Psychoneuroendocrinology 24 (1999) 363–370

A determinant factor in the efficacy of GHRHadministration in promoting sleep: high peak

concentration versus recurrent increasing slopes

Lisa Marshall a,*, Inge Derad a,b, Christian J. Strasburger c,Horst L. Fehm a,b, Jan Born a

a Clinical Neuroendocrinology, Medical Uni6ersity of Lubeck, H23a, Ratzeburger Allee 160,23538 Lubeck, Germany

b Department of Internal Medicine, Medical Uni6ersity of Lubeck, Ratzeburger Allee 160,23538 Lubeck, Germany

c Medical Clinic, Klinikum Innenstadt, Ludwig-Maximilians-Uni6ersity, Ziemssenstr. 1,80336 Munich, Germany

Received 24 August 1998; accepted 23 October 1998

Abstract

A previous experiment indicated a greater efficacy of episodic than continuous growthhormone (GH)-releasing hormone (GHRH) administration in enhancing sleep. The greaterefficacy of episodic administration could principally result from two factors, i.e. the greaterpeak concentration reached after episodic administration or the recurrence of increasingslopes in GHRH concentration. In order to investigate which factor essentially determinesthe pharmacodynamics of sleep promotion after GHRH, effects after a transient high peakin GHRH concentration were compared with those of repetitive increases in GHRHconcentration. Sleep, plasma concentrations of GH, and GHRH were examined in healthysubjects after evening administration of a ‘single’ IV bolus of 50 mg GHRH, after five‘repetitive’ boluses of 10 mg GHRH, and after placebo. Compared with placebo, singleGHRH significantly increased time spent in stage 4 sleep (pB .01) and in stage 2 sleep,reduced time spent in wakefulness and onset latency of stage 4 sleep (pB .05, for each), whilerepetitive GHRH remained without effects. GH secretory activity also tended to be higherafter single than repetitive GHRH. Thus, results suggest the relevance of a transiently highconcentration of GHRH in blood as an essential factor in enhancing the central nervoussleep process. © 1999 Elsevier Science Ltd. All rights reserved.

* Corresponding author. Tel.: +49-451-500-3644; fax: +49-451-500-3640; e-mail: marshall@kfg.muluebeck.de.

0306-4530/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.PII: S0306 -4530 (98 )00085 -7

364 L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

Keywords: Central nervous system; Growth hormone; Growth hormone-releasing hormone; Human;Non-rapid-eye-movement sleep; Slow-wave sleep

1. Introduction

The present experiment was conducted as a supplement to a previous finding inwhich a greater efficacy of growth hormone-releasing hormone (GHRH) was shownafter episodic administration as compared to a continuous GHRH infusion on thecentral nervous process of sleep (Marshall et al., 1996). At least two factorscharacterizing the episodic GHRH administration may have contributed to itsgreater efficacy. First, a much higher concentration of GHRH in blood wastransiently induced than during any time period of the infusion. Second, with theepisodic administration increasing slopes in GHRH concentration were repeatedlyinduced, in contrast to the presumably steady plasma GHRH level produced by theinfusion.

In order to compare effects on sleep of these two factors (high concentrationversus repetitive increases) an equal dose of GHRH was given IV, either as a singlehigh-dose bolus injection or as multiple low-dose repetitive injections. Both admin-istrations were to occur within a comparable time range. Dosages and time ofadministration were selected based on previous data regarding the pharmacody-namics of the GHRH induced promotion of sleep as well as on considerations ofthe pharmacokinetics of GHRH (von Werder et al., 1985; Steiger et al., 1992;Kerkhofs et al., 1993). Taken together, those data indicate that a minimum dose of50 mg GHRH given prior to sleep is required for distinct effects on sleep to occur.Based on a half-life for GHRH in blood of 6.8 min, and an ability to quicklypenetrate the blood-brain barrier (Banks and Kastin, 1985; Frohman et al., 1986),for the repetitive administration five doses of 10 mg spaced by an interval of 15 minwere considered appropriate, and should avoid gradual accumulation of the sub-stance in blood. It was hypothesized, that: (i) if recurrent increasing slopes inGHRH concentration in blood contributed most to the promotion of sleep, therepetitive administration would enhance sleep more than the single dose; and (ii) ifthe peak concentration of GHRH in blood was most relevant, the single boluswould have a greater sleep enhancing effect.

