Longitudinal Brain Magnetic Resonance ... - Wiley Online Library

7 downloads 0 Views 301KB Size Report
Chronic Alcohol Consumption. Adolf Pfefferbaum, Elfar Adalsteinsson, Rohit Sood, Dirk Mayer, Richard Bell,. William McBride, Ting-Kai Li, and Edith V. Sullivan.
ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH

Vol. 30, No. 7 July 2006

Longitudinal Brain Magnetic Resonance Imaging Study of the Alcohol-Preferring Rat. Part II: Effects of Voluntary Chronic Alcohol Consumption Adolf Pfefferbaum, Elfar Adalsteinsson, Rohit Sood, Dirk Mayer, Richard Bell, William McBride, Ting-Kai Li, and Edith V. Sullivan

Background: Tracking the dynamic course of human alcoholism brain pathology can be accomplished only through naturalistic study and without opportunity for experimental manipulation. Development of an animal model of alcohol-induced brain damage, in which animals consume large amounts of alcohol following cycles of alcohol access and deprivation and are examined regularly with neuroimaging methods, would enable hypothesis testing focused on the degree, nature, and factors resulting in alcohol-induced brain damage and the prospects for recovery or relapse. Methods: We report the results of longitudinal magnetic resonance imaging (MRI) studies of the effects of free-choice chronic alcohol intake on the brains of 2 cohorts of selectively bred alcoholpreferring (P) rats. In the companion paper, we described the MRI acquisition and analysis methods, delineation of brain regions, and growth patterns in total brain and selective structures of the control rats in the present study. Both cohorts were studied as adults for about 1 year and consumed high doses of alcohol for most of the study duration. The paradigm involved a 3-bottle choice with 0, 15 (or 20%), and 30% (or 40%) alcohol available in several different exposure schemes: continuous exposure, cycles of 2 weeks on followed by 2 weeks off alcohol, and binge drinking in the dark. Results: Brain structures of the adult P rats in both the alcohol-exposed and the water control conditions showed significant growth, which was attenuated in a few measures in the alcohol-exposed groups. The region with the greatest demonstrable effect was the corpus callosum, measured on midsagittal images. Conclusion: The P rats showed an age–alcohol interaction different from humans, in that normal growth in selective brain regions that continues in adult rats was retarded. Key Words: Alcohol, Rat, MRI, Corpus Callosum, Voluntary Drinking, Brain.

A From the Neuroscience Program, SRI International, Menlo Park, California (AP); the Department of Psychiatry & Behavioral Sciences, Stanford University School of Medicine, Stanford, California (AP, EVS); the Harvard-MIT Division of Health Sciences and Technology, MIT, Cambridge, Massachusetts (EA); the Department of Electrical Engineering and Computer Science, University of New Mexico Health Center, Albuquerque, New Mexico (EA); the Department of Neurology, University of New Mexico Health Center, Albuquerque, New Mexico (RS); the Department of Radiology, Stanford University School of Medicine, Stanford, California (DM); Institute of Psychiatry, Indiana University Medical Center, Indianapolis, Indiana (RB); the Department of Psychiatry, Indiana University Medical Center, Indianapolis, Indiana (WM); and the National Institute on Alcohol Abuse and Alcoholism, Bethesda, Maryland (TKL). Received for publication November 29, 2005; accepted March 5, 2006. Support for this project was provided by the Integrative Neuroscience Initiative on Alcoholism (INIA) from the National Institute on Alcohol Abuse and Alcoholism (AA13521 and AA13522) and by NIAAA AA05965. Reprints requests: Adolf Pfefferbaum, MD, Neuroscience Program, SRI International 333 Ravenswood Avenue, Menlo Park, CA 94025; Fax: 650-859-2743; E-mail: [email protected] Copyright r 2006 by the Research Society on Alcoholism. DOI: 10.1111/j.1530-0277.2006.00146.x 1248

LCOHOLISM—THE CHRONIC excessive consumption of alcohol—produces widespread shrinkage of brain tissue (Jernigan et al., 1991; Pfefferbaum et al., 1992) and attendant deficits in motor and cognitive function that persist even after months of abstinence (e.g., Eckardt et al., 1998; Parsons, 1993; Sullivan et al., 2000c). The brain of the detoxified alcoholic can appear as ravaged as that of a patient with Alzheimer’s disease, although the brain dysmorphology associated with alcoholism is at least partially reversible with prolonged sobriety (Carlen et al., 1978; Gazdzinski et al., 2005; Mann et al., 1999; O’Neill et al., 2001; Parks et al., 2002; Pfefferbaum et al., 1995; Pfefferbaum et al., 1998). Both cortical gray matter and white matter sustain volume loss (Jernigan et al., 1991; Pfefferbaum et al., 1992; but see Cardenas et al., 2005), greatest in the prefrontal cortex in older alcoholic individuals (Pfefferbaum et al., 1997). Nonamnesic alcoholic individuals also have notable volume shrinkage of the mammillary bodies (Davila et al., 1994; Shear et al., 1996; Sullivan et al., 1999), anterior hippocampus (Agartz et al., 1999; Sullivan et al., 1995b), thalamus (Sullivan et al., 2003), corpus callosum (Hommer et al., 1996; Pfefferbaum et al., 1996), and cerebellum (Sullivan et al., 2000a, 2000b). Alcohol Clin Exp Res, Vol 30, No 7, 2006: pp 1248–1261

ALCOHOL EFFECTS ON RAT BRAIN

Longitudinal analyses reveal that, with short-term abstinence from alcohol, gray matter increases and lateral ventricles decrease in volume (Pfefferbaum et al., 1995). With longer-term abstinence, white matter expands and the third ventricle shrinks, but with relapse, the opposite occurs (Pfefferbaum et al., 1995; Shear et al., 1994). Over a 5-year interval, the degree of excessive drinking in alcoholic individuals is related to the degree of cortical gray matter loss, especially in the frontal lobes (Pfefferbaum et al., 1998). When alcohol is used excessively and chronically, not only do alterations in the brain itself occur but sustained alterations in consummatory behavior also ensue (Li, 2000b). A combination of genetic predisposition, including the relative tendency to act impulsively or the relative inability to maintain behavioral control (Li, 2000a), and environmental factors, such as stressful life events and access to alcohol, can lead to the initial stages of high alcohol consumption. An animal model of alcoholism, in which animals consume large amounts of alcohol and drink excessively following repeated cycles of alcohol access and deprivation (Rodd-Henricks et al., 2001b), is proposed for use to test hypotheses about the degree and nature of alcoholinduced brain damage and the prospects for recovery or relapse. To parallel the human condition, an animal model would sustain the widespread brain damage, manifest primarily by tissue shrinkage, after excessive alcohol exposure as is observed in humans. Neurobiological studies of brains of rats selectively bred to consume large amounts of alcohol (McBride and Li, 1998) have generally examined focal rather than global brain changes. In rodent studies, primarily using a liquid diet animal model, neurodegenerative evidence has been observed in the frontal cortex and corpus callosum (Savage et al., 2000), cerebellum (e.g., Dlugos and Pentney, 1997; Pentney and Dlugos, 2000; Pentney and Quackenbush, 1990; Rintala et al., 1997), and locus coeruleus (Lu et al., 1997). Such neuropathology is accelerated with the dietary thiamine antagonist pyrithiamine (Langlais and Savage, 1995; Langlais and Zhang, 1997; Lee et al., 2001, 1995). A magnetic resonance imaging (MRI) study of relevance to alcoholism used thiamine deficiency and thiamine antagonist models of Wernicke’s encephalopathy (Pentney et al., 1993) and reported an increased volume of lateral ventricles following treatment. Despite the capability of current imaging protocols to encompass the whole brain, delineate separate structures, and differentiate gray matter and white matter (e.g., Fiel et al., 1991; Pfefferbaum et al., 2004a; Ting and Bendel, 1992), such protocols and analysis approaches have not yet been used in the study of animal models of alcoholism, where the animal freely self-administers alcohol, uncomplicated by nutritional deficiencies. Whole-brain imaging would permit testing of hypotheses about the specificity of regional abnormalities within the context of the entire brain and would yield a

1249

permanent library of each animal’s brain to be analyzed to address current study hypotheses and future hypotheses as they evolve from new findings and theories. A valid animal model of human alcoholism would permit full control over the amount and timing of alcohol consummatory, nutritional and hormonal factors, sex differences, systemic influences from non–central nervous system (CNS) substrates, such as hepatic and cardiovascular function, and novel pharmacological therapies, not possible in naturalistic studies of the human condition. Longitudinal studies of animals with in vivo MRI modalities during planned courses of alcohol dependency, withdrawal, and reentry into dependency could identify neural systems most vulnerable to alcohol and their capacity for structural and functional recovery. To the extent that each animal serves as its own control, the sensitivity for detection of change is greatly enhanced. The P rat meets all of the criteria proposed by Cicero et al. (1971) for a valid model of alcoholism (cf. Bell et al., 2005; McBride and Li, 1998; Murphy et al., 2002). Briefly, the P rat will (a) voluntarily consume 5 to 8 g of alcohol per kg body weight per day and attain blood alcohol concentrations of 50 to 200 mg%; (b) work to obtain alcohol when food and water are freely available; (c) drink alcohol for its CNS pharmacological effects and not solely because of its taste, smell, or caloric properties; (d) develop metabolic and functional tolerance; (e) demonstrate physical signs of dependence upon withdrawal of alcohol; and (f) demonstrate robust relapse drinking (cf. Bell et al., 2005; McBride and Li, 1998; Murphy et al., 2002). Thus, this animal model offers a means to study alcohol consumption characteristics that parallel more closely the human condition than previous animal models using a forced liquid diet paradigm. Further, a genetic model parallels the human condition because of the evidence indicating the influence of genetic factors on alcohol-drinking behavior (Begleiter and Porjesz, 1999; Cloninger, 1987; Ehlers et al., 1991; Heath et al., 1991a, 1991b; Hill, 1992; Kendler et al., 2003; Porjesz et al., 2005; Schuckit, 1985). Here we report the results of a longitudinal MRI study of the effects of free-choice chronic alcohol intake on the brain of 2 cohorts of selectively bred alcohol-preferring (P) rats (Li et al., 1979). In the companion paper (Sullivan et al., 2006), we described the MRI acquisition and analysis methods, manual delineation of brain regions on MR images, and growth patterns in total brain and selective structures in the rats, which served as controls never exposed to alcohol for the present study. Applying quantitative analysis, we expected that (1) the P rats exposed to high doses of alcohol over extended periods would demonstrate reduction in tissue volume in brain structures known to be affected in human alcoholism, including total brain tissue, corpus callosum, hippocampus, and cerebellum; and (2) amount of alcohol consumed would correlate with the extent of its effect on brain structure and would accumulate over time.

