OREXINERGIC NEURON NUMBERS IN THREE SPECIES OF AFRICAN MOLE RATS WITH RHYTHMIC AND ARRHYTHMIC CHRONOTYPES

Abstract

In the present study, orexinergic cell bodies within the brains of rhythmic and arrhythmic circadian chronotypes from three species of African mole rat (Highveld mole rat—Cryptomys hottentotus pretoriae, Ansell’s mole rat—Fukomys anselli and the Damaraland mole rat—Fukomys damarensis) were identified using immunohistochemistry for orexin-A. Immunopositive orexinergic (Orx+) cell bodies were stereologically assessed and absolute numbers of orexinergic cell bodies were determined for the distinct circadian chronotypes of each species of mole rat examined. The aim of the study was to investigate whether the absolute numbers of identified orexinergic neurons differs between distinct circadian chronotypes with the hypothesis of elevated hypothalamic orexinergic neurons in the arrhythmic chronotypes compared with the rhythmic chronotypes. We found statistically significant differences between the circadian chronotypes of F. anselli, where the arrhythmic group had higher mean numbers of hypothalamic orexin neurons compared with the rhythmic group. These differences were observed when the raw data was compared and when the raw data was corrected for body mass (M_b) and brain mass (M_br). For the two other species investigated, no significant differences were noted between the chronotypes, although a statistically significant difference was noted between all rhythmic and arrhythmic individuals of the current study when the counts of orexin neurons were corrected for M_b—the arrhythmic individuals had larger numbers of orexin cells.

It is well known that mole rats, subterranean, visually regressed rodents, have unusual patterns of circadian rhythmicity (Lovegrove and Papenfus, 1995; Lovegrove and Muir, 1996; Oosthuizen et al., 2003). Locomotor activity studies have shown that within a species of mole rat, there are individuals that have a rhythmic chronotype, and others that are distinctly arrhythmic, and this is seen in both social and solitary species of mole rat (Oosthuizen et al., 2003). A study of sleep patterns in the giant Zambian mole rat (Cryptomys mechowi) showed some differences in sleep patterns between rhythmic and arrhythmic chronotypes of this species, where arrhythmic individuals spend more time in waking with a longer average duration of a waking episode, and less time in non-rapid eye movement sleep (NREM) sleep with a shorter average duration of a NREM episode, although with a greater NREM sleep intensity (Bhagwandin et al., 2011b).

The neurotransmitter orexin is known to promote wakefulness, particularly during motor activities (Estabrooke et al., 2001; Mileykovskiy et al., 2005; Lee et al., 2005; Saper et al., 2005). Orexinergic neurons, which are restricted to the hypothalamus, have ascending projections to the cerebral cortex and descending projections to the cholinergic (basal forebrain, lateral dorsal tegmental nucleus, pedunculopontine nucleus) and monoaminergic (locus coeruleus, ventral tegmental area, raphe) nuclear groups involved in the sleep–wake cycle (Peyron et al., 1998; Chemelli et al., 1999; Baumann and Bassetti, 2005). Loss of orexinergic neurons has been associated with narcolepsy (Chemelli et al., 1999; Lin et al., 1999; Siegel, 1999), narcolepsy with cataplexy in humans (Peyron et al., 2000; Thannickal et al., 2000), narcolepsy without cataplexy in humans (Thannickal et al., 2009), and the narcoleptic like symptoms in patients with Parkinson’s disease (Thannickal et al., 2007).

It is also known that orexins are involved in appetite regulation through activation of the leptin sensitive neurons of the arcuate nucleus and its subsequent feeding associated signaling to orexinergic neurons of the lateral hypothalamus (LH) and henceforth interaction between LH and the hypothalamic ventromedial nucleus (VMH) (Stellar, 1954; Rodgers et al., 2002; Sakurai, 2003, 2005). Several studies have demonstrated increased orexinergic activity during states of vigilance and foraging and as a result have paired waking with hunger sensations and optimal feeding (Sakurai et al., 1998; Edwards et al., 1999; Haynes et al., 1999; VanItallie, 2006). Conversely, anorexinergic activity has been associated with sleep states, thereby pairing satiety with sleep (Rodgers et al., 2002; Nicolaidis, 2006; VanItallie, 2006). These studies are indicative of a coordinated orexinergic control of appetite regulation with the sleep–wake cycle that is responsive to the suprachiasmatic nucleus (SCN) circadian clock and nutritional status (Sakurai, 2005; VanItallie, 2006), plus, it has been suggested that disruption of the circadian cycle and sleep deprivation can affect energy balance resulting in changes in body composition (Taheri et al., 2004; VanItallie, 2006).

