Animals

Male mice from three different strains were used at 10 weeks old. For initial temperature experiments, C57 mice (C57BL/6 J (#000664)) were used for all the treatments, and for the assessment of ASD-like models, C58 (C58/J (#000669)) and Shank3B- (B6.129-Shank3 < tm2Gfng > /J; heterozygous (#017688)) were used. All three strains were from The Jackson Laboratory. Animals were housed in cages of four at 21 °C with a 12 h light/dark cycle. Food and water access was ad libitum. All animal experimentation was performed under licence granted by the Health Products Regulatory Authority, Ireland, with approval from the local ethical committee and in compliance with the Cruelty to Animals Act, 1876, and the European Community Directive, 86/609/EEC. Every effort was made to minimize stress to the animals.

Subcutaneous temperature transponders implantation:

Subcutaneous temperature transponders 14 mm wide and 2 mm diameter (IPTT-300 (BMDS) were implanted following the manufacturer’s instructions, under light isoflurane anaesthesia, at least three days before being subjected to the WBH protocol. The correct functioning of the transponder was checked before and after the implantation. Temperature monitoring was taken with the IPTT-300 thermoreader (BMDS).

Brain-implanted thermocouples and real-time temperature recordings:

For thermocouple recordings of brain temperature, mice (n = 6) underwent stereotaxic surgery to implant MBR-5 intracerebral guide cannulae (ID 457 μm, OD 635 μm; BASi Research Products, USA) into the striatum. Mice were anaesthetized with isofluorane (4% for induction, 1.5–3.0% for maintenance; IsoFlo®, Abbott, UK). The surgical site was shaved and disinfected, and lidocaine was administered subcutaneously for local anaesthesia. The skull was exposed, and three stainless steel support screws were implanted into the skull. A burr hole was made over the striatum, and the guide cannula was lowered into the caudate putamen (0.3 mm A/P; ± 2.0 mm M/L; 3.0 mm D/V), allowing room for the thermocouple to extend into the striatum once inserted into the guide cannula. The guide cannula was cemented into place (Dentalon® Plus, Heraeus-Kulzer, Germany), and once set, the scalp was sutured to close the wound. All animals were given saline (0.9%; 3 ml/kg body weight) and perioperative analgesia was provided (0.3 mg/kg body weight; Buprecare®, AnimalCare Ltd., UK) before animals were allowed to recover in an incubator set to 28 °C, with access to a food gel and hydrogel. Following 7 days of recovery, mice underwent three days of brain temperature recordings in an animal recovery chamber (Vet-Tech. model: HE010). There were 4-day rest periods between each recording day. On each recording day, the plug was removed from the implanted guide cannula and the thermocouple probe was inserted into the striatum. The thermocouple was an ultrafast T-type implantable device (IT-23; ADInstruments Ltd., UK) which was modified to fit into the implanted MBR-5 guide cannula. Briefly, a 2 cm length of deactivated fused silica tubing (ID 320 μm, OD 430 μm; Trajan Scientific Europe Ltd., UK) was cut and then carefully inserted under a microscope into a BR microdialysis probe head (BASi Research Products) so that it protruded 7 mm beyond the end of the probe head shaft. A small amount of glue (WEICON Epoxy Minute Adhesive) was then applied to the end of the shaft to fix the silica in place. After ca. 1 h, the thermocouple was inserted through a 1.8 cm length of PEEK tubing (ID 860 μm, OD 1270 μm; Plastics One, Roanoke, VA, USA) which was pushed well up the probe so that it was out of the way until later gluing. The thermocouple was then carefully inserted into the silica tubing under a microscope until it protruded ca. 1 cm. A small amount of epoxy glue was applied to the end of the silica and the thermocouple gently pulled back until it was 1 mm from the end of the silica as confirmed using a digital calliper. This was then left to dry for ca. 1 h before placing some epoxy glue around the silica at the top of the BR probe head. The PEEK tubing was then immediately pulled down and carefully inserted into the epoxy. Following overnight storage, the modified thermocouple was connected to a T-type Thermocouple Pod (ADInstruments Ltd.), which was in turn connected to the pod port of an e-Corder (eDAQ Pty Ltd, Australia), and the operational characteristics (temperature vs. voltage output) tested using the suppliers’ guidelines (ML312 T-type Pod Manual, AD Instruments Ltd.). This involved placing the thermocouple in a jacketed cell (ALS Ltd, IJ Cambria Scientific Ltd, Llanelli, UK) attached to a thermostatically controlled circulating water bath (Julabo Corio CD-BC4, Fisher Scientific, Dublin, Ireland) and recording the temperature vs. voltage output in Chart TM (Version 5, eDAQ Pty Ltd) at a sampling rate of 1 Hz. Raw temperature recordings in vivo were made in Chart TM (Version 5, eDAQ Pty Ltd, Australia) at a sampling rate of 10 Hz. On each recording day, baseline brain temperature recordings were made for 1 h. Following baseline recordings, the mice went through 3 interventions: (a) Day 1: room temperature protocol; (b) Day 5: whole-body hyperthermia (WBH) protocol; and (c) Day 9: LPS protocol (250 mg/Kg; i.p.). During brain temperature recording experiments, core body temperatures were measured using subcutaneous temperature transponders (IPTT-300 (BMDS) and the IPTT-300 thermoreader (BMDS) every 20 min without removing the mouse from the test chamber. All activity was marked on the real-time brain temperature recording.

