Underdeveloped Extraocular Muscles in the Naked Mole-Rat (Heterocephalus glaber)



The extraocular muscles (EOM), the effector arm of the ocular motor system, have a unique embryological origin and phenotype. The naked mole-rat (NMR) is a subterranean rodent with an underdeveloped visual system. It has not been established if their ocular motor system is also less developed. The NMR is an ideal model to examine the potential codependence of oculomotor and visual system development and evolution. Our goal was to compare the structural features of NMR EOMs to those of the mouse, a similar sized rodent with a fully developed visual system. Perfusion-fixed whole orbits and EOMs were dissected from adult NMR and C57BL mice and examined by light and electron microscopy. NMR orbital anatomy showed smaller EOMs in roughly the same distribution around the eye as in mouse and surrounded by a very small Harderian gland. The NMR EOMs did not appear to have the two-layer fiber distribution seen in mouse EOMs; fibers were also significantly smaller (112.3 ± 46.2 vs. 550.7 ± 226 sq μm in mouse EOMs, *P < 0.05). Myofibrillar density was less in NMR EOMs, and triad and other membranous structures were rudimentary. Finally, mitochondrial volume density was significantly less in NMR EOMs than in mouse EOM (4.5% ± 1.9 vs. 21.2% ± 11.6, respectively, *P < 0.05). These results demonstrate that NMR EOMs are smaller and less organized than those in the mouse. The “simpler” EOM organization and structure in NMR may be explained by the poor visual ability of these rodents, initially demonstrated by their primitive visual system. Anat Rec, 2010. © 2010 Wiley-Liss, Inc.

The naked mole-rat (Heterocephalus glaber) is a subterranean rodent that rarely experiences natural environmental light. Its visual system is underdeveloped, the naked mole-rat retina contains most of the cell types of visually guided mammals but the structural organization is rudimentary (Mills and Catania, 2004). The neural structures, which arbitrate vision are also atrophied (Crish, SD et al., 2006).Why nonvisually guided mammals retain a visual system at all remains uncertain. Potential reasons include that circadian rhythms and other metabolic activities could still rely on subtle environmental visual cues. Light avoidance, without fine visual details, could also form the sensory arm of an escape behavior in the event of a tunnel breech. In any event, it is unclear whether the regression of the visual system is accompanied by a corresponding change in the ocular motor system.

The extraocular muscles (EOMs), responsible for voluntary and reflexive movements of the eyes, are arguably the fastest and most active skeletal muscles (Porter et al., 1995). These small muscles express mostly fast myosins, including an ultrafast tissue-specific isoform, and contain abundant mitochondria and sarcoplasmic reticulum (SR) (Mayr, 1971). The visual system is anatomically and functionally immature at birth; key properties, such as binocularity and depth perception develop postnatally during a species-specific window called the "critical period" (Berardi et al., 2000). Previously, we demonstrated in mice that dark rearing impairs EOM function and the neural mechanisms underlying compensatory eye movements (McMullen et al., 2004). This indicates that visual experience early in life is necessary for the normal development of EOMs (Cheng et al., 2004, McMullen et al., 2004). The study of the ocular motor system of a fossorial mammal such as the naked mole-rat provides the opportunity to assess the role of visual experience on the development of the extraocular muscles and corresponding central motor pathways. Therefore, this project was designed to compare the orbital anatomy and EOM morphology of the naked mole-rat and the C57BL mouse.



This study was approved by the Institutional Animal Care and Use Committees at the University of Kentucky and Vanderbilt University. We used six naked mole-rats and 13 C57BL mice for histology and immunocytochemistry. Upon arrival, the mice were kept in microisolator cages with Harlan Teklad rodent food and water provided ad libitum. Naked mole-rats came from a colony maintained at Vanderbilt University. Naked mole-rats were kept in a temperature -controlled room housed in chambers connected with plastic tubing.


Before the collection of tissues, C57BL mice and naked-mole rats mice were anesthetized with ketamine hydrochloride/xylazine hydrochloride (100 mg/8 mg per kg body weight injected i.p.) exsanguinated and perfused with physiological saline, followed by 2% paraformaldehyde and 2% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4. Whole orbits were dissected and embedded in paraffin, 10 μm thick sections were stained with hematoxylin and eosin (H&E) to examine morphology. Sections were imaged with a Nikon E600 microscope equipped with a Spot RT Slider camera and Spot RT software (v 4.0). Fiber size was measured using Image J software from NIH (http://rsb.info.nih.gov/ij/). Quantitative analyses were done by personnel blinded to the experimental conditions.

