A range of 3–11 mice per sex per strain from new and historical BXD RI lines were characterized for each phenotype (Table 1). Testing occurred at 8–9 weeks of age. Within each strain, mice came from at least two litters, with some, but never all, males and females from the same litter. Litter information is stored in the MouseTrack system and can be obtained for further modeling.
Approximately 70 strains were available for phenotyping, allowing improved power for QTL mapping and genetic correlation. BXD 1–42/TyJ strains were obtained from the Jackson Laboratory (Bar Harbor, ME, USA). Recent BXD RI lines (Peirce et al. 2004) were provided by Dr Lu Lu and Dr Robert W. Williams (University of Tennessee Health Science Center, Memphis, TN, USA).
Housing and testing environment conditions were maintained throughout testing. Except where noted, BXD RI lines were imported into the Russell Vivarium at Oak Ridge National Laboratory (ORNL) for environmentally controlled, year round breeding and distribution for all assays except for handling-induced convulsion (HIC) and footshock vocalization, for which mice were bred at University of Tennessee Health Science Center (UTHSC) and housed as described in Matthews et al. (2008). Litters were weaned at about 3 weeks of age and shipped to various test sites in the TMGC's climate controlled, specific pathogen-free (SPF) mouse transport van at about 6–7 weeks of age, allowing at least 1 week of acclimation to their new home colony. Animals were transported in static micro-isolators. Housing conditions, apparatus information and testing protocols specific to each battery are summarized in Table 2. All mice were housed in rooms lit with fluorescent ceiling lights and had Harlan Softcob bedding. No other species were present in the room and mice received daily health checks.
Table 2. Housing conditions, apparatus information and testing protocols across test sites
|Cage distributor, type, material, size||Optimice Single (75 sq. in.), Thoren Single (75 sq. in.), Thoren Duplex (52 sq. in.). All poly-carobnate||Optimice Single (poly-carbonate, 75 sq. in.)||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Alternative Design Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Standard mouse shoebox cages 7.25 × 11.5 in. of poly-carbonate/ polysulfone||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||54 sq. in. plexiglass box made at University of Kentucky||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate|
|Lid distributor, type, material, size||Optimice poly-carbonate top, Thoren filter top||Optimice poly-carbonate top||Laboratory products poly-carbonate ‘One Cage’ Micro-isolator filter top||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Alternative designs, poly-carbonate||Standard mouse cage grilles of stainless steel||Standard mouse cage grilles of stainless steel||Same as cage||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate|
|Filter top, distributor, type||Optimice poly-carbonate top, Thoren filter top||Optimice poly-carbonate top||Laboratory products poly-carbonate ‘One Cage’ Micro-isolator filter top||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate||Alternative designs, poly-carbonate||Standard mouse micro-isolator tops 7.75 × 12 in. of poly-carbonate/ polysulfone||Standard mouse micro-isolator tops 7.75 × 12 in. of poly-carbonate/ polysulfone||None||Generic Static Micro-isolators 18.5 × 29.5 cm poly-carbonate|
|Enrichment||Nestlets, igloos, PVC pipes||Nestlets||None||None||None||None||None||None||None||None||None||None|
|Light/dark (LD) pattern||14:10||14:10||14:10||14:10||14:10||14:10||14:10||14:10||12:12||12:12||12:12||14:10|
|Light on:light off||0600 h, 2000 h||0600 h, 2000 h||0600 h, 2000 h||0600 h, 2000 h||0600 h, 2000 h||0600 h, 2000 h||0600 h, 2000 h||0400 h, 1800 h||0600 h, 1800 h||0600 h, 1800 h||0600 h, 1800 h||0600 h, 2000 h|
|Light-intensity (light phase)||30 FC at 1 m||30 FC at 1 m||30 FC at 1 m||30 FC at 1 m||30 FC at 1 m||30 FC at 1 m||30 FC at 1 m||30 FC at 1 m||35 FC activity chamber, 20 FC elevated plus maze||30 FC Dowel, Porsolt, rotarod, 50 FC sleep||50 FC at 1 m||30 FC at 1 m|
|Direct animal contact||Caretakers + husbandry technicians||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Technician only||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter||Caretakers + husbandry technicians + experimenter|
|Human presence (time)||Maximum 8 h||Maximum 8 h||Maximum 8 h||Maximum 8 h||Maximum 8 h||Maximum 8 h||Maximum 8 h||Maximum 8 h||8 h||3 h||2 h||Maximum 8 h|
|Handling method (hands, transfer box, restrainer)||Hands||Hands, denim pockets||Hands||Hands||Hands||Hands||Hands||Hands||Hands||Hands||Hands||Hands|
|Gloves, mask, suit/ labcoat, shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes||Gloves, mask, gown, hairnet, shoe covers||Gloves, masks, disposable gowns, shoe covers, hair bonnets||Gloves, masks, disposable gowns, shoe covers, hair bonnets||Gloves, masks, disposable gowns, shoe covers, hair bonnets||Gloves, masks (for allergy sufferers), barrier-dedicated scrubs + shoes|
