Likelihood decreasing in $\theta$ picks the smallest feasible value, $X_{(n)}/5$

Let $X_1,X_2,\dots,X_n$ be a random sample from uniform $(\theta,5\theta)$. Define $X_{(1)}=\min\{X_1,\dots,X_n\}$ and $X_{(n)}=\max\{X_1,X_2,\dots,X_n\}$. Maximum likelihood estimator of $\theta$ is: