The paper presents a novel end-to-end neural model for extracting entities and relations between them, using bidirectional tree-structured LSTM-RNNs on top of bidirectional sequential LSTM-RNNs. This approach captures both word sequence and dependency tree substructure information, allowing the model to jointly represent entities and relations with shared parameters. The model improves over state-of-the-art feature-based models on end-to-end relation extraction tasks, achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and ACE2004, respectively. It also performs favorably against state-of-the-art CNN-based models on nominal relation classification (SemEval-2010 Task 8). The paper includes extensive ablation studies to evaluate the contributions of various model components.The paper presents a novel end-to-end neural model for extracting entities and relations between them, using bidirectional tree-structured LSTM-RNNs on top of bidirectional sequential LSTM-RNNs. This approach captures both word sequence and dependency tree substructure information, allowing the model to jointly represent entities and relations with shared parameters. The model improves over state-of-the-art feature-based models on end-to-end relation extraction tasks, achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and ACE2004, respectively. It also performs favorably against state-of-the-art CNN-based models on nominal relation classification (SemEval-2010 Task 8). The paper includes extensive ablation studies to evaluate the contributions of various model components.