2. Methods

A total of 16 healthy, normal weight men (aged 22–36 years) with no history ofsleep disturbances participated in the study after having given written informedconsent. Data of two subjects were excluded for technical reasons. The experimen-tal protocol was approved by the Ethics Committee of the Medical University ofLubeck.

365L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

Subject constraints, experimental procedure and analysis were as described in aprevious study (Marshall et al., 1996). After awakening, subjects were given thesleep questionnaire-A (CIPS, 1981). Experiments were designed according to adouble-blind within-subject cross-over comparison, with order of treatments bal-anced across subjects. In condition ‘repetitive GHRH’ subjects were injected IVwith boluses of 10 mg GHRH (GHRH1–44, Ferring GmbH, Kiel, Germany) at21:45, 22:00, 22:15, 22:30 and 22:45 h. In condition ‘single GHRH’, subjectsreceived a single 50 mg GHRH bolus at 22:00 h, and were injected with salinesolution at 21:45, 22:15, 22:30 and 22:45 h. On the placebo night, saline solution(0.9% NaCl) was injected at all five times. Boluses containing 10 and 50 mg GHRH(dissolved in 1 and 5 ml saline solution, respectively) were injected within 5 and25 s, respectively.

Blood samples for analyzing plasma GH concentration were drawn at 15 minintervals. Plasma concentrations were measured by commercial radioimmunoassay(GH; DPC Biermann, Bad Nauheim, Germany). Assay sensitivity, intraassay andinterassay coefficients of variation for the relevant range of concentration were,respectively, 0.9 mg/l, and below 4 and 9%. In a subgroup of subjects (n=5) bloodsamples for analyzing plasma GHRH concentration were drawn into syringespretreated with EDTA and trasylol, and stored at −40°C until assay. GHRHplasma concentrations were measured by competitive radioimmunoassay (Schopohlet al., 1991); lower limit of detection was 20 pg/ml, intra- and interassay coefficientsof variation were below 11 and 15%, respectively). Blood sampling times arerevealed in Fig. 1(B). Analysis of GHRH plasma concentration were conducted toconfirm the time courses of GHRH levels that were predicted from the pharmacoki-netics of the peptide as documented in foregoing studies (von Werder et al., 1985).

Polysomnographic recordings were scored according to the criteria ofRechtschaffen and Kales (1968). Times spent in different sleep stages were assessedfor the entire night as well as separately for each half of the night. Also, thecumulative amount of time spent in a sleep stage was calculated time-locked tolights off at 23:00 h.

Statistical analysis for sleep parameters and GH relied on analysis of variance(ANOVA) including a repeated measures factor representing the three experimentalconditions. To specify significant treatment effects pairwise contrasts were per-formed. A p-value B .05 was considered significant.

3. Results

Analysis of the first half of sleep time revealed distinct signs of improved sleepafter the single GHRH injection: Time spent in stage 4 sleep after single GHRHwas increased by 30% as compared with placebo (pB .01) and by 25% as comparedwith repetitive GHRH (pB .10; F [2,26]=3.73, pB .05, for main effect of treat-ment). After single GHRH subjects, after falling asleep, reached stage 4 sleepearlier, (Table 1). Cumulating the time spent in a certain sleep stage across the nightindicated that by the end of the third hour after lights off, the amount of time spent

366 L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

in stage 4 sleep after single GHRH became significantly greater than after placebo(23:00–02:00 h: 22.894.9 vs. 18.293.9, F [1,13]=5.74, pB .05). By the end of thefourth hour after lights off, cumulative time in stage 4 sleep after single GHRHtended to be enhanced as compared to repetitive GHRH (23:00–03:00 h: 27.195.1

Fig. 1. (A) Mean (9SEM) GH plasma concentrations for single GHRH (thick solid line), repetitiveGHRH (dotted line) and placebo (thin solid line). Solid and dotted arrows indicate times of single andrepetitive GHRH bolus injections, respectively. N=14. (B) Mean plasma GHRH concentration(9SEM) after single GHRH (�) and repetitive GHRH (�) delivered in a subgroup of five men. Note,abscissa reveals time relative to the single or the first repetitive GHRH injection, respectively, whereby0 min corresponds to 3 min after GHRH administration. Areas under the curve were comparable for thetwo conditions, i.e. single bolus (204918 ng×min/ml) and repetitive boluses (199957 ng×min/ml).Peak concentrations after the single bolus in each subject were distinctly higher than peaks afterrepetitive GHRH boluses. On average GHRH concentrations after the single GHRH bolus peaked at16.593.1 ng/ml, while peaks after repetitive boluses averaged 5.292.0 ng/ml (pB .05, Wilcoxon pairedt-test, one-sided).

