Abstract

We present an MRI-based anatomic analysis of a series of seven human brains with the semilobar form of holoprosencephaly. The analysis defines a set of common descriptors for a pattern of topological anomaly which is uniform for the set of seven brains. The core of the anomaly is a rostro-caudally aligned midline gray matter ‘seam’ that extends from the telencephalic-suprachiasmatic junctional region to abut the posterior aspect of the callosal commissure. The seam forms the ventricular roof throughout its extent. Rostrally it is formed by the conjoined heads of caudate/accumbens nuclei. It continues caudally as a gray matter bridge in the fundus of the interhemispheric fissure, where it bridges right and left neocortex. Fornix, septal nuclei and septal limb of the choroid plexus are absent, and the telencephalic ventricles communicate with the diencephalic via open septal limbs of the choroid fissures. By contrast, the temporal limb of hippocampal formation and the choroid plexus are normal and the temporal limb of the choroid fissure is closed. This topological anomaly of conjoined left and right cortical and nuclear gray matter into a midline seam and absent septal structures is thus confined to the region of the midline telencephalic hemisphere evagination. Total telencephalic growth is strongly correlated with the length of this topologically abnormal midline telencephalic segment. The set of findings is consistent with graded failure of induction of rostral to caudal specification in the midline rostral telencephalic zone.

Introduction

Holoprosencephaly (HPE) is the designation given to a range of developmental malformations of the vertebrate CNS in which structures that are normally bilaterally represented in right and left halves of the forebrain are instead conjoined across the midline (DeMyer, 1977; Kundrat, 1882; Siebert et al., 1990; Cohen, 2001). In all vertebrates such conjunction variably involves subcortical telencephalic and diencephalic structures. In the mammalian brain, it variably also involves the cerebral neocortex and fiber systems. Current taxonomy of human forms of HPE emphasizes the degree of separation versus conjunction of the cerebra (DeMyer, 1977; Cohen, 1989a,b; Siebert et al., 1990). Thus, the cerebra of the minimal (lobar) form is represented by two hemispheres that are relatively normal in appearance. Separation of the hemispheres by the interhemispheric fissure is almost complete. In an intermediate (semilobar) form there is cerebral conjunction rostrally, but caudally there is hemispheric pairing with respect to an interhemispheric fissure. Extreme (alobar) forms have no interhemispheric fissure but rather are a single cerebral vesicle. This classification system has been usefully correlated with gravity of disability. Thus, the alobar form is usually not compatible with survival beyond the early postnatal period and is more closely associated with grave craniofacial anomaly than the less extreme forms of the disorder (DeMyer and Zeman, 1963; DeMyer, 1977; Leech and Shuman, 1986; Kobori et al., 1987; Cohen, 1989a,b; Cohen and Sulik, 1992). The most extreme of the alobar forms is the cyclopic malformation where the alobar, fully conjoined telencephalon and diencephalon are associated with a single midline eye or paired but conjoined ocular structures within a single orbit.

Extreme forms have been recognized in human abortuses from the early second gestational month indicating that HPE has its formal origin as early as the period of hemispheric vesicle evagination and perhaps even earlier (Muller and O’Rahilly, 1989). This dating of origin accords with classical experimental manipulations in a range of vertebrates which have implicated a failure of the inductive mechanisms by which signals arising in tissues of the prechordal mesenchyme initiate and direct the course of specification of the ventral forebrain (Shimamura et al., 1995; Golden, 1998, 1999; Cohen and Shiota, 2002). Contemporary molecular genetic analysis and cell biological studies in man and experimental animals have implicated the sonic hedgehog (Shh) and related signal transduction pathways as mediators of the inductive signals (Ming et al., 1998, 2002; Roessler and Muenke, 1998; Briscoe and Ericson, 1999; Nanni et al., 1999, 2000; Wallis et al., 1999; Kelley, 2000; Muenke and Cohen, 2000).

