"The tryptophan metabolic pathway collectively displays the largest perturbations in ASD [autism spectrum disorder]."
So said the findings reported by Federica Gevi and colleagues  (open-access) who provide yet more 'metabolomic' data when it comes to autism to add to the already quite voluminous peer-reviewed matter on this topic (see here for example).
Just in case you aren't analytical chemistry-saavy, metabolomics is basically the study of the various chemical fingerprints that the multitude of cellular processes going on in the body leave behind. It's the technology available these days that makes metabolomics the discipline that it is, as words such as mass spectrometry and nuclear magnetic resonance (spectroscopy) fill the metabolomic airwaves coupled with some rather smart statistics and software to translate all that captured data into something meaningful.
Gevi et al report results based on the analysis of urine samples from a small-ish group of children diagnosed with an ASD ("idiopathic ASD") compared with samples from a similar number of not autism controls. The aim was to focus on "autistic and unrelated typically developing children 2–8 years old, tightly matched by age, sex, Italian ancestry, and city of origin within the country" and look-see whether a particular HPLC-mass spec technique "hydrophilic interaction chromatography (HILIC)-LC-electrospray ionization (ESI)-MS" might provide some important data on autism vs. not autism.
Results: well, it's always nice to get a research mention in such studies as per the line: "Data were normalized by urinary specific gravity, because creatinine excretion may be abnormally reduced in ASD children" with reference to some work published a few years back . Indeed, this is not the first time creatinine has cropped up in autism metabolomic studies (see here) and is perhaps worthy of quite a bit more study itself (see here).
The authors report that urine samples from those with autism vs. those with not-autism are "largely distinguishable" based on some nifty analysis of the compounds examined from those groups. They even provide a 'top 25 discriminating metabolites' summary to illustrate this fact. Before venturing further into this list, I would perhaps advise some caution however. Caution based on the fact that urine contains many hundreds/thousands of small molecules or chemical entities as a function of being a waste product and carrying waste products from a multitude of different biological processes. It's not outside the realms of possibility that with such a huge number of metabolites, any two groups could be separated out, not just those based on the appearance of autism or not...
Anyhow: "The “metabolome overview” obtained through metabolic pathway analysis (MetPA) shows tryptophan metabolism, purine metabolism, vitamin B6 metabolism, and phenylalanine-tyrosine-tryptophan biosynthesis as the four most perturbed metabolic pathways in ASD." The reference to the aromatic amino acid called tryptophan (the stuff that eventually ends up as serotonin and melatonin) used in the title of this post kinda points to where the money might be when it came to these particular results. I've been interested in tryptophan metabolism and autism for quite a while now (see here for example) and how, outside of the whole serotonin/melatonin bit, there is quite a lot more to see besides. Mention of something called the kynurenine pathway by Gevi is interesting; not least because this pathway overlaps with other conditions/labels too (see here). This pathway might also have some important implications when it comes to epilepsy (see here) as a comorbidity to autism too.
It's also interesting (to me at least!) to note that the authors found something related to the indoles in their analyses too. So: "we also detect a significant increase in indole derivatives of bacterial tryptophan including indolyl 3-acetic acid, indoxyl sulfate, and most prominently, indolyl lactate." Indoxyl sulfate, a uremic toxin - something that is not great for the kidneys - crops up yet again  and importantly, highlights how bacteria can also 'go to work' on tryptophan in the gut. Indole -3-acetic acid also brings back research memories in relation to an indole compound close to my research heart, indolyl-3-acrylolyglycine (IAG)  that has received a bit of a research bruising quite recently  (the authors of that study and another one  however, really need to rethink their paper titles insofar as them not actually testing whether dietary intervention actually 'affects' levels of IAG or related metabolites but nonetheless implying so).
There are a range of other findings reported by Gevi and colleagues but I don't want to bore you with all the details. Suffice to say that metabolomics continues its research rise with autism in mind, and provides some rather interesting results. Of course there is more to do in this area; not least the focus on subgroups in these days of 'the autisms' and perhaps a little more metabolomic inquiry when it comes to the myriad of intervention options put forward 'for autism'. Who for example, wouldn't like to see metabolomic profiles pre- and post-folinic acid for example alongside the myriad of other interventions detailed in the peer-reviewed literature? Indeed, I might also advocate a little more investigation on whether specific patterns of urinary compounds might also be related to specific behavioural facets of autism. Given the move towards gut bacteria as potentially showing involvement in some of the results obtained by Gevi et al, it would also be interesting to see if 'altering' certain types of gut bacteria (see here for example) might also have some interesting knock-on effects when it comes to the metabolites detected too? There is quite a bit more to do.
Music and more bad lip reading.... sick of blue milk?
 Gevi F. et al. Urinary metabolomics of young Italian autistic children supports abnormal tryptophan and purine metabolism. Molecular Autism. 2016l 7: 47.
 Whiteley P. et al. Spot urinary creatinine excretion in pervasive developmental disorders. Pediatr Int. 2006 Jun;48(3):292-7.
 Diémé B. et al. Metabolomics Study of Urine in Autism Spectrum Disorders Using a Multiplatform Analytical Methodology. J Proteome Res. 2015 Dec 4;14(12):5273-82.
 Bull G. et al. Indolyl-3-acryloylglycine (IAG) is a putative diagnostic urinary marker for autism spectrum disorders. Med Sci Monit. 2003 Oct;9(10):CR422-5.
 Wilson J. et al. Can urinary indolylacroylglycine (IAG) levels be used to determine whether children with autism will benefit from dietary intervention? Pediatr Res. 2016 Nov 23.
 Dalton NR. et al. Measurement of urine indolylacroylglycine is not useful in the diagnosis or dietary management of autism. Autism Res. 2016 Aug 29.
Gevi, F., Zolla, L., Gabriele, S., & Persico, A. (2016). Urinary metabolomics of young Italian autistic children supports abnormal tryptophan and purine metabolism Molecular Autism, 7 (1) DOI: 10.1186/s13229-016-0109-5