Human parainfluenza viruses represent a leading cause of lower respiratory tract disease in children, with currently no available approved drug or vaccine. The viral surface glycoprotein haemagglutininneuraminidase (HN) represents an ideal antiviral target. Herein, we describe the first structure-based study on the rearrangement of key active site amino acid residues by an induced opening of the 216-loop, through the accommodation of appropriately functionalised neuraminic acid-based inhibitors. We discovered that the rearrangement is influenced by the degree of loop opening and is controlled by the neuraminic acid's C-4 substituent's size (large or small). In this study, we found that these rearrangements induce a butterfly effect of paramount importance in HN inhibitor design and define criteria for the ideal substituent size in two different categories of HN inhibitors and provide novel structural insight into the druggable viral HN protein.Human parainfluenza virus (hPIV) is one of the leading causes of respiratory tract disease in infants and children 1, 2 and is estimated to result in over 1.5 million cases per year in the United States alone 3 . Despite continuous efforts 4, 5 , there are neither specific antiviral therapy nor vaccines available against hPIV-3 to date. The viral surface glycoprotein haemagglutinin-neuraminidase (HN) represents an ideal target for the development of urgently needed antiviral agents. The viral HN protein encompasses three key functions in virus infection and spread. The hPIV HN recognizes and attaches to N-acetylneuraminic acid-containing glycoconjugates present on the host cell and subsequently activates the fusion machinery, facilitating infection. Upon virus replication, hPIV HN enzymatically cleaves the neuraminic acid (Neu), N-acetylneuraminic acid (Neu5Ac, 1) from host cell receptors allowing viral spread to uninfected cells 3 . A number of Neu2en (2)-based inhibitors (e.g. BCX 2798, 3) 6, 7 , as well as a novel approach that uses two well-established drugs 8 in a combinatorial manner, have been used to target the hPIV-3 HN protein. However, none of these agents have progressed to the clinic.Recently, we have reported 9 the first structural investigation into the catalytic mechanism of the hPIV-3 HN protein using the 2,3-difluoro-N-acylneuraminic acid derivative, 4. This study demonstrated that the protein forms a covalent adduct with the substrate as a result of a nucleophilic attack at the Neu moiety's anomeric carbon (C-2) by the hydroxyl group of the key catalytic amino acid, Tyr530. Hence, it was verified 9 that hPIV-3 HN can be targeted by reactive substrate-like inhibitors such as 4.We have also recently described the design and synthesis of novel potent 4-deoxy-4-triazolo-Neu2en-based inhibitors (5 and 6, Fig. 1) 10 . These inhibitors carry bulky C-4 substituents on the Neu2en template and target the proposed 216-cavity formed by movement of the flexible 216-loop 11 , a unique feature in hPIV-3 HN. Taken together these developments inspired further structural ...