367L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

Table 1Mean (9SEM) of sleep onset latency (with reference to lights off at 23:00 h), total sleep time, andpercentage of time spent in wakefulness (W) and in sleep stages 1 (S1), 2 (S2), 3 (S3), 4 (S4), and inREM sleep (REM) for the three conditions: placebo (P), single GHRH (S), and repetitive GHRH (R)relative to sleep onseta

SignificanceParameter Placebo (P) Single (S) Repetitive (R)

17.2 (5.2) NSSleep onset (min) 14.6 (2.5)17.3 (6.8)459.5 (4.2) 456.8 (5.4)Sleep time (min) NS456.8 (7.3)

% (Total)NS6.3 (1.7)W 3.7 (0.9)6.8 (1.2)

8.9 (0.8)S1 9.3 (1.3) 8.6 (1.1) NSNS46.4 (1.9)S2 47.0 (2.3)42.9 (2.0)

13.6 (1.3) 13.4 (1.5) NSS3 14.8 (1.8)P-S†5.6 (1.1)S4 7.0 (1.4)5.2 (1.0)

19.5 (1.5)REM 20.9 (1.3) NS20.2 (1.1)

% (1st half )S-R†W 4.6 (1.2) 1.4 (0.5) 6.1 (2.2) P-S*

7.6 (1.2)S1 7.2 (1.2) 7.4 (1.3) NS48.0 (3.0) NSS2 43.5 (2.4)44.2 (3.1)

NSS3 21.8 (3.1) 21.6 (2.1) 18.2 (2.0)S-R†S4 8.9 (1.7) 11.7 (2.2) 9.3 (2.2) P-S**

10.8 (1.2) NSREM 13.3 (1.9) S-R*14.4 (1.5)

% (2nd half )5.9 (1.6) 6.4 (1.7)W NS8.9 (2.1)

NS10.0 (2.3)S1 9.7 (1.6)11.4 (1.7)44.8 (2.3) P-S*S2 S-R†41.7 (2.7) 50.4 (3.0)

S-R*NSS3 8.6 (1.3)7.8 (1.0) 5.6 (1.3)NSS4 1.5 (0.9) 2.3 (1.1) 2.0 (0.6)NS26.1 (1.9)REM 28.2 (2.3)28.5 (2.1)

Latency (min)3.5 (0.6)S2 3.4 (0.5) 4.1 (0.6) NS20.6 (3.5) NSS3 18.8 (5.4)18.6 (4.4)48.0 (12.9) P-S*S4b 50.0 (17.4) S-R*29.3 (5.4)85.0 (5.2)REM P-S†116.4 (14.9) P-R†91.8 (7.3)

a Percentages are provided for the entire night as well as separately for the first and second half ofsleep time. Bottom rows indicate latencies with reference to sleep onset of S2, S3, S4 and REM sleep.Right column indicates significant differences between any two of the conditions. N=14, NS, nonsignificant.

b Since normal distribution could not be confirmed, pairwise tests were conducted with Wilcoxon(two-sided).

** pB.01.* pB.05.† pB.10.

vs. 21.295.0, F [1,13]=3.87, pB .08). Effects of repetitive GHRH as compared toplacebo remained without any statistical significance at all times (p\ .2). Comple-mentary to the effects on stage 4 sleep time spent awake during the first half ofsleep was reduced by 70% after single GHRH as compared with placebo (pB .05)

368 L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

and by 77% as compared with effects of repetitive GHRH (pB .07; F [2,26]=3.24,pB .08, for the main effect of treatment). Cumulative amount of time spent inwakefulness became lower after single GHRH than placebo and repetitive GHRHwith effects approaching significance by the end of the fourth hour after lights off.

The main effect on sleep in the second half of the night relative to sleep onset wasthe enhancement of time spent in stage 2 sleep after single GHRH as comparedwith placebo (pB .05) as well as with repetitive GHRH (pB .10, F [2,26]=4.14,pB .05, for the main effect of treatment).

Subjective ratings of the recuperative value of sleep were significantly higher aftersingle GHRH (2.890.2) than after repetitive GHRH (3.190.2), but did not differfrom placebo (2.790.1, pB .05 for pairwise comparisons; F [2,26]=6.37, pB .05for the main effect of treatment). The rating for subjective sleep quality was higherfor single GHRH than for placebo (3.190.4 vs. 4.390.7, pB .05).