Given this scenario on the action of the orexinergic neurons, and the clearly distinct circadian chronotypes of the African mole rats, along with differences in the neuropeptide population of the SCN of the mole rats (Negroni et al., 2003), it was thought that an investigation of orexinergic neuronal numbers may yield results of interest in relation to function, especially in the mole rats. In the current study, orexinergic neuronal numbers of three species of African mole rat, the Highveld mole rat (Cryptomys hottenttotus pretoriae [CHP]), Ansell’s mole rat (Fukomys anselli [FA]) and the Damaraland mole rat (Fukomys damarensis [FD]) were quantified. It should be noted that Ansell’s and Damaraland mole rats were previously considered to be members of the genus Cryptomys. All three species of mole rat have a reduced visual system, live a subterranean lifestyle, and have unusual patterns of circadian rhythmicity (Lovegrove and Papenfus, 1995; Lovegrove and Muir, 1996; Oelschläger et al., 2000; Cernuda-Cernuda et al., 2003; Negroni et al., 2003; Oosthuizen et al., 2003; Gutjahr et al., 2004; Nemec et al., 2004, 2008). Brains from circadian distinct rhythmic and arrhythmic individuals of each species were examined. The unusual circadian patterns of individuals within a species provide interesting models to examine whether the absolute numbers of identified orexinergic neurons differs between circadian chronotypes among the species examined.

MATERIALS AND METHODS

A total of 18 brains (15 males and 3 females—see Table 1) from three species of wild caught African mole rat, the Highveld mole rat CHP (average body mass: 127.9 g; average brain mass: 1.5 g), Ansell’s mole rat FA (average body mass: 84.5 g; average brain mass: 1.2 g), and the Damaraland mole rat FD (average body mass: 107.8 g; average brain mass: 1.8 g) were used in this study. For the purposes of the current study, three rhythmic and three arrhythmic individuals per species were used. A single mole rat was housed in an enclosure (30×25 cm²) and wood shavings were used to line the floor. The mole rats used in the present study were caught from wild populations in South Africa under permission and supervision from the appropriate wildlife directorates. All animals were treated and used according to the guidelines of the University of Pretoria Animal Ethics Committee, which parallel those of the NIH for the care and use of animals in scientific experimentation.

Table 1.

Stereological parameters used for rhythmic and arrhythmic individuals of Cryptomys hottentotus pretoriae, Fukomys anselli, and F. damarensis

Animal ID	Circadian chronotype	Counting frame size (μm)	Sampling grid size (μm)	Cut thickness (μm)	Average species mounted thickness (μm)	Vertical guard zones (top and bottom, μm)	Section interval	Number of sampling sites	Average CE (Gunders, m=0)	Average CE (Gunders, m=1)
C. hottentotus pretoriae
CHP2	R	275×185×12	400×300	50	20	5	2	263	0.13	0.06
CHP8	R	275×185×12	400×300	50	20	5	2	307	0.04	0.04
CHP12	R	275×185×12	400×300	50	20	5	2	344	0.14	0.05
CHP1	AR	275×185×12	400×300	50	20	5	2	421	0.04	0.04
CHP3	AR	275×185×12	400×300	50	20	5	2	347	0.06	0.04
CHP7	AR	275×185×12	400×300	50	20	5	2	358	0.05	0.04
F. anselli
FA2	R	275×185×10	400×300	50	14	5	2	521	0.07	0.05
FA3	R	275×185×10	400×300	50	14	5	2	438	0.09	0.05
FA7	R	275×185×10	400×300	50	14	5	2	352	0.10	0.05
FA4	AR	275×185×10	400×300	50	14	5	2	446	0.12	0.05
FA5	AR	275×185×10	400×300	50	14	5	2	458	0.07	0.04
FA6	AR	275×185×10	400×300	50	14	5	2	348	0.04	0.04
F. damarensis
FD4	R	275×185×10	400×300	50	15	5	2	331	0.05	0.04
FD6	R	275×185×10	400×300	50	15	5	2	476	0.05	0.04
FD7	R	275×185×10	400×300	50	15	5	2	376	0.08	0.04
FD1	AR	275×185×10	400×300	50	15	5	2	436	0.04	0.04
FD8	AR	275×185×10	400×300	50	15	5	2	443	0.04	0.04
FD13	AR	275×185×10	400×300	50	15	5	2	371	0.08	0.05

Determination of rhythmicity patterns

Infrared detectors were used to detect activity patterns in the mole rats. The mole rats were exposed to differing light conditions that included, light (L) and dark (D) cycles, DD cycles and LL cycles. Initially, the experimental protocol started with a 12 L: 12 D cycle, and once entrainment was achieved, the light cycle was switched to constant darkness for 1 month. Recorded behavioral activity data were analyzed using Actiview and Clocklab software which were also used to generate actograms. The sensitivity of the y-axis remained the same during analysis of each mole rat. For the purposes of the current study, it was of importance to determine whether the mole rats were able to re-entrain to light cycles and switch their activity according to the light. If a mole rat was able to re-entrain, it was deemed “rhythmic” and the converse were applied to “arrhythmic” mole rats (Fig. 1). For details regarding the specifics of circadian rhythm determination see Oosthuizen et al. (2003).

Fig. 1.

Actigrams illustrating circadian patterns of locomotor activity in rhythmic and arrhythmic individuals from C. hottentotus pretoriae (CHP), F. anselli (FA), and F. damarensis (FD). These actigrams indicate that the rhythmic mole rats have a predictable period of activity during the light period, whereas locomotor activity in the arrhythmic animals is irregular.