At its most fundamental level, this signal transduction pathway involves multiple sequential and parallel molecular actions linking Shh and other ligands to transcription (Cooper et al., 1998; Golden, 1998, 1999). Accordingly, at least a dozen discrete genetic loci in addition to Shh have been implicated to date (Muenke and Cohen, 2000; Cohen and Shiota, 2002). Yet it appears that those identified account for <20% of the probands which have undergone molecular genetic analysis (Bullen et al., 2001; Cohen and Shiota, 2002).

The histogenetic processes which link transcription to the pairing of forebrain structures and hemispheric evagination pose yet a second level of great complexity (Chiang et al., 1996; Walsh, 1999; Walterhouse et al., 1999; Roux et al., 2000; Gofflot et al., 2001). Primarily these processes must determine the map of structural topology. They must also enable growth behaviors which assure the emergence of form. Current clinical taxonomy and the descriptive approach upon which it rests are central to diagnosis and are usefully predictive with respect to neurological adaptive expectations. However, they are not concerned in a systematic way with the topological patterns of anomaly and their relation to the formal origin or malformation (i.e. with the most direct and explicit molecular and cell biological consequences of the primary inductive process).

We present here an MRI-based anatomic analysis of a series of seven brains with the semilobar holoprosencephalic form of the malformation. The analysis defines a set of common descriptors for a pattern of topological anomaly which is uniform for the set of seven brains. Certain of the descriptors are readily treated quantitatively and are found to correlate with more general reductions in brain and forebrain volumes.

Materials and Methods

HPE Subjects

Patient recruitment was carried out through referral to the Carter Centers for Brain Research in Holoprosencephaly and Related Malformations, a national consortium funded by a nonprofit private foundation. The entry diagnosis of HPE was made on the basis of previous MRI studies. In most this had been confirmed by prior review of the diagnostic images by one of two pediatric neuroradiologists involved in the consortium (E.M. Simon and A.J. Barkovich, University of California, San Francisco, CA). The present subset includes images from all subjects judged to have the classical semilobar form of HPE from a total 16 subjects with HPE in the total study population. Three of the complement were judged to be either lobar or a variant semilobar form with features like the lobar brains. The other six were alobar. The topological features of the remaining nine brains will be the subject of a separate study. The seven comprising the present set, designated the telencephalic 1 (or T1) series, ranged in age from 0.3 to 6.1 years at the time of imaging.

The MRI scans for the present analysis were obtained only for clinical indications, such as new neurological symptoms and/or signs, if there had been no previous diagnostic MRI, or if there were concerns about increased intracranial pressure. The study was approved by the Johns Hopkins University School of Medicine Institutional Review Board. Informed consent was obtained in advance with respect to utilization of the neuroimaging data and the subsequent investigative purposes by the research team of the Carter Centers. Patients were evaluated by a physician and sedation nurse prior to sedation (chloral hydrate at 75 mg/kg, with 25 mg/kg extra dosage if needed) prior to MRI scanning. Patients were rescheduled if there were symptoms of respiratory distress or if there were any active infection. The sedation nurse monitored the patient throughout the imaging session, with the physician nearby. If a patient awoke the imaging session was terminated.

All confirmed HPE cases were also offered the opportunity to participate in an IRB approved study of genetic causes of HPE. Although all subjects included in the present study underwent screening for mutations at the Shh, Zic2, Six3 and TGIF loci, no mutations were found in this cohort.

Normative Subjects

Normal infant and preschool children could not be imaged without clinical indication for the purposes of this study. However, a limited series of normative brain images were available with IRB approval where the studies had been obtained for clinical indications. In all instances the images were judged to be anatomically normal after expert neuroradiological review. The normative series included six children, four male, varying in age from 15 months to 6 years at time of imaging. For all but one, images were obtained because of seizures. In one a seizure had been suspected to have occurred after a mild, possibly concussive head injury. In one a seizure with fever occurred after MMM and DPT inoculations. A sister of the last mentioned, 4 years older and otherwise normal, had had a seizure under the same circumstance and was otherwise judged to be normal. For a single child in the normative series, the indication was mild developmental delay and impaired socialization.

