Thalamus of Reptiles and Mammals: Some Significant Differences

AC

anterior commissure

AD

anterodorsal thalamic nucleus

AM

anteromedial thalamic nucleus

APC

anterior pallial commissure

AV

anteroventral thalamic nucleus

c

caudal

CC

corpus callosum

CL

centrolateral thalamic nucleus

d

dorsal

DC

dorsal cortex

Dla

nucleus dorsolateralis anterior

dp

nucleus of the dorsal peduncle of the lateral forebrain bundle

DThal

dorsal thalamus

DVR

dorsal ventricular ridge

fr

fasciculus retroflexus

Gd

dorsal geniculate nucleus

iam

interanteromedial thalamic nucleus

IC

internal capsule

iN

interstitial nucleus

l

lateral

LFB

lateral forebrain bundle

m

medial

MC

medial cortex

MD

mediodorsal thalamic nucleus

MFB

medial forebrain bundle

mi

massa intermedia

mt

mammillothalamic tract

Pf

parafascicular thalamic nucleus

Po

posterior thalamic nuclear group

PR

perireticular nucleus

Pt

parataenial thalamic nucleus

r

rostral

Rt

reticular thalamic nucleus

SOD

supraoptic decussation

v

ventral

VL

ventrolateral thalamic nucleus

VM

ventromedial thalamic nucleus

VPl

ventral posterolateral thalamic nucleus

VPm

ventral posteromedial thalamic nucleus

A thalamus has been identified in all vertebrates where it is located on either side of the third ventricle [1‒3]. Because of its strategic location in the center of the brain, the thalamus influences information transmitted between the brainstem and telencephalon as well as independently processing these inputs [4].

Any comparison between reptilian and mammalian brains acknowledges several facts. One is that the sheer volume of information on mammals far exceeds that known for reptiles. Second, despite this wealth of information for either class, certain species contribute more to these data than do others. Commonly, these observations in one or more species are then generalized to other species in their respective class which are assumed to share this character. Third, comparisons between these two classes are heavily biased toward mammals. Specifically, data are usually first documented in mammals. Then, comparable information on reptiles is sought and interpreted within the mammalian framework.

Viewed in this light, most comparisons of the thalamus between reptilian and mammalian brains have focused on possible homologous structures shared between these two classes (see for example, [5‒7]. This approach is partly based on the idea that vertebrate brains follow a common organizational scheme, a bauplan. This blueprint is present both in embryos [8‒11] as well as in adults [12, 13]. Much less frequently, differences are highlighted [14]. This review has chosen that less traveled pathway.

This short review is not a discussion of the longitudinal or neuromeric models of forebrain organization as they relate to reptiles and mammals nor is it a critical summary of the development of individual structures within this part of the brain. Recent reports critically address some of these topics [15, 16] and provide the most updated version of the prosomeric model [17]. What this review does highlight are certain key differences in thalamus organization between representative species in these two classes which have not been well recognized. These contrasts fall into two major categories. One is forebrain circuitry while the other is the organization of nuclei in the thalamus. Specific examples are provided for each part.

The dorsal thalamus originates from the alar portion of prosomere 2, whereas the ventral thalamus (also termed prethalamus) arises from the alar part of prosomere 3 [10, 11]. With few exceptions, most reports on adult animals have focused on structures developing from the alar region of prosomeres 2 and 3. In this review, ventral thalamus and prethalamus are used interchangeably.

In crocodiles, nuclei that comprise the alar part of the diencephalon, exclusive of the pretectum and epithalamus, have been described [18, 19]. Since it was unknown which thalamic nuclei in crocodiles arose from the alar part of either prosomere 2 or 3, a proxy for these alar nuclei originating from each prosomere was used. This logic was based on mammalian data which indicated that nuclei that projected to the cortex were dorsal thalamic, whereas those that did not were considered ventral thalamic. Developmental aspects were not considered [3]. Based on this framework, nuclei that had connections with the telencephalic pallium were thought to have originated from alar prosomere 2, whereas those that lacked this circuit were considered to have arisen from alar prosomere 3. This accounted for all of the described diencephalic nuclei in crocodiles exclusive of the pretectum and epithalamus [19]. Studies similar to those described above to distinguish thalamic nuclei in other reptiles are lacking. Along with the absence of these data in other reptiles, differences in nomenclature and clear identification of nuclei and their borders make comparisons between nuclei in various reptiles uncertain. For some nuclei located in the dorsal thalamus, comparisons are straightforward. These include nuclei rotundus, reuniens (medialis), dorsolateralis anterior, dorsomedialis anterior, and the dorsal geniculate nucleus. A similar situation applies to ventral thalamic nuclei. Included in this group are nucleus ovalis, area triangularis, and the ventral geniculate nucleus [19]. However, additional comparisons for the remaining nuclei and those yet to be delineated are problematic.