Orexin immunohistochemistry

Individuals of each species of mole rat were perfused between 17:00 and 20:00 on three consecutive days, that is, a day per species. The mole rats were euthanazed (Euthanaze, 200 mg sodium pentobarbital/kg, i.p.) and then perfused i.c. The perfusion was initially done with a rinse of 0.9% saline solution at 4 °C, followed by a solution of 4% paraformaldehyde in 0.1 M phosphate buffer (PB) (approximately 1 L/kg of each solution). Brains were removed from the skull and postfixed overnight in 4% paraformaldehyde in 0.1 M PB, and then allowed to equilibrate in 30% sucrose in PB. The brains were blocked and the diencephalon was frozen and sectioned coronally (50-μm section thickness). Tissue from rhythmic and arrhythmic chronotypes of all three species of mole rat in the present study was processed concurrently. A one in two series of stains was made for Nissl and orexin-A immunohistochemistry. Sections used for the Nissl series were mounted on 0.5% gelatine-coated glass slides, cleared in a solution of 1:1 chloroform and absolute alcohol, then stained with 1% Cresyl Violet.

For immunohistochemical staining, the sections were first treated for 30 min with an endogenous peroxidase inhibitor (49.2% methanol: 49.2% of 0.1 PB: 1.6% of 30% H₂O₂) followed by three 10 min rinses in 0.1 M PB. This was followed by a 2-h preincubation, at room temperature, in a solution (blocking buffer) containing 3% normal goat serum (NGS, Chemicon), 2% bovine serum albumin (BSA, Sigma), and 0.25% Triton X-100 (Merck) in 0.1 M PB. The sections were then incubated in a primary antibody solution containing the appropriately diluted antibody in blocking buffer for 48 h at 4 °C under gentle agitation. To reveal orexinergic neurons, we used the anti-orexin-A antibody AB 3704 (Chemicon, raised in rabbit) at a dilution of 1:1500. This step was followed by three 10-min rinses in 0.1 M PB, after which the sections were incubated in a secondary antibody solution for 2 h. The secondary antibody solution contained a 1:1000 dilution of biotinylated anti-rabbit IgG (BA-1000, Vector laboratories) in 3% NGS and 2% BSA in 0.1 M PB. After three 10-min rinses in 0.1 M PB, the sections were incubated for 1 h in A.B. solution (Vector laboratories) and again rinsed three times. The sections were then treated in a solution of 0.05% diaminobenzidine in 0.1 M PB for 5 min, following which 3 μl of 30% H₂O₂ was added to the 1 ml of solution in which each section was immersed. Staining development was monitored visually and checked under a low power stereomicroscope. Development was allowed to continue until the immunopositive neurons became readily identifiable from the background stain. Development was arrested by placing the sections in 0.1 M PB, and then rinsed twice more in the same solution. Sections were mounted on glass slides coated with 0.5% gelatine and left to dry overnight. They were then dehydrated in a graded series of alcohols, cleared in xylene, and coverslipped with Depex. Two controls were used in the immunohistochemistry, including the omission of the primary antibody and the omission of the secondary antibody in selected sections, which eliminated staining.

Quantitative analysis

The number of orexinergic (Orx+) positive cells was determined with stereological techniques on a one in two series of sections through the complete hypothalamus in all 18 brains. All analyses were done blind to the rhythmicity status of the individual. A Nikon E600 microscope with three axis motorized stage, video camera, Neurolucida interface, and Stereo-Investigator software (Microbrightfield Corp., Colchester, VT, USA) was used for the stereological counts. In an attempt to achieve the most accurate estimation of Orx+ neurons, a pilot study was first conducted in an individual of each species. The pilot study determined the best counting frame size and grid size, and these parameters were then used for all individuals of each species investigated. A 275×185 μm² counting frame and a 300×400 μm² sampling grid were used in each individual. Only orexinergic neurons with clearly visible nuclei were marked in the sampling grids (Table 1). For the calculation of total number of neurons, we measured section thickness in a random sample of 50 sections from each species in the regions where orexinergic neurons were present and used these measurements to calculate the species’ average mounted thickness. The “optical fractionator probe” function of the software computationally determined the total number of Orx+ cell in the hypothalamus of each individual using the following formula:

$$\mathrm{N}=\mathrm{Q}∕(\text{SSF}\times \text{ASF}\times \text{TSF})$$

Where N—was the total estimated neuronal number, Q—was the number of neurons counted, SSF—was the section sampling fraction (in the current study this was 0.5), ASF—was the area sub fraction (this was the ratio of the size of the counting frame to the size of the sampling grid), and TSF—was the thickness sub fraction (this was the ratio of the dissector height relative to 50 μm). To determine TSF, we used the average mounted section thickness calculated for each species (as described earlier in the text, Table 1), subtracted the total vertical guard zones (10 μm) to give dissector height, and used the ratio of dissector height to cut section thickness (50 μm) to provide TSF for each species and applied this to all individuals within a species.

A function in the stereology program called the “nucleator probe” facilitated the estimation of the mean cross-sectional area, volume, and length of the orexinergic positive cells. Only neurons with a distinct nucleus were chosen for analysis. The “nucleator probe” was used in conjunction with the optical fractionator and stereology procedures for systematic random sampling to identify cells (Gunderson, 1988). In total, seven probes were used in the current study, namely: the optical fractionators, optical fractionator using number-weighted section thickness, physical dissector, physical fractionator, Schmitz nearest neighbor, Cavalieri estimator for area and volume, and combination of planes with the optical fractionator for absolute length.

