Michael W. Pankhurst and Ian S. Mclennan1

Inhibin B and anti-Müllerian hormone/ Müllerian Inhibiting Substance may contribute to the male bias in autism

Michael W. Pankhurst and Ian S. McLennan1

Supplementary Information

InhB and AMH can confound each other

InhB and AMH are both derived from the Sertoli cells of the testes, and the levels of these hormones thus correlate in non-autistic boys 15 and in the current cohort of autistic spectrum boys (Fig. S3, R=0.66). This implies that approximately 38% of the variation of InhB and AMH relate to common influences, such as the number of Sertoli cells in a boy’s testes. AMH and InhB have opposite effect on autistic traits (Table 3 44). This will diminish the apparent relationship between each hormone and the autistic traits, as 38% of the variation in each hormone relates to an opposing affecter. For this reason, we suggest that the Sertoli cell hormones should always be examined together, with their relative influences analysed using linear regression models.

The correlates relate to the hormones and not the testes

InhB and AMH are used clinically to estimate Sertoli cell number, which is a determinant of the size of the testes. The current observations do not link testes size to autistic spectrum disorders, as InhB and AMH have opposing effects. The levels of hormones are affected by their rate of production and release. Thus, it is probable that the boy-boy variation in the levels of InhB and AMH relate in part to the function of their testes. However, hormone levels are also influenced by the half-life of the hormone in blood, which may vary between boys. Thus, the current study only establishes a link between hormones (InhB/AMH) and autistic traits, and does not prove that autistic symptoms are linked to the function of the testes, per se.

When do the hormones act?

The current study examined 4- to 9-year-old boys, as the assessment of autistic symptoms is difficult at a younger age. However, the autistic pathology begins at an early age 2, and both InhB and AMH are present during the entire period of male development 19, 21. Furthermore, the variation between boys is stable over time 15, 21. It is therefore possible that the correlation between the hormones and autistic symptoms arise because the levels of hormones in the boys reflect their levels at earlier stages of development.

Do the hormones affect severity of symptoms?

The relationship between InhB and AMH and the ADI-R scores related to an effect on the number of autistic traits rather than to the severity of the traits (Table 2). Whilst this is correct for ADI-R scores and clinical diagnosis, it does not necessarily imply that the severity of symptoms experienced by a boy is independent of his Sertoli cell hormones. The power of regression analyses decrease when the scale is compressed. The severity scale for each trait in the ADI-R is 0-3, with the distinction between 2 and 3 being irregularly used: the scores of 2 and 3 are included as “2” for clinical diagnosis 16. Consequently, the reported experiments have a greater ability to detect associations with the breadth of symptoms than with the severity of symptoms. This technical limitation notwithstanding, mild pathology is more likely to be influenced by modulators, and the data as a whole is consistent with this.

Girls with autistic spectrum disorders

AMH is profoundly dimorphic during development 19. However, Inhibin has two isoforms, A and B, both of which are hormones. InhA has a female bias, whereas InhB has a male bias. InhA and InhB have similar, but not identical activities 45, and are expressed at different times in development (Fig. S1). Consequently, the current study points to the need for InhA to be examined in girls

Direction of causality

The direction of causality cannot be determined from a correlation analysis. AMH and InhB are both plausible regulators of brain development, suggesting that the direction is hormone to brain. Alternatively, one function of hormones is to transduce environmental influence on to physiological functions. We therefore do not exclude the possibility that autistic pathology alters the levels of male hormones. However, the behaviour of boys with an autistic spectrum disorder is highly divergent from the general population, as is the growth trajectory of their brains. Hence, if either autistic behaviour or growth trajectory regulates the levels of AMH and InhB, then the levels of AMH and InhB should be different between autistic and non-autistic boys. The current data is inconsistent with this possibility (Fig. 2), although we note that the number of non-autistic boys in the current study is insufficient to exclude a minor influence of autistic behaviour on hormone levels.