Whole-body hyperthermia and LPS treatment

Mice were exposed to whole-body hyperthermia (WBH) for 4 h by transferring them to a small animal recovery chamber (Vet-Tech. model: HE010) set at 38.5 °C (30 ± 1% humidity) to reach the target body temperature of 39.5 ± 0.5 °C. This was adapted for C58 strain (see results section). WBH protocol was invariably performed at the same time of day (8 a.m.) to minimize the effects of the circadian rhythm that occurs in the body temperature of rodents. Prior to starting the WBH protocol, the animals were transferred from their home cage to an empty new cage without bedding, food or water to allow them to habituate to a new space (T−1 h) and their body temperature and weight were taken. One hour later (T0h), their body temperature was recorded again using the thermoreader and subcutaneous temperature transponder (IPTT-300, BMDS) and they were injected with 500 µl either saline for control and WBH groups or LPS (250 µg/kg) before being transferred to the WBH chamber or the room temperature (RT) chamber (21 ± 1 °C; 50 ± 1% humidity). This LPS dose was chosen in order to ensure lethargy (as was observed in human studies of fever and ASD) but in the absence of robust hyper- or hypothermia (see Additional file 1: Fig. 2). Animals were left to freely move and explore, and their temperatures were taken every 20 min from T0h to T4h. Two hours after starting WBH (T2h) the temperature and weight of the animals was taken and every animal, independent of treatment, was injected with 500 µl of saline to avoid dehydration and quickly returned to the chamber. Temperature recording every 20 min was resumed. At 4 h, temperature and weights were taken and every animal, independent of treatment, was again injected with 500 µl of saline. Although some WBH animals did have short episodes of jumping in the heating chamber, they were largely lethargic for the 4-h period and, on removal from the heating chamber, did rapidly re-establish normal locomotor behaviour, while LPS animals remained lethargic for many hours. For molecular experiments, animals were euthanized immediately after the WBH protocol and for behavioural studies, they underwent a ‘step-down’ cooling protocol to avoid the rebound hypothermia that occurs after WBH [11]. The ‘step-down’ protocol was as follows: the heated animals were returned to the WBH chamber, which was set, sequentially, at 32, 28, 23 °C and RT, for 20 min at each point in the sequence. Their temperature was also monitored, and after the step-down protocol, they were either returned to their home cage or started a panel of behavioural tasks (T5h), depending on the requirements of the experiment. (See Fig. 1 for protocol schematic).