Statistical analysis

In addition, the quantitative values generated using stereological techniques; data for body mass (M_b), brain mass (M_br) and encephalization quotient (EQ) were compiled for analysis. EQ in each individual was determined using the following formula (from Manger, 2006):

$$EQ=M_{br}∕0.069M_b^{0.718}$$

Comparisons were conducted between species (CHP vs. FA, CHP vs. FD and FA vs. FD) and within species (CHP rhythmic vs. CHP arrhythmic, FA rhythmic vs. FA arrhythmic, and FD rhythmic vs. FD arrhythmic). These comparisons were made for M_b, M_br, EQ, number of Orx+ cell bodies (counts), estimated volume of an Orx+ cell body, estimated cross-sectional area of an Orx+ cell body, and estimated length of an Orx+ cell body. The values obtained for the counts, volume, area, and length of Orx+ cell bodies in each mole rat were divided by the M_b of each individual and the M_br of each individual thereby creating an index that corrected for M_b and M_br. These indexes were statistically analyzed between species and within species for the counts, volume, area, and length. The indexes also facilitated comparisons between all rhythmic individuals and all arrhythmic individuals. All data mentioned later in the text are presented as the mean with its standard error. All statistical tests were performed using PAST (Paleontological Statistics, v2.02; Hammer et al., 2001). Initially, the Kruskal–Wallis test was used for the aforementioned comparisons. When significance was observed (P<0.05), multiple Bonferroni-corrected Mann–Whitney U-tests were used as post hoc to the Kruskal–Wallis for the detection of specific differences. P-values for statistically significant differences (again where P<0.05) are provided in the results

RESULTS

In all three species of mole rat, orexinergic immunopositive (Orx+) cell bodies were located within the hypothalamus, as previously reported in all mammals studied to date, including mole rats (Bhagwandin et al., 2011a). Stereological counts of Orx+ cell bodies were found to be statistically significantly different only between the distinct circadian chronotypes of Ansell’s mole rat (FA) and between all rhythmic and arrhythmic individuals of the current study when the counts of Orx+ cell bodies were corrected for M_b. Estimation of cell volume, cross-sectional area, and length showed statistically significant differences between species without correction for M_b and M_br and statistically significant differences between species when these variables were corrected for M_b and M_br, although no statistically significant differences were reached for these parameters within species when similar comparisons were made.

Orexinergic cell body distribution

All three species of mole rat expressed Orx+ neurons only within the hypothalamus and were observed as sharing common neuronal locality within the lateral hypothalamic area (LHA), perifornical region (PFR), and the lateral ventral hypothalamic supraoptic area (LVHA), similar to those reported in two other species of mole rat (Bhagwandin et al., 2011a). Thus, there appears to be three distinct clusters of Orx+ neurons in the hypothalamus of the mole rats studied, a large homogeneous cluster spanning the lateral and perifornical regions (the main cluster), a distinct cluster extending into the region of the zona incerta (the zona incerta cluster), and a final cluster in the ventral lateral hypothalamus adjacent to the diminutive optic tracts (the optic tract cluster). The mole rats, from both chronotypes and all three species, exhibited neuronal cell bodies that were morphologically homogenous in all three clusters, that were ovoid in shape, a varying mixture of bi- and multipolar types and that showed no specific dendritic orientation (Figs. 2 and 3).

Fig. 2.

Photomicrographs showing the orexin-A immunoreactive neurons within the hypothalamus of rhythmic and arrhythmic chronotypes of CHP (A, B), FA (C, D), and FD (E, F). Scale = 1 mm and applies to all.

Fig. 3.

High power photomicrographs showing the morphology of orexin-A immunoreactive neurons within the hypothalamus of rhythmic and arrhythmic chronotypes of CHP (A, B), FA (C, D), and FD (E, F). Scale=100 μm and applies to all.

Comparisons of M_b, M_br, and EQ

Mean M_b in CHP was 132.8±22.4 g for the rhythmic group and 123±17.6 g for the arrhythmic group. FA had a mean M_b of 90.6±2.8 g for the rhythmic group and 78.5±10.9 g for the arrhythmic group (Table 2). FD had a mean M_b measurement of 121.9±22.7 g for the rhythmic group and 93.7±9.6 g for the arrhythmic group. M_b comparisons between species were only statistically significant (Kruskal–Wallis P-value: 0.0064; Mann–Whitney P-value: 0.008) between CHP and FA where it was observed that CHP had a higher mean M_b (Table 3).

Table 2.

Table presenting data for rhythmic and arrhythmic individuals of C. hottentotus pretoriae, F. anselli, and F. damarensis and the respective weighted means for M_br, M_b, EQ, estimated cell counts, average estimated volume, average estimated cell area, and average estimated cell length

