T-Cell Stimulation by Melanoma RNA-Pulsed Dendritic Cells

In situations where well-established approaches such as surgery,
radiation therapy and chemotherapy fail to help cancer patients,
immunotherapy has the potential to be an effective alternative.
Tumour cells can sometimes be distinguished from corresponding
normal cells due to their expression of tumour-associated antigens
(TAAs), most of which are unaltered self-molecules. These molecules
must be presented to the immune system in the context of danger in
order to achieve their specific recognition. If dendritic cells
(DCs), the most potent professional antigen-presenting cells, are
loaded with RNA, they will translate the RNA into protein, process
the protein into peptides and present the peptides within MHC
molecules (pMHC) on their surface to cytotoxic T-lymphocytes (CTLs)
and T-helper cells in a stimulatory manner. These effector cells
can, in turn, recognise tumour cells. The goal of these studies was
to find optimal conditions for producing a DC-based vaccine for
cancer patients using TAAs in the form of RNA. The studies were
designed to quantitate RNA transfer into DCs, to determine the
intracellular stability of transfected RNA in DCs and to analyse
the kinetics of protein expression and the generation of functional
pMHC ligands that could activate effector memory CTLs. Simultaneous
activation of CTLs with specificities for different antigens
minimises the potential for tumour escape through immune selection
of tumour variants showing loss of individual antigens. Thus,
generation of multiplex pMHC ligands for CTLs may improve clinical
efficiency. On the other hand, peptide competition for MHC
molecules within the DC may limit pMHC ligand generation. This
central immunological question was addressed by comparing DCs
loaded with total cellular tumour RNA, amplified total cellular
tumour mRNA and pools of defined single-species tumour-antigen
cRNAs versus individual single-species tumour-antigen cRNAs for
their capacity to display various pMHC ligands and activate CTLs of
corresponding specificities. Experiments performed with RNA
encoding the enhanced green fluorescence protein (EGFP), a reporter
protein, showed that the highest efficiency of RNA transfection
into DCs was achieved with electroporation, reaching levels of 90%
positive cells. The fact that mature DCs expressed more EGFP than
immature DCs suggests that this stage of DC maturation will be
optimal for vaccine development. Importantly, electroporation and
RNA transfer did not alter the expression of antigen-presenting and
co-stimulatory molecules on the surface of DCs. The melanoma model
was chosen for extensive analyses because its characterisation at
the cellular and molecular levels has made it a very informative
model for understanding cancer immunity. In addition to total
cellular melanoma RNA, single-species cRNAs were used encoding the
melanoma-associated antigens, tyrosinase, Melan-A and CDK4-R24C.
Antigen presentation was detected with the help of effector memory
CTL clones specific for each of these antigens. The CTL stimulatory
capacity of RNA-transfected DCs was higher if they were allowed one
day to recuperate from electroporation and to produce pMHC
complexes. Tyrosinase cRNA dose-finding showed that more RNA would
indeed result in higher stimulatory capacities of transfected DCs.
Kinetics of tyrosinase cRNA degradation, similar to kinetics of
EGFP cRNA degradation, revealed that the amount of transfected RNA
rapidly decreased inside the DCs within 1.5 hr after
electroporation. The smallest decrease was observed with the
highest amount of RNA applied in electroporation. The kinetics of
RNA degradation and protein half-life will be important parameters
to consider in defining the right time-frame for T-cell activation
by engineered DCs. When reverse transcription PCR (RT-PCR) was
performed with total cellular melanoma RNA samples to generate
amplified mRNA, the Melan-A, tyrosinase and CDK4/CDK4-R24C message
was amplified 62-fold, 24-fold and 2-fold, respectively. These
differences likely reflect variations in the expression levels of
the corresponding antigen message in melanoma cells from which the
RNA was isolated. Approximately 17240-fold more CDK4-R24C message,
at least 500-fold more tyrosinase message and at least 480-fold
more Melan-A message was found in single-species cRNAs when the
same masses of single-species cRNA and amplified melanoma mRNA
samples were compared. This explained why electroporation of
single-species cRNAs into DCs yielded the highest DC stimulatory
capacities. Combinations of tyrosinase, Melan-A and CDK4-R24C cRNAs
were studied for their capacity to induce satisfactory levels of
T-cell stimulation when presented by DCs. Here it was demonstrated
that antigen competition was not a critical factor, since CTL
responses to pooled RNAs were not inhibited even though competition
for MHC class I molecules may have occurred within the DCs. DCs
also developed CTL stimulatory capacities, but at much lower
levels, using amplified melanoma mRNA. Two antigen-specific CTL
clones displayed higher reactivities upon exposure to pMHC produced
naturally by RNA-transfected DCs than to synthetic peptides pulsed
onto DCs. In one case, this could be explained by a
post-translational modification of the peptide, which normally
occurs within cells. Since this particular modification was not
represented in the synthetic peptide, which was chosen from the
protein sequence, the synthetic peptide was not well recognised.
This demonstrated that the use of RNA technology eliminates the
need to know the correct sequences of immunogenic peptides.
Thereby, DCs are better than scientists at choosing antigens and
their epitopes for presentation to T-cells. These data provided a
better understanding of antigen presentation by DCs based on the
use of RNA, giving insight into antigen competition and paving the
way for the use of pooled RNAs of defined species for the
development of a multiplex vaccine. They also allowed a precise
protocol for efficient T-cell activation to be defined. Further
experiments will demonstrate whether quantitative differences
detected in antigen presentation between DCs loaded with total
cellular tumour RNA and amplified total cellular tumour mRNA versus
single-species tumour-antigen cRNAs have an impact on de novo
T-cell priming in vitro and in vivo.