Serum testosterone and electroencephalography spectra in developmental male rhesus Macaca mulatta monkeys

Archives of Medical Research 35 (2004) 406–410

ORIGINAL ARTICLE

Serum Testosterone and Electroencephalography Spectra in Developmental Male Rhesus Macaca mulatta Monkeys Adria´n Poblano,a Braulio Herna´ndez-Godı´nez,b Arturo Arellano,b Carmina Arteaga,a Yolanda Elı´as,a Jose´ Morales,a Roma´n Poblanoa and Adriana Poblano-Alcala´a a

Laboratorio de Neurofisiologı´a Cognoscitiva, Instituto de la Comunicacio´n Humana, Centro Nacional de Rehabilitacio´n, Mexico City, Mexico b Centro de Investigacio´n, Proyecto CAMINA, A.C., Mexico City, Me´xico Received for publication January 21, 2004; accepted June 4, 2004 (04/021).

Background. The main objectives in this work were to determine whether the relationship between serum testosterone concentration and electroencephalography (EEG) developmental change observed in human males is also present in Macaca mulatta and, if so, to determine which frequency bands are involved and which regions change in pre-pubertal monkeys as a function of serum testosterone concentration. Methods. Nine healthy monkeys were divided into three groups according to age. Serum testosterone was measured using immunoenzymatic chemiluminescent assay. EEG results were processed using Fast Fourier transform; average relative spectral power analysis was calculated and separated into delta and theta bands. Results. The main findings were higher delta relative power in temporal area of the youngest group. Significant positive correlations were observed between serum testosterone levels and theta relative power across the entire scalp, and between theta relative power at frontal and temporal locations and in negative direction between delta relative power in temporal areas. Partial correlations controlling for cephalic perimeter remained significant between testosterone and total theta relative power and theta relative power in temporal areas. Partial correlations remained significant for theta relative power controlling for age at temporal locations. Conclusions. Our data show that testosterone may be a significant covariate in EEG development in Macaca mulatta males. 쑖 2004 IMSS. Published by Elsevier Inc. Key Words: Electroencephalography, Testosterone, Development, Macaca mulatta.

Introduction Electroencephalography (EEG) is the recording of brain electric activity in the scalp. Generators of continuous sinusoidal brain waves are currently being investigated (1). In humans, development of EEG signals begins in the newborn with mixed rhythms mainly in delta band; at 3 years of age, theta rhythms are the main frequencies of EEG, and at 7 years of age alpha band appears as the main component

Address reprint requests to: A. Poblano, M.D., Ph.D., Calzada Me´xicoXochimilco #289, Col. Arenal-Guadalupe, Deleg. Tlalpan, 14889 Me´xico, D.F., Me´xico. Telefax: (⫹52) (55) 5489-6599. E-mail: drdyslexia@ starmedia.com

0188-4409/04 $–see front matter. Copyright d o i : 10 .1 0 1 6/ j.ar c med .2 0 04 .0 6 .0 0 3

of electrical activity (2). Although the majority of researchers agree that there are EEG changes with chronologic age in puberty, scant literature exists on why this is so. We reported in a previous paper (3) that 9- to 11-year-old children showed decrease in delta relative power and increase in alpha relative power with higher salivary testosterone concentrations when compared with groups with lower concentrations, suggesting that increased testosterone in pre-pubertal children can predict the same EEG changes usually associated with increased chronologic age. The main objectives of this paper were 1) to determinate whether testosterone concentration is associated with male EEG developmental change in other primate species such as Macaca mulatta, 2) to measure which frequency bands of EEG spectra are

쑖 2004 IMSS. Published by Elsevier Inc.

Testosterone Effects in EEG in Monkeys

involved, and 3) to identify which regions change in prepubertal monkeys as a function of testosterone concentration.

