Maximum Principle for Infinite-horizon Optimal Control Problems under Weak Regularity Assumptions