**4.2 18O labeling of bilins**

In another attempt to quantitate stercobilin, an isotopologue standard was created based off of the work by Bergmann et al. [46]. Isotopologues are often ideal internal standards for quantitation using mass spectrometry because the isotopologue is itself chemically similar to the analyte of interest but is shifted in mass due to the incorporation of another isotope. In the structure of stercobilin, the four oxygens within the carboxylic acid groups on the inner pyrroles are labile and able to be exchanged with the oxygen atoms of H2 18O (as can be seen in red in **Figure 3**). The reaction works under an acidic environment utilizing trifluoroacetic acid (TFA) as a proton donor to aid in the protonation of the carbonyl oxygen for the nucleophilic substitution reaction. Currently, we have scaled up our original procedure [47] by using ca. 5 × 10<sup>−</sup><sup>6</sup> mol of stercobilin mixed with 10 μL 5% (v/v) TFA and 95 μL of H2 18O in an LC autosampler vial with screw cap lid. The vial is placed in an incubator at 70°C for 8 h [43]. Following the reaction, the sample is dried down under air and reconstituted in 100 mL of 20:80 (v/v) ACN/H2O.

**Figure 3.** *The structure of stercobilin with the labile oxygen sites highlighted in red.*


**Table 1.**

*Corresponding m/z peaks from labeled stercobilin with the percentage of labeling of each peak denoted.*

To improve upon our previous 18O-stercobilin isotopologue yield, the reaction was conducted at a higher temperature [47] than what was reported by Bergmann et al. [46]. To further push the reaction toward full labeling, the reaction is carried out a second time under the same initial conditions except for allowing it to react for 22 h instead of 8. With this a labeling efficiency of 72.1 ± 0.3% was observed with minimal original stercobilin left in the reaction (**Table 1**). The results of this experiment have allowed for the quantitation of stercobilin within fecal samples.

#### **4.3 Deuterated stercobilin**

Next, we synthesized a more stable isotopologue of stercobilin through the use of deuterium-carbon bonds, which were achieved by the incorporation of deuterium across several of the carbon-carbon double bonds of bilirubin as described by Putzbach et al. [48]. This protocol incorporated deuterium atoms into bilirubin affording stercobilin with a mass increase of more than 12 atomic mass units. The conversion of bilirubin to stercobilin was previously reported (**Figure 4**) [49].

#### **Figure 4.**

*Step (1) involves the reduction of bilirubin into stercobilinogen. Step (2) involves partial oxidation of stercobilinogen into stercobilin (3), our desired product.*

**57**

**Figure 5.**

9.59 × 10<sup>−</sup><sup>9</sup>

of 1.13 × 10<sup>−</sup><sup>8</sup>

± 4.1 × 10<sup>−</sup><sup>9</sup>

± 1.1 × 10<sup>−</sup><sup>8</sup>

*Stercobilin: A Putative Link between Autism and Gastrointestinal Distress?*

**5. Biomarker validation: connection to the microbiome**

This method allowed us to reduce the six non-pyrrole C〓C double bonds with hydrogen gas (control reaction) or with deuterium gas (isotopologue) producing stercobilinogen and labeled stercobilinogen, respectively [44, 50]. For the deuterated isotopologue, bilirubin (200 mg) was combined with 25 mL deuterated glacial acetic acid (CD3COOD) and 200 mg of palladium on carbon, and deuteration was allowed to proceed for 1.5 h at 65°C to produce stercobilinogen. Stercobilinogen is subsequently aerated in the presence of copper sulfate, resulting in the final product, stercobilin. Combined nuclear magnetic resonance (NMR) and MS/MS analysis indicated incorporation of deuterium at all 12 sites, with no evidence of unreacted bilirubin.

Utilizing the 18O isotopologue standard, the amount of stercobilin could be quantified within the fecal samples of a murine model of ASD. In the described study, a population of mice with Timothy Syndrome (TS) was utilized; these mice have been previously described as exhibiting autistic behaviors. In particular, the mice used herein had a more severe case of TS, TS2-NEO, caused by a missense mutation in exon 8 at G406R in tandem with a flipped neomycin cassette, allowing for the mice to survive to adulthood [51]. Fourteen pairs of mice that were age- and

Response factor calculations were first completed in order to quantify the amount of labeled stercobilin in the fecal samples as well as account for the amount of unlabeled stercobilin that would be present in the sample from the isotopologue standard. Calculations of the concentration of stercobilin were determined utilizing the *m/z* 601 peak from the labeled stock. Concentrations were then normalized per gram of fecal material. From these calculations, box and whisker plots were created and are shown in **Figure 5** for both stercobilin and its precursor, stercobilinogen. An unpaired *t*-test was utilized to determine *p*-values and to establish whether the populations' mean bilin levels were statistically significantly different, or not, from each other.