PFEFFERBAUM ET AL.

1250

Cohort A

9 wk alcohol

on

age:

88 days

M R I

M R I

2-wk on-off 3 cycles

o o f 1 n f

o f f

o n

o f f

o o n f f

o f f

o 2 n

o n

o f f

o o n f f

296 days

186 days

M R I

2-wk on-off 3 cycles

M R I

2-wk on-off 4 cycles

o 3 n

o o f n f

o o f n f

o f f

o n

458 days

o f f

4

578 days

Cohort B M R I

M R I

9 wk alcohol

on

1

age: 88 days

o f f

o 2 n

179 days

M R I

2-wk on-off 3 cycles

o f f

o n

o f f

o o n f f

o 3 n

284 days

M R I

2-wk on-off 3 cycles

o f f

o n

o o f n f

o f f

o 4 f f

369 days

8 wk binge

M R I

on

o f f 5

452 days

Fig. 1. Timelines for magnetic resonance imaging and alcohol exposure schedules for the 2 cohorts of P rats.

METHODS Subjects The study groups comprised 2 cohorts of adult male P rats (17 rats in Cohort A; 18 rats in Cohort B), selectively bred at the Indiana University Medical Center (Indianapolis, IN) for a preference for 10% (v/v) alcohol over concurrently available water. As young adults, rats were shipped by air to SRI International (Menlo Park, CA), where they remained thereafter and were available for brain imaging. Rats were singly housed in plastic tubs, with water and food (vitamin and mineral enriched mouse and rat diet #7,001; Teklad, Madison, WI) freely available in the home tubs at all times. Group Assignment and Alcohol Exposure Cycles Timelines of the alcohol exposure schedules for each rat cohort are presented in Fig. 1, and alcohol consumption and body weight data are presented in Fig. 2A for Cohort A and Fig. 2B for Cohort B. Cohort A. Cohort A included 9 alcohol-exposed rats derived from 3 litters and 8 control rats from a single litter of the S52 generation and studied in 4 MRI sessions at ages 186, 296, 458, and 578 days. The rats were in their new housing environment for about 3 weeks before the first MRI scan. Because the initial alcohol exposure for Cohort A was conducted in Indiana, pre–alcohol exposure baseline MRI data are not available. All rats were weighed twice a week, and fluid intake (water and alcohol) was measured by weighing bottles every other day, with the volumes and/or weights for the missing or outlying (typically because of spillage) values being the average of the 2 nearest measurements. The 9 alcohol-exposed rats were first presented alcohol for 9 weeks continuously and then underwent MRI scanning about 1 month later. Between MRI 1 and 2 and then again between MRI 2 and 3, these rats experienced 3 cycles of 2 weeks of access to alcohol, followed by 2 weeks of deprivation from alcohol access; between MRI 3 and 4, they experienced 4 cycles of 2 weeks of access to alcohol, followed by 2 weeks of deprivation from alcohol access. Alcohol was freely available at all alcohol exposure times under 3-bottle-choice conditions, with 15% (v/v) alcohol, 30% (v/v) alcohol, and 100%

water available concurrently. The use of multiple concentrations of alcohol has been shown to markedly increase intakes of alcohol both under continuous (e.g., Bell et al., 2003, 2004b) and following periods of abstinence (cf. Bell et al., 2005; Li et al., 2001; Rodd et al., 2004). Additionally, chronic alcohol drinking (4 or more weeks) produces marked long-range behavioral (Rodd-Henricks et al., 2001a) and CNS metabolic activity (Smith et al., 2002) changes. Cohort B. Cohort B included 9 alcohol-exposed rats and 9 non– alcohol-exposed rats; each group was derived from the same 5 litters of the S53 generation and studied 5 times over 1 year at ages 88, 179, 284, 369, and 452 days. At the outset of the study, the rats were 88 days old and were habituated to their new housing environment for 4 weeks before the baseline MRI scan and initial alcohol exposure. After baseline scanning, rats were assigned to 1 of 2 groups, alcoholnaı¨ ve (n 5 9) and alcohol drinking (n 5 9), based on litter, body weight, and brain size. The resulting groups started with the following characteristics before alcohol exposure: on average, the alcohol group weighed 415.4  37.7 g and had a brain volume of 2.20  0.07 cm3, and the control group weighed 406.9  39.5 g and had a brain volume of 2.22  0.10 cm3. The 5 litters were distributed between the 2 study groups as follows: the alcohol and control groups each had 2 animals from 3 litters; the remaining 2 litters each had 3 animals, where 1 rat from 1 litter and 2 rats from the other were assigned to each group. Between MRI 1 (prealcohol baseline scanning) and MRI 2, the 9 rats in the alcohol group were exposed to alcohol for 9 weeks continuously. Between MRI 2 and 3 and then again between MRI 3 and 4, these rats were exposed to alcohol for 2 cycles of 2 weeks on followed by 2 weeks off alcohol. For these 2 sets of alcohol exposure cycles, alcohol was freely available during ‘‘on’’ weeks in a 3-bottlechoice paradigm: 15% (v/v) alcohol, 30% (v/v) alcohol, and 100% water. Between MRI 4 and 5, they were placed on a multiple scheduled access drinking-in-the-dark (DID-MSA) ‘‘binge’’ schedule (e.g., Bell et al., 2004a), where alcohol was placed in the cages for 4, 1-hour periods in the 12-hour dark cycle for 5 consecutive days each week for 8 weeks. The first hour of access began 1 hour after the onset of the dark cycle, and each hour of access was separated by 2 hours of no-access in the dark cycle. The dark/light cycle was reversed, and

ALCOHOL EFFECTS ON RAT BRAIN

a

1251

b 20

Alcohol Consumption 20

Alcohol g/kg/day

Alcohol g/kg/day

Cohort A 15 10 5 0

Cohort B Cohort B (7 day)

15 10 5 0

0

10

20 30 40 50 Week of Alcohol Exposure

60

70

0

Body Weight

10

600

500

400

20 30 Week of Alcohol Exposure

40

50

Body Weight

700 Body Weight g

700 Body Weight g

Alcohol Consumption

600

500

400 0

10

20 30 40 Week during Alcohol Exposure

50

60

0

10

20 30 40 Week during Alcohol Exposure

50

60

Fig. 2. Alcohol consumption and body weight during the exposure periods for the 9 rats in Cohort A (A) and the 9 rats in Cohort B (B). Note that the open circles represent the last 8-week binge cycles prorated to 7 days of drinking; the closed circles represent the values based on the actual 5-day exposure period. Given the excessive weight gain in all animals over the course of the experiment, the alcohol consumption at the exposure cycle was even higher than reflected in raw values when considered in terms of lean body mass.

dim indirect red lighting was on during the 12 hours of the dark cycle, to facilitate alcohol administration and measuring bottle weights after each 1-hour session. The DID-MSA alcohol exposure regimen also used a 3-bottle choice but with higher alcohol concentrations than was used in the earlier regimens: 20% (v/v) alcohol, 40% (v/v) alcohol, and 100% water available concurrently. With this protocol, P rats have been reported to consume as much alcohol during the four 1-hour sessions as they would consume over a 24hour period, with bouts of 1 to 2 g/kg occurring during each 1-hour session (Bell et al., 2004a). MRI Scanning Procedures Cohort A animals underwent 4 MRI sessions, and Cohort B underwent 5 MRI sessions. Figure 1 presents the timeline for alcohol exposure and MRI scanning. Details of the MRI procedures have already been described (Sullivan et al., 2006) and are summarized here. All animal preparation, monitoring, and recovery were conducted according to procedures approved by the Administrative Panel on Laboratory Animal Care and performed by 2 highly experienced veterinary technicians (R.V.T. and L.A.Tg.). The rats were anesthetized using 1% to 3% isoflurane and monitored continually. Rats were scanned in an animal positioning system (John Houseman, Thurso, Caithness, Scotland) to position and immobilize the head. Brain MRI acquisition was conducted on 2 different platforms at different locations. For Cohort A, all 4 sessions were performed on a 4.7-T animal system (Varian, Palo Alto, CA) at the R.M. Lucas Center (Stanford University, Stanford, CA). For Cohort B, the first 3 sessions were also acquired at the 4.7-T scanner, and sessions 4 and 5 were acquired on a 3-T human scanner (GE Healthcare, Milwaukee, WI) fitted with custom RF coils for rats, located at SRI International in Menlo Park, CA. Each MRI session took about 1.5 to 2 hours. Although suboptimal to change scanners while conduct-

