the epidermis may have a variety
of therapeutic uses, including the treatment of skin diseases by the expression
of cytokines and other biologically active molecules for the treatment
of skin lesions. For example, genes encoding IFN-alpha could be injected
to treat skin tumors and viral lesions (Kaposi's sarcoma, basal cell carcinoma,
cutaneous squamous cell carcinoma, papilloma). In addition, gene-gun vaccination
with plasmid encoding cytokines may be advantageous for the treatment
of cancer, rather than intravenous cytokine delivery which is often inefficient
and frequently accompanied by systemic cytotoxicity.
It should be emphasized that in contrast to mouse skin, in which injected
DNA is expressed not only in the epidermis but also in the dermis and
the underlying fat and muscle tissue and is expressed at low levels, DNA
injected into the superficial dermis of human or pig skin organ cultures
and in human skin grafted onto immunocompromised mice is taken up and
expressed in the epidermis [33]. These data further indicate that the
pig is an appropriate model for preclinical studies on DNA vaccination
and therapeutics.
In addition to skin vaccination, mucocutaneous gene therapy offers new
potential approaches for local treatment of various skin lesions. The
feasibility of gene-gun delivery of DNA through the oral mucosa or the
epidermis has been tested in dogs, which develop spontaneous oral and
epidermal tumors. Pilot studies using the reporter gene
ß-galactosidase inserted into a CMV plasmid showed that direct injection
of 20 µg of plasmid into the oral mucosa induced 35 fold higher local
expression of ß-galactosidase as compared to injection of the same
dose in the epidermis. Due to the accelerated turn over of mucosal epithelium,
ß-galactosidase positive cells were detected in the basal and suprabasal
layers as early as 3 hrs after injection, whereas only the most superficial
mucosal layers demonstrated ß-galactosidase activity at 24 hrs post-injection
[34]. It was observed that particle-mediated gene transfer of either ß-galactosidase,
luciferase, IL-2, IL-6 or GM-CSF cDNA into the oral mucosa and the epidermis
of healthy dogs generated effective and localized transgene expression,
without signs of toxicity. Other studies confirmed and extended these
observations by showing that injection of DNA encoding IL-8 directly into
the epidermis is able to recruit neutrophils into the underlying dermis
[35]. Thus, the gene-gun approach should be considered for potential clinical
applications in cancer immunotherapy.
Nucleic acid inoculation through the skin represents a promising tool
in the field of vaccinology and treatment of skin lesions. Plasmid DNA
immunization has been used in a number of animal models as a novel strategy
to induce both antibodies and CTL responses as well as long term protection
against infectious agents or tumor development. Moreover, vaccines consisting
of plasmid DNA have several important advantages over alternate approaches
such as purified or recombinant proteins and live-attenuated or recombinant
viruses. They can be easily constructed for any suitable antigen and produced
economically in large quantities with a high degree of purity and stability
and thus are appropriate for mass immunization.
REFERENCES
1. Banchereau J, Steinman RM. 1998. Dendritic cells and the control of
immunity. Nature 392: 245-52.
2. Caux C. Pathways of development of human dendritic cells. Eur
J Dermatol 1998; 6: 375-84.
3. Brossart P, Bevan MJ. Presentation of exogenous protein antigens
on major histocompatibility complex class I molecules by dendritic cells:
pathway of presentation and regulation by cytokines. Blood 1997;
90: 1594-9.
4. Donnelly JJ, Ulmer JB, Shiver JW, Liu MA. DNA vaccines. Ann Rev
Immunol 1997; 15: 617-48.
5. Fynan EF, Webster RG, Fuller DH, Haynes JR, Santoro JC, Robinson
HL. DNA vaccines: protective immunizations by parenteral, mucosal and
gene-gun inoculations. Proc Natl Acad Sci 1993; 90: 11478-82.
6. Cardoso A, Sixt N, Vallier A, Fayolle J, Buckland R, Wild F. Measles
virus DNA vaccination antibody isotype is determined by the method of
immunization and by the nature of both the antigen and the coimmunized
antigen. J Virol 1998; 72: 2516-8.
7. Pertner TM, Roberts TR, Haynes JR. Influenza virus nucleoprotein-specific
immunoglobulin G subclass and cytokine responses elicited by DNA vaccination
are dependent on the route of vector DNA delivery. J Virol 1996;
70: 6119-25.
8. Condon C, Watkins SC, Celluzzi CM, Thompson K, Falo LD. DNA-based
immunization by in vivo transfection of dendritic cells. Nature
Medicine 1997; 2: 1122.
9. Andree C, Swain WF, Page CP, Macklin MD, Slama J, Hatzis D, Eriksson
E. In vivo tansfer and expression of a human epidermal growth factor
gene accelerates wound repair. Proc Natl Acad Sci 1994; 91: 12188.
10. Barry MA, Johnston SA. Biological features of genetic immunization.
Vaccine 1997; 8: 788.
11. Torres CA, Iwasaki A, Baber AH, Robinson HL. Differential dependence
on target site tissue for gene gun and intramuscular DNA immunizations.
J Immunol 1997; 158: 4529.
12. Iwasaki A, Torres CAT, Ohashi PS, Robinson HL, Barber BH. The dominant
role of bone-marrow-derived cells in CTL induction following plasmid DNA
immunization at different sites. J Immunol 1997; 155: 2039.