For this review, the relevant areas of the pallial telencephalon are simplified for the discussion that follows. For a recent detailed description and critical comparison among reptiles, other more comprehensive articles should be consulted [15, 16]. Regions termed cortex were organized into layers and were located between the lateral ventricle and the pial surface. They were simply divided into medial and dorsal areas based on their location. Pallial telencephalic structures located internal to the lateral ventricle where they were bounded by the lateral ventricle on one side and neural tissue on the other were organized as nuclei. Those areas situated dorsal to the dorsal medullary lamina (zona limitans of [20]; pallial-subpallial boundary of [16]) in the transverse plane represented the dorsal ventricle ridge (DVR) [21].

This review focuses on the following two aspects of thalamic organization between reptiles and mammals. One is its extrinsic organization. This refers to the circuitry between the dorsal thalamus and the telencephalon (Fig. 1). The other concerns the intrinsic organization of the thalamus. This latter perspective has two parts. One is the numbers and location of local circuit neurons in the dorsal thalamus. Here, neuronal immunoreactivity to gamma amino butyric acid (GABA) or glutamic acid decarboxylase (GAD) is used as a proxy for local circuit neurons. However, immunohistochemistry, transcriptomics, or some other yet to be described technique may also mark local circuit neurons and differ from the proxy used in the present analysis. This possibility is acknowledged and could influence the interpretation of these data. The other intrinsic character is the morphology of individual thalamic reticular neurons and their relation to the surrounding fiber bundles (Fig. 2, 3).

One of the most striking features of forebrain organization in mammals is the set of reciprocal, ipsilateral connections between individual thalamic nuclei and specific cortical areas. A given dorsal thalamic nucleus projects to a specific part of cortex which then projects back to this same thalamic nucleus. The internal capsule is the fiber tract used in either direction [3]. Bilateral interconnections between the thalamus and cortex have been described in various mammals. Except for hedgehogs [25‒27], the majority of contralateral connections are related to structures located close to the midline [28‒34]. The thalamic massa intermedia [29‒31] and the corpus callosum [28] are suggested to be the commissural sites (Fig. 1).

In contrast to mammals, reptiles have three routes from the dorsal thalamus to the telencephalon (Fig. 1). One is similar to the ipsilateral, mammalian circuit described above. In turtles, the dorsal geniculate nucleus projects to an ipsilateral pallial area [35‒42] which, in turn, sends axons back to the dorsal geniculate nucleus [36, 38]. The lateral forebrain bundle (internal capsule of mammals) contains the axons extending in either direction [35‒39, 41]. A second pathway goes from the nucleus dorsolateralis anterior bilaterally to the medial cortex and uses the medial forebrain bundle [43‒45]. To reach the contralateral target, three commissures have been identified depending on the species: anterior pallial commissure (lizards; [43]), anterior commissure (lizards; [44, 45]), and the supraoptic decussation (crocodiles; unpublished observations). Reciprocal connections from the medial cortex to nucleus dorsolateralis anterior have not been described. A third path connects the vast majority of dorsal thalamic nuclei with the DVR. Individual dorsal thalamic nuclei project to restricted parts of the ipsilateral DVR with their axons utilizing the lateral forebrain bundle [35, 43, 46‒52]. Reciprocal connections from the DVR back to any of these dorsal thalamic nuclei have yet to be documented. In reaching the DVR, fibers from these individual dorsal thalamic nuclei pass through some of the basal nuclei on route to their termination in the DVR. A reciprocal circuit from the basal nuclei to these dorsal thalamic nuclei that project to the DVR has been identified [53]. This fiber tract represents at least one source of feedback to the thalamic nuclei that project to the DVR (Fig. 1).

Individual nuclei in the dorsal thalamus contain both relay and local circuit neurons. The latter are sometimes referred to as interneurons [3, 54]. Relay neurons are commonly glutaminergic (use glutamate as their transmitter), excitatory, and their axons terminate outside their nucleus of origin [3, 54]. Local circuit neurons are commonly GABAergic (use GABA as their transmitter) and inhibitory. Their axons remain within the confines of that given nucleus [3, 54].

In mammals, most dorsal thalamic nuclei have varying numbers of GABA/GAD (+) neurons depending on the nuclear group in question [3]. In reptiles, GABA/GAD (+) neurons are present in the dorsal geniculate nucleus of snakes [55], lizards [55, 56], turtles [55, 57, 58], and crocodiles [59, 60]. In crocodiles, area ventralis anterior, and nucleus diagonalis, possess GAD (+) cells, although these immunopositive neurons are infrequent [19]. Although the data are incomplete, the dorsal geniculate nucleus, area ventralis anterior, and nucleus diagonalis are thought to project to the pallium [19]. Nuclei that project to the DVR and nucleus dorsolateralis anterior, which projects bilaterally to medial cortex, do not have GAD (+) neurons [19, 61]. In addition, scattered GABA (+) cells have been noted surrounding several of the nuclei that project to the DVR as well as in the interconnecting fiber tracts in the turtle thalamus [57]. The significance of this is uncertain.