|Brand, type, % fat, % protein||Irradiated Purina 5053: 5% fat, 20% protein||Irradiated Purina 5053: 5% fat, 20% protein||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640||Harlan Teklad #8640|
|pH||Chlorinated 3–5 p.p.m.||Chlorinated 3–5 p.p.m.||Chlorinated 3–5 p.p.m.||Tap water||Tap water||Tap water||Tap water||Tap water||Tap water||Tap water||Tap water||Tap water|
|Watering system||Automatic: Edstrom||Automatic: Edstrom||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles||Standard cage bottles|
|Ventilated cages (vcs)||Thoren/ Optimice||Thoren/ Optimice||None||None||None||None||None||None||None||None||None||None|
|Total air/h within vcs||Thoren 50 c.p.h.||Thoren 50 c.p.h.||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A|
| ||Optimice 20–30 c.p.h.||Optimice 20–30 c.p.h.||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A|
|Total air/h||10–20 c.p.h.||10–20 c.p.h.||None||None||None||None||None||10–20 c.p.h.||None||None||None||None|
|Temperature||70 ± 2°F||70 ± 2°F||70 ± 2°F||70 ± 2°F||70 ± 2°F||70 ± 2°F||70 ± 2°F||71 ± 3°F||72 ± 3°F||72 ± 3°F||72 ± 3°F||70 ± 2°F|
|Maximum animals/cage||Five adults||Five adults||4||1||1||5||1||Five adults||1||1||1||5|
|Room space (m2)||464 sq. ft. small, 672 sq. ft. large||464 sq. ft. Small, 672 sq. ft. Large||229 sq. ft.||N/A||N/A||N/A||N/A||Approximately 400 sq. ft.||Testing: 198 sq. ft., housing: 256 sq. ft.||Testing: 198 sq. ft., housing: 256 sq. ft.||88 sq. ft.||N/A|
|Acoustic deprivation||None||Insulation/ dampening||None||N/A||N/A||N/A||N/A||None||None||None||None||N/A|
|White noise||Caging system||Caging system||Ventilated caging system||N/A||N/A||N/A||N/A||Subzero freezers||Testing equipment||Testing equipment||Testing equipment||N/A|
|Health and Hygiene|
|Parasitology||Quarterly||Quarterly||Quarterly||Quarterly||Quarterly||Quarterly||Quarterly||Bi-anually||As required||As required||As required||Quarterly|
|Bacteriology||Quarterly||Quarterly||Quarterly||Quarterly||Quarterly||Quarterly||Quarterly||No||As required||As required||As required||Quarterly|
To avoid seasonal and other cohort-type effects on the correlation of trait values, each of the batteries was run in parallel with the exception of footshock vocalization and HIC which were collected independently in a related project. For all other test batteries in this study, all strains were tested in all batteries over the same period of several years by the collaborating laboratories. Confounding environmental variation with strain variation was minimized by sampling individuals from the various strains across this multi-year project. Because phenotype analyses can be influenced by fluctuations in laboratory environment that interact with genotype (Chesler et al. 2002a,b; Crabbe et al. 1999), we recorded laboratory variables such as experimenter, age and test date, which can be matched to records pertaining to the animal colonies and test rooms. These remained largely consistent through the course of this study.
Neurobehavioral testing procedures
Each mouse was assigned to one and only one battery of testing, and transported by ground in a dedicated SPF van to the appropriate test site. The housing conditions at the testing sites and overall testing protocols are summarized in Tables 1 and 2. Specific methods for each battery of tests were as follows:
Naïve mice were housed in a fume hood up to 24 h before dissection. On the day of dissection, cages were removed from the hood one at a time. The individual cages were placed on a cart on the far end of the room, separated from the dissection area. Individual mice were removed from each cage one at a time, while their body weight, coat color, sex, birth and histology dates were noted. This was performed calmly so as to minimize the possibility of sympathetic nervous system activation, which might ultimately affect adrenal weights (Ulrich-Lai et al. 2006). The individual mice were subject to cervical dislocation and their abdominal cavities opened. Whole kidneys were removed one at a time with the adrenal glands attached. The adrenal glands were identified with the naked eye, as they are lighter flesh-colored compared with the surrounding tissue and are often either attached directly to the kidney or within the connective tissue just anterior to the organ. The adrenal glands of males are generally smaller than those of females. The sample was transferred to the stage of a Zeiss dissecting microscope to facilitate clean dissection of the adrenal gland from the surrounding tissue. Once the adrenal glands had been separated they were weighed on a Mettler Toledo scale to a 10th of a milligram, fixed and stored for subsequent histological analysis.