13. Klinman DM, Secgler JMG, Conover J, Gu M, Rosenberg AS. Contribution
of cells at the site of DNA vaccination to the generation of antigen specific
immunity and memory. J Immunol 1998; 160: 2388-92.
14. Porgador BA, Irvine KR, Iwasaki A, Barber BH, Restifo NP, Germain
R. Predominant role for directly transfected dendritic cells in antigen
presentation to CD8+ T cells after gene gun immunzation. J
Exp Med 1998; 188: 1075-82.
15. Sato Y, Roman M, Tighe H, Lee D, Corr M, Nguyen MD, Silverman GJ,
Lotz M, Carson D, Raz E. Immunostimulatory DNA sequences necessary for
effective intradermal gene immunization. Science 1996; 273: 352-4.
16. Roman M, Martin-Orozco H, Goodman JS, Nguyen MD, Sato Y, Ronhaghy
A, Kornbluth A, Richman DD, Carson D, Raz E. Immunostimulatory DNA sequences
function as T helper-1-promoting adjuvants. Nature Medicine 1997;
3: 849-54.
17. Jacob T, Walker PS, Krieg AM, Udey MC, Vogel JC. Activation of cutaneous
dendritic cells by CpG-containing oligonucleotides: a role for dendritic
cells in the augmentation of Th1 responses by immunostimulatory DNA. J
Immunol 1998; 161: 3042-9.
18. Krieg AM, Wu T, Weeratna R, Efler SM, Love-Homan L, Yang L, Yi AK,
Short D, Davis HL. Sequence motifs in adenoviral DNA block immune activation
by stimulatory CpG motifs. Proc Natl Acad Sci 1998; 95: 12631-6.
19. Porgador A, Gilboa E. Bone marrow generated dendritic cell pulsed
with a class I-restricted peptides are potent inducers of cytotoxic T
lymphocytes. J Exp Med 1995; 182: 255-8.
20. Czernick FI, Brezofsky JA, Carbone DP. Induction of cytotoxic T
lymphocytes and anti-tumor immunity with DNA vaccines expressing single
T cell epitopes. J Immunol 1996; 156: 2369-75.
21. Thompson SA, Sherritt MA, Medveczky J, Elliott SL, Moss DL,
Fernando GJ, Brown LE, Suhrbier AA. Delivery of multiple CD8 cytotoxic
T cell epitopes by DNA vaccination. J Immunol 1998; 160: 1717-23.
22. Xiang Z, Ertl HCJ. Manipulation of the immune response to a plasmid-encoded
viral antigen by co-inoculation with plasmids expressing cytokines. Immunity
1995; 2: 129-35.
23. Boyle JS, Brady JL, Lew AM. Enhanced responses to a DNA vaccine
encoding a fusion antigen that is directed to sites of immune induction.
Nature 1998; 392: 408-11.
24. Hedley ML, Curley J, Urban R. Microspheres containing plasmid-encoded
antigens elicit cytotoxic T cell responses. Nature Medicine 1998;
4: 365-8.
25. Lu B, Federoff HJ, Wang Y, Goldsmith LA, Scott G. Topical application
of viral vectors for epidermal gene transfer. J Invest Dermatol
1997; 108: 803-8.
26. Glenn GM, Matyas GR, Alving CR. Skin immunization made possible
by cholera toxin. Nature 1998; 391: 851.
27. Glenn GM, Scharton-Kersten T, Vassell R, Mallett CP, Hale TL, Alving
CR. Transcutaneous immunization with cholera toxin protects mice against
lethal mucosal toxin challenge. J Immunol 1998; 161: 3211-4.
28. Thestrup-Pedersen K. Which factors are of relevance in the pathogenesis
of atopic dermatitis? Eur J Dermatol 1997; 7: 549-53.
29. Castelain M. Atopic dermatitis and delayed hypersensitivity to dust
mites. Clin Rev Allergy Immunol 1995; 13: 161-72.
30. Wang LF, Lin JY, Hseih KH, Lin RH. Epicutaneous exposure of protein
antigen induces a predominant Th2-like response with high IgE production
in mice. J Immunol 1996; 156: 4079-82.
31. Spergel JM, Mizoguchi E, Brewer JP, Martin TR, Bhan AK, Geha RS.
Epicutaneous sensitization with protein antigen induces localized allergic
dermatitis and hyperresponsiveness to methacholine after single exposure
to aerosolized antigen in mice. J Clin Invest 1998; 101: 1614-22.
32. Musu T, Gregoire C, David B, Dandeu JP. The relationship between
the biochemical properties of allergens and their immunogenicity. Clin
Rev Allergy Immunol 1997; 15: 485-98.
33. Hengge UR, Walker PS, Vogel JC. Expression of DNA in human, pig
and mouse skin. J Clin Invest 1996; 12: 2911-6.
34. Hengge UR, Pfutzner W, Williams M, Goos M, Vogel JC. Efficient expression
of naked plasmid DNA in mucosal epithelium: prospective for the treatment
of skin lesions. J Invest Dermatol 1997; 111: 605-8.
35. Hengge UR, Chan EF, Foster RA, Walker PS, Vogel JC. Cytokine gene
expression in epidermis with biological effects following injection of
naked DNA. Nature Genetics 1995; 10: 161-6.
|
Printable version