Animal ID	Sex	Circadian chronotype	Body mass (g)	Brain mass (g)	EQ	Estimated cell counts	Average estimated cell volume (μm3)	Average estimated cell area μm2)	Average estimated cell length (μm)
C. hottentotus pretoriae
CHP2	F	R	174.5	1.6	0.6	4875	1701. 4	163.2	6.9
CHP8	F	R	126.1	1.3	0.6	12,344	2438.3	198.9	7.6
CHP12	M	R	97.9	1.4	0.8	9922	1936. 5	174.4	7.2
Mean/weighted mean R			132.8±22.4	1.4±0.09	0.6±0.06	9047±2200	2068.0±98	180.9±4.8	7.3±0.09
CHP1	M	AR	110.4	1.4	0.7	11,794	2441.3	203.9	7.8
CHP3	M	AR	157.8	1.4	0.5	12,753	2520.2	204.1	7.7
CHP7	M	AR	100.7	1.7	0.9	10,095	1797. 2	167.8	7.1
Mean/weighted mean AR			123.0±17.6	1.5±0.10	0.7±0.10	11,547±777	2281.2±116.7	193.3±5.9	7.5±0.11
F. anselli
FA2	M	R	95.9	1.2	0.7	5693	2837.0	225.3	7.9
FA3	M	R	86.3	1.1	0.6	5944	2875.9	223.6	8.1
FA7	M	R	89.6	1.2	0.7	5719	3407.7	259.6	8.8
Mean/weighted mean R			90.6±2.8	1.2±0.03	0.7±0.01	5786±80	3228.6±172.2	247.7±8.1	8.6±0.14
FA4	M	AR	69.3	1.2	0.8	8242	3036.8	238.6	8.4
FA5	M	AR	100.1	1.1	0.6	10,686	3233.0	243.7	8.5
FA6	M	AR	66.0	1.2	0.9	8969	3792.3	271.9	9.0
Mean/weighted mean AR			78.5±10.9	1.2±0.03	0.8±0.09	9299±724	3337.3±128.7	250.7±5.8	8.6±0.10
F. damarensis
FD4	M	R	76.9	1.6	1.0	8803	2428.4	204.8	7.8
FD6	M	R	149.2	1.8	0.7	8024	2318.9	191.1	7.5
FD7	M	R	139.7	2.0	0.8	12,723	2524.6	209.7	7.9
Mean/weighted mean R			121.9±22.7	1.8±0.11	0.9±0.09	9850±1454	2429.5±116.2	202.4±5.4	7.8±0.09
FD1	M	AR	102.2	1.9	1.0	11,704	2289.6	198.3	7.7
FD8	F	AR	104.4	1.9	1.0	9638	1988. 0	180.3	7.4
FD13	M	AR	74.5	1.7	1.0	6482	2211.6	188.8	7.4
Mean/weighted mean AR			93.7±9.6	1.8±0.07	1.0±0.04	9275±1518	2163.2±88.9	189.3±4.9	7.5±0.10

Table 3.

Table showing results of statistical comparisons between various data sets

Between species
	Body Mass	Brain Mass	EQ	Counts	Volume	Area	Length
CHP vs. FA	* CHP	* CHP	–	–	* FA	* FA	* FA
CHP vs. FD	–	–	* FD	–	–	–	–
FA vs. FD	–	* FD	* FD	–	* FA	* FA	–
Within species
CHP R vs. NR	–	–	–	–	–	–	–
FA R vs. NR	–	–	–	* AR	–	–	–
FD R vs. NR	–	–	–	–	–	–	–

Between species corrected for body mass
	Counts	Volume	Area	Length
CHP vs. FA	–	* FA	* FA	* FA
CHP vs. FD	–	–	–	–
FA vs. FD	–	* FA	* FA	–
Within species corrected for body mass
CHP R vs. NR	–	–	–	–
FA R vs. NR	* AR	–	–	–
FD R vs. NR	–	–	–	–
Between species corrected for brain mass
	Counts	Volume	Area	Length
CHP vs. FA	–	* FA	* FA	* FA
CHP vs. FD	–	* CHP	* CHP	* CHP
FA vs. FD	–	* FA	* FA	* FA
Within species corrected for brain mass
CHP R vs. NR	–	–	–	–
FA R vs. NR	* AR	–	–	–
FD R vs. NR	–	–	–	–

All R vs. NR corrected for body mass
	EQ	Counts	Volume	Area	Length
All R vs. NR	–	* AR	–	–	–
All R vs. NR corrected for brain mass
	EQ	Counts	Volume	Area	Length
All R vs. NR	–	–	–	–	–

^*Statistical significance as determined using the post hoc Mann–Whitney U and Bonferroni correction factor when P<0.05 (P-values included in-text). Letters in subscript indicate the species or chronotype of mole rat where the higher mean was observed. CHP, C. hottentotus pretoriae; FA, F. anselli; FD, F. damarensis; AR, arrhythmic.

Mean M_br in CHP measured 1.4±0.09 g for the rhythmic group and 1.5±0.10 g for the arrhythmic group. FA had a mean M_br measurement of 1.2±0.03 g for the rhythmic group and 1.2±0.03 g for the arrhythmic group. F. damarensis had a mean M_br measurement of 1.8±0.11 g for the rhythmic group and 1.8±0.07 g for the arrhythmic group (Table 2). Statistically significant differences were noted between the CHP and FA (where the CHP had a higher mean M_br, Kruskal–Wallis P-value: 0.0039; Mann–Whitney P-value: 0.004) and FA and FD (where FD had a higher mean M_br, Kruskal–Wallis P-value: 0.0039; Mann–Whitney P-value: 0.004) (Table 3).