ing a longitudinal study, change was necessitated by a shim coil failure on the 4.7 T, causing a substantial decline in image quality. The acquisition protocols were designed to match signal intensity across scanners, and postacquisition image processing used phantoms to control for scanner spatial scaling differences. 4.7-T System. A localizer scan was acquired transaxially with respect to the scanner bore, producing coronal sections of the rat brain [multislice spin echo (SEMS), TE/TR 5 14/500 ms, 1 NEX, 256  128, field of view (FOV) 5 80 mm, 21 slices, 1 mm thick, no slice gap]. A sagittal localizer scan (SEMS, TE/TR 5 13/370 ms, 2 NEX, 256  128, FOV 5 80 mm, 21 slices, 0.5 mm thick, no slice gap) was then acquired and used for a graphical prescription of coronal sections that covered the entire brain. The dual-echo coronal sequence comprised 60 contiguous, 0.5-mm-thick slices, with echo times of 22 and 44 ms, with a 3-second TR covering the entire brain of the animal. The acquired spatial resolution was 256256 pixels over a 4-cm, square FOV. To obtain the 0.5-mm-thick slices, an RF pulse designed by Dr. Peter Jezzard (FMRIB Centre, University of Oxford, UK) was used instead of the Varian-supplied sinc waveform. 3.0-T System. A clinical 3-T GE Signa human MRI scanner (gradient strength 5 40 mT/m; slew rate 5 150 T/m/s; software version VH3) was used for data acquisition with standard product sequences. An 8-rung high-pass birdcage transmit/receive coil was constructed from cylinders of clear plexiglass with the same dimensions as that used on the 4.7-T system for imaging at 3 T. The acquisition protocol closely followed that used at 4.7 T to achieve similar tissue contrast. A 21-slice per plane, 3-plane localizer (TE 5 3 ms, TR, FOV 5 80 mm, 256  128 pixel resolution, 0.5-mm slice thickness) was used to define the rostral and caudal extremes of each rat brain; the sagittal view was used for graphical prescription. A dual-echo spin-echo acquisition was graphically prescribed to cover the entire rat brain (3.3 cm) in 66 coronal slices (TE1 5 16 ms, TE2 5 50 ms, TR 5 5 s, FOV 5 60 mm, 256  256 pixel resolution,

PFEFFERBAUM ET AL.

1252

0.5-mm slice thickness, 22-min acquisition time) and run 3 times for improved signal-to-noise ratio. The resulting native spatial resolution was 234  234 500 mm.

Image Postprocessing Brain images were transferred to a workstation where they were analyzed using in-house software routines, written in IDL (Interactive Data Language, Boulder, CO). The images were interpolated from the acquired 256  256 resolution, which corresponds to a 156 mm (at 4.7 T) and 234 mm (at 3 T) in-plane linear pixel size to a 512  512 pixel grid (a 78-mm voxel size). The center of the brain at the center slice was identified, and interactively, a midsagittal line was drawn. The data were then rotated such that the midsagittal orientation was vertical. The middle 256  256 pixels of this centered, rotated volume were extracted, as they were sufficient to contain all brain voxels. Before averaging the separate acquisitions for each animal within a session, the data were spatially aligned on a slice-by-slice basis to maximize a cross-correlation metric, with no slice shifts larger than 1.2 mm. This procedure reduced blurring and increased the conspicuity of tissue margins compared with simple averaging without alignment, thereby reducing the adverse effects of motion during the scan. The 2 acquired echoes were combined in a weighted average. The brain was then extracted based on nonparametric thresholding and 2-dimensional contour filling. For each animal, the data were also interpolated to an isotropic volume at a 78-mm voxel size. For each cohort, a well-aligned animal from a single session was chosen as the template dataset to which each individual animal for each session was aligned using a 6parameter rigid body function (AIR 5.0), thus preserving brain size. To compare the data from Cohort B, which were acquired across 2 different scanners, phantoms were used to correct for within-scanner drift and across-scanner scaling.

Whole and Regional Brain Measures The isotropic data were used for computation of whole-brain volume; midsagittal corpus callosum area, length, and height; and cortical thickness and volume of the cerebellum and dorsolateral ventricles for all MRI sessions. The hippocampus was measured on the native 0.5-mm-thick data using anatomical landmarks on MRI 1, 4, and 5. A rostral-to-caudal grand average profile of coronal slice areas across the entire brain was constructed for each session for each cohort. Whole-brain volume was computed by summing all voxels beginning at the demarcation of the olfactory bulb/frontal cortex junction rostrally and proceeding caudally for 23 mm [300 (78-mm-thick) isotropic coronal slices]. Total brain volume and the area of the corpus callosum were measured in Cohorts A and B. Additional regions were measured in Cohort B: dorsal cortical thickness; area, length, and height of the corpus callosum; and volumes of cerebellum, dorsal and ventral hippocampus, and dorsolateral ventricles. Examples of the outlining of these regions are presented in Part 1 of this pair of papers (Sullivan et al., 2006).

Statistical Analysis Data from each cohort were analyzed separately. Group differences were examined with t-tests (a 5 0.05, 2-tailed) and repeatedmeasures analysis of variance (ANOVA) and follow-up t-tests or Scheffe tests (a 5 0.05, 2-tailed) for multiple comparisons. Depending on the distribution of the measures, either Pearson correlations or Spearman Rank correlations were used to test relationships between variables.

RESULTS

Longitudinal Changes Measured in Cohort A The examination of Cohort A must be considered exploratory because baseline scanning was not acquired and the treatment groups were not matched in litter distribution. All 8 of the water control rats were derived from a single litter, and of the 9 rats in the alcohol group 4 came from 1 litter (Litter 1) and 5 from another litter (Litter 2). One rat in the water control group died of a splenic tumor between MRIs 2 and 3. The remaining rats were scanned at all 4 MRI sessions. Litter and Alcohol Effects Across MRI Sessions. Because of the confound of litter and alcohol exposure, group analyses conducted for weight and total brain volume were with repeated measures ANOVAs in the water versus the alcohol group across the 4 MRIs. Follow-up 1-way ANOVAs used the 3 litters as the grouping factor. Both the water and the 2 alcohol litter groups had significant body weight gain across sessions [F(3,4 2) 5 133.202, p 5 0.0001]. Although the treatment group effect was not significant [F(1, 42) 5 2.805, p 5 0.1161], the treatment group-by-session interaction was [F(3, 42) 5 5.681, p 5 0.0023], indicating that on average the water group showed a greater body weight gain than the alcohol group. One-way ANOVAs indicated a significant litter effect across the sessions (p 5 0.007), where 1 litter (Litter 1) in the alcohol group was consistently smaller than the other alcohol litter (Litter 2) and the water group, and the latter 2 groups did not differ from each other (Fig. 3). A followup repeated-measures ANOVA, therefore, included only the water group and the alcohol Litter 2. This ANOVA yielded a trend toward a difference in the rate of body weight gain, suggesting that the gain in the alcohol Litter 2 was attenuated relative to the water group [group-bysession interaction F(1, 30) 5 2.37, p 5 0.0905]. The results of the total brain volume analysis were similar to those characterizing growth in body weight. The 3-litter repeated measures ANOVA yielded a litter effect [F(2, 39) 5 8.207, p 5 0.005], a growth effect [F(3, 39) 5 49.388, p 5 0.0001], and a trend toward significance for the interaction [F(6, 39) 5 2.245, p 5 0.059]. A second repeated-measures ANOVA included only the water group and alcohol Litter 2 and yielded a trend toward a difference in the rate of brain volume growth [group-bysession interaction F(1, 30) 5 2.606, p 5 0.07]. Although both groups exhibited brain growth over the first 3 MRIs, the brain volume of the Litter 2 alcohol group declined at MRI 4 (Fig. 3). The litter-group effects on corpus callosum area were more modest than those observed for total brain volume. In particular, the 3-litter-group repeated-measures ANOVA yielded only a trend for a litter-group effect [F(2, 39) 5 2.804, p 5 0.0972] and a growth effect [F(3, 39) 5 49.855, p 5 0.0001], but no interaction [F(6, 39) 5 1.525, p 5 0.20]. A follow-up ANOVA based on the water group and

ALCOHOL EFFECTS ON RAT BRAIN

1253

Body Weight

2.9

Weight (g)

650 600 550

7.0

2.8 2.7 2.6

500

200 300 400 500 MRI Session (days old)

600

2.5 100

Water (N=7) Alcohol Litter 1 (N=4) Alcohol Litter 2 (N=5)

6.5 Area (mm2)

700 Volume (mm3)

3.0

450 100

Corpus Callosum

Brain Volume

750

6.0 5.5 5.0 4.5

200

300

400

500

MRI Session (days old)

600

4.0 100

200

300

400

500

600

MRI Session (days old)

Fig. 3. Mean  SEM of body weight, brain volume, and corpus callosum area at each of the 4 magnetic resonance imaging sessions for the 2 alcoholexposed litters and the 1 water-control litter of Cohort A.