MR Image Acquisition

Subjects with HPE

All subjects with HPE underwent brain MR imaging examination at the Kennedy-Krieger Institute on a 1.5 T MR system (ACS NT Power Trak 6000, Philips Medical Systems, Best, Netherlands) using a standard quadrature head coil. The conventional brain imaging protocol consisted of sagittal T1-weighted spin-echo (SE) (550 ms/20 ms TR/TE), axial T2-weighted double-echo SE (3000 ms/20 ms, 80 ms TR/TE1, TE2), axial fast fluid-attenuated inversion-recovery (FLAIR) (6000 ms/120 ms/2000 ms TR/Teff/TI), coronal three dimensional spoiled gradient recalled echo (3-D SPGR) (35 ms/6ms/45° TR/TE/flip angle, 1.2–1.5 mm slice thickness), coronal fast phase-sensitive inversion recovery (PSIR) (3000 ms/40 ms/200 ms TR/Teff/TI) imaging.

Normative subjects

MR image sets were acquired at the Massachusetts General Hospital on a 1.5 T MR system (Signa; General Electric, Milwaukee, WI) using a standard quadrature head coil. The imaging protocol included a coronal 3-D SPGR (25 ms/6–9 ms/2535° TR/TE/flip angle, 1.5 mm slice thickness) as well as an axial T2-weighted fast spin echo (4000–6000 ms/95–102 ms TR/TEeff).

Our prior experience with morphometry based upon T1 weighted images acquired on the same subject from various high performance imaging systems in different institutions has been that there is excellent accord in volume estimates (Filipek et al., 1994). Given the great volumetric variances encountered among the HPE subjects in the present series and the great difference in volumes when these are compared with the normative brains, any small variance contributed by differential volume estimates from the different imaging systems will be trivial compared to the variances due to differential brain growth.

Anatomic Analysis

The routine of image analysis which precedes the anatomic analysis operation from 3-D T1 weighted image sets has been previously detailed (Filipek et al., 1994). Positional normalization, a step in the processing of the normative brains, could not be done with HPE brains because anomalies of interhemispheric fissure and other mid-line landmarks that are essential to the normalization procedure.

The brain was primarily divided into forebrain, brainstem and cerebellum, and the forebrain was segmented by standard algorithms into cerebral cortex, cerebral white matter, caudate, putamen, pallidum, thalamus, hippocampus, and amygdala (Filipek et al., 1994; Caviness et al., 1996). For the normative brains no adjustment of the routine standard to the laboratory was required. However, there were anomalous anatomically variant features of the HPE brains that were separately segmented by hand driven cursor. At the completion of the segmentation operations, volumes for brain regions were computed from the imaging parameters and the number of voxels assigned to that structure in the course of the segmentation routines (Kennedy et al., 1989).

Morphometry

We seek to characterize both the anatomic volumetric consequences of the disorder as well as to the nature, degree and spatial distribution of the anatomic anomaly. To this end, we make the following classes of observation.

Normative Age-to-volume Growth Function

The age range of the HPE and normative subjects is one of rapid brain growth of the normal brain. Moreover, normative and HPE subjects are not directly matched for age although they distribute through the same age range. For this reason comparisons and correlations to be performed are based upon a ‘normal growth function’ as characterized by a logarithmic function of age derived from our normative series. In general, this is represented as follows:

graphic

where y is brain volume and x is age, and α and β are the fitting coefficients. With this normative function for each anatomic structure, we can then calculate the observed ‘deviation’ that a particular subject has relative to this function (see forebrain volume example in Fig. 1). This formulation of the relationship between brain volume and age is similar to other non linear formulations which have attempted to define this relationship throughout the full fetal through postnatal interval of rapid brain growth (Lemire et al., 1975; Kretschmann et al., 1976; Kretschmann et al., 1986; Pfefferbaum et al., 1994). Here this expression is used as the basis for estimates of the normalized percentage volumes of the brains with HPE.