Mean EQ in CHP was calculated as 0.6±0.06 for the rhythmic group and 0.7±0.10 for the arrhythmic group. FA had a mean EQ calculation of 0.7±0.01 for the rhythmic group and 0.8±0.09 for the arrhythmic group. FD had a mean EQ calculation of 0.9±0.09 for the rhythmic group and 1 ±0.04 for the arrhythmic group (Table 2). Statistically significant differences were obtained when FD was compared with CHP (where FD had a higher mean EQ, Kruskal–Wallis P-value: 0.02; Mann–Whitney P-value: 0.02) and when FD was compared with FA (where the FD showed a higher mean EQ, Kruskal–Wallis P-value: 0.02; Mann–Whitney P-value: 0.02) (Table 3).

Stereological counts of Orx+ cell bodies

Stereological counts of Orx+ cell bodies in the CHP revealed a mean of 9047±2200 for the rhythmic group and 11,547±777 for the arrhythmic group. Counts from FA revealed a mean of 5786±80 for the rhythmic group and 9299±724 for the arrhythymic group. FD revealed a mean of 9850±1454 for the rhythmic group and 9275±1518 for the arrhythmic group (Fig. 4) (Table 2).

Fig. 4.

Graphs showing the various parameters of stereological data for arrhythmic and arrhythmic individuals (gray bars) of CHA, FA, and FD and the respective weighted means (black bars). The graphs indicate average estimated values for somal number, volume, area, and length.

Comparisons of Orx+ cell body counts between rhythmic and arrhythmic groups in each species revealed a statistically significant difference only between the circadian chronotypes of FA (where the mean number of Orx+ neurons in the hypothalamus of the arrhythmic group was higher than the rhythmic group, Kruskal–Wallis P-value: 0.008; Mann–Whitney P-value: 0.009). Even though statistical significance was not reached in the other two species investigated, the mean number of Orx+ neurons in the hypothalamus of CHP was higher for the arrhythmic group than the rhythmic group. For FD, stereological counts of Orx+ cell bodies yielded similar totals for both the rhythmic and arrhythmic groups. When the counts of Orx+ cell bodies were corrected for M_b and M_br, statistically significant differences were noted only between the circadian chronotypes of FA. In both cases, the arrhythmic groups had a higher mean numbers of Orx+ neurons in the hypothalamus than the rhythmic groups (Kruskal–Wallis P-value: 0.0028; Mann–Whitney P-value: 0.003). Additionally, when all rhythmic individuals of all species studied here were compared with all arrhythmic individuals, a statistically significant difference was only reached in counts that had been corrected for M_b (where the arrhythmic group had a higher mean number of Orx+ neurons in the hypothalamus than the rhythmic group, Kruskal–Wallis P-value: 0.02; Mann–Whitney P-value: 0.03) (Table 3).

Stereological estimation of volume, area, and length of Orx+ cell bodies

Stereological estimation of the volume of an Orx+ cell body in CHP presented a weighted mean of 2068±98 μm³ for the rhythmic group and 2281.2±116.7 μm³ for the arrhythmic group. FA presented a weighted mean of 3228.6±172.2 μm³ for the rhythmic group and 3337.3±128.7 μm³ for the arrhythmic group. FD presented a weighted mean of 2429.5±116.2 μm³ for the rhythmic group and 2163.2±88.9 μm³ for the arrhythmic group (Fig. 4) (Table 2). Statistically significant differences were noted: (1) between species (where FA showed a higher mean volume of an Orx+ cell body than the other two species, Kruskal–Wallis P-value: 0.003; Mann–Whitney P-value: 0.0005); (2) when volume was corrected for M_b between species (where FA had a higher mean volume of an Orx+ cell body than CHP, Kruskal–Wallis P-value: 0.004; Mann–Whitney P-value: 0.0008), and where FA had a higher mean volume of an Orx+ cell body than FD (Kruskal–Wallis P-value: 0.004; Mann–Whitney P-value: 0.008); and (3) when volume was corrected for M_br between species (where FA had a higher mean volume of an Orx+ cell body than the other two species, Kruskal–Wallis P-value: 0.003; Mann–Whitney P-value: 0.005, and where CHP had a higher mean volume of an Orx+ cell body than FD, Kruskal–Wallis P-value: 0.03; Mann–Whitney P-value: 0.03) (Table 3).

Stereological estimation of the cross-sectional area of an Orx+ cell body in the CHP revealed a weighted mean of 180.9±4.8 μm² for the rhythmic group and 193.3±5.9 μm² for the arrhythmic group. FA revealed a weighted mean of 247.7±8.1 μm² for the rhythmic group and 250.7±5.8 μm² for the arrhythmic group. FD revealed a weighted mean of 202.4±5.4 μm² for the rhythmic group and 189.3±4.9 μm² for the arrhythmic group (Fig. 4) (Table 2). Statistically significant differences were noted: (1) between species (where FA showed a higher cross-sectional area of an Orx+ cell body than the other two species, Kruskal–Wallis P-value: 0.003; Mann–Whitney P-value: 0.005); (2) when cross-sectional area was corrected for M_b between species (where FA had a higher cross-sectional area of an Orx+ cell body than CHP, Kruskal–Wallis P-value: 0.004; Mann–Whitney P-value: 0.005, and where FA had a higher cross-sectional area of an Orx+ cell body than FD, Kruskal–Wallis P-value: 0.01; Mann–Whitney P-value: 0.01); and (3) when area was corrected for M_br between species (where FA had a higher cross-sectional area of an Orx+ cell body than the other two species, Kruskal–Wallis P-value: 0.003; Mann–Whitney P-value: 0.005, and where CHP had a higher cross-sectional area of an Orx+ cell body than FD, Kruskal–Wallis P-value: 0.04; Mann–Whitney P-value: 0.03) (Table 3).

