Urinary Porphyrin Excretion in Neurotypical and Autistic Children

James S. Woods; Sarah E. Armel; Denise I. Fulton; Jason Allen; Kristine Wessels; P. Lynne Simmonds; Doreen Granpeesheh; Elizabeth Mumper; J. Jeffrey Bradstreet; Diana Echeverria; Nicholas J. Heyer; James P.K. Rooney


Environ Health Perspect. 2010;118(10):1450-1457. 

In This Article


Urinary Porphyrins are Naturally Elevated in Young Children

We describe here mean urinary porphyrin concentrations for children in the age range of 2–12 years who participated in the present study. Of particular note is the observation that younger children have inherently higher porphyrin concentrations, particularly of uro-, hepta-, and coproporphyrins, which decline by as much as 2.5 times over the 2- to 12-year age range. Also of interest is the finding that precoproporphyrin, an atypical porphyrin previously identified only in adult humans and animals with prolonged exposure to Hg or Hg compounds, is present in substantial concentrations in urine of younger children. This is a novel and unexpected finding in light of previous observations from studies in animals (Woods et al. 1991) showing that precoproporphyrin is formed as a consequence of Hg inhibition of uroporphyrinogen decarboxylase in the kidney during prolonged exposure, producing excess pentacarboxylporphyrinogen, which then competes with coproporphyrinogen as a substrate for coproporphyrinogen oxidase as the basis of precoproporphyrin formation (Woods et al. 2005). The etiology of this atypical porphyrin in the urine of young children in the presumed absence of prolonged Hg exposure as observed here remains unknown. One possible explanation may be the consequence of accelerated hepatic heme biosynthesis that occurs during the period of perinatal development (Woods 1976). In this respect, formation of precoproporphyrin would be consistent with the observation that the specific activity of hepatic uroporphyrinogen decarboxylase in perinatal rat liver greatly exceeds that of the adult (Woods and Kardish 1983), likely generating comparably greater amounts of pentacarboxylporphyrinogen to compete with coproporphyrinogen as a substrate for coproporphyrinogen oxidase, as proposed in the etiology of precoproporphyrin in the presence of Hg exposure in adults (Woods et al. 2005). Further research is required to confirm this prospect.

Comparison of Urinary Porphyrins in NT and AU Children

Our findings suggest that mean concentrations of uro- and precoproporphyrins are comparable between NT and AU children of the same age ranges. In contrast, the concentrations of all remaining porphyrins, particularly hexacarboxyl-, pentacarboxyl-, and coproporphyrins, were significantly higher in AU children than NT children, especially in older age groups. Several possibilities might account for these differences. Of initial concern, Hg exposure appears unlikely to play a role in this effect, because no significant differences were observed between NT and AU subjects for indices of past exposure to Hg from dental or medical sources, as reported by parents/caregivers. Additionally, urinary Hg concentrations, measures of recent Hg exposure, were very low among all subjects in this study (Table 2), and no significant differences between diagnostic groups were observed. As noted recently (Woods et al. 2009a), incipient although statistically nonsignificant changes in urinary porphyrin concentrations were seen among children with urinary Hg concentrations derived from prolonged dental amalgam Hg exposure on the order of 3.2 µg/g creatinine. This is nearly 10 times the mean urinary Hg concentration observed among children in this study. Similar findings describing very low blood Hg levels and insignificant differences between NT and AU children have recently been reported (Hertz-Picciotto et al. 2010). These observations do not preclude a possible role of Hg exposure from sources not measured or validated in the present study, especially during the perinatal period, in the etiology of autism or related neurodevelopmental disorders in some children, particularly in relation to genetic variation that may predispose to increased risk of the neurotoxic effects of Hg as Hg0 as reported in adults (Echeverria et al. 2005, 2006, 2010; Heyer et al. 2009). Our findings indicate instead that porphyrin metabolism, particularly in preadolescent children, may be too disordered or differently regulated to permit detection of the Hg-mediated changes in urinary porphyrin excretion apparent in adult subjects. Further studies using a substantially larger population, such as the National Children's Study now in progress (National Children's Study 2010), are required to resolve this question.