Topologic Segments

The analysis includes a quantitative treatment of topological divisions in the HPE brain with correlations between these and growth. This required that we first set up parameters by which to express quantitatively the grades and extents of the grades of topological abnormality. For comparative quantitative treatment of anomaly localization we have partitioned the prosencephalon into three segments (Fig. 2) that are normally represented bilaterally: (i) a diencephalic (DD) segment extending from the subthalamus to the suprachiasmatic junction with telencephalon; (ii) a telencephalic (TT) rostral interhemispheric midline segment extending from the DD segment through the septal limb of the prosencephalic ventricles to the hippocampal commissure; and (iii) the DT segment which diverges laterally from the midline to follow the temporal limb of the choroid fissure to the amygdala. This segment is throughout in immediate lateral relation to the diencephalon.

In order to quantify the grades of topological anomaly we simply calculate the rostral to caudal linear extents of the anomaly. Thus each grade of topologic abnormality can be represented either in terms of absolute anterior to posterior length or as a percentile of the total length or length of the DD, TT or DT subregion in which it is observed. For these purposes, anterior-to-posterior length is estimated from the number of coronal image planes for which the grade is present multiplied by the coronal image plane thickness.

Results

The topological anomaly in these seven semilobar cases of HPE is limited to the forebrain (Figs 35). Capsules, diencephalon including subthalamic region, cerebral peduncles, brain stem and cerebellum are normal in this respect. We emphasize in this description two classes of forebrain topological abnormality. (i) Conjunction at the midline of normally paired and separate structures: neocortex, claustrum, heads of caudate and accumbens nuclei (Figs 46), and variably the suprachiasmatic anterior hypothalamus. We have designated as ‘seam’ a gray matter continuity formed by the midline conjunction of these structures. (ii) Absence (or at least non-differentiation) of all septal structures, the fornix and the septal limb of the choroid plexus. These anatomic abnormalities notwithstanding, a substantial set of forebrain structures are present, normally paired and separate in all brains: putamen, pallidum, amygdala and the temporal limbs of the hippocampal formation, including the hippocampal commissure and choroid plexus. The subcortical white matter includes axonal systems that span the midline of the conjoined rostral telencephalon. It also includes the anterior and posterior limbs of the internal capsule and external and extreme capsules, axonal systems that discretely delineate nuclear masses. Finally, it includes still other axonal systems that traverse posterior paired telencephalic regions via the rudimentary posteriorly located corpus callosum. In none of this set of seven brains was there a dorsal cyst, evidence of hydrocephalus or encephalomalacia, secondary processes typically associated with the severer forms of HPE (Brocklehurst, 1973; Probst, 1979; Yokota et al., 1984; Siebert et al., 1990; Simon et al., 2001).

General Cerebral Form and Interhemispheric Fissure

The seven brains with HPE are all semilobar by current diagnostic criteria. Figure 3 shows the surface configuration of mild and severe examples relative to a control subject. We recognize three grades of abnormality with respect to cerebral form (Fig. 7). The first and most severe is where the right and left cortical surfaces are continuous without interhemispheric fissure. This is encountered rostrally, including orbital and polar regions. The appearance of the frontal region is such as to suggest that insular regions are rotated from their normal lateral position to a more medial and orbital position at the expense of orbitofrontral and prefrontal cortical areas (see also Yakovlev, 1959). The claustra, intervening between putamen and insular cortex are conjoined across the midline. The second grade, intermediate in degree, is where there emerges a shallow midsagittal interhemispheric fissure. Still there remains a recognizable cortex continuity crossing the midline at the fundus of the shallow fissure. This is variably encountered as one progresses posteriorly toward the midcoronal level. The fissure deepens to full depth with further progression posteriorly. An orderly slope as the fissure deepens is somewhat disrupted by irregularities of the cortical buckling, particularly rostrally where the fissure begins and cortex is tucked under the more rostral non-fissured cortex. The third and mildest grade is where the fissure reaches full depth and its fundus forms the midline roof of the ventricular system. This is encountered further posteriorly. Where the fundus is the ventricular roof, it is part of a midline gray matter continuity or ‘seam’ to be characterized more fully in the following section.

