Effects of Divalproex Sodium on Food Intake, Energy Expenditure, and Posture Allocation

Depakote (Abbott Laboratories, Abbott Park, IL) is an anti-convulsant medication used to
treat epilepsy [1] and mania associated with bipolar disorder [2, 3]. Depakote also is used
as a prophylaxis for migraine headache [4]. One side effect of Depakote that negatively
influences its appeal to health care professionals and consumers is weight gain. It is
unknown if changes in energy intake, energy expenditure, or a combination of both are
responsible for this side effect. The purpose of the proposed study is to: 1) test if
Depakote increases body weight by increasing food intake or decreasing energy expenditure,
possibly through changes in posture allocation, and 2) identify methods to prevent and treat
weight gain in people taking Depakote. This randomized placebo-controlled trial will measure
both sides of the energy balance equation (energy intake and expenditure), and posture
allocation (the time spent in active and sedentary behaviors, and the energy cost of these
behaviors). Measurement of both total daily energy expenditure (TEE) and posture allocation
provides a powerful tool to determine if Depakote decreases energy expenditure by increasing
the time spent in sedentary behavior. These data, along with the food intake data, provide a
test of the mechanisms responsible for weight gain associated with Depakote, and these data
can identify methods to prevent or treat this weight gain.

Divalproex sodium is a co-ordination compound consisting of sodium valproate and valproic
acid. The exact mechanism of action of Depakote is unknown, but it is believed to increase
brain concentrations of gamma aminobutyric acid (GABA). Similarly, the mechanism by which
Depakote increases body weight is not understood. The balance between energy intake and
expenditure influences body weight; therefore, Depakote likely alters energy intake, energy
expenditure, or both. Alterations in energy intake or expenditure can result from, or be
associated with, changes in biological mechanisms, including hormone and peptide levels. To
our knowledge, only one study has tested the effects of a compound containing valproic acid
on energy intake and expenditure, but this study was small (n = 8), uncontrolled, and used
methods to measure energy intake (food records and recall) that were unlikely to detect
changes in such a small sample [5]. The study proposed herein is the first study to test the
effects of Depakote on eating behavior, measured in the laboratory, and energy expenditure
and posture allocation, measured with the Intelligent Device for Energy Expenditure and
Activity (IDEEA™; MiniSun LLC, Fresno, CA).

In vitro, leptin secretion and mRNA levels in adipocytes decrease in response to valproic
acid, and it is believed that altered leptin levels might influence weight gain in people
taking compounds containing valproic acid [6]. Other researchers have reported that
postprandial insulin and proinsulin levels increase in people taking valproic acid, and BMI
is positively related to two-hour postprandial levels of insulin, proinsulin, and C-peptide
[7]. Luef and colleagues indicate that treatment with valproic acid might increase glucose
stimulated pancreatic section, which could be related to higher body weight due to two
factors that are related to pancreatic beta-cell regulation and insulin secretion. First,
valproic acid is a free-fatty acid (FFA) derivative that competes with FFAs for albumin
binding, and, second, valproic acid is a GABA agonist. Thus, valproic acid treatment might
increase glucose stimulated pancreatic secretion and contribute to weight gain. Evidence
from other laboratories suggests that valproic acid treatment in children increases insulin
levels and decreases glucose levels, which might stimulate appetite [8]. Importantly, Demir
and Aysun note that carnitine levels did not correlate with weight gain, suggesting that
valproic acid induced weight gain is not due to impairments in beta-oxidation of fatty
acids. Carnitine is involved in the transfer of fatty acids into the mitochondria for
beta-oxidation. Due to the potential role of FFA and valproic acid in influencing pancreatic
beta-cell regulation and insulin secretion, FFA will be measured in the present study at
baseline (day 0) and day 21, ½ hour before and one hour after the food intake tests at
lunch.