BrdU administration and perfusion. Body weight and coat color were recorded for each animal. Mice were injected with BrdU solution (Sigma, St. Louis, MO, USA; Cat B5002; see below) at 11:00 and were put back in their home cage with a tail mark to indicate test order. Typically, one mouse was injected every 10 min. Fresh BrdU solution 0.5% (5 mg/ml) was prepared before each day's perfusions. BrdU is dissolved in 0.007 N NaOH in 0.9% NaCl. Each mouse was injected with BrdU (50 µg/g body weight or 0.1 ml/10 g body weight) and perfused 1 h after injection. Approximately 5 min before starting the perfusion, the mouse was anaesthetized with Avertin. Mice were perfused transcardially first with 0.1 M phosphate-buffered saline (PBS) and then alcohol–acetic acid solution (1:3, 95% EtOH:acetic acid). Brains were removed and post-fixed in the same fixative overnight with one brain in each vial. The following day, the brains were put into 70% EtOH where they sat, at room temperature, until they were embedded in paraffin. Immediately before embedding, the brains were cut at midline into two hemispheres, dehydrated/defatted in an ethanol–xylene series and placed in 64°C paraffin over night. The following day, brains were transferred twice into fresh paraffin. Brains were then embedded in a mold and cooled for sectioning. Each embedded half brain was serially sectioned in the sagittal plane at 8 µm and every 10th section was mounted on Superfrost/Plus slides. Slides were allowed to dry at 37°C overnight.
Anti-BrdU immunohistochemistry. On the first day of BrdU immunostaining for paraffin-embedded 8 µm sections, a xylene–ethanol series is used for deparaffinization. Brains were immersed in distilled water and rinsed in a series of PBS, HCl, 8.C00.404, PBS and hydrogen peroxide PBS. Slides were incubated with mouse anti-BrdU (Sigma, Cat B8434)×200 primary antibody in 5% normal horse serum overnight. On the second day, slides were incubated with horse anti-mouse immunoglobulin G (IgG)×200 secondary antibody for 1 h. Finally, a diaminobenzidine tetrahydrochloride (DAB) reaction is performed using the Vectastain Elite ABC Kit (Vector Laboratories, Burlingame, CA, USA). After development, slide-mounted sections were rapidly dehydrated and cover-slipped.
Counts of BrdU-labeled cells. For each animal, adult neurogenesis in the rostral migratory stream (RMS) was evaluated as the number of BrdU-positive cells was calculated for the full length of the structure. In an ideal case, these data were obtained from a single section. However, when necessary, data were retrieved from two sections and very rarely from three sections (the number of sections used for the analysis is recorded). BrdU-positive cells in the RMS were counted using a 40× objective. The RMS length was measured using AnalySIS Opti Version 3.3.776 software (Soft Image System). Only clearly labeled cells were counted in the analysis. The number of BrdU-positive cells per millimeter was calculated by dividing the cell number in each section by its corresponding RMS length. These data were expressed as a total number of BrdU+ cells, and also as the total number of cells divided by the number of sections analyzed to obtain a per section average.
Habituation to a novel environment. Mice were individually placed into a bank of eight activity chambers (43.2 cm L × 43.2 cm W × 30.4 cm H, ENV-515, Med Associates, St Albans, VT, USA) that contained two sets of 16 photocells placed at 2.5 and 5 cm above the chamber floor. Activity was measured as photocell beam breaks and converted into horizontal distance traveled (cm), and the number of rears was also recorded. Rears were automatically counted when a mouse broke the upper set of photocell beams. In addition, the test chambers were subdivided into a peripheral zone that encompassed a corridor adjacent to each wall that was 7.6 cm wide and central zone (28 cm2). Total distance traveled and rears were also separately compiled for both zones. All measures were collected at 15 min intervals during the 1 h test and also expressed as totals over the hour. As an indicator of the distribution of activity, the novelty ratio was calculated as (distance traveled in the periphery/total distance traveled) × 100.
Locomotor response to saline or cocaine injections. The same methods, apparatus and dependent measures were used with the exception that the activity chambers were not subdivided into peripheral and central zones. On successive test days, mice were injected (i.p.) with isotonic saline (10 ml/kg) or cocaine (10 mg/kg in isotonic saline at a volume of 10 ml/kg) and immediately placed in the activity chambers. An additional measure of cocaine sensitization was calculated by subtracting total distance traveled after the first cocaine injection from total distance traveled after the second cocaine exposure. Positive values indicated sensitization.