Fig. 1

Whole-body hyperthermia produces elevation of body and brain temperature. A Schematic timeline of the whole-body hyperthermia (WBH) protocol, starting at the −60 min point for habituation, ending at 240 min of WBH treatment and refined by a step-down period to slowly cool down the animals up to 300 min. B Body temperature time course (°C) over the five hours duration of the protocol, measured by subcutaneous temperature transponders. Data are shown as Mean ± SEM (n = 24 for each group) and are analysed by repeated measures two-way ANOVA (* vs. RT group; # vs. LPS. Bonferroni post hoc, p < 0.05 °C). C Real-time body and brain temperature (°C) over the five hours duration of the protocol. Brain temperature was monitored using brain-implanted thermocouples in a small subset of those animals monitored by subcutaneous transponders (n = 5 for LPS and 6 for other groups). Abbreviations: RT room temperature; LPS intraperitoneally injected lipopolysaccharide, 250 µg/kg; WBH whole-body hyperthermia

Tissue preparation

For the analyses of transcriptional changes, animals were terminally anaesthetized with sodium pentobarbital at 4 h post-WBH protocol (Euthatal; Merial Animal Health) and rapidly transcardially perfused with heparinized saline before the dissection of hypothalamus, hippocampus and amygdala that were snap frozen in liquid nitrogen and stored at − 80 °C until use. Animals for immunohistochemical examination were terminally anaesthetized with sodium pentobarbital (Euthatal; Merial Animal Health) and transcardially perfused with heparin–saline followed by 4% paraformaldehyde (PFA). Brains were gently removed and postfixed in 4% PFA. Coronal sections, 50 μm thick, were obtained using a Vibratome (Leica, Laboratory Instruments and Supplies, Ashbourne) to perform immunohistochemistry.

RNA extraction, cDNA synthesis and quantitative PCR

Total RNA was isolated using the RNeasy Plus Mini method (Qiagen, Limburg, the Netherlands) following the manufacturer’s instructions. The RNA yield and quality of each sample were quantified based on optical density (OD) using the NanoDrop ND-1000 UV–Vis spectrophotometer (Thermo Fisher Scientific). cDNA synthesis was carried out using a high-capacity cDNA Reverse Transcriptase Kit (Applied Biosystems, Warrington, UK). Primer and probe sets were designed using NCBI Nucleotide tool and amplified a single sequence of the correct amplicon size, as verified by SDS-PAGE. Primer pair/probe sequences are shown in Table 1. Samples for RT-PCR were run in duplicate using FAM-labelled probes or SYBR green dsDNA-intercalating fluorescent dye (Roche) in a StepOne Real-Time PCR system (Applied Biosystems, Warrington, UK) under the cycling conditions: 95 °C for 10 min followed by 95 °C for 10 secs and 60 °C for 30 secs for 40–45 cycles. Quantification was achieved by exploiting the relative quantitation method. We used cDNA, prepared from isolated RNA that was pooled from the brains of WBH-treated and LPS-injected mice as a standard that expressed all genes of interest. Serial 1 in 4 dilutions of this cDNA were prepared in order to construct a linear standard curve relating cycle threshold (CT) values to relative concentrations, as previously described [12]. Gene expression data were normalized to the housekeeping gene 18S and expressed as relative concentration.

