The central dogma of molecular biology hinges on messenger RNA (mRNA), which presents a blueprint of the genetic information encoded in the DNA and serves as a template for translation into proteins. In addition to its fundamental importance in basic research, this class of biomolecules has recently become the first approved Covid vaccine, underscoring its utility in medical applications.Eukaryotic mRNA is heavily processed, including the 5' cap as the primary hallmark. This 5' cap protects mRNA from degradation by exoribonucleases but also interacts specifically with several proteins and enzymes to ensure mRNA turnover and processing, like splicing, export from the nucleus to the cytoplasm, and initiation of translation. The absence of a 5' cap leads to a strong immune response, and the methylation status contributes to distinguishing self from non-self RNA.Non-natural modifications of the 5' cap provide an avenue to label mRNAs and make them accessible to analyses, which is important to study their cellular localization, trafficking, and binding partners. They bear potential to engineer mRNAs, e.g., more stable or immunogenic mRNAs that are still translated, by impacting select interactions in a distinct manner. The modification of the 5' cap itself is powerful as it can be applied to make long mRNAs (∼1000 nt, not directly accessible by solid-phase synthesis) by in vitro transcription.This Account describes our contribution to the field of chemo-enzymatic modification of mRNA at the 5' cap. Our approach relies on RNA methyltransferases (MTases) with promiscuous activity on analogues of their natural cosubstrate S-adenosyl-L-methionine (AdoMet). We will describe how RNA MTases in combination with non-natural cosubstrates provide access to site-specific modification of different positions of the 5' cap, namely, the N2 and N7 position of guanosine and the N6 position of adenosine as the transcription start nucleotide (TSN) and exemplify strategies to make long mRNAs with modified 5' caps.We will compare the chemical and enzymatic synthesis of the AdoMet analogues used for this purpose. We could overcome previous limitations in methionine adenosyltransferase (MAT) substrate scope by engineering variants (termed PC-MATs) with the ability to convert methionine analogues with benzylic and photocaging groups at the sulfonium ion.The final part of this Account will highlight applications of the modified mRNAs. Like in many chemo-enzymatic approaches, a versatile strategy is to install small functional groups enzymatically and use them as handles in subsequent bioorthogonal reactions. We showed fluorescent labeling of mRNAs via different types of click chemistry in vitro and in cells. In a second line of applications, we used the handles to make mRNAs amenable for analyses, most notably next-generation sequencing. In the case of extremely promiscuous enzymes, the direct installation of photo-cross-linking groups was successful also and provided a way to covalently bind protein-interaction partners. Finally, the non-natural modifications of mRNAs can also modulate the properties of mRNAs. Propargylation of Am as the transcription start nucleotide at its N6 position maintained the translation of mRNAs but increased their immunogenicity. The installation of photocaging groups provides a way to revert these effects and control interactions by light.