Conditioned place preference. Eight chambers were used (ENV-3013, Med Associates). Each chamber (46 cm L × 14 cm W × 20 cm H) was subdivided into a center chamber (10 cm L, painted gray with a solid floor) separated by guillotine doors from two conditioning chambers (18 cm L). One conditioning chamber was painted black and had a wire-mesh floor while the other was painted white with a stainless steel grid floor, in order to provide distinctive visual and tactile cues. Three sets (transmitter and receiver) of infrared photocells were spaced equidistantly along the long wall of the place preference conditioning boxes (2 cm above the floor) in order to record the time spent in each conditioning chamber. Mice received either injections of saline or cocaine [3.2 mg/kg (i.p.) in saline vehicle]. Testing was conducted over 5 days using procedures similar to those of Seale and Carney (1991): Day 1—Each mouse was introduced into the middle of the place conditioning apparatus. The guillotine doors were raised and the time spent on each side was automatically recorded during this 20 min baseline session. Days 2–4—There were a total of three conditioning days that totaled 40 min in duration. The guillotine doors remained closed during conditioning sessions. All mice were injected with saline and placed in the black compartment. After 20 min, each mouse was removed from the apparatus and briefly returned to its home cage. The animals were then injected with cocaine (3.2 mg/kg) and placed into the white compartment for an additional duration of 20 min. Day 5—Place preference was evaluated in a 20 min test. Each mouse was introduced into the middle of the apparatus and the guillotine doors were raised. The time spent on each side of the apparatus was recorded. The dependent measures included the time (seconds) spent on the drug- and saline-paired sides at baseline (Day 1) and test (Day 5). In addition, change in preference was calculated as time spent on the drug-paired side at test minus the time spent on the drug-paired side at baseline. Positive numbers indicated an increased preference.
Habituation to a novel environment. Habituation to a novel environment was conducted using the same procedure as in the cocaine tests.
Locomotion in response to an injection of morphine. For the test of locomotion in response to an injection of morphine, the methods of Kest et al. (2002a,b) and Schulteis et al. (1997) were used. Mice received a single injection (i.p.) of morphine sulfate (50 mg/kg in isotonic saline at a volume of 10 ml/kg) and were immediately placed into the activity chambers. Horizontal distance traveled and rearing were recorded in 15 min intervals throughout the 3 h test and also expressed as totals over the 3 h.
Behavioral (morphine withdrawal) response to an injection of naloxone. Mice were briefly removed from the activity chambers and injected (i.p.) with naloxone (30 mg/kg in isotonic saline at a volume of 10 ml/kg), and immediately returned to the chambers for an additional 15 min. Horizontal distance traveled was automatically recorded. The effect of naloxone on activity was calculated as distance traveled between 165 and 180 min post-morphine minus total distance traveled following naloxone. In addition, over the 15 min post-injection of naloxone, a trained observer counted the number of jumps, fecal boli and urine puddles that each mouse produced. Jumps were defined as all 4 ft out of contact with the floor of the chamber and the mouse in an upright posture. In addition, between 5 and 10 min post-naloxone, somatic signs of withdrawal intensity were rated by a trained observer similar to the weighted scale of Gellert and Holtzman (1978). This scale consisted of graded ranking (range = 1–3) of wet dog shakes, instances of abdominal contraction, salivation, ptosis and abnormal posture.
The general behavior phenotyping battery was performed as previously described (Cook et al. 2001, 2007). The open field, light–dark and fear conditioning tests were performed using the Hamilton-Kinder SmartFrame system (Hamilton-Kinder, Poway, CA, USA) and test-specific inserts described below.
Zero-maze. Briefly, animals were brought into a darkened testing area a minimum of 30 min before testing and allowed to acclimate. The test apparatus is a plexiglass maze placed 108.9 cm off the floor with 40 cm outer diameter and 30 cm inner diameter, and closed arm walls at 28.5 cm H (AccuScan Instruments, Inc., Columbus, OH, USA) were dimly illuminated by 15 W red bulbs suspended above the maze. To begin the test, animals were placed into a closed quadrant of one of the three identical mazes. The test session is 5 min in duration. Once an animal has been tested, it is placed in a holding cage until all animals from the home cage have been tested.
Open field. Animals were brought into the testing area a minimum of 30 min (but ideally 45 min to an hour) before testing and allowed to acclimate. The open field session was 20 min in length. To begin the test, individual animals were removed from the cages and placed into the center of an open field apparatus (24.13 cm L × 45.72 cm H). Once an animal had been tested, it was placed in a holding cage until all animals from the home cage completed testing.