Materials and Methods Animals. EEG recordings were made and testosterone measurements were taken of healthy young male rhesus monkeys (Macaca mulatta). We studied nine animals in three age and weight groups: the first group was made up of newborns to 1 year of age (0–12 months); the second group consisted of animals 2–3 years of age (13–36 months), while the third group comprised animals 4–6 years of age (37–60 months). This Cercopithecinae monkey is considered an infant from 0 to 12 months of age, a juvenile from 12 to 36 months, a sub-adult from 48 to 72 months and is sexually mature at 42–48 months of age (4). Animals were housed in special rooms in groups of 15 subjects with sufficient space and allowing social hierarchy. Environmental enrichment was performed with soft wood and plastic artifacts to promote this, rooms were cleaned once or twice daily, animals were fed twice daily with a total feeding dose of 6 kg of Harlan Teklad pellets (20% proteins) and water ad libitum, with light-dark cycles of 12 h. Monkeys were periodically screened for veterinary care to assure their health status. Animals were pre-anesthetized for electrophysiologic studies with ketamine 10 mg/kg body weight and anesthetized with a light dose of tyletamine-zolacepam at 3 mg/kg of body weight in younger animals and 6 mg/kg in juvenile and sub-adults (5). This protocol was approved by the ethics committees of the participating institutions, and animals were treated gently according to the principles of the rights of primates (6). Testosterone measurement. Testosterone was measured in serum. Samples were collected in clean assay tubes after taking EEG recordings between 9:00 a.m. and 12:00 p.m. to control for diurnal variation. A sample volume of 2 mL of blood was taken. Testosterone concentration was estimated using immunoenzymatic chemiluminescent assay (Immulite DPC, Los Angeles CA, USA). A solid-phase total testosterone, ligand-labeled, competitive chemiluminescent enzyme immunoassay coated with a polyclonal rabbit testosteronespecific antibody was used. The sample and ligand-labeled testosterone were incubated for 30 min at 37⬚C with intermittent agitation; unbound material was removed by centrifugation. Alkaline phosphatase-labeled anti-ligand was added and incubated for an additional 30-min cycle. The unbound enzyme conjugate was removed by centrifugal wash. Substrate was then added and incubated for a further 10 min. The chemoluminescent broken substrate resulted in sustained emission of light that was measured by luminometer, yielding the concentration of testosterone in the sample. Accuracy of equipment measurement was assured by weekly calibration of the apparatus (7).

407

Electroencephalographic recordings. EEG was recorded on a digital Vector electroencephalograph (Bioscience, Buenos Aires, Argentina) in a semi-dark, silent room. Gold cup skin electrodes filled with electrode cream were applied according to the 10–20 system (8) modified for monkeys. Sites were cleaned of dry skin and oil using a cotton swab and alcohol. All impedances were ⬍10 Kilo-ohms. EEG was recorded from Fp1, F3, C3, P3, O1, Fp2, F4, C4, P4, O2, F7, T3, T5, F8, T4, T6, Fz, Cz, Pz and Oz sites. Band pass filters were set between 0.1 and 35 cycles/sec. Sampling frequency was 256 Hz. Sixty four seconds of EEG were recorded in epochs of 8 sec as recommended by IFCN guidelines (9). Artifacts of muscle, eye, and body movement and respiratory and electrocardiogram were eliminated. EEG was then processed with Fast Fourier transform in the amount 30 sec after artifact rejection averages of relative power spectral analyses were calculated and presented as cartographic maps of brain electric activity. Mean values were calculated from data divided into delta, theta, alpha, and beta bands by averaging epochs and calculating relative power data in each band for each electrode. Data analyses. Mean and standard deviations (SDs) from continuous variables were calculated. We calculated mean of relative power of the bands, including delta (0.1–4 Hz), theta (4.1–8 Hz), alpha (8.1–12 Hz), and beta (12.1–35 Hz) across the entire scalp and at the following four locations: frontal (average of F3, F4, C3, C4, F7, and F8); posterior (average of P3, P4, O1, and O2); temporal (average of T3, T4, T5, and T6), and central (average of Fz, Cz, Pz, and Oz). We used one-way analysis of variance (ANOVA) to determinate differences in overall characteristics, testosterone, and EEG bands among the three age groups; post hoc analyses were done by Tukey test (10). Afterward, we used Pearson correlation coefficient to measure the association between testosterone serum levels and EEG relative power only in delta and theta bands to avoid inflation due to multiple testing and because these rhythms have highest power in monkeys. Significant partial correlations were controlled for age and cephalic perimeter (11); p value for statistical significance was p ⬍0.05.