When calculating the average moles of stercobilin utilizing the peak area of the

± 3.7 × 10<sup>−</sup><sup>9</sup>

tively. These values show a depletion of ca. 48% in stercobilin levels of TS2-NEO mice (*p* ≤ 0.001). In comparison, calculating the average moles of stercobilinogen utilizing the peak area of the *m/z* of 601 of WT and TS2-NEO populations, values

the two populations, respectively. These values showed a depletion of 51% in

*A comparison of the average concentration of both stercobilin and stercobilinogen found in wild type (WT) as opposed to mice with TS2-NEO per gram of fecal material. The p-values calculated from the unpaired t-test* 

mol/g feces were found for the two populations, respec-

± 7.1 × 10<sup>−</sup><sup>9</sup>

mol/g feces were found for

and

*m/z* of 601 of WT and TS2-NEO populations, values of 1.84 × 10<sup>−</sup><sup>8</sup>

and 5.55 × 10<sup>−</sup><sup>9</sup>

*are shown in the upper right-hand corner of the box and whisker plots.*

*DOI: http://dx.doi.org/10.5772/intechopen.84791*

gender-matched were utilized in this study.

*Stercobilin: A Putative Link between Autism and Gastrointestinal Distress? DOI: http://dx.doi.org/10.5772/intechopen.84791*

*Autism Spectrum Disorders - Advances at the End of the Second Decade of the 21st Century*

*m/z* **peak (18On) % labeling** 595 (18O0) 0.16 ± 0.04 597 (18O1) 7.2 ± 0.9 599 (18O2) 25.6 ± 0.8 601 (18O3) 38.6 ± 0.7 603 (18O4) 28.5 ± 0.5

To improve upon our previous 18O-stercobilin isotopologue yield, the reaction was conducted at a higher temperature [47] than what was reported by Bergmann et al. [46]. To further push the reaction toward full labeling, the reaction is carried out a second time under the same initial conditions except for allowing it to react for 22 h instead of 8. With this a labeling efficiency of 72.1 ± 0.3% was observed with minimal original stercobilin left in the reaction (**Table 1**). The results of this experiment have allowed for the quantitation of stercobilin within fecal samples.

*Corresponding m/z peaks from labeled stercobilin with the percentage of labeling of each peak denoted.*

Next, we synthesized a more stable isotopologue of stercobilin through the use of deuterium-carbon bonds, which were achieved by the incorporation of deuterium across several of the carbon-carbon double bonds of bilirubin as described by Putzbach et al. [48]. This protocol incorporated deuterium atoms into bilirubin affording stercobilin with a mass increase of more than 12 atomic mass units. The conversion of bilirubin to stercobilin was previously reported (**Figure 4**) [49].

*Step (1) involves the reduction of bilirubin into stercobilinogen. Step (2) involves partial oxidation of* 

**56**

**Figure 4.**

*stercobilinogen into stercobilin (3), our desired product.*

**4.3 Deuterated stercobilin**

**Table 1.**

This method allowed us to reduce the six non-pyrrole C〓C double bonds with hydrogen gas (control reaction) or with deuterium gas (isotopologue) producing stercobilinogen and labeled stercobilinogen, respectively [44, 50]. For the deuterated isotopologue, bilirubin (200 mg) was combined with 25 mL deuterated glacial acetic acid (CD3COOD) and 200 mg of palladium on carbon, and deuteration was allowed to proceed for 1.5 h at 65°C to produce stercobilinogen. Stercobilinogen is subsequently aerated in the presence of copper sulfate, resulting in the final product, stercobilin. Combined nuclear magnetic resonance (NMR) and MS/MS analysis indicated incorporation of deuterium at all 12 sites, with no evidence of unreacted bilirubin.

### **5. Biomarker validation: connection to the microbiome**

Utilizing the 18O isotopologue standard, the amount of stercobilin could be quantified within the fecal samples of a murine model of ASD. In the described study, a population of mice with Timothy Syndrome (TS) was utilized; these mice have been previously described as exhibiting autistic behaviors. In particular, the mice used herein had a more severe case of TS, TS2-NEO, caused by a missense mutation in exon 8 at G406R in tandem with a flipped neomycin cassette, allowing for the mice to survive to adulthood [51]. Fourteen pairs of mice that were age- and gender-matched were utilized in this study.