Table 1 Primer and probe sequences for quantitative PCRImmunohistochemistry

Immunohistochemistry was carried out on free-floating sections under moderate shaking. All washes and incubations were done in 0.1 M phosphate buffer pH 7.4, containing 0.3% bovine serum albumin and 0.3% Triton X-100. The endogenous peroxidase activity was quenched in a solution of 3% hydrogen peroxide in 30% methanol. Sections were incubated overnight at 4 °C with anti-c-Fos (C-10) mouse monoclonal IgG2 (Santa Cruz Biotechnology, Heidelberg, Germany), diluted 1:1000 in the presence of 5% normal horse serum (Vector Laboratories Inc., Burlingame, CA). Next day, sections were incubated for 2 h with biotinylated horse anti-mouse secondary antibody (1:300, Vector). After several washes in phosphate-buffered saline (PBS), ABC method was used (Vectastain, PK6100, Vector) and the reaction product was revealed using 3, 3’ diaminobenzidine as chromogen (Sigma-Aldrich) and H2O2 as substrate. Finally, sections were dehydrated, mounted on gelatinized slides, coverslipped and photographed using an Olympus DP25 camera (Mason) mounted on a Leica DM3000 microscope (Laboratory Instruments and Supplies, Ashbourne), captured using CellA™ software (Olympus, Mason).

cFos analyses

The assessment of cFos activation was by a qualitative analysis using series of sections separated by 300 µm, beginning from the olfactory bulbs and continuing to the brain stem. Microscope images were taken at 10 × and 20x (n = 5–6). Regions of interest were selected based on prior studies used to capture the two main components of fever: temperature and inflammation. Thus, regions were chosen based on previous works that analysed cFos activation under cold/warm temperature protocols and after LPS treatment [13,14,15]. However, we also performed an anterior to posterior screen to positively identify any brain regions that showed particularly robust activation with respect to RT controls. Therefore, selection was not performed blind to treatment. Images for brain regions of interest, at positions along the anterior–posterior axis (with reference to the Allen Brain Atlas), were taken at either 10 × or 5 × magnification (n = 5–8), and for each photographed region, all sections were captured under the same conditions of light intensity, exposure, colour balance and saturation. Labelled cFos cells were analysed using Fiji (an open-source image processing package built off ImageJ2 software). Images were converted to 8 bit and thresholded before setting particle size (> 50) and circularity values (0.5–1). Throughout this process, cells detected by ImageJ automatic counts were compared visually to the original photograph and sample manual quantitative counts of cFos cells were compared to automatic counts to verify the validity of the methodology. Cell counts are expressed as cells/mm2.

Blood glucose measurements

Two different methods were used to assess blood glucose. For serial sampling, the blood glucose levels were measured via serial tail vein microsampling (less than 10 μl) one hour before starting the protocol and then at 40 min, 2 h, 4 h, 7 h and 24 h. Animals were bled using a 30G lancet, and glucose was determined in the blood drop with the precision Xtra glucometer (Abbott). In a different set of experiments, the glucose levels were determined in the first drop of blood from the right atrium immediately before performing transcardial perfusion (4 h or 24 h after the heating protocol).

Plasma ELISA assays

Animals were terminally anaesthetized at 4 h post-WBH protocol with sodium pentobarbital (Euthatal, Merial Animal Health). The thoracic cavity was opened, and blood was collected in heparinized tubes directly from the right atrium of the heart. This whole blood was spun at 1.5 × g for 15 min to remove cells; the plasma was then aliquoted and stored at − 20 °C until use. These samples were then diluted appropriately and analysed for IL-1β, TNF-α and IL-6 by sandwich-type ELISA, using ELISA MAX Mouse IL-6, ELISA MAX Mouse IL-1β (Biolegend, San Diego, USA) and DuoSet Mouse TNF-α (R&D Systems, Minneapolis, USA). The required capture and detection antibodies, cytokine standard and avidin–HRP (IL-6) or streptavidin–HRP (TNF-α) were supplied with each respective kit; however, for IL-1β a more sensitive streptavidin poly-HRP (Sanquin, Amsterdam, the Netherlands) was used in place of the supplied one. Optical density was read at 450 nm with correction at 570 nm. Standard curves for each antibody were used, and samples were quantified only if the absorbance fell on the linear portion of the standard curve. Reliable quantification limits for the assays used were IL-1β 31.25 pg/ml, TNF-α 15.6 pg/ml and IL-6 15.6 pg/ml.