alcohol Litter 2 yielded a significant difference in group [F(1, 30) 5 12.096, p 5 0.006] and a trend toward significance for a group-by-session interaction [F(1, 30) 5 2.591, p 5 0.0711]. Although both groups exhibited brain growth over the first 3 MRIs, the water group brain volume, which had lagged that of the alcohol Litter 2 group, increased faster than that of the alcohol group between MRIs 3 and 4 (Fig. 3). Within each litter group, the correlations between body weight and brain volume or callosal area on a session-bysession basis were at best modest, none being statistically significant. Body weight-brain volume correlations based on the combination of all 16 rats with all 4 MRIs were significant for each session (r 5 0.51–0.66). The only significant correlation between brain volume and callosal area was for MRI 5 and only when all 16 rats were included (r 5 0.50, p 5 0.05). Longitudinal Changes Measured in Cohort B All 18 rats in Cohort B survived for the duration of the study and were examined 5 times at 3-month intervals over 12 months. Given the planned litter distribution between the alcohol and water study groups, ANOVAs considered litter and alcohol as grouping factors initially and for brain regions not showing a significant litter effect follow-up ANOVAs were conducted using alcohol treatment as the sole grouping variable. Litter Effects at Baseline MRI. One-way ANOVAs for body weight and each brain regional measure at baseline across the 5 litters revealed litter effects for volume of total brain [F(4, 17) 5 8.616, p 5 0.0013], ventral hippocampus [F(4, 17) 5 6.868, p 5 0.0034], total hippocampus [F(4, 17) 5 3.343, p 5 0.0433], and dorsolateral ventricles [F (4, 17) 5 4.006, p 5 0.0248] and trends toward significance for body weight [F(4, 17) 5 2.448, p 5 0.0986] and cerebellar volume [F(4, 17) 5 2.641, p 5 0.082]. Litter effects were not significant for callosal area, length, or height; cortical thickness; or dorsal hippocampal volume. Litter size var-

ied 4-fold (4, 10, 11, 14, and 16) and brain size and body weight correlated significantly with litter size, such that animals from smaller litters had larger brain volumes and body weight: total brain r 5 0.728, p 5 0.0027; ventricles r 5 0.656, p 5 0.0068; corpus callosum area r 5 0.394, p 5 0.1045; total hippocampus r 5 0.665, p 5 0.0061; cerebellum r 5 0.457, p 5 0.0596; body weight r 5 0.378, p 5 0.1189. Scatterplots of these data are presented in Fig. 4. Alcohol Consumption and Effect on Growth in Body Weight. During the DID-MSA 8-week binge protocol, the amounts of alcohol consumed, averaged across weeks, during the first hour of the previous Friday was 2.4  0.1 g/kg (mean  SEM) and 3.2  0.2 g/kg on the subsequent Monday, indicating that the rats significantly [t(8) 5 4.96, p 5 0.001] increased their drinking after a weekend hiatus. Over the 1-year study period, the water rats gained 217  7 g (53.9%) and the alcohol rats gained less weight, 188.9  44.0 g (45.3%), but the difference was not significant [t(16) 5 1.289, p 5 0.22] (Fig. 5). However, when expressing each rat’s weight as a percentage change from baseline, the proportional weight gain over the course of the study was significant in both groups [F(3, 48) 5 106.58, p 5 0.0001] but significantly less in the alcohol than water group [F(1, 48) 5 5.932, p 5 0.0269]. Alcohol Effect on Change in Whole and Regional Brain Measures. The mean  SEM values for each regional brain measure for each of the 5 MRI sessions are presented in Fig. 5. Statistics were based on 2-group (alcohol vs water) by 5 MRI session ANOVAs conducted for percentage change from baseline for each brain measure. Irrespective of alcohol exposure, a significant change was detected in total brain volume [F(3, 48) 5 337.428, p 5 0.0001], callosal area [F(3, 48) 5 10.677, p 5 0.0001], callosal length [F(3, 48) 5 96.131, r 5 0.0001], and cerebellar volume [F(3, 48) 5 140.204, p 5 0.0001]; the ventricles showed a trend toward significance for expansion [F(3, 48) 5 2.386, p 5 0.0807]. Analyses based on raw scores uncorrected for baseline measures yielded the same

PFEFFERBAUM ET AL.

1254

Weight p =.099

500

350

30 Volume (mm3)

400

2200

2100

300 10

11 14 Litter Size

16

0 4

Ventral Hippocampus p=.0034

33

10

11 14 Litter Size

16

4

Total Hippocampus p=.0433

54

32

29 28

50 48 46 44

27 26 10

11 14 Litter Size

16

16

Cerebellum p =.082

260 250 240 230

42 4

11 14 Litter Size

270 Volume (mm3)

Volume (mm3)

30

10

280

52

31

20

10

2000 4

Dorsolateral Ventricles p =.0248

40

2300 Volume (mm3)

Weight (g)

450

Volume (mm3)

Total Brain p =.0013

2400

220 4

10

11 14 Litter Size

16

4

10

11 14 Litter Size

16

Fig. 4. Body weight and regional brain measures for each of the 18 rats in Cohort B plotted by litter size at baseline, that is, before alcohol exposure. The pvalues refer to the group effect from analysis of variances.

results, with the exception of the ventricles, for which the change over time was significant (p 5 0.0054). The alcohol effect on proportional change showed trends for callosal area [F(1, 48) 5 3.398, p 5 0.0839] and total hippocampal volume [F(1, 16) 5 3.588, p 5 0.0764] to be smaller in the alcohol than the water group. The alcohol effect in the lateral ventricles was significant, but in this case, the ventricles shrank in the alcohol rats and expanded in the water rats [F(1, 48) 5 4.756, p 5 0.0445]. A similar unpredicted effect occurred in the cerebellum, which showed greater expansion in the alcohol than water rats [F(1, 48) 5 4.460, p 5 0.0508]. Although ANOVAs using raw brain measures uncorrected for baseline failed to yield significant simple alcohol effects, alcohol-by-session interactions were significant for callosal area [F(4, 64) 5 3.893, p 5 0.0068] (Fig. 6) and callosal height [F(4, 64) 5 3.011, p 5 0.0244], with a trend noted for changes in ventral hippocampal volume [F(4, 64) 5 2.751, p 5 0.079]. For these regions, growth was attenuated in the alcohol group. A statistical trend for cortical thickness increases in alcoholconsuming rats [F(4, 64) 5 2.420, p 5 0.0574] was in the unexpected direction. Relations Between Body Weight and Brain Measures. At baseline, body weight was a predictor of several brain measures, including total brain volume (r 5 0.56, p 5 0.0148), callosal length (r 5 0.53, p 5 0.0225), ventral

hippocampus (r 5 0.53, p 5 0.0252), total hippocampus (r 5 0.50, p 5 0.0344), and cerebellum (r 5 0.50, p 5 0.0345). Body weight gain and a few brain measures expressed as a proportion of change from baseline correlated positively in each group (Table 1). Significant correlations with body weight gain were found for total brain volume (water r 5 0.77, p 5 0.0155; alcohol r 5 0.88, p 5 0.0018), cortical thickness (water r 5 0.76, p 5 0.0178; alcohol r 5 0.79, p 5 0.011), hippocampal volume (alcohol r 5 0.71, p 5 0.0307), and ventricular volume (water r 5 0.88, p 5 0.0019; alcohol r 5 0.68, p 5 0.0434). Relations Between Alcohol Consumption and Brain Measures. Our hypothesis that the amount of alcohol consumed would correlate negatively with the size of regional brain measures was not supported for any measure. DISCUSSION

Both cohorts of P rats were studied as adults for about 1 year and consumed high doses of alcohol for much of the duration of each study. Rather than exhibiting the human pattern of tissue volume decline that occurs in normal aging with an exacerbated decline in alcoholism, brain structures of the adult P rats in both the alcohol and the water control conditions showed significant growth, which was attenuated in a few measures in the alcohol-exposed

ALCOHOL EFFECTS ON RAT BRAIN

1255

Weight

650

Brain Volume

2500

300

600 550 500 450

290

Water (N=9)

Volume (mm3)

Weight (g)

2400

2300

250 2100

350 0

100

200

300

400

500

100

200

300

400

500

0

MRI Session (days old)

Corpus Callosum Group-by-MRI Interaction p=.0068

6.5

4.5

8 7.5

200

300

400

500

0

MRI Session (days old)

100

200

300

400

0

54

19

31 30

200

300

400

500

0

MRI Session (days old)

100

200

300

400

500

2.3

200

300

400

MRI Session (days old)

500

100 200 300 400 MRI Session (days old)

500

Ventricle-to-Brain Ratio

16

0.007

14

0.006

12

0.005

10

0.004

8

0.003

6

2.25 100

0

Ratio

Water (N=9)

2.35

0

50

Dorsolateral Ventricles

2.4

500

Total Hippocampus

MRI Session (days old)

Cortical Thickness Group-by-MRI Interaction p=.0574

2.45

400

46

28

17

300

48

29

18

200

52

Volume (mm3)

Volume (mm3)

20

100

MRI Session (days old)

32

100

1.3

500

Ventral Hippocampus Group-by-MRI Interaction p=.079

33

21

0

1.35

MRI Session (days old)

Dorsal Hippocampus

22

500

1.2

6.5 100

400

1.25

7

4

300

1.4

Height (mm2)

Length (mm2)

5

200

Callosal Height Group-by-MRI Interaction p=.0244

1.45

8.5

5.5

100

MRI Session (days old)

Callosal Length

9

6

0

Alcohol (N=9)

240 0

MRI Session (days old)

Area (mm2)

270 Water (N=9)

400

Volume (mm3)

280

260

2200

Area

Cerebellum

310

0.002 0

100

200

300

400

MRI Session (days old)

500

0

100 200 300 400 MRI Session (days old)

500

Fig. 5. Mean  SEM for body weight and each brain measure at each of the 5 magnetic resonance imaging sessions for the 9 water control rats and the 9 alcohol-exposed rats of Cohort B.

1256

PFEFFERBAUM ET AL.

