Pair-copula constructions are flexible models for the dependence in a random vector that have attracted a lot of interest in recent years. In this paper, we use generalized additive models to extend pair-copula constructions to allow for effects of covariates on the dependence parameters. We let each pair-copula parameter depend directly on the covariates in a parametric, semi-parametric or non-parametric way. We propose a sequential estimation method that we study by simulation, and apply our method to investigate the time-varying dependence structure between the intraday returns on four major foreign exchange rates.