Electron Microscopy

Mice and naked mole-rats were perfusion-fixed as described earlier. Individual EOMs were dissected and postfixed in 1% osmium tetroxide, stained en bloc in uranyl acetate, dehydrated in a methanol series and propylene oxide, and embedded in epoxy resin. Thin (70 nm) sections were examined and photographed with a Philips Tecnai 12 transmission electron microscope (UK Imaging Core). Mitochondria volume density (% of muscle fiber volume occupied by mitochondria) was determined from 104 extraocular muscle fibers (sampled from both global and orbital layers) from digital pictures obtained from six naked mole-rats and 43 extraocular fibers obtained from 13 C57BL mice using a standard point-counting method (144-point grid) with systematic sampling (Weibel, 1979).

Data Analysis

All results are presented as the mean ± SE of n observations. Mitochondrial volume density and fiber area were compared with Student's t-tests. The significance level for rejection of the null hypothesis was set at P ≤ 0.05 for all comparisons.



The orbits of naked mole-rats were analyzed for overall anatomical features. The basic pattern of six EOMs; superior, lateral, medial, and inferior rectus muscles; superior and inferior oblique muscles (Fig. 1a) are shared among all vertebrate classes (Porter et al., 2003). There are, of course a few difference, for example, in some rainbow trout, the lateral rectus is oriented vertically instead of horizontally as in other species (Noden and Fraancecis-West, 2006). Also, in marlins, swordfish and billfish, the superior rectus is exaggerated and is used as a heat-generator to warm the brain and eyes above low water temperatures (Fritsches et al., 2005). Naked mole-rats have the same number of EOMs in the same basic arrangement around the optic nerve and globe, such as the superior oblique, retractor bulbi, and rectus muscles as the mouse (Fig. 1a–d). The eye is protected by an oily substance from the Harderian glands that coats the cornea and prevents dryness (Buzzell, 1996). The Harderian gland from the naked mole-rat is noticeably smaller compared to mouse (Fig. 1e,f).

Figure 1.

Representative light micrographs of cross-sections of mouse and naked mole rat orbits (H&E stain). (A) Anatomy of the mouse orbit showing optic nerve (arrowhead), Harderian gland (black arrow), retractor bulbi (RB), superior oblique, o, orbital layer; g, global layer; *, rectus muscle. (B) Mouse orbital layer, global layer, retina, retractor bulbi, levator palpebral superioris (lps), and rectus muscles. (C) Naked mole rat orbit showing optic nerve, muscle, Harderian gland. (D) Naked mole rat orbit with optic nerve, EOMs, Harderian gland. (E and F) Naked mole rat retina, EOMs. (scale bars = 100 μm Figures A, 50 μm Figures B–D, and 25 μm Figures E and F).

In most mammals, EOMs exhibit a distinctive layered organization known as the orbital and global layers. These two layers are characterized by their fiber content: the orbital layer consists of smaller fibers and is typically c-shaped (Fig. 1b). These two layers maintain rotational stability of the eyes (Demer et al., 2000). The EOMs from naked mole-rats do not have the two-layer distribution of fibers (orbital and global layers) typically seen in mice, cats, dogs, and primates (Fig. 1e).

Muscle Fiber Comparisons

EOMs in other mammals have smaller fibers with greater mitochondrial content and unique metabolic and fiber type compositions which are distinct from typical skeletal muscles (Porter et al., 2001, Cheng et al., 2004, Cheng and Porter 2002). The EOMs from naked mole-rats were noticeably smaller compared to mouse (Fig. 2a,b). EOM fiber size was significantly smaller in naked mole-rat than mouse (112.3 ± 46.2 vs. 550.7 ± 226 μm2, respectively) (Fig. 2c). A characteristic feature of EOMs is a small myofibril size compared to other skeletal muscles. Myofibril density was greatly reduced in the EOMs from naked mole rats, leading to more space between myofibrils (Fig. 3,4a).

Figure 2.

Naked mole-rat fiber size is smaller. Comparison of mouse (A) and naked mole rat (B) EOM fiber size (H&E stain) (scale bars = 50 μm). (C) Mean fiber size in EOMs from mouse and naked mole rat. *, significant at P < 0.05.

Figure 3.

EOMs of naked mole-rats are not as compact as mouse EOMs. Representative electron micrographs of EOM fibers from mouse (left) and naked mole rat (right). Ultrastructure profiles of mitochondria (m), sarcoplasmic reticulum (sr), myofibrillar organization in the A-band (A) and arrows denote large spaces. Scale bars = 2 mm for top and middle rows, and 500 nm for bottom row.

Figure 4.