Experimental design of behavioural studies

Given the heating protocol, it was necessary to test several behaviours in a short time after the animal had emerged from the heating protocol. Therefore, we prioritized those behaviours that were most important to assess in these distinct strains. The criteria for prioritizing individual tasks for each strain were: 1) That those behavioural indices were demonstrably altered (at baseline) in the ASD strain of choice with respect to normal controls. 2) That the assembled test battery contained tasks that could be regarded as ‘positive symptoms’, i.e. higher scores are present in the ASD strain with respect to controls, and ‘negative symptoms’, i.e. lower scores are present in the ASD strain with respect to controls. This was done so that ‘improvements’ in these ASD-relevant behaviours could not be confounded by a general suppression of all spontaneous activity. Therefore, careful phenotyping was essential at baseline, even though assessment of those was not novel per se, in order to develop test batteries that allowed us to select for tasks that showed impairments in our chosen strains, in our hands (see Figs. 6 and 8). From those baseline measures, we then selected the most sensitive battery of tests that could be conducted in a short time frame after ASD strains were removed from the heating chamber. Importantly, the selected behavioural tasks included behaviours that are well known as hallmarks of these strains, but not all tasks listed below could be performed in WBH experiments with each strain. Experiments were performed using independent cohorts of mice since strains were obtained at different times and because the size of cohorts that were required to perform these analyses could not accommodate experiments with > 2 strains. In the first instance, C57 mice were used to illustrate that WBH and LPS had quite distinct effects on spontaneous behaviour in the hours after treatment and independent cohorts of C57 mice were used subsequently when acting as controls in experiments with C58 and Shank3B mice since each of these experiments had to be conducted and presented independently.

Burrowing

Burrowing is a species typical behaviour and was performed following Deacon’s protocol [16, 17]. Burrows were made from a 200-mm-long, 68-mm-diameter black plastic tube. One end of the tube was closed with a piece of plastic from the same material, and the other end was open and raised 30 mm above the cage floor by two 50 mm screws. Animals were placed into individual opaque cages with fresh bedding and provided with water and the burrowing tube filled with 300 g of food pellets as substrate. The food pellets remaining in the burrowing tubes were measured after 2 h and at 24 h, and this weight was subtracted from 300 g to calculate the burrowing activity for each mouse. At 24 h, the mice were returned to their home cages.

Open-field activity

The open-field test was used to assess spontaneous activity in a novel environment, and it also served as the habituation period for the social interaction test that immediately followed it (detailed below). Briefly, mice were allowed to freely and individually explore an open-field arena (58 × 33 × 19 cm; divided into squares of 10 × 10 cm) for 10 min. Activity was monitored via an overhead camera and recorded using AnyMaze software (version 4.99). The mice were assessed on parameters such as distance travelled, mean speed, rotations, time spent in the outer and inner zone and time freezing.

Social interaction

We assessed social interaction using a rectangular arena 58 × 33 × 19 cm. Two wire mesh cages (9 cm diameter) were placed in the middle of each half of the arena. The test began with a habituation period to the arena for 10 min. (This was used to take measurements of locomotor activity as described in ‘Open-field activity’.) Thereafter the mouse was allowed to adapt, for 5 min to the placement of two small empty wire mesh cages for 5 min. Finally, the social preference test was conducted for 10 min, during which an unfamiliar (stranger) conspecific mouse of the same age, weight and sex was placed into one of the mesh cages, whereas the other mesh cage contained an inanimate object. AnyMaze software was linked via an overhead camera and used to recorded test mice, tracking the movement of the test mice, and recording parameters such as time spent sniffing the novel mouse and the number of sniffs, time spent in the area of the mouse mesh cage vs. the empty cage and time freezing. After the trial, the arena and wire cages were cleaned with 10% ethanol.

