Genetic immunization, also known as DNA or polynucleotide immunisation, is well documented to induce broad-based immunity in various animal models of infectious and non-infectious diseases. However, the low potency of DNA vaccines has to date precluded the development of commercial vaccines. The aim of this study was to systematically investigate a number of parameters to improve the potency of DNA vaccines for use in chickens, using a low pathogenic avian influenza (LPAI) virus as a proof-of-concept for their ability to produce a humoral immune response. The index virus used in the study was avian influenza virus A/coot/WA/2727/79 (H6N2), isolated from an apparently healthy Eurasian coot in 1979. Prior to any DNA experiments the virus was rigorously characterized. The virus strain was shown to be an H6 subtype by haemaglutination inhibition (HI) testing and as an N2 subtype by gene sequence analysis. The isolate was shown to be able to grow on MDCK cells in the absence of exogenous trypsin. It was further biologically characterized as LPAI with an intravenous pathogenicity index (IVPI) of 0.15 and a motif of 321PQAETRG328 at the cleavage site of the haemagglutinin (HA) protein. It was capable of infecting domestic chickens under experimental conditions with a low level of virus excretion via the cloaca and oropharynx following intravenous or oral and oculonasal inoculation. The full-length HA and nucleoprotein (NP) genes of this H6N2 virus were subsequently cloned into the eukaryotic expression vector VR1012 to generate VR-HA and VR-NP constructs. Six-week-old Hy-Line chickens were intramuscularly injected with either the VR-HA or VR-NP vaccine at different dose rates, with or without lipofectin as adjuvant. Minimal or no detectable antibody was produced, as measured by HI, ELISA and Western blotting-based assay, but high titres of H6-specific HI antibodies appeared 10 days after homologous virus challenge. In contrast to the empty vector controls, there was a significant difference in HI antibody titre between pre- and post-challenge in vaccinated birds, indicating some evidence for the priming effect of the DNA vaccines. Using the frequency of virus shedding as an indicator of protection, lower doses (50 or 100 ¦Ìg per chicken) of either adjuvanted VR-HA or VR-NP vaccine significantly reduced virus shedding in oropharyngeal and cloacal swabs compared to higher doses (300 or 500 ¦Ìg per chicken ) or empty vector control chickens. Although two vaccinations with naked VR-HA alone were not sufficient to induce an effective immune response against a homologous virus challenge, further repeat vaccinations and incorporation of adjuvant did lead to the generation of low to moderate HI antibody titres in some chickens and resulted in no or reduced virus shedding after challenge. Next, to examine the effect of expression vector, three different DNA vectors, pCI, pCI-neo and pVAX1 were used to clone the same HA gene and generate three DNA vaccine constructs. Once again, direct intramuscular injection of the three DNA constructs did not elicit measurable H6-specific HA antibody response in Hy-Line chickens but the 100 µg pCI-HA lipofectin adjuvanted vaccine group showed a significant increase in post-challenge HI titres from the naive control group, indicating that an anamnestic antibody response had been induced by the pCI-HA DNA vaccination. Compared with the controls, the three DNA constructs showed significantly reduced virus shedding in cloacal swabs post virus challenge, suggesting that the three DNA vaccines induced some level of immune response in vaccinated chickens. As with the VR-HA construct, the lower dose groups for each vaccine (50 or 100 g) were more effective at reducing virus shedding from the cloaca than the higher dose group (300 g). To further investigate why the DNA vaccines did not elicit a measurable antibody response, the HA gene incorporating a Kozak enhancer sequence was cloned into an alternative expression vector, pCAGGS, to produce the pCAG-HAk construct. Three-week-old SPF chickens were immunized with this construct either by the intramuscular route (IM) or electroporation (EP). H6 HI antibodies were present in some chickens by 3 weeks after the first IM vaccination and 75% of the chickens vaccinated with 10, 100 or 300 µg pCAG-HAk were antibody positive by 2 weeks after the second IM vaccination. For EP immunization, 87.5% of vaccinated birds seroconverted after the first vaccination and 100% seroconverted after the second vaccination and the H6 HI antibody titres were significantly higher than for chickens vaccinated by IM inoculation. Another group was given a single dose IM vaccination with 100 µg of the pCAG-HAk construct and showed a maximum sero-conversion rate of 53.3% with a peak H6 HI titre of 27 at 5 weeks post-vaccination. This demonstrated that optimization of the expression vector and insertion of a Kozak sequence could synergistically enhance expression of the H6 HA gene and result in a measurable H6 antibody response in SPF chickens. EP was also compared with IM inoculation with the 100 g pCI-HA construct in SPF chickens, resulting in a 50% sero-conversion rate and mean HI titre of 21.3 at 2 weeks after the second vaccination by EP. By comparison, only 25% chickens had trace HI titres by IM inoculation. This indicated that EP was more efficient than IM delivery for both constructs. A codon-optimized complete HA gene from A/coot/WA/2727/79 (H6N2) was then chemically synthesized and cloned into a pCAGGS vector to generate the pCAG-optiHAk construct. SPF chickens immunized twice with either 10 µg or 100 µg of pCAG-optHA showed 37.5% and 87.5% sero-conversion rates respectively, with a mean H6 HI tire of 21.4 and 22.6 at 3 weeks after the second immunization, but the differences were not statistically significant. There were also no significant differences in either the sero-conversion rate or the H6 HI titre between the pCAG-HAk and pCAG-optiHAk groups, suggesting that a codon-optimized HA DNA vaccine did not achieve significantly better immunogenicity than the pCAG-HAk vaccine. In vitro expression of the developed DNA constructs in chicken-, hamster-, monkey- and human-origin cells, as measured by Western blotting and immunofluorescence testing (IFT), showed the strength of H6 HA expression in the following descending order - pCAG-optiHAk/pCAG-HAk, pCI-HAk, VR-HA, pCI-HA, pCIneo-HA and pVAX-HA. The in vivo chicken vaccinations also showed that the pCI-HA construct was more effective than the pCI-neo-HA, and that the pCAG-optiHA or pCAG-HAk constructs were better than pCI-HAk in term of reduction in virus shedding after H6N2 virus challenge. Thus, in vitro HA gene expression directly correlated with the generation of immune responses in vivo, indicating that in vitro studies can be used for pre-selection of expression plasmids prior to development of avian influenza DNA vaccines. Lipofectin as a chemical adjuvant was shown to enhance the DNA-induced immune response but is prohibitively expensive for routine use in poultry vaccines. Thus, an experimental adjuvant for poultry DNA vaccines (Essai) and a new nanoparticle (Phema) adjuvant used for the first time in poultry were compared with conventional aluminum salts (alum) adjuvant in the present study. No HI antibody was detected in any adjuvant-vaccinated Hy-Line chickens following two immunizations. However, in comparison with the naive control group, the alum- and Phema adjuvanted pCAG-HAk groups significantly reduced the frequency of virus shedding in oropharyngeal swabs, but Essai adjuvant was not effective in augmenting the pCAG-HAk vaccine efficacy. This pilot study also emphasised that the traditional aluminum hydroxide adjuvant, either DNA binding or non-binding, may be useful as an adjuvant for enhancing DNA-induced immune responses in chickens owing to its low price and safety record. Overall, DNA immunization with various HA-expressing constructs was shown to be variably effective in inducing immune responses in chickens. The efficacy of DNA vaccines could be synergistically improved by taking appropriate approaches. With continuing research DNA vaccines have the potential to become an important tool for disease prevention and control.