Stereological estimation of the length of an Orx+ cell body in CHP revealed a weighted mean of 7.3±0.09 μm for the rhythmic group and 7.5±0.11 μm for the arrhythmic group. FA revealed a weighted mean of 8.6±0.14 μm for both rhythmic and arrhythmic groups. FD revealed a weighted mean of 7.8±0.09 μm for the rhythmic group and 7.5±0.10 μm for the arrhythmic group (Fig. 4) (Table 2). Statistically significant differences were noted: (1) between species (where FA showed a longer length of an Orx+ cell body than the other two species, Kruskal–Wallis P-value: 0.003; Mann–Whitney P-value: 0.005); (2) when length was corrected for M_b between species (where FA had a longer length of an Orx+ cell body than CHP, Kruskal–Wallis P-value: 0.004; Mann–Whitney P-value: 0.005); and (3) when volume was corrected for M_br between species (where FA had longer length of an Orx+ cell body than the other two species, Kruskal–Wallis P-value: 0.001; Mann–Whitney P-value: 0.005, and where CHP had a longer length of an Orx+ cell body FD, Kruskal–Wallis P-value: 0.02; Mann–Whitney P-value: 0.03) (Table 3).

DISCUSSION

The initial aim of the current study was to determine whether rhythmic and arrhythmic chronotypes had different numbers of hypothalamic orexin immunopositive neurons with the hypothesis of elevated hypothalamic orexinergic neurons in the arrhythmic chronotypes as a result of the involvement of orexins in the sleep–wake cycle. The results of the current study indicated statistically significant differences between the circadian chronotypes of FA, where the arrhythmic group had higher mean numbers of hypothalamic orexin neurons compared with the rhythmic group. No statistically significant differences in orexinergic neuronal numbers were observed in the other two species studied. A statistically significant difference was noted between all rhythmic and arrhythmic individuals of the current study when the counts of orexin neurons were corrected for M_b.

Orexins, sleep–wake cycle, and circadian rhythmicity

The results of the present study indicated statistically significant differences between circadian chronotypes of FA for the counts of hypothalamic Orx+ neurons when the raw data were compared. In this case, the arrhythmic chronotype consistently displayed higher mean counts of hypothalamic Orx+ neurons than the rhythmic chronotype. A similar tendency, although not statistically significant, was observed when the raw data were compared between the chronotypes of CHP; however, a similar comparison between chronotypes of FD showed that the mean count of Orx+ neurons in the rhythmic group was higher than the rhythmic group, but statistical significance was not reached. The analysis of FA supports the hypothesis of elevated Orx+ neurons in the hypothalamus. Given the previous suggestions for the role of orexins in the sleep–wake cycle (Siegel, 2004; Baumann and Bassetti, 2005), the results of the current study indicate that the arrhythmic chronotypes of FA and possibly CHP would theoretically have an enhanced maintenance of waking (despite the reduced motor activity of the arrhythmic chronotype compared with the rhythmic, as indicated by their actograms, with a net result that possibly enhances vigilance during quiet waking) and a reduced amount of time spent in NREM relative to the rhythmic chronotypes. The study of sleep in rhythmic and arrhythmic chronotypes of the Giant Zambian mole rat (Fukomys mechowii) (Bhagwandin et al., 2011b) has shown that the arrhythmic group statistically spent more time awake and spent less time in NREM compared with the rhythmic group.

It is known that the bifurcated ascending arousal system incorporates first, a pathway to the thalamus which excites thalamic relay neurons that transmit information to the cerebral cortex, and second, a branch that bypasses the thalamus to activate neurons in the LHA, basal forebrain, and cerebral cortex (Saper, 1985; Hallanger et al., 1987; Saper et al., 2001, 2005; Jones, 2003). Orexinergic neurons, with cell bodies restricted to the hypothalamus, have ascending projections to the cerebral cortex and descending projections to the cholinergic (basal forebrain, lateral dorsal tegmental nucleus, pedunculopontine nucleus) and monoaminergic (locus coeruleus, ventral tegmentum area, raphe) nuclear groups of arousal systems (Peyron et al., 1998; Chemelli et al., 1999; Baumann and Bassetti, 2005; Bhagwandin et al., 2011a). The exact role of orexins in REM is unclear, as some studies suggest increased orexinergic activity during REM, whereas others report the contrary (Lee et al., 2005; Mileykovskiy et al., 2005); however, a recent study has postulated that orexin neurons mediate the circadian timing of REM by suppressing REM during the active period (Kantor et al., 2009). In light of the results of the current study, it would appear that the increased numbers of hypothalamic orexinergic neurons observed within FA arrhythmic chronotypes, and potentially CHP, through possible increased activity and the resulting effect on wakefulness and motor activities, may facilitate a higher level of vigilance allowing for extended foraging and feeding activity.