Hot plate. This test occurs approximately 2–3 h after the completion of the open field test. The lights in the testing area were turned off at least an hour prior to testing and animals were allowed to sit undisturbed in the darkened room. A lamp (15 W bulb) behind the hot plate (Hotplate Analgesia Meter, Model 39, IITC, Inc.) was faced away from the hot plate surface. A mirror was placed behind the hot plate so that the experimenter can observe the animal. The hot plate was maintained at 52°C. The mouse was placed on the center of the hot plate in a smoke gray Plexiglas bottomless cube and the built-in timer started. As soon as the animal elicited a pain response (i.e. paw licking, guarding, shaking or jumping), the timer was immediately stopped and the animal removed from the hot plate surface. If the animal did not show a response within 30 seconds, the test was stopped and the animal was assigned the 30 seconds maximum time as its response latency. Once an animal had been tested, it was placed in a holding cage until all animals from the home cage have been tested.
Light/dark. The animals were acclimated to the darkened room for a minimum of 30 min. A lamp, with 15 W bulb was located directly above the light portion of the light/dark box which had total dimensions of 24.13 cm L × 45.72 cm W. To begin the 10-min test, animals were placed in the light half of the box. The guillotine door was then removed to allow the animals to freely move between the two halves of the box. The amount of time spent in the light vs. dark compartment was measured. Once an animal had been tested, it is placed in a holding cage until all animals from the home cage have been tested.
Startle/pre-pulse inhibition. The startle and pre-pulse inhibition tests were performed using a Hamilton-Kinder SM100 startle chamber inside a 14 in L × 10.875 in W × 19.5 in H sound-attenuating chamber. Animals were placed in the chamber with a 65 dB background white noise and allowed to habituate. Over an approximately 15 min session, 55 pseudo-random trials were given. A 120 dB white noise burst was used as the acoustic startle stimulus. Pre-pulses were 70, 80 and 85 dB white noise bursts which preceded the startle stimulus by 10 milliseconds. Startle responses to the startle stimulus and to each of the pre-pulse dB levels were measured.
Fear conditioning. The first part of fear conditioning (Training) was carried out approximately 1–1/2 to 2 h after the startle and pre-pulse inhibition test. Animals were placed in the fear conditioning chambers (24.13 cm × 22.86 cm, with a grid floor) and allowed to habituate for 2.5 min. Animals were then presented with three pairings of an 85 dB tone and 0.36 mA footshock. The tone was 30 seconds in duration, and the shock was presented during the last 2 seconds of the tone. There was a 2.5 min interval between each of the tone plus shock pairings.
Contextual conditioning. On the day following the training session, animals were placed back into the same chambers where they underwent training. During the 6 min session, activity (beam breaks) per 30-second bin was measured and compared with activity during the habituation period on the training day.
Cued conditioning. Approximately 2 h later, the behavior of the mice was tested in an altered context. The fear conditioning chambers were altered by placing a gray, square tile over the grid floor, placing a black Plexiglas insert over the walls of the chambers, and attaching a small cup containing orange oil diluted in water in the upper corner of the box. Animals were allowed to explore the altered environment for 2.5 min, after which, the conditioned stimulus (tone) is presented for 2.5 min. Activity (beam breaks) was evaluated in 30-second bins.
Tail suspension. All animals were weighed to the nearest one-tenth of a gram prior to tail suspension testing. The body weight for each animal was entered into the Med Associates tail suspension program. On the basis of the body weight, a threshold force of movement for each animal is automatically calculated. Mice were suspended by the tail with generic sports tape attaching them to the transducer. During the 6 min test, force of movement or lack thereof was recorded and reflected the time the animal spent immobile during the test.
Overview of test sequence. Each mouse was singly housed on arrival and was given at least 1 week to acclimate before testing. The testing was carried out over a period of 3 days. Half of the mice were given saline on Day 2 and an ethanol injection (2.25 g/kg) on Day 3. For the other half of the mice, the order of injection was reversed with ethanol on Day 2 and saline on Day 3. On Day 1, all mice were trained on the rotarod, as described below. On Day 2, mice were weighed and given the appropriate injection. Ten minutes after the injection, mice were given a 5-min test in the elevated plus maze followed immediately by a 20-min test in the activity chamber. After the activity chamber, mice were tested on the rotarod and blood was taken to measure blood ethanol concentration (BEC). Mice were then returned to the home cage and animal room until testing the next day. On Day 3, order of testing was identical to the second except that the elevated plus maze was not conducted.