Fig. 6. Grand average late echo images in the sagittal (left), coronal (middle), and axial (right) planes for controls (top) and alcohol-exposed (bottom) animals at magnetic resonance imaging session 5. The white arrows on the sagittal and axial images indicate the slice position of the coronal view. The arrow on the sagittal image points to the corpus callosum.

groups. The region with the greatest demonstrable alcohol exposure effect was the corpus callosum, which was measured on midsagittal images. The P rats, therefore, showed an age–alcohol interaction different from humans, in that normal growth in selective brain regions that continues in adult rats was retarded. Examination of the growth in brain volume of the first cohort study which was initiated in later adulthood and extended toward 2 years of age, provided a suggestion of age-related volume decline between the third and fourth MRIs, that is, between age 458 and 578 days old. This hint of a volume decrease with age suggests that extending longitudinal study to very old rats is necessary to produce an adequate model of the human age-by-alcohol interaction. Our brain growth analysis (Sullivan et al., 2006) revealed that the brain structures measured in the water control rats in the current study showing the greatest growth were the corpus callosum, cerebellum, and hippocampus. Of these regions, the effect of alcohol consumption was most evident on the corpus callosum and modestly on the hippocampus. In humans, white matter, including the corpus callosum (Giedd et al., 1999), continues growth throughout adolescence and into young adulthood (Yakovlev and Lecours, 1967) and is a principal determinant of ultimate intracranial volume (Pfefferbaum et al., 1994). One speculation is that structures that continue substantial growth in adulthood may be the most vulnerable to effects of environmental factors and insult. Alternatively, given continued growth, these regions may be able to regenerate when a neurotoxin or other unhealthy agent is removed (cf. Nixon and Crews, 2002, 2004). The alcohol-exposed rats of both cohorts gained less weight than did their water control counterparts, whether matched for litter (Cohort B) or not (Cohort A). In Cohort B, body weight at the last MRI session was significantly related to the total lifetime consumption of alcohol. Body

weight loss, as commonly occurs in alcoholic humans, or lack of normal weight gain, as occurred in these 2 cohorts, may signal a nutritional–alcohol interaction, putting alcohol-exposed individuals at risk for nutritional deficiencies (e.g., Harper and Butterworth, 1997; Martin et al., 2003; Zimatkin and Zimatkina, 1996) and alcoholismrelated brain dysmorphology (e.g., Pfefferbaum et al., 2004b). Nonetheless, in the present study, all rats, regardless of alcohol history, were obese and thus even the alcohol-drinking rats ate the vitamin-fortified chow and apparently had adequate nutrition while drinking, which may have mitigated potential brain damage (cf. Harper and Butterworth, 1997; Langlais and Savage, 1995; Martin et al., 2003; Pentney et al., 1993; Zimatkin and Zimatkina, 1996). Litter differences are well known to exert highly significant effects on brain structure (e.g., Crabbe, 1997) and on consummatory, cognitive, and motor behaviors (e.g., Crabbe et al., 1999). In the present study, litter was as significant a factor in determining group differences as was alcohol exposure history. Of the measurements taken, litter effects were significant for volumes of total brain, hippocampus, cerebellum, and ventricles. The genesis of the litter differences can be genetically or environmentally based. A current study reported that differential maternal grooming and nursing behavior was adequate to modify action of the glucocorticoid receptor gene promoter in the hippocampus that was previously assumed to be genetically determined (Weaver et al., 2004). Although the rat siblings in the current study were treated similarly during weaning and thereafter, it is possible that even slight differences in early maternal behavior or caging differences were predisposing factors resulting in differential growth rats and vulnerability. We did observe that litter size correlated with brain structure size, and we speculate that litter size may have contributed significantly to the

ALCOHOL EFFECTS ON RAT BRAIN

1257

Table 1. Correlations Between Percentage Changes from Baseline in Weight and Brain Measure Within Each Study Group Percentage change in weight correlated with Control rats (N 5 9) Total brain volume r p Corpus callosum Area r p Length r p Height r p Cortical thickness r p Hippocampus Dorsal r p Ventral r p Total r p Cerebellum r p Dorsolateral ventricles r p Alcohol rats (N 5 9) Total brain volume r p Corpus Callosum Area r p Length r p Height r p Cortical thickness r p Hippocampus Dorsal r p Ventral r p Total r p Cerebellum r p Dorsolateral ventricles r p

MRI 1–2

MRI 2–3

MRI 3–4

MRI 4–5

0.77 0.02

0.09 NS

0.50 NS

0.44 NS

0.21 NS

0.08 NS

0.19 NS

0.32 NS

0.28 NS

0.06 NS

0.09 NS

0.44 NS

0.31 NS

0.35 NS

0.21 NS

0.32 NS

0.76 0.02

0.43 NS

0.04 NS

0.06 NS





0.62 0.08

0.20 NS





0.74 0.02

0.59 0.09





0.75 0.02

0.55 NS

0.48 NS

0.09 NS

0.41 NS

0.09 NS

0.26 NS

0.44 NS

0.88 .00

0.25 NS

0.88 0.00

0.50 NS

0.45 NS

0.11 NS

0.30 NS

0.84 0.00

0.20 NS

0.44 NS

0.48 NS

0.22 NS

0.17 NS

0.43 NS

0.63 0.07

0.71 0.03

0.49 NS

0.03 NS

0.42 NS

0.79 0.01

0.28 NS

0.44 NS





0.38 NS

0.13 NS





0.65 0.06

0.78 0.01





0.61 0.08

0.71 0.03

0.59 0.09

0.09 NS

0.14 NS

0.35 NS

0.68 0.04

0.36 NS

0.49 NS

0.39 NS

MRI, magnetic resonance image.

observed differences across litters. This effect may be genetic or may represent the earliest environmental differences among animals. As has been documented in human twin studies, the size of certain brain structures is more heritable than others, with high genetic contribution to intracranial volume, lateral ventricles, and corpus callosum (Pfefferbaum et al., 2004c, 2000) and also the frontal lobes (Carmelli et al., 2002), but there is less than 50% genetic contribution to the size of the hippocampus (Sullivan et al., 2001). To the extent that litter differences reflect genetic variance, the litter differences in brain morphology may also indicate genetic influences within selectively bred lines of rodents and a need for identifying genes underlying the brain structural differences. Indeed, age and litter variability highlight the need for longitudinal study of litter-matched controls even in rodent strain selectively bred for a feature, such as voluntary alcohol consumption. In vivo MR imaging permits noninvasive, longitudinal study of the effects of age and alcoholism. To the extent that particular mechanisms of aging can be generalized across species (cf. Finch and Kirkwood, 2000), the P rat, which has been selectively bred to consume alcohol in large quantities voluntarily, holds promise for modeling the age–alcohol interaction. While not all features of rodent aging follow patterns of normal human aging, some do, including lack of decline in regional hippocampal cell number in old rats (Rasmussen et al., 1996) and humans (West et al., 1994; West and Gundersen, 1990) in normal aging and the susceptibility to alcohol-related loss of selective hippocampal cell lines in rats (Walker et al., 1980) and humans (Bengochea and Gonzalo, 1990; but see Harding et al., 1998, 1997). These postmortem findings mirror in vivo MRI human studies reporting lack of hippocampal volume shrinkage in older healthy adults (Raz et al., 2004; Sullivan et al., 1995a, 2001; but see Jernigan et al., 2001) and presence of volume shrinkage in alcoholic individuals (Agartz et al., 1999; De Bellis et al., 2000; Sullivan et al., 1995b). The opportunity to survey the entire brain repeatedly and to archive longitudinal data for continued analysis provides a study method impossible with cross-sectional examination, including postmortem analysis. Nonetheless, MRI is limited in its resolution of small structures that may be of particular interest to studies of alcoholism, for example, selective nuclei of the accumbens, thalamus, and amygdala (e.g., Koob et al., 1998; Lu et al., 1997; Porrino et al., 1998; Rintala et al., 1997; Smith et al., 2001; Strother et al., 2005) and awaits improvements in scanning technology that may enhance current detection limits. In light of the alcohol effect on the corpus callosum noted herein, of particular promise will be the application of MR diffusion tensor imaging (DTI), which provides an in vivo measure of the microstructural integrity of white matter (for a review, see Pfefferbaum and Sullivan, 2005). Unlike conventional MRI, which is used to determine size and