Due to the likelihood that hormones and peptides influence the weight gain associated with
Depakote, blood samples will be archived for later analysis, pending availability of funds.
Specifically, a portion of the archives will be used to measure gastric inhibitory
polypeptide (GIP) and oxyntomodulin. Oxyntomodulin and GIP will not be assayed immediately
because commercially available assay kits are not yet available. Oxyntomodulin has been
implicated in the control of food intake and satiety in both humans and animals [9], and GIP
modulates insulin secretion and might provide new treatments for diabetes [10].
Additionally, levels of PYY3-36, glucagon-like peptide-1 (GLP-1), leptin, and ghrelin will
be measured. PYY3-36 and GLP-1 are distal gut hormones that have been found to reduce food
intake [9, 11]. PYY3-36 decreases food intake by 30% when infused into humans [11], and
obese humans and rodents have attenuated fasting and post-prandial PYY3-36 levels that are
likely associated with their obesity [12]. GLP-1 also decreases food intake in rodents and
man, and it works synergistically with PYY3-36 [13]. Leptin reduces food intake [14] and
increases energy expenditure in humans [15], while ghrelin increases food intake and is
likely an important regulator of food intake [16].

The aforementioned hormones will be sampled ½ hour before and one hour after the start of
the food intake tests at lunch on days 0 and 21. The timing of these samples is based on the
finding that PYY3-36 and GLP-1 peak one hour after a meal [17, 18]. Therefore, we will
compare levels prior to meals and the change after the meals between the Depakote and
placebo groups. Different pre-meal levels of these hormones and peptides can be correlated
with food intake between the groups, and differential change in these hormones in response
to the meal will be evaluated between the groups and correlated with food intake.

Two factors that likely affect the amount of weight gain with Depakote treatment are gene
sequencing and gene expression. Although these analyses can be costly, DNA material,
including RNA, can be archived from whole blood samples and stored for later analysis. This
method allows cost to be controlled since analyses can be conducted on a sub-sample of the
participants who displayed specific reactions to treatment during the study. For example,
genomic studies can be conducted on participants who fall within the lower and upper
quartiles of weight gain, or a sub-sample of participants from the placebo and treatment
groups. This provides empirical tests of a priori stated hypothesis without the need to
endure the costs of running genomic analyses on the entire sample of participants. In the
present study, both gene sequencing and gene expression studies will be made possible by
archiving blood samples, as outlined in the Methods section. These analyses will depend on
the availability of funds and the development of specific testable hypotheses (e.g., weight
gain in the Depakote group will be associated with the peroxisome proliferator-activated
receptor-gamma gene).

Inclusion Criteria:

1. Healthy male or female, age 18 to 54 years

2. 20 < Body Mass Index (BMI, kg/m2) < 30

3. Willing to have a blood sample stored for possible future genetic testing

Exclusion Criteria:

1. For females, pregnant or unwilling to use an effective form of contraception while on
this study (hormonal methods like birth control pills, implants or shots; barrier
methods like condoms or diaphragms with foam; surgical sterilization; or abstinence)

2. For females, use of any other oral contraceptive other than monophasic oral
contraceptives

3. For females, irregular menstrual cycles

4. For females, history of partial hysterectomy

5. For females, nursing

6. For females, history of polycystic ovarian syndrome

7. Aspirin use or the refusal to abstain from aspirin use during the study

8. Current or history of urea cycle disorders

9. Tobacco users

10. Use of anti-convulsant medication

11. Use of barbiturates, such as Phenobarbital

12. Use of tranquilizers, such as Xanax and Valium

13. Use of blood thinners, such as Coumadin

14. Use of anti-depressant medication

15. Liver disease or impaired liver function

16. History of pancreatitis

17. Regular (4 or more days per week) consumption of 3 or more alcoholic beverages a day

18. Refusal to abstain from alcohol intake during the study

19. Dietary restraint score > 14 or disinhibition score > 12 on the Eating Inventory [19]

20. Dislike or allergy to the food used during the food intake tests

21. Unwilling to wear an IDEEA™ during the study

22. Unwilling to wear an accelerometer during the study