Rotarod. For each trial, each mouse was placed on its own segment of the rotarod facing the back wall. The rod was spinning at five revolutions per minute (r.p.m.) at the beginning of the test and accelerated to 25 r.p.m. The mouse remained on the rotarod until it fell off. Both the length of time and the speed of the rotarod when the mouse falls were recorded. On Day 1 (training day), each mouse was given 10 trials on the rotarod and data were recorded. Because the behavior of the mice reached a plateau after five trials, the last three were used to calculate the mean for the training day for both speed and time on rotarod in seconds. On Days 2 and 3, only three trials were given and all were used to compute the mean for each condition (saline vs. ethanol). All mean times were recorded. In addition, several computed measures were also determined: training mean–saline mean (to determine the effects of the injection, repeated exposure and prior tests on the rotarod score), training minus ethanol (to determine the effects of ethanol on motor incoordination) and saline–ethanol (this measure may ultimately not be required depending on whether differences were seen between training and saline).
Elevated plus maze. Each mouse was placed in the center of the maze facing an open arm of the plus maze which has been previously described (Hamre et al. 2007). Mice were given 5 min to explore the maze. Both the amount of time in the various arms and the number of entries into the arms were recorded. The amount of time provides a measure of the degree of anxiety while the number of entries provides a measure of the activity level and insures that mice entered more than one arm. In addition, the amount of time in the middle of the maze was recorded. The percentage of entries into the open and closed arms were computed and recorded.
Activity chamber. Mice were placed in the activity chamber for 20 min on Days 2 and 3 as previously described (Hamre et al. 2007). Horizontal distance traveled was recorded in centimeters. Activity was recorded in 5-min bins as well as computed for the total time. Differences between saline and ethanol were computed and recorded.
Blood ethanol concentration. A nick was made in the end of the tail and 10–20 ml of blood was drawn from each mouse. Blood was drawn both on Day 2 and Day 3, although it was not saved or analyzed from the saline injected animals. The blood was centrifuged and the BEC determined using the Analox ethanol analyzer (Analox, USA).
Sleep/wake analysis. For the entire period of monitoring, each mouse was placed in its own chamber of a Piezo-electric grid and chamber system (Donohue et al. 2008). The Piezo chamber detects movement, and software analysis of respiration rates determine sleep or wake for each mouse. The mice had access to food and water ad lib while in the chamber. The room was maintained on a 12:12 light:dark cycle. Mice were placed in the chambers between 0900 and 1000 h on Day 1 and were removed on Day 5 at the same time. Each day, the computer, food and water were checked. Otherwise, the mice remained undisturbed.
Porsolt forced swim test. Mice were tested in the Porsolt forced swim test at least 72 h after completion of the sleep analysis. The water in the chamber was heated to 25°C. Each mouse was placed in the chamber for 5 min and videotaped. The water was changed for every third mouse. The behavior of each mouse was scored from the videotapes. The total time immobile as well as the time immobile for the last 3 min were scored and analyzed.
Dowel test. Mice were tested for the Dowel test at least 48 h after completion of the Porsolt analysis. At baseline, each mouse was placed on the dowel for a maximum of 2 min. If the mouse remained on the dowel for the entire 2 min, it was removed and injected with 2.0 g/kg of ethanol. Each mouse was then placed back on the dowel for a maximum of 5 min immediately after the injection. The test was repeated 30 min after injection. For all three tests, the length of time until the mouse fell off the dowel was recorded.
Locomotor response to MDMA. On Day 1, all mice were weighed, injected with saline (10 ml/kg, s.c.) and placed in the Open Field Chamber (MED-OFA-510, Med Associates). Activity was recorded for 90 min. On Day 2, mice were habituated to the Open Field chamber for 60 min. They were then injected with saline (10 ml/kg, s.c.) or MDMA (10 mg/kg, s.c.). Activity was recorded for 90 min.
Hargreaves' paw withdrawal test. Mice were placed on a 3/16th-in. thick glass floor within small (9 cm L × 5 cm W × 5 cm H) Plexiglas cubicles and allowed a habituation period of 120 min. A focused, high-intensity projector lamp beam (IITC Model 336 Plantar Test and Tail-flick Analgesia Meter, Woodland Hills, CA, USA) was shone from below onto the mid-plantar surface of the hindpaw (Hargreaves et al. 1988). The beam was aligned to the mid-plantar surface of the left paw with the projector lamp set to 10% idle intensity (II10). The lamp was then switched to 25% active intensity (AI25) and the latency to respond with withdrawal of the paw from the light or licking of the paw was recorded using the internal timer. To avoid tissue damage, if no response occurred by 30 seconds, the lamp was returned to the idle intensity and removed from the paw. This process was repeated for all mice in the enclosure, and then migrated to the right hindpaw, following an intra-trial period of at least 5 min. Mice were tested for three to six trials depending on the variance observed on each paw, and the three tightest latencies averaged.