1258

morphology of whole brain and selective structures, DTI is sensitive to the microstructural constituents of white matter fibers, including myelin, and has successfully identified white matter abnormalities of the corpus callosum in alcoholic women undetectable with conventional MRI measures (Pfefferbaum et al., 2002; Pfefferbaum and Sullivan, 2002). Each rat cohort underwent a series of different alcohol administration schedules, including free access for extended periods (at least 2 months), on/off access commonly producing an alcohol deprivation effect (cf. Bell et al., 2005; Li et al., 2001; Rodd et al., 2004), and the binge schedule, which results in equivalent amounts of alcohol consumed during the four 1-hour access periods as that seen over the entire 24 hours (Bell et al., 2004a). In all cases, water was available ad libitum to the alcohol rats and alcohol drinking was voluntary. Animals were given a 3-bottlechoice paradigm—15% v/v alcohol, 30% v/v alcohol, and water—with the DID-MSA cycle for Cohort B, using higher concentrations of alcohol (20 and 40% v/v). With these alcohol-exposure schedules, the P rats voluntarily consumed vast amounts of alcohol, even after adjusting for the 2-fold greater alcohol metabolism rate of rats than humans (a rat’s drinking 9 g/kg/d alcohol is equivalent to consuming a fifth of a 100-proof alcoholic beverage by a 70-kg person per day). These animals had high-sustained basal levels of alcohol consumption, which was even greater in terms of lean body mass in their later months when they were obese. Nonetheless, following alcohol exposure, which had continued for much of their lives, the P rats in neither cohort showed the extensive alcoholrelated changes in brain morphometry reported in human alcoholism (but see Fein and Landman, 2005). Although behavioral tolerance was not tested in the current study, prior studies have reported that lower amounts of alcohol consumed across shorter periods than used herein have resulted in the development of metabolic and functional tolerance to the motor impairing and aversive effects of alcohol in P rats (Gatto et al., 1987; Lumeng and Li, 1986; Stewart et al., 1991). Nonetheless, the substantial alcohol consumed by these animals induced only modest focal effects on brain morphology. Indeed, the only significant one seen was attenuated growth of the corpus callosum. Selectively bred animals would be expected to have less morphological variance than unrelated humans. For example, the range of neuron number in the hippocampus in adult humans is upward of 50% of the mean, which is twice the variability of inbred rats (Finch and Kirkwood, 2000). The lesser morphological variance in selectively bred or inbred animals compared with humans should enhance the ability to detect treatment effects. Nonetheless, even in the P rat, we found substantial pretreatment morphological variability, attributable in part to litter differences, which may represent genetic variability of the earliest environmental effects. Moreover, the between species difference—from rat to human—in brain mass is

PFEFFERBAUM ET AL.

approximately 700-fold. Brain-to-body mass ratios also vary across species; a 2-g brain in a 400-g rat is 1:200, whereas a 1,400-g brain in a 70-kg man is 1:50. Proportional differences in brain-to-body ratio between species may be a relevant variable to consider when translating newly developed pharmacological treatments from use in animals to application in humans (Cosson et al., 1997). The current set of studies provides methods and essential baseline data for longitudinal examination of the myriad factors involved in the development and maintenance of alcohol dependence and potential neural and behavioral improvement with abstinence. As the first longitudinal and quantitative examinations of total and regional brain morphology in the adult development and alcohol exposure in the P rat, these MRI studies set the stage for further investigations. Follow-up studies should plan brain morphological comparison of P rats, NP rats, and wild-type Wistar rats to determine differences attributable to selective breading. Similarly, MRI examination should compare developmental differences related to sex. Behavioral testing targeted at functions affected by aging and alcoholism, such as balance, spatial working memory, and response inhibition, should be conducted concurrent with MRI to test the functional ramifications of brain structural growth with age or shrinkage with alcohol exposure. Expansion of the age range to younger and older animals will also maximize the model’s translational utility. Inducement of dramatic and widespread dysmorphology, as occurs in the human condition or alteration in local cerebral glucose metabolism as occurs in P rats chronically exposed to alcohol but voluntarily consuming less alcohol than did the rats in the present study (Porrino et al., 1998; Smith et al., 2001), may require higher and more prolonged brain alcohol levels than those used herein, introduction of alcohol at earlier stages of development, and reduction or removal of essential nutrients from food. In a thiamine deficiency challenge given the rats in Cohort B of the present study, with MRI we observed profound focal effects on brain tissue in the thalamus, mammillary bodies, and inferior colliculi with modest additional effects in thiamine-deficient compared with thiamine-enriched rats and that prior alcohol exposure modestly attenuated recovery of thalamic tissue (Pfefferbaum et al., 2006). The high rate of alcohol metabolism, or metabolism in general, of the rodent, (e.g., the selectively bred P rat; Edenberg et al., 2005; Lumeng et al., 1982; Schreiber and Freund, 2000; Smith et al., 2001), compared with humans, also contributes to the extensive quantities of alcohol these rats are able to drink. Identification of species or other genetic protective factors of brain integrity (e.g., Worst et al., 2005) may provide novel clues for understanding the heterogeneity of chronic alcoholism’s impact on brain structure and function in humans.

ALCOHOL EFFECTS ON RAT BRAIN

ACKNOWLEDGMENTS

We would like to thank Dr. Peter Jezzard of the FMRIB Centre, University of Oxford, for provision of his pulse sequence and user interface code. The authors would also like to thank Diane Howard, Wendy Baumgarden, Shay Cook, Luke Garcia, Robert Williams, Joanne Blum, Andrea Spadoni, Daniel J. Pfefferbaum, and Ted Sullivan for the invaluable care and time they devoted to various aspects of animal care, transportation, and scanning. At the inception of the study, the affiliation of Dr. T.-K. Li was Indiana University; his current affiliation is the National Institute on Alcohol Abuse and Alcoholism.

REFERENCES Agartz I, Momenan R, Rawlings RR, Kerich MJ, Hommer DW (1999) Hippocampal volume in patients with alcohol dependence. Arch Gen Psychiatry 56:356–363. Begleiter H, Porjesz B (1999) What is inherited in the predisposition toward alcoholism? A proposed model. Alcohol Clin Exp Res 23:1125–1135. Bell RL, Rodd ZA, Boutwell CL, Hsu CC, Lumeng L, Murphy JM, Li TK, McBride WJ (2004b) Effects of long-term episodic access to ethanol on the expression of an alcohol deprivation effect in low alcohol-consuming rats. Alcohol Clin Exp Res 28:1867–1874. Bell RL, Rodd ZA, Hsu CC, Lumeng L, Li TK, Murphy JM, McBride WJ (2004a) Effects of concurrent access to a single concentration or multiple concentrations of ethanol on ethanol intake by periadolescent high-alcohol-drinking rats. Alcohol 33:107–115. Bell RL, Rodd ZA, Murphy JM, McBride WJ (2005) Use of selectively bred alcohol-preferring rats to study alcohol abuse, relapse and craving, in Comprehensive Handbook of Alcohol Related Pathology, Vol. 3 (Preedy VR, Watson RR eds), pp 1515–1533. Academic Press/ Elsevier Science, New York. Bell RL, Rodd-Henricks ZA, Kuc KA, Lunneng L, Li TK, Murphy JM, McBride WJ (2003) Effects of concurrent access to a single concentration or multiple concentrations of ethanol on the intake of ethanol by male and female periadolescent alcohol-preferring (P) rats. Alcohol 29:137–148. Bengochea O, Gonzalo LM. (1990) Effect of chronic alcoholism on the human hippocampus. Histol Histopathol 5:349–357. Cardenas VA, Studholme C, Meyerhoff DJ, Song E, Weiner MW (2005) Chronic active heavy drinking and family history of problem drinking modulate regional brain tissue volumes. Psychiatry Res 138:115–130. Carlen PL, Wortzman G, Holgate RC, Wilkinson DA, Rankin JG (1978) Reversible cerebral atrophy in recently abstinent chronic alcoholics measured by computed tomography scans. Science 200:1076–1078. Carmelli D, Swan GE, DeCarli C, Reed T (2002) Quantitative genetic modeling of regional brain volumes and cognitive performance in older male twins. Biol Psychol 61:139–155. Cicero TJ, Snider SR, Perez VJ, Swanson LW (1971) Physical dependence on and tolerance to alcohol in the rat. Physiol Behav 6:191–198. Cloninger CR (1987) Neurogenetic adaptive mechanisms in alcoholism. Science 236:410–416. Cosson VF, Fuseau E, Efthymiopoulos C, Bye A (1997) Mixed effect modeling of sumatriptan pharmacokinetics during drug development. I: interspecies allometric scaling. J Pharmacokinet Biopharm 25: 149–167. Crabbe JC (1997) Where does alcohol act in the brain? Mol Psychiatry 2:17–20. Crabbe JC, Wahlsten D, Dudek BC (1999) Genetics of mouse behavior: interactions with laboratory environment. Science 284:1670–1672.