Response factor calculations were first completed in order to quantify the amount of labeled stercobilin in the fecal samples as well as account for the amount of unlabeled stercobilin that would be present in the sample from the isotopologue standard. Calculations of the concentration of stercobilin were determined utilizing the *m/z* 601 peak from the labeled stock. Concentrations were then normalized per gram of fecal material. From these calculations, box and whisker plots were created and are shown in **Figure 5** for both stercobilin and its precursor, stercobilinogen. An unpaired *t*-test was utilized to determine *p*-values and to establish whether the populations' mean bilin levels were statistically significantly different, or not, from each other.

When calculating the average moles of stercobilin utilizing the peak area of the *m/z* of 601 of WT and TS2-NEO populations, values of 1.84 × 10<sup>−</sup><sup>8</sup> ± 7.1 × 10<sup>−</sup><sup>9</sup> and 9.59 × 10<sup>−</sup><sup>9</sup> ± 4.1 × 10<sup>−</sup><sup>9</sup> mol/g feces were found for the two populations, respectively. These values show a depletion of ca. 48% in stercobilin levels of TS2-NEO mice (*p* ≤ 0.001). In comparison, calculating the average moles of stercobilinogen utilizing the peak area of the *m/z* of 601 of WT and TS2-NEO populations, values of 1.13 × 10<sup>−</sup><sup>8</sup> ± 1.1 × 10<sup>−</sup><sup>8</sup> and 5.55 × 10<sup>−</sup><sup>9</sup> ± 3.7 × 10<sup>−</sup><sup>9</sup> mol/g feces were found for the two populations, respectively. These values showed a depletion of 51% in

#### **Figure 5.**

*A comparison of the average concentration of both stercobilin and stercobilinogen found in wild type (WT) as opposed to mice with TS2-NEO per gram of fecal material. The p-values calculated from the unpaired t-test are shown in the upper right-hand corner of the box and whisker plots.*

stercobilinogen levels of TS2-NEO mice (*p* = 0.07). A larger sample set will be necessary to determine the significance of depletion in stercobilinogen. Furthermore, the *p*-values determined were improved upon since our last report with a study of nine pairs of mice [43].

The depletion of stercobilin in the ASD model of mice relative to controls at a greater than 99.9% confidence level suggests that stercobilin depletion in fecal material may have potential value as a biomarker for ASD in humans. Although less statistically significant, stercobilinogen, the metabolic precursor to stercobilin, is also depleted in fecal samples. The observation of these depletions suggests that there may be interference in the metabolic pathway that allows for the differences. As shown in **Figure 6**, stercobilin and stercobilinogen are products of heme catabolism. As bilirubin glucuronides enter the intestines, the action of enzyme systems by anaerobic bacterial flora converts the glucuronides to mesobilirubinogen, which is further converted to stercobilinogen.

Our results are also intriguing in the context of a discovery decades ago by Gustafsson and Lanke in which they observed no bilins present in the feces or urine of germfree rats [52]. Once the germfree animals were exposed to fecal matter from control animals, they too began to produce bilins to the same extent as the controls (when both groups were given identical diets). Moreover, they observed that the negative urobilin test (note: urobilin is a metabolic product derived from urobilinogen and is primarily excreted through urine, as shown in **Figure 6**) turned positive in germfree animals infected with a single *Clostridium*-like microorganism that had been isolated from the intestinal contents of rats that showed the presence of bilins in fecal matter. The bilin output increased in these animals after infection

#### **Figure 6.**

*A depiction of the catabolism of heme into stercobilin. The enterohepatic cycle in which stercobilin can be recirculated back and excreted instead through the urine is also shown. The line shows the point in which bacterial interaction takes over in the metabolic pathway to create stercobilin.*

**59**

#S10-RR029517).

**6. Conclusions**

suggests [57].

