This study provides a comprehensive, quantitative analysis of retinal projections to nuclei of the subcortical visual system in congenic Fisher F344-c/+ albino and pigmented rats. Previous studies of albino and pigmented rats described an enhanced contralateral and a reduced ipsilateral retinal projection to thalamic targets (Lund, 1965
; Lund et al., 1974
; Guillery, 1971
; Giolli and Creel, 1973
; Klooster et al., 1983
; Guillery et al., 1984
; Dreher et al., 1985
). This study extends previous work by showing the absence of a coherent pattern in retinal input to subcortical visual nuclei in albino vs. pigmented rats. As predicted, some nuclei of the subcortical visual system show an enhanced contralateral retinal projection in albino rats (VLH, PLi, CPA, IGL), but others do not (SC, VLPO, OPN, vLGN). With the exception of the superior colliculus, no reduction in ipsilateral retinal input was detected in any of the subcortical visual nuclei in albino rats. However, as ipsilateral label was sparse in several of these nuclei, with the methods used in this study small differences would be difficult to detect. The lack of a coherent pattern of augmented/reduced retinal input to subcortical nuclei in albino rats is intriguing. Several factors may be involved including: (1) the spatial origin of retinal ganglion cells giving rise to the aberrant projection, (2) the type of retinal ganglion cell, (3) presence or absence of visuotopic organization within the retinorecipient nucleus, (4) the degree of branching of retinal axons giving rise to the abnormal projection.
Previous studies of albino cats show that visuotopic maps within the primary visual pathway are misaligned (Guillery et al., 1974
). Geniculate layers that receive the normal crossed input from the nasal retina are normal in albinos. Layers that receive input from the temporal retina have normal and abnormal segments. The abnormal segment receives a mirror-reversed input from the central part of the ipsilateral visual field, whereas the normal part receives a normal input from the peripheral part of the contralateral visual field (Guillery et al., 79; Guillery, 1986
). Geniculocortical projections are also misaligned, with the reversed pattern of input also present in the visual cortex (Montero and Guillery, 1978
; Huang and Guillery, 1985
; Akerman et al., 2003
). Nonetheless, the behavior of these animals appears to be normal.
The superior colliculus is also retinotopically organized with a strong projection from the contralateral nasal retina, and a weak but overlapping projection from the ipsilateral temporal retina. Central visual fields are represented more rostrally and peripheral fields more caudally (Huerta and Harting, 1984
). In the Siamese cat which has similarities to albinos with respect to retinal pathways, the ipsilateral projection is reduced to a small region at the caudal pole of the colliculus (Kalil et al., 1971
). Albino ferrets also have a reduced ipsilateral retinocollicular projection (Zhang and Hoffmann, 1993
). However, it is not clear from these studies whether the misdirected contralateral projections replace the missing ipsilateral projections in their locus of termination. We were unable to detect an enhanced contralateral projection in the present study, although this may reflect the sensitivity of the method used. Because the ipsilateral projection is patchy, it would be difficult to detect small increases in misrouted ipsilateral label embedded within the homogeneous contralateral label. Although the role of patches in the superior colliculus is poorly understood, they are thought to represent functional modules (Illing, 1996
). Thus, the absence of patches in the ipsilateral SC albino rats may be a consequence of a reduced retinal contribution to shaping this compartmental organization during development.
The functional consequence of a reduced ipsilateral retinotectal projection is not clear. Previous studies in mice reported that the dark threshold of single units recorded in the superficial layers of the superior colliculus were higher in albino than in pigmented mice (Balkema and Drager, 1991
). Pigmented mice also performed better than albinos in a water maze escape paradigm under the same illuminance conditions, confirming a difference in visual sensitivity (Hayes and Balkema, 1993
). A reduced ipsilateral retinocollicular projection in albino rats may impair visual function and subsequent escape behavior, especially under low illuminance conditions (Overton and Dean, 1988
; Dean et al., 1989
; Brandao et al., 1999
). However, in addition to the SC, other retinorecipient nuclei in the subcortical visual system may be involved in the water maze escape paradigm.