Hot plate. After 30 min of habituation to the testing room, mice were placed on a metal surface (IITC Inc., Hotplate Analgesia Meter, Woodland Hills, CA, USA) maintained at 54°C (±0.2°C) (HP54) within a transparent Plexiglas cylinder (15 cm D; 22.5 cm H) with Plexiglas lid. The latency to respond with a jump, or hindpaw lick or shake/flutter was measured to the nearest 0.1 s with a stopwatch. Two latencies were recorded per mouse with intra-trial separation of 30 seconds and maximum trial duration of 30 seconds. If no response occurred within 30 seconds, the mouse was removed from the hot plate. The apparatus was thoroughly cleansed with MB-10 (QuipLabs, Wilmington, DE, USA) between each mouse tested.
Tail withdrawal. As with the hot plate, mice were allowed 30 min of habituation. Although lightly restrained in a denim pocket, the distal half of the mouse's tail was dipped into a bath of water thermostatically controlled at 47.0°C (±0.1°C) (TW47) by Boekel Scientific/Grant Optima Immersion Circulator Model GR150 (Boekel Scientific, Feasterville, PA, USA). Latency to respond to the heat stimulus by vigorous flexion of the tail was measured. Mice received three to five trials separated by 10 seconds, with maximum trial duration of 30 seconds. If no response occurred by 30 seconds, the mouse's tail was removed from the hot water. The last three trials were recorded.
Tail clip. As in hot plate and tail withdrawal tests, mice were allowed 30 min of habituation. The enclosure is a Plexiglas-bound arena measuring 13.5 in L × 16 in W × 15 in H. The front of the arena was open and aligned with the leading edge of the table, such that the experimenter could easily restrain and release mice quickly. All mice were lightly restrained in a denim pocket and an alligator clip with a rubber cuff around each jaw (exerting ≈600 g of force) was applied to the tail 1 cm from the base and vertically oriented with respect to the table. The mouse was immediately removed from the holder, and the latency to lick, bite or grab the clip or bring the head within 1 cm of the clip was measured with a stopwatch to the nearest 0.1 second, after which the clip was immediately removed. Each mouse was tested only once with maximum trial duration of 60 seconds. If no response occurs by 60 seconds, the tail clip was removed. The enclosure and clip were thoroughly cleansed with MB-10 (QuipLabs) between each mouse.
von Frey test. Mice were allowed 120 min of habituation to the same Plexiglas enclosure (9 cm L × 5 cm W × 5 cm H) as used in Hargreaves' test on a wire-mesh floor (aquarium/vivarium top) instead of a glass floor. For assessment of mechanical sensitivity thresholds, mice were tested with von Frey type nylon monofilaments. A set of eight calibrated von Frey fibers (Stoelting, IL, USA), ranging from 0.067 to 9.33 g of force, were applied to the plantar surface of the hindpaw until they bowed. The threshold force required to elicit withdrawal of the paw (median 50% paw withdrawal) was determined using the up–down method (Chaplan et al. 1994). A maximum of nine trials were required for each paw with four trials performed after the first response cross-over. As in the Hargreaves test, trials began with the left hindpaw and then switched to the right after all animals in the enclosure had received the stimulus. An intra-trial period of 150 seconds was used between left and right paws, such that each paw's stimulus was separated by 5 min.
Stress vocalization. The footshock stimulus to which animals audibly vocalized were assessed following the generation of a mild footshock by Med Associates, Inc. Shock Titration Package for Mice (model ENV-307 W, St Albans, VT, USA). Specifically, on each test day, mice were moved from the mouse colony room to a holding room adjacent to the footshock chambers. Following at least a 25-min wait period (to allow for acclimation to the move), audible vocalization thresholds was assessed. Mice were placed individually in a shock chamber and allowed to adapt to the chamber for 5 min. Each mouse then received a mild footshock via the floor grid every 30 seconds for 500 milliseconds. The intensity of the first footshock was 0.05 mA and each subsequent footshock increased in increments of 0.05 mA until the mouse vocalized as determined by a single technician who performed all assays, positioned within 1 m of the shock chamber. Once the mouse vocalized, the experiment was terminated and the mouse was subsequently removed from the chamber. Each chamber was cleaned between test subjects. To control for experimenter-related variation in audible vocalization detection, the same technician collected every data point and was blind to the subjects' genotype. Naïve mice were held in the adjacent room during the time other subjects were being tested so that they could not hear or otherwise be influenced by the vocalizations of other subjects (Matthews et al. 2008).
Handling-induced convulsions were determined using a modified scoring paradigm from Buck et al. (1997). Briefly, each mouse was gently picked up by the tail and spun, if required, to determine a baseline convulsion score. Mice were then injected, i.p., with 4.0 g/kg ethanol (20%) and HIC scores re-determined at 4, 6 and 7 h post-injection. HIC scores were the combined difference scores (i.e. baseline score subtracted from the later scores) for each value following the injection.