The most prominent nucleus in the ventral thalamus of mammals is the thalamic reticular nucleus [3]. In mammals, it was originally identified in fiber-stained material because of its reticulated appearance (see [3] for history). Its other prominent features relevant to this review are the following. First, its neurons are homogeneous. The long axis of the soma lies perpendicular to thalamo-cortical and cortico-thalamic fibers and parallel to the orientation of the primary dendrites [62, 63] (see Fig. 2, 3). Second, all neurons are immunoreactive to both GABA/GAD and parvalbumin [3].

Similar to mammals, a thalamic reticular nucleus has been identified in reptiles by its appearance in fiber-stained material [64]. However, it was further characterized by its input from the dorsal thalamus [65‒68]. In crocodiles, its neurons are heterogeneous in their morphology. The long axis of all neurons is not oriented perpendicular to traversing fibers nor are all primary dendrites positioned parallel to the long axis of the soma [24]. Furthermore, crocodiles have two separate nuclear areas in the ventral thalamus that are connected with nuclei of the dorsal thalamus. One, the dorsal peduncular nucleus, is associated with dorsal thalamic nuclei that utilize the lateral forebrain bundle and send axons to end in the DVR. The other, the interstitial nucleus, is associated with nucleus dorsolateralis anterior [24], which uses the medial forebrain bundle to terminate in medial cortex (see Fig. 2, 3). Similar to mammals, its neurons are immunoreactive to both GABA/GAD [65, 67, 69] and parvalbumin [67, 69‒71]. Unlike mammals, not all thalamic reticular neurons are parvalbumin (+) in crocodiles [72].

The information discussed above points to several significant differences in the organization of the thalamus between reptiles and mammals. These are summarized below.

First, reptiles have at least three circuits that transmit information between the thalamus and the telencephalon, whereas mammals have just one. One of these pathways, so far only described for turtles, possesses the reciprocal connections that is a fundamental property of mammalian thalamo-cortical interconnections. A second circuit in reptiles transmits information bilaterally from thalamus to medial cortex, uses a different tract from the one employed in mammals for bilateral projections, and lacks reciprocal connections. Third, the pathway between thalamus and DVR also lacks reciprocal connections from the DVR back to the dorsal thalamus. Its feedback circuit arises, at least in part, from the basal nuclei.

Second, reptilian nuclei that project ipsilaterally to the DVR and bilaterally to medial cortex lack local circuit neurons. Comparable nuclei in mammals possess at least some interneurons.

Third, the morphology of neurons in the thalamic reticular nucleus differs between reptiles and mammals. Furthermore, the thalamic reticular nucleus, at least in crocodiles, is composed of two separate nuclei.

The dearth of functional studies in reptiles has hampered any attempt to explain the consequences of these differences in thalamic organization between reptiles and mammals. One possibility is that these morphological differences merely represent two different ways to solve the same problem. Here is one good example. Owls have entirely crossed retinal projections to the dorsal geniculate complex [73, 74] but have bilateral projections from this nucleus to a visual pallial area known as the visual Wulst [75]. In contrast, primates have bilateral retinal projections to the dorsal geniculate nucleus but only have ipsilateral projections from the dorsal geniculate nucleus to visual cortex [3]. Yet, electrophysiological recordings from the dorsal geniculate complex and the visual Wulst in owls yield response properties very similar to those found in the dorsal geniculate nucleus and the striate cortex in primates [76].

While functional studies will provide important information, I suggest that explanations for these anatomical differences will likely come from two different approaches applied to reptiles: transcriptomics and development. Transcriptomics can answer such questions as the following. One is whether “new” neuron types and their respective circuits have been added or lost during evolution. The other is whether cell types and/or circuits have been re-purposed to be used in a new and different fashion. Some of these questions have already begun to be addressed [54, 77, 78]. The other is through examination of the development of the thalamus and its connections. Both older [79, 80] and more recent developmental studies in reptiles are scarce [81‒86]. Unanswered is the possibility that thalamic development in reptiles differs from that of mammals. Experiments using these two different approaches will provide critical information to understand the organization and evolution of the thalamus in both reptiles and mammals.

I thank Prof. G.F. Striedter whose suggestions and comments greatly improved this manuscript.

The author has no conflict of interest to declare.

No funding was received for this study.

The author analyzed the data, wrote the paper, and made the figures.