Another factor that may account for the differences in urinary porphyrin levels between AU and NT children is mitochondrial dysfunction, a disorder commonly associated with autism (Correia et al. 2006; Oliveira et al. 2005; Pons et al. 2004). Of particular interest in this respect is the prospect of deficient mitochondrial porphyrin uptake mediated by the recently identified mammalian mitochondrial porphyrin transporter Abcb6 (Krishnamurthy et al. 2006). Abcb6, one of several identified ATP-binding cassette transporters, is located in the outer mitochondrial membrane and has a particularly high affinity for coproporphyrinogen III. Defects in Abcb6 gene expression or in Abcb6 activity could predispose to impaired mitochondrial porphyrin uptake, leading to cellular accumulation and aberrant porphyrin metabolism and excretion. Similarly, defects in the mitochondrial transmembrane domain that mediates the binding of porphyrins with the Abcb6 transporter could restrict normal porphyrin metabolism, contributing to the disordered porphyrin excretion observed (Krishnamurthy et al. 2006, 2007). Although the association of Abcb6 with autism has yet to be investigated, numerous validated missense mutations of the Abcb6 gene have been reported (National Center for Biotechnology Information 2010).

Finally, although genetic susceptibility studies were not included as part of the present investigation, previous studies identified a polymorphism in the gene encoding corproporphyrinogen oxidase (CPOX, EC (Li and Woods 2009; Woods et al. 2005) that may predispose to impaired heme biosynthesis and subsequent heme-dependent neurological functions (Chernova et al. 2006; Echeverria et al. 2006). Genotyping studies of 100 DNA samples from autistic children acquired through the University of Washington Autism Center revealed more than double the expected frequency of the homozygous variant of this polymorphism (CPOX4) (rs1131857). An intriguing notion rests on the possibility that mitochondrial respiratory chain disorder associated with CPOX4, which itself is linked to the mitochondrial inner membrane (Grandchamp et al. 1978), could account for exaggerated porphyrin excretion as observed here among at least a subgroup of those with autism. Future studies involving a larger cohort of subjects are required to confirm these findings and to define the genetic and/or metabolic factors associated with altered porphyrin excretion in autism.

Strengths and Limitations

A principal limitation of this exploratory study is the relatively small population of NT and AU subjects among whom we sought to define and differentiate excretion levels of metabolites (urinary porphyrins) that can exhibit substantial intraindividual (e.g., diurnal) and interindividual variability, especially in children. Despite this shortcoming, the findings demonstrate significant differences both with age among NT and between NT and AU of the same age, suggesting disordered porphyrin excretion as a metabolic characteristic among at least some AU subjects. Although these findings must be confirmed through larger studies, these preliminary observations provide a context for better understanding and interpreting altered porphyrin excretion among children with AU/ASD.

An additional limitation is potential misclassification of case (AU) or control (NT) status for subjects enrolled through the ARI. We view the possibility of AU misclassification as unlikely, however, because of the multidisciplinary evaluation protocol employed by the neurodevelopmental diagnostic and treatment centers from which subjects were recruited. We note also likely continuity in the diagnostic procedures employed for most subjects enrolled in the study, supporting homogeneity in the AU, PDD-NOS, and NT diagnoses. Nonetheless, our inability to confirm each diagnosis by outside review of subjects' records is a potential limitation of the present exploratory investigation. Future studies in which subjects are identified through established registries such as the Autism Treatment Network (2010) or that developed by the CHARGE study (Hertz-Picciotto et al. 2006) will minimize prospects of this concern.

Although we made direct measurements of urinary Hg levels as indices of recent Hg exposure, there was the potential for measurement error based on participants' recall of past Hg exposures from dietary, medical, and dental sources, which could not be fully validated in this study. Such exposures may be of concern in relation to perinatal events that can be accurately assessed only in the context of a prospective study design. Moreover, potential exposures to Hg from other environmental or occupational sources were not assessed and therefore could not be controlled in the present analysis. Thus, although we found few significant differences between AU and NT in reported measures of past Hg exposure, the possibility of exposure misclassification remains a limitation of this study.

Principal strengths of this study were the availability of urine samples for porphyrin and Hg analyses from all study participants and our established capabilities for accurately measuring and interpreting these constituents in the context of this study. Notably, the urinary porphyrin levels reported herein among NT children ≥ 8 years of age are comparable with normative values recently described for children and adolescents of the same ages who were participants in a large clinical trial (Woods et al. 2009b), supporting the generalizability of these findings. The urinary Hg levels measured in this study were also comparable with those reported for a nationally representative sample of children 6–11 years of age acquired as part of the 2003–2004 U.S. National Health and Nutrition Examination Survey [geometric mean = 0.245; 95% CI, 0.213–0.304) (Centers for Disease Control and Prevention 2007). Finally, despite the small number of cases, the results were consistent across age groups with significant differences in porphyrin levels between the diagnostic groups, suggesting that further consideration of this observation may be warranted.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.