1259

Davila MD, Shear PK, Lane B, Sullivan EV, Pfefferbaum A (1994) Mammillary body and cerebellar shrinkage in chronic alcoholics: an MRI and neuropsychological study. Neuropsychology 8:433–444. De Bellis MD, Clark DB, Beers SR, Soloff PH, Boring AM, Hall J, Kersh A, Keshavan MS (2000) Hippocampal volume in adolescentonset alcohol use disorders. Am J Psychiatry 157:737–744. Dlugos CA, Pentney RJ (1997) Morphometric evidence that the total number of synapses on Purkinje neurons of old F344 rats is reduced after long-term ethanol treatment and restored to control levels after recovery. Alcohol Alcohol 32:161–172. Eckardt M, File S, Gessa G, Grant K, Guerri C, Hoffman P, Kalant H, Koob G, Li TK, Tabakoff B (1998) Effects of moderate alcohol consumption on the central nervous system. Alcohol: Clini Exp Res 22:351–359. Edenberg HJ, Strother WN, McClintick JN, Tian H, Stephens M, Jerome RE, Lumeng L, Li TK, McBride WJ (2005) Gene expression in the hippocampus of inbred alcohol-preferring and—nonpreferring rats. Genes Brain Behav 4:20–30. Ehlers CL, Chaplin RI, Lumeng L, Li TK (1991) Electrophysiological response to ethanol in P-rats and NP-rats. Alcohol Clin Exp Res 15:739–744. Fein G, Landman B (2005) Treated and treatment-naive alcoholics come from different populations. Alcohol 35:19–26. Fiel RJ, Alletto JJ, Severin CM, Nickerson PA, Acara MA, Pentney RJ (1991) MR imaging of normal rat brain at 0.35 T and correlated histology. J Magn Reson Imaging 1:651–656. Finch CE, Kirkwood TBL (2000) Chance, Development, and Aging. Oxford University Press, New York. Gatto GJ, Murphy JM, Waller MB, McBride WJ, Lumeng L, Li TK, (1987) Chronic ethanol tolerance through free-choice drinking in the P line of alcohol-preferring rats. Pharmacol Biochem Behavior 28: 111–115. Gazdzinski S, Durazzo TC, Meyerhoff DJ (2005) Temporal dynamics and determinants of whole brain tissue volume changes during recovery from alcohol dependence. Drug Alcohol Depend 78:263–273. Giedd JN, Blumenthal J, Jeffries NO, Rajapakse JC, Vaituzis AC, Liu H, Berry YC, Tobin M, Nelson J, Castellanos FX (1999) Development of the human corpus callosum during childhood and adolescence: a longitudinal MRI study. Prog Neuropsychopharmacol Biol Psychiatry 23:571–588. Harding AJ, Halliday GM, Kril JJ (1998) Variation in hippocampal neuron number with age and brain volume. Cereb Cortex 8: 710–718. Harding AJ, Wong A, Svoboda M, Kril JJ, Halliday GM (1997) Chronic alcohol consumption does not cause hippocampal neuron loss in humans. Hippocampus 7:78–87. Harper C, Butterworth R (1997) Nutritional and metabolic disorders, in Greenfield’s Neuropathology (Graham DI, Lantos PL eds), pp 601– 642. Arnold, London. Heath AC, Meyer J, Eaves LJ, Martin NG (1991a) The inheritance of alcohol consumption patterns in a general population twin sample.1. Multidimensional scaling of quantity/frequency data. J Stud Alcohol 52:345–352. Heath AC, Meyer J, Jardine R, Martin NG (1991b) The inheritance of alcohol consumption patterns in a general population twin sample.2. determinants of consumption frequency and quantity consumed. J Stud Alcohol 52:425–433. Hill SY (1992) Is there a genetic basis of alcoholism. Biol Psychiatry 32:955–957. Hommer D, Momenan R, Rawlings R, Ragan P, Williams W, Rio D, Eckardt M (1996) Decreased corpus callosum size among alcoholic women. Arch Neurol 53:359–363. Jernigan TL, Archibald SL, Fennema-Notestine C, Gamst AC, Stout JC, Bonner J, Hesselink JR (2001) Effects of age on tissues and regions of the cerebrum and cerebellum. Neurobiol Aging 22:581–594. Jernigan TL, Butters N, DiTraglia G, Schafer K, Smith T, Irwin M, Grant I, Schuckit M, Cermak L (1991) Reduced cerebral grey matter

1260

observed in alcoholics using magnetic resonance imaging. Alcohol Clin Exp Res 15:418–427. Kendler KS, Prescott CA, Myers J, Neale MC (2003) The structure of genetic and environmental risk factors for common psychiatric and substance use disorders in men and women. Arch Gen Psychiatry 60:929–937. Koob GF, Roberts AJ, Schulteis G, Parsons LH, Heyser CJ, Hyytia P, Merlo-Pich E, Weiss F (1998) Neurocircuitry targets in ethanol reward and dependence. Alcohol Clin Exp Res 22:3–9. Langlais PJ, Savage LM (1995) Thiamine deficiency in rats produces cognitive and memory deficits on spatial tasks that correlate with tissue loss in diencephalon, cortex and white matter. Behav Brain Res 68:75–89. Langlais PJ, Zhang SX (1997) Cortical and subcortical white matter damage without Wernicke’s encephalopathy after recovery from thiamine deficiency in the rat. Alcohol Clin Exp Res 21:434–443. Lee H, Holburn GE, Price RR (2001) In vivo and in vitro proton NMR spectroscopic studies of thiamine-deficient rat brains. J Magn Reson Imaging 13:163–166. Lee H, Tarter J, Holbourn GE, Price RR, Weinstein DD, Martin PR (1995) In vivo localized proton NMR spectroscopy of thiaminedeficient rat brain. Magn Reson Med 34:313–318. Li TK (2000a) Clinical perspectives for the study of craving and relapse in animal models. Addiction 95 (suppl 2): S55–S60. Li TK (2000b) Pharmacogenetics of responses to alcohol and genes that influence alcohol drinking. J Stud Alcohol 61:5–12. Li TK, Lumeng L, McBride WJ, Waller MB (1979) Progress toward a voluntary oral consumption model of alcoholism. Drug Alcohol Depend 4:45–60. Li TK, Spanagel R, Colombo G, McBride WJ, Porrino LJ, Suzuki T, Rodd-Henricks ZA (2001) Alcohol reinforcement and voluntary ethanol consumption. Alcohol Clin Exp Res 25:117S–126S. Lu W, Jaatinen P, Rintala J, Sarviharju M, Kiianmaa K, Hervonen A (1997) Effects of lifelong ethanol consumption on rat locus coeruleus. Alcohol Alcohol 32:463–470. Lumeng L, Li TK (1986) The development of metabolic tolerance in the alcohol-preferring P rats: comparison of forced and free-choice drinking of ethanol. Pharmacol Biochem Behav 25:1013–1020. Lumeng L, Waller MB, McBride WJ, Li TK (1982) Different sensitivities to ethanol in alcohol-preferring and -nonpreferring rats. Pharmacol Biochem Behav 16:125–130. Mann K, Gunther A, Stetter F, Ackermann K (1999) Rapid recovery from cognitive deficits in abstinent alcoholics: a controlled test-retest study. Alcohol Alcohol 34:567–574. Martin PR, Singleton CK, Hiller-Sturmhofel S (2003) The role of thiamine deficiency in alcoholic brain disease. Alcohol Res Health 27:134–142. McBride WJ, Li TK (1998) Animal models of alcoholism: neurobiology of high alcohol-drinking behavior in rodents. Crit Rev Neurobiol 12:339–369. Murphy JM, Stewart RB, Bell RL, Badia-Elder NE, Carr LG, McBride WJ, Lumeng L, Li TK (2002) Phenotypic and genotypic characterization of the Indiana University rat lines selectively bred for high and low alcohol preference. Behav Genet 32:363–388. Nixon K, Crews FT (2002) Binge ethanol exposure decreases neurogenesis in adult rat hippocampus. J Neurochem 83:1087–1093. Nixon K, Crews FT (2004) Temporally specific burst in cell proliferation increases hippocampal neurogenesis in protracted abstinence from alcohol. J Neurosci 24:9714–9722. O’Neill J, Cardenas VA, Meyerhoff DJ (2001) Effects of abstinence on the brain: quantitative magnetic resonance imaging and magnetic resonance spectroscopic imaging in chronic alcohol abuse. Alcohol Clin Exp Res 25:1673–1682. Parks MH, Dawant BM, Riddle WR, Hartmann SL, Dietrich MS, Nickel MK, Price RR, Martin PR (2002) Longitudinal brain metabolic characterization of chronic alcoholics with proton magnetic resonance spectroscopy. Alcohol Clin Exp Res 26:1368–1380.

PFEFFERBAUM ET AL.

Parsons O (1993) Impaired neuropsychological cognitive functioning in sober alcoholics, in Alcohol Induced Brain Damage: NIAAA Research Monograph No 22 (Hunt WA, Nixon SJ eds), pp 173–194. National Institutes of Health, Rockville, MD. Pentney RJ, Alletto JJ, Acara MA, Dlugos CA, Fiel RJ (1993) Small animal magnetic resonance imaging: a means of studying the development of structural pathologies in the rat brain. Alcohol Clin Exp Res 17:1301–1308. Pentney RJ, Dlugos CA (2000) Cerebellar Purkinje neurons with altered terminal dendritic segments are present in all lobules of the cerebellar vermis of ageing, ethanol-treated F344 rats. Alcohol Alcohol 35:35–43. Pentney RJ, Quackenbush LJ (1990) Dendritic hypertrophy in Purkinje neurons of old Fischer 344 rats after long-term ethanol treatment. Alcohol Clin Exp Res 14:878–886. Pfefferbaum A, Adalsteinsson E, Bell RL, Sullivan EV (2006) Selective distribution of thiamine-deficiency induced neural damage and evidence for recovery in a rat model of Wernicke’s encephalopathy: an MR imaging and spectroscopy study (abstract). Presented at 29th Annual Scientific Meeting of the Research Society on Alcoholism, Baltimore, MD, June 24–28, 2006. Pfefferbaum A, Adalsteinsson E, Sullivan EV (2004a) In vivo structural imaging of the rat brain with a 3 T clinical human scanner. J Magn Reson Imaging 20:779–785. Pfefferbaum A, Lim KO, Desmond J, Sullivan EV (1996) Thinning of the corpus callosum in older alcoholic men: a magnetic resonance imaging study. Alcohol Clin Exp Res 20:752–757. Pfefferbaum A, Lim KO, Zipursky RB, Mathalon DH, Lane B, Ha CN, Rosenbloom MJ, Sullivan EV (1992) Brain gray and white matter volume loss accelerates with aging in chronic alcoholics: a quantitative MRI study. Alcohol Clin Exp Res 16:1078–1089. Pfefferbaum A, Mathalon DH, Sullivan EV, Rawles JM, Zipursky RB, Lim KO (1994) A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood. Arch Neurol 51:874–887. Pfefferbaum A, Rosenbloom MJ, Serventi K, Sullivan EV (2002) Corpus callosum, pons and cortical white matter in alcoholic women. Alcohol Clin Exp Res 26:400–405. Pfefferbaum A, Rosenbloom MJ, Serventi KL, Sullivan EV (2004b) Brain volumes, RBC status, and hepatic function in alcoholics after 1 and 4 weeks of sobriety: predictors of outcome. Am J Psychiatry 161:1190–1196. Pfefferbaum A, Sullivan EV (2002) Microstructural but not macrostructural disruption of white matter in women with chronic alcoholism. Neuroimage 15:708–718. Pfefferbaum A, Sullivan EV (2005) Diffusion MR imaging in neuropsychiatry and aging, in Clinical MR Neuroimaging: Diffusion, Perfusion and Spectroscopy (Gillard J, Waldman A, Barker P eds), pp 558–578. Cambridge University Press, Cambridge. Pfefferbaum A, Sullivan EV, Carmelli D (2004c) Morphological changes in aging brain structures are differentially affected by time-linked environmental influences despite strong genetic stability. Neurobiol Aging 25:175–183. Pfefferbaum A, Sullivan EV, Mathalon DH, Lim KO (1997) Frontal lobe volume loss observed with magnetic resonance imaging in older chronic alcoholics. Alcohol Clin Exp Res 21:521–529. Pfefferbaum A, Sullivan EV, Mathalon DH, Shear PK, Rosenbloom MJ, Lim KO (1995) Longitudinal changes in magnetic resonance imaging brain volumes in abstinent and relapsed alcoholics. Alcohol Clin Exp Res 19:1177–1191. Pfefferbaum A, Sullivan EV, Rosenbloom MJ, Mathalon DH, Lim KO (1998) A controlled study of cortical gray matter and ventricular changes in alcoholic men over a five year interval. Arch Gen Psychiatry 55:905–912. Pfefferbaum A, Sullivan EV, Swan GE, Carmelli D (2000) Brain structure in men remains highly heritable in the seventh and eighth decades of life. Neurobiol Aging 21:63–74.

