This is a kind of chicken or egg question, since there was never really a point at which there was no river flowing, or no hills to flow around.
My question is whether the hills preexisted the course of the river or did the river in some way cause the hills to "appear".
Most of the area which you describe along the Mississippi is formed of sand, gravel and till deposited by glaciers in the previous two glaciations. Though there would have been solid bedrock beneath it all before that, complete with hills and rivers flowing across the landscape.
Rivers flow along the path of least resistance over the landscape under the influence of gravity. This means that, in general, they will take the route with the steepest downhill gradient from their source to the sea.
Are the hills all glacial deposits which guided the course of river?
The hills formed by glacial deposits would indeed have influenced the course of river after de-glaciation occurred. Though the flowing glacier will have shaped the hills in the first place.
At the same time, the river will be eroding the material through which it flows, so it is constantly (but very slowly) carving out new formations in the landscape. Fluvial landscapes tend to produce very dendritic landscapes – that is, comprising of a dense network of channels with many small tributaries – in comparison to landscapes formed by glaciers. In this sense, the river will be causing new hills to 'appear' as it carves out new valleys and tributaries into the glacial material. Some of the smaller hills in the area may well have been formed entirely post-glacially.
Summary:
A mixture of both really. River courses are influenced by existing topography, but they also have the power to carve out new features into the landscape.
Afterthought:
If you go back far enough in Earth history, before the formation of water at the surface, there would have been some sort of topography (hills). So you could argue that ultimately the hills came first.
Sources: Bennett and Glasser (2009) Glacial Geology; Anderson and Anderson (2010) Geomorphology