The advanced text generation methods have witnessed great success in text summarization, language translation, and synthetic news generation. However, these techniques can be abused to generate disinformation and fake news. To better understand the potential threats of synthetic news, we develop a new generation method FactGen to generate high-quality news content. The existing text generation methods either afford limited supplementary information or lose consistency between the input and output which makes the synthetic news less trustworthy. To address these issues, FactGen retrieves external facts to enrich the output and reconstructs the input claim from the generated content to improve the consistency among the input and the output. Experiment results on real-world datasets show that the generated news contents of FactGen are consistent and contain rich facts. We also discuss the possible defending method to identify these synthetic news pieces if FactGen is used to generate synthetic news.