Caudally the segment of seam forming the fundus of the interhemispheric fissure terminates where it abuts a partial callosal commissure at its midline crossing (Fig. 5). In superior relation to the midline crossing of the callosal commissure, the opposing left and right neocortical faces of the full depth interhemispheric fissure become discretely separate. That is, at the callosal level there is no longer a gray matter continuity forming the fundus of the interhemispheric fissure which instead is formed of the corpus callosum.

The ‘Seam’

The core of the topological anomaly is the rostro-caudally aligned midline gray matter ‘seam’ that lies at the surface of the ventricular cavity throughout its extent (Figs 48). The relationship of seam to the exterior surface of the hemisphere and to deep nuclear structures and ventricular system are presented in the 3-D image reconstruction in Figure 8. The seam arises rostrally in continuity with the anterior suprachiasmatic hypothalamus. At its rostral extreme it is formed of caudal orbitofrontal cortex and the heads of caudate/accumbens nuclei. Tissue normally part of the septal nuclei, if also present, is not recognizable at the resolution of MRI. Anterioventrally, the head of caudate/accumbens is continuous with the anterior ventral putamen. The striatal nuclei become paired and diverge with posterior progression. The rostrum of the seam encases the rostral tip of the forebrain ventricular system. As it continues caudally above the ventricular cavity, it forms a bridge between neocortex of the two hemispheres where the interhemispheric fissure approaches full depth. Whereas its signal intensity is similar to that of cortex, its margins are irregular and fragmented, contrasting with the regularity of contour of laminate neocortex. Caudally it terminates where it abuts the rudimentary body of the corpus callosum.

Septal Structures and the Ventricular System

The topology of the ventricular system is anomalous only in its septal limb where there is continuity between telencephalic and diencephalic ventricular space (Figs 46). Thus, continuing caudally from the callosal decussation, the choroid fissures are closed in relation to fully formed choroid plexus. The ventricular space including trigone and temporal horns is normally paired and separate. Rostrally, on the other hand, the septal limb of the choroid fissure is splayed open. Structures which normally partition the two rostral telencephalic and the midline diencephalic ventricular spaces are absent. These include choroid plexus, septum pellucidum, septal nuclei, fornix and the fiber systems normally separating septal nuclei from orbital and subcallosal neocortex and other basal forebrain nuclear structures. As noted above, not excluded is the presence of undifferentiated septal components in the most rostral extreme of the seam. Foramina of Monro are not defined. Rostral to the genu of the internal capsule the telencephalic ventricular spaces converge with the diencephalic which narrows to a rounded tip at its rostral extreme. This rostral extremity of the ventricular space is ensheathed by the rostral extent of the gray matter seam.

Forebrain Volume and Topologic Anomaly

Volume

The seven HPE subjects in this series show substantially reduced total brain as well as forebrain, telencephalic and diencephalic volume with respect to a series of normative brains of corresponding age (Table 1). They are uniform with respect to the topology of the seam and other forebrain structures but vary with respect to the extent and grade of anomaly within this uniform topology. The growth functions calculated for the normal subjects for each of the volumetric structures was quite good. The estimated average age-adjusted volumetric deviation for the HPE series was ∼55% of the normal expectation for the total brain volume (52, 51 and 68% for forebrain, telencephalon and diencephalons, respectively) but with wide variation in these estimates.

Topologic Anomaly

The method used to estimate age-relative deviation of brain growth from the normative series was also used to estimate the normalized percentage lengths of DD, TT, and DT segments in the HPE brains (Figs 9 and 10). For the overall series the average of the normalized total length DD, TT and DT segments as well as the average for each of these segments is ∼90% of the average normalized values (Table 2). Whereas these average lengths are close to the predicted normal value, there is substantial variability among the individual brains with HPE with coefficients of variation 2- to 4-fold those of the corresponding measures in the normal brains.