Results Overall data. Average age in each of the three groups was 0.66 ⫾ 0.57, 2.33 ⫾ 0.57, and 5 ⫾ 1 years, respectively. Weight, height, cephalic perimeter, and testosterone serum levels are shown in Table 1; as expected, these showed significance among group differences in ANOVA comparisons. Correlation coefficients were significant for each of these measurements and testosterone serum levels (p ⬍0.05). EEG relative power spectra. Relative power spectra for delta, theta, and alpha bands for the three groups are shown

408

Poblano et al. / Archives of Medical Research 35 (2004) 406–410

Table 1. Overall characteristics of the groups Group (years of age) 0–1 2–3 4–6

Table 3. Delta relative power by region

Weight (g)

Height (cm)

Cephalic perimeter (cm)

Testosterone (ng/dL)

Group (years of age)

191.2 ⫾ 12.5 303.8 ⫾ 73.4 671.6 ⫾ 72.8

50.1 ⫾ 14.1 64.3 ⫾ 33 85.8 ⫾ 7.1

21.9 ⫾ 2 24.3 ⫾ 2 29.3 ⫾ 1.2

0.31 ⫾ 0.31 0.10 ⫾ 0.01 16.33 ⫾ 2.08

0–1 2–3 4–6

Frontal

Temporal

83.08 ⫾ 4.47 80.41 ⫾ 6.08 82.25 ⫾ 8.17 76.58 ⫾ 10.75 64.33 ⫾ 16.44 59 ⫾ 6.51

Posterior

Central

77.83 ⫾ 10.21 78 ⫾ 10.14 65.5 ⫾ 13.53

79.91 ⫾ 7.45 77.66 ⫾ 10.12 65 ⫾ 17.97

Each cell presents mean values ⫾ standard deviation (SD); differences in comparison between younger and older groups in all cases were significant.

Each cell presents mean ⫾ SD.

in Table 2. Beta rhythms are not present in the table because values are zero or ca. zero. Overall, the younger group showed higher delta and lower theta and alpha power than older groups. We observed no significant difference between groups in delta and theta relative power across the entire scalp in ANOVA test; nonetheless, when comparisons were performed by region a significant difference in delta relative power in temporal location was found (F ⫽ 6.01, p ⫽ 0.037). Tukey test showed that younger groups demonstrated higher delta relative power than the older group (see Tables 3 and 4). Difference in theta relative power at temporal location was nearly significant (F ⫽ 3.49, p ⫽ 0.09), and the older group showed higher theta relative power than the younger group.

Macaca mulatta monkeys from newborns to 6 years of age— in other words, from very young to nearly sexually mature subjects. Decreased delta relative power and increased theta relative power agree with the literature describing EEG spectral power changes with advancing development. To our knowledge, this is the first study to report testosterone as a covariate in EEG spectral characteristics in non-human primate species. Onset of puberty is marked by substantial hormonal changes that not only precede and cause development of secondary sexual characteristics but that also are associated with and the likely cause of changes in brain organization (12). Age at onset of puberty varies with the species, but changes in production of sexual hormones often precede development of obvious secondary sexual characteristics. Effects of hormones in sexual development comprise a complex and well-described event. Appearance of increased testosterone marks onset of accelerated physical growth and development of secondary sexual characteristics in pubertal and adolescent male children (13). In Macaca mulatta model, if increase in testosterone over relatively low levels observed during the majority of pre-pubertal childhood is a bio-marker for onset of a period of accelerated growth and development, we should be able to observe an association between testosterone level and spectral power and topography of EEG, as shown in this study. Experimental studies provide evidence that steroids affect neuronal size, survival and outgrowth, synapse number and organization, dendritic branching, nuclear volume, cortical thickness, and neurotransmitter systems of the brain (12). Androgens enter cells of target tissues and bind to specific receptors in cytoplasm, where their effects include increase of RNA and protein synthesis (14). It is not well known

Correlations between serum testosterone levels and EEG power spectra. We found significant positive correlation between total theta relative power across the entire scalp and testosterone levels (r ⫽ 0.74, p ⫽ 0.02). Negative significant correlation was observed between testosterone serum levels and delta relative power at temporal location (r ⫽ ⫺0.82, p ⫽ 0.006) and positive in theta relative power at frontal location (r ⫽ 0.76, p ⫽ 0.01) and at temporal site (r ⫽ 0.83, p ⫽ 0.006) (see Figure 1). Partial correlations controlled for cephalic perimeter remained significant between testosterone and total theta relative power (r ⫽ 0.68, p ⫽ 0.05) and theta relative power at temporal location (r ⫽ 0.75, p ⫽ 0.02). For age, partial correlations remained significant for theta relative power at temporal location (r ⫽ 0.75, p ⫽ 0.02). Discussion The main finding in the study was that presence of increased testosterone was associated with EEG change in male rhesus

Table 4. Theta relative power by region

Table 2. Relative power of EEG band by group Group (years of age) 0–1 2–3 4–6

Delta

Theta

Alpha

80.05 ⫾ 7.45 79.21 ⫾ 7.98 64.78 ⫾ 12.15

12.73 ⫾ 5.37 10.56 ⫾ 3.24 19.53 ⫾ 5.58

3.73 ⫾ 5.37 10.56 ⫾ 3.24 19.53 ⫾ 5.58

Each cell presents mean ⫾ SD.

