How about play 1. Nc6 against d4.  Now you got the kevitz defense and there colle formation will not work.

Actually, you might find more Colle players taking this route against 3...c5 because two recent books on the Colle suggest it.

Note, a key improvement in that line is that in the position below, White does not need to protect his pawn structure with 9.Qc2

If You want dynamics, you play c5, d6, e5 against white Pawnstructure c3,d4,e3
(optional b6, bb7, g6, bg7, ngf6, nbd7)

No serious modern colle player would respond to c5/d6/e5 with typical moves (I would hope).

Small world. I've only lived in C'ville a few years.

In Zuke 'Em I point out that 4...cxd4 is only likely to avoid that line I mention if Black follows it up with 5...g6 (which can be twisted into a type of Sicilian) or 5...Nc6 6.Nc3 Bg4 7.cxd5 Nxd5

This latter is a fine try for Black [and would be the one I suggest for you based on what I've read]... but it is quite sharp after 8.Qb3!? (as you almost certainly already know). If you trust your defensive skills, that line is as good as any and better than most.

By the way, the 4...g6 play (which used to strike fear in the hearts of Colle Players) is now (relatively speaking) busted, so I would definitely affirm your 3...c5 choice. Of those lines where Black waits until after 3.e3 to deviate, it is easily the most challenging to Colle players. (Not counting the exceptional situation where you are playing up agaisnt a Colle Player and are happy with a draw...in that case I could probably come up with a more annoying defense for Black.)

Another good alternative is Alekhine's 3 ... Bf5, intending after 4 Bd3 e6 and if 5 Bxf5 exf5 and Black will use the e-file and have the somewhat better Bishop, while the doubled button on f5 locks down the e4 square.  Since the whole theme of the Colle is to play e3-e4 at the proper time, White is suddenly forced to find a new plan.

My experience has been that Colle players absolutely hate this defense, which makes it a real pleasure to play against them.

Estragon, the slav lines that transposes to (after 4.c4 e6 5.Nc3 c6) are not particularly inspiring for Black. Furthermore, I found what I consider some real improvements for White in those backwaters.

Also keep in mind that against a C-Z player, locking down on e4 does no good at all. Colle-Zukertort players almost never play e4.

Actually that line has been pretty much analysed to a draw:

8. Qb3 Bxf3 9. gxf3 e6 10. Qxb7 Nxd4 11. Bb5+ Nxb5 12. Qc6+! Ke7 13. Qxb5 Qd7! 14. Nxd5+ Qxd5 15. Qxd5 exd5 16. Bg5+ (or the immediate Be3) f6 17. Be3 Ke6 18. O-O-O Bb4! 19. a3 Rhc8+ 20. Kb1 Bc5 with an equal ending.

Black can also try 9. ..Nb6, which is definitely risky and probably somewhat better for white if he knows his stuff.
White on the other hand doesn't seem to have many other options in the 8. Qb3 line.

It's the special charm of this forum that no query about the details a particular opening goes without a response recommending something completely different.

I think you might be overstating things.

White has several options on move 16 in that variation, for example, castling or playing 16.Bf4!? are both interesting options. Against the latter it is hard to find more than a couple games where Black has won, while there are several White wins at master level or above. In my database at least, White scores 59% after castling; Ivanov beat Seirawan in the 91-92 US championship in that line.

It might be drawish, and White may have only a slight advantage (at least based on practice), but that is not the same as saying it is a draw.

I don't like the statistical argument. Can you show how white has a slight advantage after 16. Bf4 ?

Here's what GM Wells has to say about the position after 9. ..e6:
" The text-move in fact provokes a rather forcing sequence resulting in an endgame generally felt to be quite tenable for Black. Re-examining this ending, I am in fact inclined to think that while most of the top players now regard this line as rather too drawish, for lesser mortals there is sufficient potential for play for both sides in the coming ending. After all, there is no shortage of structural imbalance in the position!"

No, I don't have anything I would want to hang too many clothes on. I would say that there is plenty of play left after Bf4 or O-O. In my book, I simply say that White has a comfortable position. There is still plenty of game for the person with the better technique to win.

Well, transposing into the Slav isn't a big endorsement of the Colle, and a lot of top GMs will happily take the Black side of the Slav, so . . .

If e4 is not the strategic aim of the Colle, then what exactly is the plan?

This is ofcourse true.

It isn't just "transposing to the Slav," but rather transposing to a very specific line [and only after 6 moves have been played on both sides.] I don't see this as a slight against the Colle. Black deviates from the mainline, so White responds. If White plays 1.e4 e5 2.Nf3 and Black responds 2...Nf6, is it a slight against the Ruy if White does not play 3.Bb5?

Anyways, to answer your second question. Please note I said "Colle-Zukertort." There are two Colle Variations. In the Colle-Koltanowski, White plays c3 and plans to unlock his Bishop with a later e4 [though even in this variation, I think the mainline is taking another direction.] In the Colle-Zukertort, White plays b3 and has no need to play e4...and practically never does (for Black will eventually exchange on d4 to save himself from fatal tactics on the long diagonal after a possible dxc5/Nxc6/Bxf6 combination. See "Understanding the Cosmic Exchange" in an intro article I wrote.) Instead, White actually prefers to keep the game closed and build up a K-side attack, leveraging Black's transportation problems.

Hope that helps.