Peak and mean plasma GH concentrations were higher after both schedules ofGHRH administration as compared with placebo (F [2,26]=33.62, pB .001 and(F [2,26]=5.94, pB .01, respectively). After single as compared to repetitiveGHRH, mean plasma GH concentrations tended to be higher in the first half ofsleep time, as well as from the time of the first GHRH bolus (21:45 h) until 03:00h (8.191.9 vs. 5.391.3 ng/ml, F [1,13]=3.96, pB .07, Fig. 1(A)). Individuallydetermined peak concentrations were significantly higher after single GHRH(25.993.0 ng/ml vs. 19.792.9 ng/ml, pB .01, F [2,26]=33.62, pB .001, for maineffect of treatment). Peak GH latencies did not differ between the two GHRHadministration schedules, but both latencies were shorter in comparison withplacebo (F [2,26]=4.27, pB .05). Pharmacokinetics of GHRH after both adminis-tration schedules are indicated in Fig. 1(B), estimations of peak concentration andarea under the curve are given in the legend.

4. Discussion

A single bolus of 50 mg GHRH administered IV 1 h before lights off was effectivein enhancing stage 4 sleep, shortening onset latency of this sleep stage andsuppressing wakefulness in the first part of the night. Furthermore, the single IVbolus of GHRH increased stage 2 sleep in the second half of the night and alsoimproved signs of subjective sleep quality. A repetitive schedule of 10 mg GHRHadministered five times, once every 15 min did not exert these effects on sleep. Also,mean plasma GH concentrations increased to a somewhat greater extent after singlethan repetitive GHRH. Thus, although doses of GHRH administered were identi-cal in both conditions, effects on sleep and GH release were not. A transiently veryhigh peak concentration proved more effective than the induction of repetitivelyincreasing slopes in GHRH concentration over time.

Results confirm and extend previous reports indicating increased slow-wave sleepand stage 2 sleep, but reduced episodes of shallow sleep or awakenings after GHRHadministration in humans as well as in animals (Ehlers et al., 1986; Steiger et al.,1992; Kerkhofs et al., 1993; Marshall et al., 1996; Obal et al., 1996). The present

369L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

results indicate that one factor contributing to the greater efficacy of GHRH inpromoting sleep after episodic than continuous administration as demonstrated ina foregoing study (Marshall et al., 1996), resulted from the higher plasma concen-trations of GHRH attained with the episodic mode of administration.

The overall moderate size of the sleep promoting effects even with the singlebolus of GHRH in the present study can be explained by the comparatively lowtotal dose of the peptide administered (Steiger et al., 1992). This lower dose waschosen to enable a comparison, within a limited time of 1 h before sleep initiation,between the effects of a short heightened level of GHRH concentration in bloodand those of recurrent, but distinctly lower increases in GHRH concentration.Determination of plasma GHRH concentrations confirmed a higher (on average3-fold) GHRH peak concentration in systemic blood after the single bolus thanafter repetitive administration, with comparable areas under the curve. It may besupposed that the high plasma concentration of GHRH after the single bolus ismore effective for the entrance of GHRH into the brain, e.g. by surpassing theblood-brain barrier (Begley, 1994), where it might act via receptors in the preopticanterior hypothalamic area (Takahashi et al., 1995).

This conclusion, however, is challenged by the finding that also GH responseswere distinctly higher after single as compared with repetitive GHRH administra-tion. Thus, suggesting that the transiently high GHRH peak concentration wasbiologically more potent than the increasing slopes in the peptide concentration,both on the central nervous sleep process, as well as at the peripheral, pituitarylevel. However, it cannot be excluded that a central nervous action on hypothala-mic receptors contributed to the relatively higher GH secretory response to thesingle GHRH bolus (Lanzi and Tannenbaum, 1992). Growth hormone per se hasnot been shown to enhance slow-wave sleep, and thus probably did not contributedirectly to the differential effects of GHRH administration schedules on sleep (Kernet al., 1993; Obal et al., 1996). Conversely, it is rather unlikely that differentialeffects of GHRH administration schedule on GH concentrations were a conse-quence of the more efficacious sleep enhancement after single GHRH, since onaverage differential effects on GH peak concentration occurred already before anychanges in sleep parameters became significant.