Group (years of age) 0–1 2–3 4–6

Frontal

Temporal

Posterior

Central

11.0 ⫾ 3.03 10.16 ⫾ 3.95 19.66 ⫾ 6.84

11.66 ⫾ 3.44 12.08 ⫾ 4.88 20.83 ⫾ 5.76

13.58 ⫾ 8.32 11.5 ⫾ 3.96 19.83 ⫾ 10.56

13.41 ⫾ 5.5 9.25 ⫾ 2 18.66 ⫾ 8.2

Each cell presents mean ⫾ SD.

Testosterone Effects in EEG in Monkeys

Figure 1. Typical scatterplot between distribution of delta relative power (%) at temporal location and serum testosterone (nanograms/mL). Significant negative correlation is evident, and as testosterone serum levels decrease delta relative power increases.

whether the effect of testosterone is to alter 1) electrical activity of neurons, producing changes in synthesis secretion, or 2) uptake of neurotransmitters or changes in their connections (15). Changes that occur as a result of high or low levels of sex hormones present early in development are considered organizational; they are associated with permanent changes in the wiring of the brain. Behavior may be affected directly through changes to brain regions involved in behavior; for example, gonadal hormones may affect hypothalamic areas involved in sexual behavior and hippocampal areas related with aspects of spatial learning (16). Effects of hormones in adult life are different. Hormones do not produce long-lasting permanent changes, but instead have activational effects, that is, they activate neural systems organized earlier in life (12). The majority of our subjects of the younger group showed the lower levels found in monkeys ⬍2 years of age. While older subjects had testosterone concentration substantially above this range, our monkeys with higher salivary testosterone levels showed signs of sexual development. In this study, testosterone effects on EEG appear to be activational. Increasing testosterone levels were significantly associated with decreasing relative delta power and increasing effect in relative theta power. Nevertheless, they were not associated with changes in alpha and beta power; in other words, testosterone accelerates neurophysiologic maturation. In humans, we see decrease of relative delta power and increase of alpha; this difference between species may be attributed to the monkey’s lack of alpha rhythm development. Fast rhythms (beta frequency) have lower relative

409

power in monkeys because this occurs mainly in frontal area, which in monkeys is a restricted cortical area. Another reason is related to the fact that beta rhythm is characteristic of attentive immobility (17) and our animals were anesthetized; thus, for both reasons we were unable to perform any comparisons with this frequency. In adult humans, androgen administration has many effects, including the following: acute administration of testosterone slows alpha rhythm in waking-eye closed state (18), but dehydroepiandrosterone increases rapid-eye-movement sleep and sigma rhythm during rapid-eye-movement sleep (19). Long-term administration results in EEG resistance to photic stimulation (20). In children, 9- to 11-year-olds showed decrease in delta relative power and increase in alpha relative power with higher salivary testosterone concentrations when compared with lowest concentration groups, as previously mentioned (3) and are in agreement with results from this study. Although no testosterone reactivity on cerebral cortex by means of immunohistochemical studies has been observed, its effects upon EEG may be indirect due to its action in profound cerebral structures; nonetheless, because previously mentioned studies have different designs or were performed on different species no comparisons can be made with our results. It is unknown whether similar effects may be mediated by estrogen and progesterone in female monkeys. The relationship between maturation of EEG and sexual hormones should be studied longitudinally in both male and female subjects in several primate species. The main limitation of our study was low number of subjects studied; however, the effect of testosterone was sufficiently robust to detect its interactions with EEG bands. Small group size and restricted age range in this study could have prevented detection of the expected change in EEG with age, notwithstanding that it is clear that for delta and theta bands at least, testosterone may play a key role at these ages. In summary, our data show that testosterone may be a significant covariate in EEG development in Macaca mulatta males. This merits additional attention in future work related with ontogeny of EEG rhythms during this period of life as well as in other primates.

Acknowledgments The authors wish to thank Emma Alquicira, B.Sc., for her generous help in testosterone determinations, Juan Poblano, D.Ed. and Jovita Luna, D.Ed. for partial support of the study, and Robert Burns, Ph.D., Raquel Bialik, Ph.D., and Maggie Brunner, M.A. for help in the preparation of the manuscript in English.