Figure S1. The pattern of gonadal hormones varies during development. The levels of AMH (red line), inhibin B (Inhb, blue line) and testosterone (green line) are dimorphic during development. 1The level of InhB in the plasma of women varies considerably during the ovarian cycle. This figure is generated from multiple sources, including references 19, 21, 38, 39, 42, 43, 46, 47.

Figure S2. A boy’s level of AMH does not correlate with his ADI-R scores.The boy’s level of AMH was measured by ELISA and correlated with the scores for each of the 3 ADI-R domains. The level of AMH was not significantly correlated with the boy’s social score (A) R=0.07, n=82, (B) communication and language score (B) R=0.09, n=63 or with his repetitive and stereotypic behaviour score (C) R=0.11, n=82.

Figure S3. A boy’s level of AMH and InhB positively correlate. The levels of AMH and InhB were measured by ELISA. The values for the non-autistic boys are shown as black squares (n=16), with the data from autistic spectrum boys being illustrated as green circles (n=82). The levels of AMH and InhB positively correlated (R=0.66, p=0.000, n=82 for the autistic spectrum boys; R=0.66, p=0.006, n=16 for the non-autistic boys).

44.Pankhurst M, McLennan IS. Inhibin B and anti-Müllerian hormone/ Müllerian Inhibiting Substance may contribute to the male bias in autism. Current Manuscript 2012.

45.Brown CW, Houston-Hawkins DE, Woodruff TK, Matzuk MM. Insertion of Inhbb into the Inhba locus rescues the Inhba-null phenotype and reveals new activin functions. Nat Genet 2000; 25: 453-457.

46.Chada M, Prusa R, Bronsky J, Kotaska K, Sidlova K, Pechova M et al. Inhibin B, follicle stimulating hormone, luteinizing hormone and testosterone during childhood and puberty in males: changes in serum concentrations in relation to age and stage of puberty. Physiol Res 2003; 52: 45-51.

47.Chada M, Prusa R, Bronsky J, Pechova M, Kotaska K, Lisa L. Inhibin B, follicle stimulating hormone, luteinizing hormone, and estradiol and their relationship to the regulation of follicle development in girls during childhood and puberty. Physiol Res 2003; 52: 341-346.

Sertoli cell hormones and autism p1/1

Domain / SOC1T / SOC2T / SOC3T / SOC4T / COM1T / COM2T / COM3T / COM4T / BEHT1T / BEHT2T / BEHT3T / BEH4T / n
SOC1T_CS / 1.00 / 0.29 / 0.20 / 0.30 / 0.26 / 0.15 / 0.00 / 0.31 / 0.02 / 0.01 / 0.17 / 0.12 / 82
SOC2T_CS / 1.00 / 0.31 / 0.32 / 0.33 / 0.37 / 0.00 / 0.41 / 0.03 / 0.07 / 0.14 / 0.10 / 82
SOC3T_CS / 1.00 / 0.34 / 0.36 / 0.15 / 0.00 / 0.26 / 0.01 / 0.00 / 0.07 / 0.00 / 82
SOC4T_CS / 1.00 / 0.40 / 0.24 / 0.00 / 0.42 / 0.00 / 0.01 / 0.37 / 0.05 / 82
COM1T_CS / 1.00 / 0.20 / 0.03 / 0.39 / 0.00 / 0.00 / 0.11 / 0.01 / 82
COM2T_CS / 1.00 / 0.00 / 0.16 / 0.04 / 0.02 / 0.12 / 0.21 / 63
COM3T_CS / 1.00 / 0.02 / 0.09 / 0.14 / 0.04 / 0.05 / 63
COM4T_CS / 1.00 / 0.00 / 0.00 / 0.27 / 0.01 / 82
BEH1T_CS / 1.00 / 0.20 / 0.04 / 0.12 / 82
BEH2T_CS / 1.00 / 0.04 / 0.21 / 82
BEH3T_CS / 1.00 / 0.06 / 82
BEH4T_CS / 1.00 / 82

Table S1. The ADI-R subdomains are measuring traits that are largely independent of each other.

The independence of the traits measured by each subdomain was examined using partial correlates. The values are correlation coefficients (R2).

Sertoli cell hormones and autism p1/1