In conclusion, with the same total dose of GHRH administration, induction of asingle transiently high peak concentration in GHRH proved more effective inpromoting (stage 4) sleep than repetitively increasing slopes in GHRH concentra-tions (attaining lower peak levels). High peak concentrations may facilitate penetra-tion through the blood-brain barrier thereby increasing the potency of the peptideto influence central nervous sleep mechanisms. Nevertheless, an action mediated viaa more potent activation of receptors located peripherally cannot be excluded.

Acknowledgements

We thank Nicole Boes, Thorsten Spielvogel, Ann-Kristin Jurs and SteffanieBaxmann for acquiring data and technical assistance. This study was supported bya grant from the DFG.

370 L. Marshall et al. / Psychoneuroendocrinology 24 (1999) 363–370

References

Banks, W.A., Kastin, A.J., 1985. Peptides and the blood-brain barrier: lipophilicity as a predictor ofpermeability. Brain Res. Bull. 15, 287–292.

Begley, D.J., 1994. Peptides and the blood-brain barrier: the status of our understanding. Ann. NewYork Acad. Sci. 739, 89–100.

Collegium Internationale Psychiatriae Scalarum, 1981. Internationale Skalen fur Psychiatrie (Interna-tional Rating Scales for Psychiatry) Weinheim, Beltz Test GmbH.

Ehlers, C.L., Reed, T.K., Henriksen, S.J., 1986. Effects of corticotropin-releasing factor and growthhormone-releasing factor on sleep and activity in rats. Neuroendocrinology 42, 467–474.

Frohman, L.A., Downs, T.R., Williams, T.C., Heimer, E.P., Pan, Y-C.E., Felix, A.M., 1986. Rapidenzymatic degradation of growth hormone-releasing hormone by plasma in vitro and in vivo to abiologically inactive product cleaved at the NH2 terminus. J. Clin. Invest. 78, 906–913.

Kerkhofs, M., Van Cauter, E., Van Onderbergen, A., Caufriez, A., Thorner, M.O., Copinschi, G., 1993.Sleep-promoting effects of growth hormone-releasing hormone in normal men. Am. J. Physiol. 264,E594–E598.

Kern, W., Halder, R., Al-Reda, S., Spath-Schwalbe, E., Fehm, H.L., Born, J., 1993. Systemicgrowth-hormone does not affect human sleep. J. Clin. Endocrinol. Metab. 76, 1428–1432.

Lanzi, R., Tannenbaum, G.S., 1992. Time-dependent reduction and potentiation of growth hormone(GH) responsiveness to GH-releasing factor induced by exogenous GH: role for somatostatin.Endocrinology 130, 1822–1828.

Marshall, L., Molle, M., Boschen, G., Steiger, A., Fehm, H.L., Born, J., 1996. Greater efficacy ofepisodic than continuous growth hormone-releasing hormone (GHRH) administration in promotingslow-wave sleep (SWS). J. Clin. Endocrinol. Metab. 81, 1009–1013.

Obal, F. Jr., Floyd, R., Kapas, L., Bodosi, B., Krueger, J.M., 1996. Effects of systemic GHRH on sleepin intact and hypophysectomized rats. Am. J. Physiol. 270, E230–E237.

Rechtschaffen, A., Kales, A., 1968. A Manual of Standardized Terminology, Techniques and ScoringSystem for Sleep Stages of Human Subjects. Washington, NIH Publication 204, US GovernmentPrinting Office.

Schopohl, J., Losa, M., Frey, C., Wolfram, G., Huber, R., Permanetter, W., von Pawel, J., Muller, O.A.,von Werder, K., 1991. Plasma growth hormone (GH)-releasing hormone levels in patients with lungcarcinoma. Clin. Endocrinol. 34, 463–467.

Steiger, A., Guldner, J., Hemmeter, U., Rothe, B., Wiedemann, K., Holsboer, F., 1992. Effects ofgrowth hormone-releasing hormone and somatostatin on sleep EEG and nocturnal hormonesecretion in male controls. Neuroendocrinology 56, 566–573.

Takahashi, T., Okimura, Y., Yoshimura, K., Shigeyoshi, Y., Kaji, H., Abe, H., Chihara, K., 1995.Regional distribution of growth hormone-releasing hormone (GHRH) receptor mRNA in the ratbrain. Endocrinology 136, 4721–4724.

von Werder, K., Losa, M., Stalla, G.K., Bock, L., Muller, O.A., 1985. Human pancreatic growthhormone releasing factor (hpGRF): GRF- and GH-levels after bolus injection and infusion ofhpGRF 1-44. Acta Neurochir. 75, 81–87.

.