or you can just dean stark the whole mess mebbe not.
you said reflux was a bad idea (puts foot in mouth)
you said reflux was a bad idea (puts foot in mouth)
It's an electrophile strength thing, I believe. IIRC:
Chloroiminium (Vilsmeier complex) > acyl chloride > iminium > aldehyde > aldimine > alkyl halide > ketone > imine
Under acidic / high temperature conditions protonation of the iminium allows it to attack the highly activated 2-position on indole. I'm not sure if a low temperature reaction will provoke a Pictet-Spengler cyclization; I'm aware it happens at reflux. The Eschweieler-Clarke reaction requires high temperatures to proceed and so it's out but novel reductants could work (even in acid) if they can reduce imines at low temperatures.
Zinc ions are Lewis acids which catalyze the alkylation of aromatic rings and so naturally a zinc-based reducing agent would probably be an issue.
It's kind of scary, but my guess would be a Strecker-type reaction with formaldehyde and sodium cyanide, followed by a reductive decyanation. Cyclisation is impeded by the fact that the imine will react with cyanide faster than it will cyclize.
Alternatively...
People have talked a lot about using ethanolamine to demethylate the trimethyltryptammonium salt. Why not do this backwards, using tetramethylammonium salts, or the tetramethylammonium precursor trimethylglycine, to methylate tryptamine? Here's a paper that says it should be possible, and in water no less!
http://pubs.acs.org/doi/abs/10.1021/ja021212u
Stop citing bullshit. STAB reduces indoles to indolines. (check Handbook of Reagents for Organic Synthesis Oxidizing and Reducing Agents)