Backflips and upright scrabbles

Backflips and upright scrabbles have been described as specific behavioural alterations of the C58/J strain [18,19,20,21,22]. The test has two parts. First, a period (5 min) of adaptation to an opaque plastic cage (19.5 × 31 × 13 cm) with a regular metal bar lid placed on top provided with fresh bedding for each mouse. The second part consisted of observation and counting of the number of backflips and upright scrabbles for a period of 30 min. Back flipping manifests as backward somersaulting, often with the assistance of the cage lid, while upright scrabbling consisted of rapidly running or climbing ‘on the spot’, usually against a wall or corner, which may be related to wall-climbing stereotypy. Neither of these stereotypies were ever observed in C57 or Shank3B mice.

Marble burying

An opaque plastic cage (45 × 23 × 13 cm) was used for this test. The cage was filled approximately 6–8 cm deep with wood chip bedding, lightly patted down to flatten the surface and make it even. A regular pattern of 20 glass marbles was placed on the surface: 5 columns 8 cm apart and 4 rows 4 cm apart. The mouse was placed in the cage and left for 30 min. After that time, the marbles that were completely buried or buried to 2/3 their depths were counted.

Grooming

One of the specific behavioural alterations of the Shank3B- strain is excessive, sometimes injurious grooming [23,24,25]. To assess this, mice were placed individually in a clean clear plastic cage with fresh bedding and the lid, and they were left for 5 min to adapt to the new environment. After that, the time spent grooming was measured over 5 min.

Elevated zero maze

Shank3B- mice have been shown to spend less time in the open arms during the Elevated-Zero-Maze (EZM) [24, 25]. The EZM is a modification of the plus-maze based on two conflicting innate tendencies: exploring a novel environment and avoiding elevated and open spaces that constitute a risky situation. The EZM consists of an elevated circular track divided into 4 equal lengths: 2 lengths of track enclosed by an opaque wall on both sides and 2 equal lengths that do not have a surrounding wall. Open and closed areas alternate. Mice were placed individually into one of the closed arms and left to explore for 5 min. The time spent in exploring enclosed versus open arms, the latency to enter the open arms and the number of risk assessment events were counted.

Horizontal bar

Shank3B- mice present mild motor abnormalities that are increased in Shank3KO [24,25,26]. The horizontal bar test was used to assess the muscular strength, motor coordination and prehensile reflex. This test consisted of a 26-cm-long, 0.2-cm-diameter metal bar, supported by a 19.5-cm-high column at each end. Each mouse was held by the tail, placed with its front paws at the central point of the bar and rapidly released. A score was assigned depending on whether and when the mouse fell, whether it held on for 60 s or whether it reached a supporting column. Animals score 1 if they fell off within 10 s, score 2 if they held on for 11–59 s, score 3 if they held on for 60 s or reached the safe platform in 60 s, score 4 if they reached the safe platform within 30 s and score 5 if they reached the platform within 10 s.

Statistical analyses

All statistical tests employed, and results obtained are compiled in Additional file 1: table 1. For two-group comparisons, data were analysed using unpaired two-tailed t-test when they were normally distributed and the Mann–Whitney test if data did not pass the assumptions for parametric analyses. For correlation analyses, data from all groups were pooled and adjusted by a linear regression model. One-way analysis of variance (ANOVA) was performed to compare RT, LPS and WBH on molecular changes and on cFos positive cell counts. Multiple groups were also analysed by repeated measures, two-way, analysis of variance (ANOVA), with treatment (room temperature (RT), LPS or whole-body hyperthermia (WBH) and time (0–48 h) as independent factors. Post hoc comparisons (Bonferroni's test or Fishers LSD where correction for multiple comparisons was not required) were made with a level of significance set at p < 0.05. Data from 2 factor experiments were not always normally distributed, and in these cases, nonparametric tests were used (Kruskal–Wallis and post hoc pairwise comparisons with Mann–Whitney U-test). Post hoc comparisons were made with a level of significance set at p < 0.05. Data are presented as mean ± standard error of the mean (SEM) or median ± interquartile range where nonparametric. Symbols in the graphs denote post hoc tests. Statistical analyses were carried out with the SPSS 22.0 software package (SPSS, Inc., Chicago, IL, USA).