Baumann and Bassetti (2005) proposed that orexins are under control of the SCN and that orexins do not significantly affect circadian rhythmicity. Saper et al. (2005) described a three stage circuitry integrator for circadian rhythms that involve a pathway from the SCN to the sub-paraventricular zone (SPZ) and finally to dorsomedial nucleus of the hypothalamus (DMH). The DMH communicates directly with the LHA, the region with the highest population of orexinergic neurons, and the ventrolateral preoptic nucleus (VLPO), the area primarily concerned with the promotion of slow wave sleep (SWS). The rationale behind the three-stage pathway lies in the observation that in both diurnal and nocturnal animals, the SCN is primarily active during the light period and the VLPO during the dark period (Sherin et al., 1996; Gaus et al., 2002). Therefore, in animals that are nocturnal vs. diurnal, there should be an intervening circuitry that allows the circadian cycle to be set at opposite phases (Saper et al., 2005). It is well known that mole rats have unusual patterns of circadian rhythmicity (Oosthuizen et al., 2003; Lovegrove and Muir, 1996; Lovegrove and Papenfus, 1995) and that mole rats have different neuropeptide populations in the SCN compared with other rodents (Negroni et al., 2003). Therefore, the increased counts of Orx+ neurons in the hypothalamus within FA arrhythmic chronotypes, and potentially CHP, may alter the circadian control of wakefulness in these individuals.

Orexins and appetite regulation

The results of the present study indicated statistically significant differences between circadian chronotypes of FA when the raw data for counts of Orx+ neurons were corrected for M_b and M_br, where the arrhythmic groups showed higher mean counts than the rhythmic groups. This can be interpreted as the arrhythmic chronotype having a higher number of Orx+ neurons per gram of M_b or M_br compared with the rhythmic chronotype. Since the initial demonstration of orexin induced hyperphagia in rats (Sakurai et al., 1998), orexins have become synonymous with mechanisms of appetite regulation and increased food intake that is driven by orexin involvement in motivated behavior (Rodgers et al., 2002). Cai et al. (2001) have reported activation of the lateral hypothalamic orexin neurons in response to low blood glucose levels and an empty stomach, signals that indicate replenishment, and concluded that this context-dependency explains the effects associated with orexin activation—increased food intake with delayed onset of satiety and increased wakefulness with vigilance and behavioral activity. It is also known that orexins are involved in appetite regulation through activation of the leptin sensitive neurons of the arcuate nucleus and its subsequent feeding associated signaling to orexinergic neurons of the LH and henceforth interaction between LH and the hypothalamic VMH (Stellar, 1954; Rodgers et al., 2002; Sakurai, 2003, 2005). Therefore, it would seem fair to infer that the elevated counts of hypothalamic orexinergic neurons in the arrhythmic chronotype of FA may yield a greater drive for motivated behavior and theoretically a higher food intake compared with the rhythmic chronotype.

Interestingly though, the results of the present study showed a statistically significant difference for the counts of Orx+ neurons that were corrected for M_b when all rhythmic individuals were compared with all arrhythmic individuals from all three species of mole rats examined. This indicated that the arrhythmic individuals have a higher number of Orx+ neurons per gram of M_b. Given the effect of orexins on motivated behavior and food intake, this may be indicative of a higher M_b in the arrhythmic individuals compared with the arrhythmic individuals. This was not the case, as it was observed in the present study that the mean M_b of the arrhythmic chronotype in each species investigated was lower than that of the rhythmic chronotype. A possible explanation for this could lie in the assessment of the impact of chronic administration of orexin-A on food intake on rats (Haynes et al., 1999; Yamanaka et al., 1999). Both these studies indicated an increased food intake during the light period and a compensatory reduction of food intake during the dark period, and both agree that chronic administration of orexin-A does not alter 24 h food intake or M_b gain. In addition, Lubkin and Stricker-Krongrad (1998) postulated that orexins may increase overall metabolic rate to support increased motor activity; however, Nicolaidis (2006) suggests that overall metabolic rate be divided into resting metabolism and locomotion-related metabolism and based on resting metabolism, a hypometabolic but physically active animal has a higher (total) metabolism than a hypermetabolic animal that remains quiet. This concept supports the observations made in the mole rats assessed in the current study, as it has been previously reported that subterranean mole rats have lower resting metabolic rates than surface dwelling mammals (Bennett and Faulkes, 2000; Zelová et al., 2007, 2009). Furthermore, it has been suggested that sleep duration tracks the metabolism of nutrients in a meal and the size of a meal, with the effect that a higher nutritional content and larger meal size produces a longer sleep duration (Danguir and Nicolaidis, 1980a,b). Despite this, brain metabolism is conserved and independent from that of the periphery, the VMH, dorsomedial nucleus, and paraventricular nucleus, metabolic strategic areas, are capable of reflecting metabolic changes resulting from bodily depletion and repletion (Nicolaidis, 2006). Therefore, it may be possible that a higher peripheral metabolism may result in a lower M_b despite the elevated Orx+ neurons observed in the arrhythmic chronotype compared with the rhythmic chronotype.

Abbreviations:

CHP

Cryptomys hottenttotus pretoriae

FA

Fukomys anselli

FD

Fukomys damarensis

LH

lateral hypothalamus

LHA

lateral hypothalamic area

M_b

body mass

M_br

brain mass

PB

phosphate buffer

SCN

suprachiasmatic nucleus

TSF

thickness sub fraction

VMH

ventromedial nucleus