**Acknowledgements**

*Stercobilin: A Putative Link between Autism and Gastrointestinal Distress?*

with *Escherichia coli*, although not reaching the levels observed for controls [52]. In context, our results showing depleted stercobilin and stercobilinogen in fecal matter of ASD model mice might suggest that *Clostridium* constitutes part of their microbiota but that *E. coli* may have been impacted such that bilin production is reduced. Clearly, microbiological testing of fecal matter from ASD model mice is needed to provide additional insight. A recent report on microbiota and fecal metabolites comparing humans with ASD *vs.* controls revealed discernible differences between the two groups [53]. How this might be applicable to our murine ASD model is a

Due to the high number of differences observed in the gut microbiome of those with ASD [54], it is possible that the bacterial population variations are making an important impact on bilin metabolism. The significance of the gut on brain activity has begun to be heavily researched. In some studies, disorders such as autism, depression, and anxiety have seen lessening of symptoms based on the introduction of different bacteria within the patient's microbiome [13]. In particular, the altered microbiome of those with ASD has developed changes in the production of shortchain fatty acids. One such fatty acid noted was propionic acid, which has been reported to be increased in those with autism [55]. Activity on propionic acid chains are important to the conversion of bilirubin to bilirubin diglucuronide and may

Based on previous knowledge of the production of stercobilin within humans,

The discovery that stercobilin, and to a lesser extent stercobilinogen, are depleted in the fecal matter from a murine ASD model gives promise of the potential of these substances to serve as clinical biomarkers for ASD. Work to understand the relationship between the depletion of these bilins and the identity of the microbiota responsible is intriguing, as is the possibility that microbiota may play a role in the etiology of ASD; if this is true, it means that fecal transplants may have impact in the treatment of ASD, as recent clinical evidence

The authors would like to thank all of their coworkers who contributed to the results described in this chapter, including Kevin Quinn, Charmion Cruickshank-Quinn, Jordan Coffey, Anthony Vadas, Thomas Puleo, Katelyn Lewis, Gregory Pirrone, Eric Helms, Alessandra Dettori, Emanuela Azara, Mauro Marchetti, Paolo Tomasi, Giuseppe Delitala, Emma Fenude, Maria Piera Demontis, Elisabetta Alberico, Vittorio Anania, Silvia Gianorso, Will Friesen, Nhu Nguyen-Dudziak, Stephen Carro, and Michael Wach. Financial support for this research has come from the United Kingdom Legal Services Commission, the Fondazione Banco di Sardegna, the California Scottish Rite, the Mark Diamond Research Fund, and the National Institutes of Health through the Center for Research Resources (Grant

the results presented herein suggest that microbiome analysis coupled to the molecular analysis of bilins from fecal material is warranted. Fecal material can be collected noninvasively and proved to give a wealth of metabolomic information. Through the combination of these techniques, a combinatory biochemical and molecular biological approach to diagnosing ASD may yet be developed.

*DOI: http://dx.doi.org/10.5772/intechopen.84791*

subject worthy of follow-up investigation.

provide insight into the potential depletion observed [56].

#### *Stercobilin: A Putative Link between Autism and Gastrointestinal Distress? DOI: http://dx.doi.org/10.5772/intechopen.84791*

with *Escherichia coli*, although not reaching the levels observed for controls [52]. In context, our results showing depleted stercobilin and stercobilinogen in fecal matter of ASD model mice might suggest that *Clostridium* constitutes part of their microbiota but that *E. coli* may have been impacted such that bilin production is reduced. Clearly, microbiological testing of fecal matter from ASD model mice is needed to provide additional insight. A recent report on microbiota and fecal metabolites comparing humans with ASD *vs.* controls revealed discernible differences between the two groups [53]. How this might be applicable to our murine ASD model is a subject worthy of follow-up investigation.

Due to the high number of differences observed in the gut microbiome of those with ASD [54], it is possible that the bacterial population variations are making an important impact on bilin metabolism. The significance of the gut on brain activity has begun to be heavily researched. In some studies, disorders such as autism, depression, and anxiety have seen lessening of symptoms based on the introduction of different bacteria within the patient's microbiome [13]. In particular, the altered microbiome of those with ASD has developed changes in the production of shortchain fatty acids. One such fatty acid noted was propionic acid, which has been reported to be increased in those with autism [55]. Activity on propionic acid chains are important to the conversion of bilirubin to bilirubin diglucuronide and may provide insight into the potential depletion observed [56].

Based on previous knowledge of the production of stercobilin within humans, the results presented herein suggest that microbiome analysis coupled to the molecular analysis of bilins from fecal material is warranted. Fecal material can be collected noninvasively and proved to give a wealth of metabolomic information. Through the combination of these techniques, a combinatory biochemical and molecular biological approach to diagnosing ASD may yet be developed.