References 1. Steriade M. Cellular substrates of brain rhythms. In: Niedermeyer E, Lopes da Silva F, editors. Electroencephalography. Basic principles, clinical applications and related fields, Philadelphia, PA, USA: Lippincott, Williams and Wilkins;1999. pp. 28–75.

410

Poblano et al. / Archives of Medical Research 35 (2004) 406–410

2. Niedermeyer E. Maturation of the EEG. Development of waking and sleep patterns. In: Niedermeyer E, Lopes da Silva F, editors. Electroencephalography. Basic principles, clinical applications and related fields, Philadelphia, PA, USA: Lippincott, Williams and Wilkins;1999. pp. 189–214. 3. Poblano A, Rothenberg SJ, Fonseca ME, Cruz ML, Flores T, Zarco I. Salivary testosterone and EEG spectra of 9- to 11-year-old male children. Dev Neuropsychol 2003;23:375–384. 4. King F. Studies on primate behavior. In: Martı´nez-Contreras J, editor. Present issues (in Spanish), Mexico: Universidad Metropolitana-Iztapalapa;1988. pp. 49–59. 5. Kohn DF. Anesthesia and analgesia in laboratory animals. New York: Academic Press;1997. pp. 233–245. 6. King F. Biomedical research in primates. In: Martı´nez-Contreras J, editor. Present issues in primatology (in Spanish), Me´xico, D.F.: Universidad Metropolitana, Campus Iztapalapa;1988. pp. 15–21. 7. Navarro MA, Juan L, Bonnin MR. Salivary testosterone: relationship to total and free testosterone in serum. Clin Chem 1986;32:231–232. 8. Jaspers H. The ten-twenty electrodes system of the international federation. Electroencephalogr Clin Neurophysiol 1958;10:371–375. 9. Nuwer M, Lehman D, Lopes da Silva F, Matsuoka S, Sutherling W, Vibert JF. IFCN guidelines for topographic and frequency analysis of EEG and EPs. Report of an IFCN Committee. Electroencephalogr Clin Neurophysiol 1994;91:1–5. 10. Dawson-Saunders B, Trapp RG. Medical biostatistics (in Spanish). Me´xico, D.F.: Manual Moderno;1997. 11. Hair JF, Anderson RE, Tatham RL, Black WC. Multivariate data analysis (in Spanish). Madrid, Spain: Prentice Hall Iberia;1999.

12. Berembaum SA. How hormones affect behavior and neural development. Dev Neuropsychol 1998;14:175–196. 13. Grumbach MM, Styne DM. Puberty: ontogeny. Neuroendocrinology, physiology, and its disorders. In: Wilson TD, Forster DW, editors. Williams textbook of endocrinology, Philadelphia, PA, USA: Saunders;1994. pp. 1139–1221. 14. McDonnell DP, Clevenger B, Dana S, Santiso-Mere D, Tzukerman MT, Gleeson MAG. The mechanism of action of steroid hormones: a new twist to an old tale. J Clin Pharmacol 1993;33:1165–1172. 15. Robel P, Baulieu EE. Neurosteroids. Biosynthesis and function. TINS 1994;5:1–8. 16. Michael RP, Zumpe D. Developmental changes in behavior and in steroid uptake by the male and female macaque brain. Dev Neuropsychol 1998;14:233–260. 17. Lopes da Silva F. Dynamics of EEGs as signal of neuronal populations: models and theoretical considerations. In: Niedermeyer E, Lopes da Silva F, editors. Electroencephalography. Basic principles, clinical applications and related fields. Philadelphia, PA, USA: Lippincott, Williams and Wilkins;1999. pp. 76–92. 18. Vogel W, Broverman DM, Klaiber EL, Abraham G, Cone FL. Effects of testosterone infusions upon EEGs of normal male adults. Electroencephalogr Clin Neurophysiol 1971;31:400–403. 19. Friess E, Trachsel L, Guldner J, Schier T, Steiger A, Holsboer F. DHEA administration increase rapid eye movement sleep and EEG power in the sigma frequency range. Am J Physiol 1995;31:E107–E113. 20. Stenn PG, Klaiber EL, Vogel W, Broverman DM. Testosterone effects upon photic stimulation of the electroencephalogram (EEG) and mental performance of humans. Percept Mot Skill 1972;34:371–378.