Analysis methods and data access
MouseTrack (Baker et al. 2004) and GeneNetwork (Chesler et al. 2004; Wang et al. 2003) are the two main resources used for data storage, sharing and analysis. MouseTrack serves as the primary data archive and analysis engine for individual mouse data, whereas GeneNetwork serves as the database and analysis engine for QTL mapping, genetic correlation of strain means and integration with other public data from the BXD RI reference population. MouseTrack consists of an ORACLE database and customized sas (version 9.1.3, Cary, NC, USA) client tools for genetic analysis. MouseTrack's RI analysis tools include univariate analysis, box plots of individual strain data, and linear models of sex, treatment and strain effects and interactions, heritability, sub-population effects, and estimation of strain means and strain means by sex for export into GeneNetwork.org. Detailed information about each of the phenotypic values and protocols used to generate them are also accessible from MouseTrack. The MouseTrack tool also performs individual outlier detection, multi-variate outlier detection and distributional evaluation including displays of the phenotypic distributions within strain. These tools can be used to identify phenotypically extreme strains for advanced study. In the present report, outlier detection tools were used exclusively for quality assurance. When extreme univariate outliers are detected (> 5 SD), possible data entry errors and other traceable sources of outliers were considered. If none could be found, the outliers were retained in subsequent analyses and in our submission to GeneNetwork.org.
Strain effects and sex differences
Sex, strain and strain × sex interaction effects were tested using non-sequential sums of squares estimation in a general linear model. The model used for testing these effects was
where yijis the phenotype being measured for Straini and Sexj. In addition to testing these effects, we estimated the magnitude of strain, sex and strain × sex effects by estimating the intra-class correlation, a partial ω2, for each effect per trait. Partial ω2 was estimated as the proportion of variance accounted for by the main or interaction effects relative to the total phenotypic variance. All variance components were estimated using the REML option of SAS PROC VARCOMP. This method can be biased due to departures from normality, a common phenomenon for behavioral traits. The percent variance accounted for by strain is considered by some to be an estimate of broad-sense heritability (Hirsch 1967; Lynch & Walsh 1998) for clones, and is formulated as the strain intra-class correlation:
Standard errors were obtained using an adjustment for unbalanced data (Swiger et al. 1964). This calculation was performed for the two sexes separately, and for the data combining both sexes. It should be noted that this measure provides an estimate of the resemblance among relatives in a population in which segregation had occurred and does not reflect the transmission of genetic material in a randomly mating population. It is nonetheless an indication of suitability and plausibility of genetic analysis for a given phenotype.
Multi-trait QTL analysis in the expanded BXD RI lines
Multi-trait QTL analysis can be performed by extracting common underlying factors from multiple behavioral phenotypes and generating strain-specific factor scores. The underlying hypothesis of this type of analysis is that behavioral measures for stress, anxiety, pain and addiction to drugs of abuse are under common genetic control and should share some degree of correlation. Brigman et al. (2009) perform a similar decomposition of anxiety and fear behavior in BXD RI mice. This approach to multi-trait QTL mapping has been undertaken in a study by Trullas and Skolnick (1993), wherein factor analysis was used to report that elevated plus maze behavior predicts anxiety-like behavior, and more recently by Henderson et al. (2004), who performed QTL mapping on multiple phenotypic assays of anxiety-like behavior. The value of performing such studies in the BXD RI population is that the data can be expanded indefinitely with additional independent phenotypic profiling which adds depth and detail to the multi-dimensional analysis.
As the phenotypic data contained missing observations, any strain with more than 25% of trait data missing was removed from this analysis. This resulted in the elimination of 32 of the 95 BXD strains, giving us an effective sample size of 63 BXD strains. The resultant data set was subjected to column mean imputation in order to fill in the missing trait values, given that the data were missing at random, i.e. not missing over all measures within a battery. Another issue that was encountered during this analysis was that of dimensionality. The dataset used in the factor analysis had more variables/traits (p) than observations/strains (n) (i.e. n < p). This results in incorrect estimation of the covariance matrix and thereby leads to singularity of the estimated covariance matrix. This issue was addressed by using the James–Stein-type shrinkage estimator (Schafer & Strimmer 2005) of the covariance matrix. R packages e1071 (for missing data imputation), corpcor (for covariance shrinkage), nFACTOR and factanal were used for the purposes of factor analysis. Factor loadings were analyzed to obtain factor interpretations. Factor scores were obtained for all interpreted latent factors. Quantitative trait locus mapping was performed on these latent factors to identify common genetic drivers of variability in factor scores.