ALCOHOL EFFECTS ON RAT BRAIN

Porjesz B, Rangaswamy M, Kamarajan C, Jones KA, Padmanabhapillai A, Begleiter H (2005) The utility of neurophysiological markers in the study of alcoholism. Clin Neurophysiol 116:993–1018. Porrino LJ, Whitlow CT, Samson HH (1998) Effects of the self-administration of ethanol and ethanol/sucrose on rates of local cerebral glucose utilization in rats. Brain Res 791:18–26. Rasmussen T, Schliemann T, Sorensen JC, Zimmer J, West MJ (1996) Memory impaired aged rats: no loss of principal hippocampal and subicular neurons. Neurobiol Aging 17:143–147. Raz N, Rodrigue KM, Head D, Kennedy KM, Acker JD (2004) Differential aging of the medial temporal lobe: a study of a five-year change. Neurology 62:433–438. Rintala J, Jaatinen P, Lu W, Sarviharju M, Eriksson CJ, Laippala P, Kiianmaa K, Hervonen A (1997) Effects of lifelong ethanol consumption on cerebellar layer volumes in AA and ANA rats. Alcohol Clin Exp Res 21:311–317. Rodd ZA, Bell RL, Melendez RI, Kuc KA, Lumeng L, Li TK, Murphy JM, McBride WJ (2004) Comparison of intracranial self-administration of ethanol within the posterior ventral tegmental area between alcoholpreferring and Wistar rats. Alcohol Clin Exp Res 28:1212–1219. Rodd-Henricks ZA, Bell RL, Kuc KA, Murphy JM, McBride WJ, Lumeng L, Li TK (2001a) Effects of concurrent access to multiple ethanol concentrations and repeated deprivations on alcohol intake of alcohol-preferring (P) rats. Alcohol Clin Exp Res, 25:1140–1150. Rodd-Henricks ZA, Bell RL, Kuc KA, Murphy JM, McBride WJ, Lumeng L, Li TK (2001b) Effects of concurrent access to multiple ethanol concentrations and repeated deprivations on alcohol intake of alcohol-preferring rats. Alcohol Clin Exp Res 25:1140–1150. Savage LM, Candon PM, Hohmann HL (2000) Alcohol-induced brain pathology and behavioral dysfunction: using an animal model to examine sex differences. Alcohol Clin Exp Res 24:465–475. Schreiber R, Freund WD (2000) Glutamate transport is downregulated in the cerebral cortex of alcohol-preferring rats. Med Sci Monit 6: 649–652. Schuckit MA (1985) Genetics and the risk for alcoholism. JAMA 253:2614–2617. Shear PK, Jernigan TL, Butters N (1994) Volumetric magnetic resonance imaging quantification of longitudinal brain changes in abstinent alcoholics. Alcohol Clin Exp Res 18:172–176. Shear PK, Sullivan EV, Lane B, Pfefferbaum A (1996) Mammillary body and cerebellar shrinkage in chronic alcoholics with and without amnesia. Alcohol Clin Exp Res 20:1489–1495. Smith DG, Learn JE, McBride WJ, Lumeng L, Li TK, Murphy JM (2001) Long-term effects of alcohol drinking on cerebral glucose utilization in alcohol-preferring P rats. Pharmacol Biochem Behav 69: 543–553. Smith DG, Learn JE, McBride WJ, Lumeng L, Li TK, Murphy JM (2002) Local cerebral glucose utilization after relapse in ethanol drinking in alcohol-preferring (P) rats. Alcohol 27:115–126. Stewart RB, McBride WJ, Lumeng L, Li TK, Murphy JM (1991) Chronic alcohol consumption in alcohol-preferring P rats attenuates subsequent conditioned taste aversion produced by ethanol injections. Psychopharmacology (Berl) 105:530–534. Strother WN, McBride WJ, Lumeng L, Li TK (2005) Effects of acute administration of ethanol on cerebral glucose utilization in adult alcohol-preferring and alcohol-nonpreferring rats. Alcohol 35: 119–128.

1261

Sullivan EV, Adalsteinsson E, Sood R, Mayer D, Bell R, McBride W, Li TK, Pfefferbaum A (2006) Longitudinal brain magnetic resonance imaging study of the alcohol-preferring rat. Part I: Adult brain growth. Alcohol Clin Exp Res. DOI: 10.1111/j1530-0277.2006.00145.x. Available at http://www.blackwellpublishing.com. Sullivan EV, Deshmukh A, Desmond JE, Lim KO, Pfefferbaum A (2000a) Cerebellar volume decline in normal aging, alcoholism, and Korsakoff’s syndrome: relation to ataxia. Neuropsychology 14:341–352. Sullivan EV, Deshmukh A, Desmond JE, Mathalon DH, Rosenbloom MJ, Lim KO, Pfefferbaum A (2000b) Contribution of alcohol abuse to cerebellar volume deficits in men with schizophrenia. Arch Gen Psychiatry 57:894–902. Sullivan EV, Lane B, Rosenbloom MJ, Deshmukh A, Desmond J, Lim KO, Pfefferbaum A (1999) In vivo mammillary body volume deficits in amnesic and nonamnesic alcoholics. Alcohol Clin Exp Res 23: 1629–1536. Sullivan EV, Marsh L, Mathalon DH, Lim KO, Pfefferbaum A (1995a) Age-related decline in MRI volumes of temporal lobe gray matter but not hippocampus. Neurobiol Aging 16:591–606. Sullivan EV, Marsh L, Mathalon DH, Lim KO, Pfefferbaum A (1995b) Anterior hippocampal volume deficits in nonamnesic, aging chronic alcoholics. Alcohol Clin Exp Res 19:110–122. Sullivan EV, Pfefferbaum A, Swan GE, Carmelli D (2001) Heritability of hippocampal size in elderly twin men: equivalent influence from genes and environment. Hippocampus 11:754–762. Sullivan EV, Rosenbloom MJ, Lim KO, Pfefferbaum A (2000c) Longitudinal changes in cognition, gait, and balance in abstinent and relapsed alcoholic men: relationships to changes in brain structure. Neuropsychology 14:178–188. Sullivan EV, Rosenbloom MJ, Serventi KL, Deshmukh A, Pfefferbaum A (2003) The effects of alcohol dependence comorbidity and anti-psychotic medication on volumes of the thalamus and pons in schizophrenia. Am J Psychiatry 160:1110–1116. Ting YL, Bendel P (1992) Thin-section MR imaging of rat brain at 4.7 T. J Magn Reson Imaging 2:393–399. Walker DW, Barnes DE, Zornfetzer SF, Hunter BE, Kubanis P (1980) Neuronal loss in hippocampus induced by prolonged ethanol consumption in rats. Science 209:711–713. Weaver IC, Cervoni N, Champagne FA, D’Alessio AC, Sharma S, Seckl JR, Dymov S, Szyf M, Meaney MJ (2004) Epigenetic programming by maternal behavior. Nat Neurosci 7:847–854. West MJ, Coleman PD, Flood DG, Troncoso JC (1994) Differences in the pattern of hippocampal neuronal loss in normal ageing and Alzheimer’s disease. Lancet 344:769–772. West MJ, Gundersen HJG (1990) Unbiased stereological estimation of the number of neurons in the human hippocampus. J Comp Neurol 296:1–22. Worst TJ, Tan JC, Robertson DJ, Freeman WM, Hyytia P, Kiianmaa K, Vrana KE (2005) Transcriptome analysis of frontal cortex in alcoholpreferring and nonpreferring rats. J Neurosci Res 80:529–538. Yakovlev PI, Lecours A-R (1967) The myelogenetic cycles of regional maturation of the brain, in Regional Development of the Brain in Early Life (Minkowski A ed), pp 3–70. Blackwell Scientific Publications, Oxford. Zimatkin SM, Zimatkina TI (1996) Thiamine deficiency as predisposition to, and consequence of, increased alcohol consumption. Alcohol Alcohol 31:421–427.