Perceptron Learning Algorithm is an algorithm or a way to train the weights of the Perceptron. This was proposed by Minsky and Papert in 1969.
Perceptron is more general computational model than McCulloch Pitts Neuron. It has Numerical Weights for inputs and a mechanism for learning these Weights.
Mcculloch Pitts Neuron can be used to represent Boolean functions which are linearly separable. This is the First Artificial Neural Model in A.I.