Previous studies of the retinopretectal projection reported a smaller ipsilateral projection to the OPN in Sprague Dawley albino rats by comparison with Lister pigmented rats (Chan et al., 1995
), and a reduced projection in albino rabbits (Klooster et al., 1983
). Additionally, Green et al. (1994)
reported that pupillary light responses (PLR) in congenic albino mice were “smaller and more sluggish” than in pigmented mice. OPN neurons are essential to the pupillary light response (Trejo and Cicerone, 1984
; Clarke and Ikeda, 1985a
; Young and Lund, 1994
; Gamlin and Clarke, 1995
; Buttner-Ennever et al., 1996
). Although our quantitative analysis did not detect differences in retinal input to either the shell or the core of the OPN in albino and pigmented rats, it is possible that the albino mutation could preferentially affect the small population of melanopsin-containing retinal ganglion cells that project to the OPN (Gooley et al., 2001
; Lucas et al., 2003
; Melyan et al., 2005
; Fu et al., 2005
). Melanopsin-containing retinal ganglion cells contribute to the pupillary light response as well as being involved in circadian photoentrainment (Lucas et al., 2003
; Gooley et al., 2003
; Hannibal and Fahrenkrug, 2004
; Freedman et al., 1999
; Berson et al., 2002
; Hatter et al., 2002; Morin et al., 2003
). As the OPN is retinotopically organized (Scalia and Arango, 1979
), even a small change in retinal input from melanopsin-containing neurons in albinos could preferentially diminish the neuronal output of the OPN to a number of targets including those associated with the pupillary light reflex.
We found an enhanced retinal projection to the contralateral IGL in albino rats by comparison with pigmented animals, which is consistent with findings in albino Sprague Dawley vs.
hooded Long Evans rats (Hickey and Spear, 1976
). The IGL has extensive connections with the circadian system, and photic information can access the SCN via the IGL (Johnson et al., 1988
; Zhang and Rusak, 1989
). Additionally, the retina innervates the IGL via
collaterals of the retinohypothalamic tract that project to the SCN (Pickard, 1985
; Morin, 1994
). Previous studies from our lab have shown an enhanced bilateral projection to the SCN in congenic albino rats as well as in albino Lewis vs.
Brown Norway rats (Steininger et al., 1993
; Miller et al., 1996
). An enhanced retinal projection to the SCN and/or the IGL in albinos might result in altered circadian rhythms, as has been reported in albino mice which showed shorter circadian rhythms than pigmented controls (Possidente et al., 1982
). Recently it has been suggested that the IGL may regulate other systems, including those related to visuomotor function and sleep/arousal, by means of its extensive anatomical connections (Morin and Blanchard, 2005
). For example, an augmented retinal projection to IGL in albino rats could preferentially influence REM sleep-promoting neurons in the lateral dorsal tegmental nucleus and pedunculopontine tegmental nucleus with which it is connected (Jones, 2004
Retinal input to the VLPO was sparse and there were no differences between albino and pigmented rats. The VLPO is a small cluster of neurons lateral to the optic chiasm at the level of the rostral SCN, rostral to the supraoptic nucleus (SON) that has a sleep-promoting function (Sherin et al., 1996
; Lu et al., 2000
; Gaus et al., 2002
). Directly caudal to the VLPO lies the VLH, a region that receives dense retinal input in Sprague Dawley and Wistar rats (Lu et al., 1999
; Hannibal and Fahrenkrug, 2004
). Retinal input to the VLH in congenic albino and pigmented F344 rats forms a column (~1 mm rostrocaudal), dorsal and lateral to the optic chiasm, with maximal label at Bregma –1.50 (Paxinos and Watson, 1997
). In contrast to the VLPO, there was a significantly larger retinal projection to the contralateral VLH in albino rats by comparison with pigmented rats (). Several observations suggest that the VLH and VLPO are closely associated, and the VLH may overlap with the “extended VLPO” (Lu et al., 2000
) which has also been implicated in sleep regulation: 1) retinal input is continuous from the rostral tip of the VLPO to the caudal pole of the VLH; 2) following injections of a retrograde tracer into the core of the VLPO, neurons in the VLH were labeled (Chou et al., 2002