EOMs of naked mole-rats contain m-lines. Electron micrograph illustrating arrangement of bands in sarcomere in NMR (A) and mouse (B). Ultrastructure profiles of the M-lines, z-lines (z), mitochondria (m), triads (t), and arrows denote large spaces in naked mole-rats. Scale = 500 nm for both images.

Sarcomeres produce the typical longitudinal binding pattern of striated muscles. They are formed by an ordered arrangement of thick and thin filaments. M-lines contain proteins that interconnect and stabilize adjacent myosin filaments. These structures, especially prominent in fast skeletal muscles, are missing in mouse extraocular muscle (Fig. 4b) (Andrade et al., 2003). They are, however, present in naked mole-rat EOMs (Fig. 4a). Triad and other membranous structures were rudimentary in naked mole- rat EOMs (Fig. 4a).

The EOMs are reported to have one of the highest mitochondrial contents of mammalian skeletal muscles (Mayr, 1971). This high mitochondrial content has been considered to reflect the metabolic demands imposed by their fast and constant activity. The mitochondrial volume density of naked mole rat EOMs was significantly less than in mouse (4.5 ± 1.9 vs. 21.2 ± 11.6% of total fiber volume) (Fig. 5). Mitochondria were also more pleomorphic in the EOM fibers from naked mole rat (Fig. 3, 4a). Naked mole rat mitochondria are more variable in shape than those of the mouse.

Figure 5.

Naked mole-rat EOMs contain less mitochondria. Mitochondrial volume density (% of fiber volume occupied by mitochondria) in EOMs from mouse and naked mole rat. *, significant at P < 0.05.


Our findings demonstrate underdeveloped EOMs in the naked mole rat. Although the naked mole-rat EOMs retain a somewhat typical organization, EOMs are remarkably smaller in size and typical sarcomere arrangement is less well defined. Contrary to adult mouse EOMS; m-lines are present in the EOMs of naked mole-rats. M-lines are present in mouse EOMs during myogenesis, but they disappear soon after birth (Porter et al., 2003). It is then possible that NMR EOMs may persist in a state of incomplete development; also m-line repression may be a characteristic of visually-guided rodents. The results are consistent with these muscles being less active and weaker in these nonvisually guided rodents. It has previously been shown that normal development of the mouse and monkey ocular motor system and its muscles requires visual experience during the critical period (McMullen et al., 2004; Cheng et al., 2004). For example, we have shown that abnormal visual experience postbirth renders mouse EOMs weaker and more fatigable (McMullen et al., 2004). While this paradigm is clearly not applicable to NMR, it does serve to illustrate the connection between the visual pathways and the motor systems serving the eyes. In the case of the NMR, vision is basically replaced by somatosensory inputs, and the visual pathways are correspondingly diminished compared to visually-guided mammals (Mills and Catania, 2004; Nikitina, et al., 2004; Hetling et al, 2005). With this in mind, it is not surprising that NMR EOMs appear underdeveloped.

Mitochondrial content in muscle is dynamic and reflects the functional demands of the fiber type (Lyons et al., 2006). Mitochondria content is also a principal determinant of aerobic capacity. Naked mole-rats show reduced mitochondrial volume density compared to mice. Reduced mitochondrial density is indicative of a reduction of energy demand in these muscles. (Moyes, 2003). As mitochondria are the main generators of ATP, it is possible that this reduction of mitochondria represents a conservation of energy from the eyes, which seem to be used only to detect light or to regulate circadian rhythms (Hetling et al., 2005; Nikitina et al., 2004), so that other systems receive this metabolic gain. The reduced mitochondrial volume density also suggests that naked mole-rat EOMs rely on anaerobic (isolated distribution of small mitochondria) metabolism. Mitochondria may also be important in regulating [Ca2+]i kinetics during the activation of extraocular muscle fibers, influencing force production and increasing the dynamic response range for this muscle group (Andrade et al., 2005). Reduced mitochondrial content suggests the naked mole-rats move their eyes less than mice; which is consistent with small fiber size.

These findings correlate with previously shown findings of an overall less well-developed central visual system; a reduced lateral geniculate nucleus, superior colliculus and visual cortex (Crish et al., 2006; Catania and Remple, 2002). The lack of architectural specialization and small size of EOMs in naked mole-rats suggests that the development of their ocular motor system parallels the visual system. The naked mole-rat provides a novel model to study the coordinated evolution and development of visual and ocular motor systems.


The authors wish to thank Denise Hatala and Gayle Joseph for technical assistance. This work was supported by National Health grant EY12998 (to F.H. Andrade).