Converging Pharmacological and Genetic Evidence Indicates a Role for Steroid Sulfatase in Attention

Background Attention-deficit/hyperactivity disorder (ADHD) is a complex neurodevelopmental disorder characterized by deficits in attention, increased motor impulsivity, and hyperactivity. Preliminary work in mice and humans has suggested the X-linked gene STS (which encodes the enzyme steroid sulfatase) as a mediator of attentional functioning and as a candidate gene for ADHD. Methods The effects of modulating the murine steroid sulfatase axis pharmacologically (through administration of the substrate dehydroepiandrosterone sulfate [DHEAS], 0–40mg/kg, or acute inhibition of the enzyme by COUMATE, 10mg/kg) or genetically (through loss of the gene in 39,XY*O mice) were assayed using the 5-choice serial reaction time task (5-CSRTT) a test of visuospatial attention and response control, and a locomotor activity paradigm. Results DHEAS administration improved 5-CSRTT performance under attentionally demanding conditions, whereas steroid sulfatase inhibition impaired accuracy under the same conditions. Loss of Sts expression constitutively throughout development in 39,XY*O mice resulted in deficits in 5-CSRTT performance at short stimulus durations and reduced anticipatory responding. Neither the pharmacologic nor the genetic manipulations affected basic locomotor activity. Conclusions These data provide converging evidence indicating a role for steroid sulfatase in discrete aspects of attentional functioning and are suggestive of a role in motor impulsivity. The findings provide novel insights into the neurobiology of attention and strengthen the notion of STS as a candidate gene for the attentional component of ADHD.


Supplemental Information
Subjects for genetic study 39,X Y* O mice were produced from two separate crosses: 1) 39,X Paf O (female) x 40,XY* (male) and 2) 39,X Paf O (female) x 39,X Y* O (male). Paf is an X-linked mutation involving a small inversion spanning the pseudoautosomal boundary which gives rise to a 'patchy fur' phenotype (1) in heterozygous females (or males), whilst Y* is a Y chromosome that has been hijacked by a non-Y centromere attached distal to the pseudoautosomal region (2,3). Cross 1 produces three male genotypes: 40,X Paf X Y* (which can be identified through their patchy fur between postnatal days 7-10), 40,X Paf Y* and 39,X Y* O; the latter two genotypes can be differentiated through a polymerase chain reaction (PCR) for the Sts gene (see below). Cross 2 produces two male genotypes: 40,X Paf X Y* (which can be identified through their patchy fur between postnatal days 7-10) and 39,X Y* O. 40,XY males were generated from a 40,XY x 40,XX cross. Care was taken to keep the genetic backgrounds for the crosses equivalent (i.e. predominantly MF1 with two C3H strainderived factors which enable fertility of 39,X Y* O males).

Animal husbandry
Upon arrival in Cardiff, the mice for the genetic study were treated with Baytril and Septrin antibiotics for one month in a negative-pressure isolator to cure a Pasteurella pneumotropica infection prior to being released onto the open racks. Mice for the pharmacological study were group housed (3-5 mice per cage) in a holding room maintained at 21°C ± 2°C and 50% ± 10% humidity, with a 12 hour light-dark cycle (lights on at 07:00hr). Mice for the genetic study were housed in an identical environment, either singly (due to the tendency of 39,X Y* O mice to fight) or in groups of up to three. Group housed mice for the genetic study were kept with mice of the same genotype. Initially, mice were allowed ad libitum access to food and water. Two weeks prior to the onset of behavioural testing mice were placed on a water restriction schedule whilst ad libitum access to food was maintained (4 hours access to water for 4 days and 3-4 hours per day thereafter according to the number of mice in the cage). Regular health checks and weighing ensured that mice were not adversely affected by this schedule. Behavioural testing was performed between the hours of 07:00hr and 12:00hr.

Training on the 5-CSRTT
Mice were trained to respond via a nosepoke to a light stimulus presented pseudorandomly in one of 5 uncovered holes of a standard 9-hole box; stimulus presentation occurred 5s (inter-trial interval, ITI) after initiation of the trial by a 'panel push' opposite the response array. A correct response (i.e. a nosepoke in the illuminated hole) resulted in delivery of 20μl of reinforcer behind the panel, the collection of which initiated a second trial. An incorrect response (i.e. a nosepoke in a non-illuminated hole), an omission (i.e. a failure to respond for the duration of the stimulus + 5s) or a premature response (i.e. a response prior to the onset of the stimulus light) resulted in a 5s 'time-out' period in which the house-lights were illuminated. This 'time-out' could be terminated through a panel push, which initiated a new trial. Mice were given one training session per day, of 20 minutes duration. During training, the stimulus duration was reduced from 32s to a baseline of 1.0s in the following sequence (32s, 16s, 8s, 4s, 2s, 1.8s, 1.6s, 1.4s, 1.2s and 1.0s). The stimulus duration for a given mouse was reduced to the next level once it had achieved stable and high performance at the previous level (i.e. >50 trials, >80% accuracy i.e. ratio of correct:total responses and <30% omissions i.e. no response to <30% of trials initiated).

Gene expression
In order to try and identify a neurobiological correlate for the deficit in accuracy in COUMATE-treated mice relative to their vehicle treated counterparts, gene expression was compared between the groups in previously drug-naïve mice (n = 7 for both groups).
We assayed two brain regions where Sts expression is highest: the frontal cortex, and the thalamus (4). We examined a number of ADHD candidate genes covering three major neurochemical systems: dopaminergic (Drd4, Drd5, Slc6a3), serotoninergic (Htr1b) and cholinergic (Chrna4, Chrna7) (5). Additionally, the expression of GABA A receptor subunit genes which are highly expressed in regions of high Sts expression, and/or whose expression may be sensitive to steroid sulfatase levels (6) was assessed (Gabra1, Gabra2, Gabra3, Gabra4, Gabra5, Gabrd, Gabrg2). Quantitative PCR was performed as described previously (7); primer sequences are available on request. There were no significant differences in expression between vehicle and COUMATE-treated mice for any of the genes assayed ( Figure S2 and Table S1). The ADHD candidate genes Dbh and Slc6a4 were not expressed in either the frontal cortex or the thalamus, whilst the dopaminergic genes Drd4, Drd5 and Slc6a3 were expressed at low, or non-existent, levels in the thalamus (C t values of >30).