), suggesting that VLH neurons project to the VLPO; 3) galanin immunoreactive neurons have been described in the VLPO (Gaus et al., 2002
), and galanin-immunoreactive axons and terminals are also present in the VLH (Behan, unpublished observations). Taken together, these data suggest that the VLPO and the VLH are reciprocally connected. Neurons in the extended VLPO show increased neuronal activity during REM sleep as determined by increased expression of the immediate early gene c-fos
(Lu et al., 2002
). Thus, an enhanced retinal input to the VLH in albino rats could contribute to the REM-sleep triggering observed in response to light to dark transitions in albino rats (Chamberlin et al., 2003
; Benca et al., 1991
; Leung et al., 1992
There were no differences in the retinal projection to the OPN between albino and pigmented rats. Previous studies in our lab have shown that chemical ablation of the pretectum results in the elimination of REM sleep triggering in albino rats following a light to dark transition (Miller et al., 1999
). The identity of the specific pretectal nucleus/nuclei responsible for this behavior is not yet known, although one candidate is the OPN as this nucleus was extensively damaged in all of rats in which REM sleep triggering was eliminated (Miller et al., 1999
). Although we failed to find any difference in CTβ label in either the shell or the core of the OPN between congenic albino and pigmented rats, this does not eliminate the OPN from a potential circuit controlling REM sleep-triggering. Rather, it suggests that other nuclei in the pretectum may also be involved in the control of sleep/wake behaviors. As there are extensive interconnections between virtually all nuclei of the subcortical visual system (Morin and Blanchard, 1998
), OPN neurons may be modulated by inputs from other pretectal nuclei that show significant differences in retinal input in albino vs.
pigmented rats. Although not statistically significant, we found differences in retinal input to the PLi and the CPA between albino and pigmented rats, with albino rats showing an enhanced contralateral projection (). Previous studies in our lab have shown that these two pretectal nuclei are particularly sensitive to acute increases in illuminance, as measured by c-fos
expression (Prichard et al., 2002
). Thus, the PLi and CPA could be involved in mediating REM sleep-triggering in albino, but not pigmented rats. Interestingly, Prichard et al. (2002)
found a significant change in Fos immunoreactivity in the PLi of albino rats following a light to dark transition at zeitgeber time 18 (ZT18; 18 hours after lights-on), but not at ZT6, suggesting possible circadian modulation. Melanopsin-containing RGCs project to the PLi (Hannibal and Fahrenkrug, 2004
), and a recent study has shown that melanopsin expression in RGCs is under circadian control (Hannibal et al., 2005
). Thus retinal input to the PLi could regulate light/dark sensitivity in this nucleus at different circadian times.
Albino animals have been widely used in studies of light mediated behaviors, and numerous studies have compared pigmented and albino strains. However, phenotypically identical albino strains can vary dramatically in their sensitivity to light (LaVail and Lawson, 1986
). The use of a congenic rat strain, in which albino and pigmented animals are genetically identical except for a mutation at the c locus for albinism, provides a powerful tool with which to investigate the structure and function of visual pathways. Albino mammals lack a functional gene for tyrosinase, a necessary enzyme in the production of melanin in the retina. Introduction of a tyrosinase gene results in a normal pigmented retina and chiasmatic pathway (Jeffery et al., 1994, 1997
). By using a congenic rat strain, the present study confirms that many of the previously described differences in retinal projections in other strains of albino and pigmented rats can be attributed to a mutation at (or near) the c locus. In summary, differences in retinal projections to subcortical visual nuclei in albino vs.
pigmented rats may underlie alterations in a range of behaviors. The present study highlights differences between albino and pigmented rats in retinal input to subcortical nuclei that are involved in visuomotor (SC), circadian (IGL, PLi) and sleep/wakefulness (VLH, PLi, CPA) functions.