The three grades of cerebral anomaly are seen to be limited within the TT segment, involving neither the diencephalon nor the DT or temporal limb of forebrain structures. The relative prominence of the three grades of forebrain abnormality within the TT segment is particularly variable in this series. Thus, the average proportion of the TT segment given to grade 1 (segment with no fissure) is 0.33, grade 2 (segment with partial fissure) is 0.20 and grade 3 (full depth fissure with seam) is 0.48, with coefficients of variation of 0.36, 0.73, and 0.31, respectively (Table 2).

Topologic to Volume Correlations. Strong and highly significant correlations hold in the T1 HPE series between the total length as well as the separate DD and TT segments and the volume of the telencephalon (Table 3). The TT and DT measures do not correlate with the volume of the diencephalon. Correlations between the DT segments and the telencephalon and diencephalon volumes are weak and not significant. In the normative brains, by contrast to the brains with HPE, there was no correlation between the topologic and volumetric parameters. We also correlate brain and forebrain volumetric deficit with length of each grade of forebrain anomaly. Strong and highly significant correlations are observed between the linear extent of the abnormal topologic grade 3 (full-depth fissure with seam) segment but not those of abnormal topologic grades 1 and 2 segments and the telencephalic and diencephalic volumes.

Discussion

Current evidence links the molecular biological origin of the HPE malformation primarily to an inductive failure implemented by signal transduction via the Shh and interactive pathways (Muenke and Cohen, 2000; Cohen and Shiota, 2002). In the first instance, the consequence of induction are the mechanisms of specification, which distinguishes cell populations which are ‘like’ from those that are ‘unlike.’ In the second instance, induction entrains a succession of histogenetic processes and events appertaining to specified populations. Collectively these result in the fundamental topology and organization as well as the topographic features of the collective specified components. These histogenetic processes and events include proliferation, the movements underlying pattern formation, pruning, growth and the assembly of circuitry, each following in a distinctive way for the separated specified cell populations. Thus, paired neocortical and striatal structures will be separate from each other only if septal, dorsal hippocampal structures and the choroid plexus and choroid fissure are specified (Edlund and Jessell, 1999; Cepko, 2001; Livesey and Cepko, 2001) and undergo the normal histogenetic sequences so as to intervene between them.

In this set of brains with semilobar HPE we find that the anomaly of specification is limited to structures lying within the TT topological segment. Specifically structures which are either not specified or anomalous in their specified characteristics are the septal nuclei, dorsal hippocampal the anlagen of the septal hippocampus (induseum griseum and forniceal projection) and rostral midline neocortical structures. Secondly the consequences of inductive failure for the conduct of histogenetic processes in paramedian structures appear to be expressed according to a rostral to caudal gradient of declining severity. In particular there appears to be a rostral to caudal failure of differential growth affecting midline and immediately paramedian (as opposed to lateral convexity) neocortex. Thus, the evagination of the cerebral hemispheres reflects massive proliferation and growth of the telencephalic cell populations where this is most prominent in the convexity as opposed to the midline (Sidman and Rakic, 1982; Caviness et al., 1995; Hebert et al., 2002). A corresponding pattern of differential proliferation and growth underlies the formation of the olfactory bulbs (Gong and Shipley, 1995; Hebert et al., 2002).

Within this framework we interpret the three grades of forebrain anomaly as follows (Figure 11).

  1. Cortex continuity and no fissure reflects total failure of the specification of septal structures normally intervening between left and right cortex or caudate/accumbens heads and also total failure of the sequence of histogenetic processes, in particular differential growth of paramedian with respect to more laterally positioned cortex.

  2. Cortex continuity and partial fissure again reflects total failure of specification of midline septal structures but only partial failure of those histogenetic processes including differential growth so that a partial fissure is achieved.

  3. Seam and full depth fissure appears to reflect only partial failure of specification. Thus, a gray matter structure does intervene between left and right cortex but it is differentiated neither with respect to a cortical nor nuclear character. Whatever their specified fate the cells of this gray matter bridge have not executed migration and other pattern forming maneuvers of normally specified cortical neurons. There has, however, been a sufficient differential in medial and lateral growth to provide a full depth fissure.

Current models of the origin of HPE hold that Shh and related signaling systems appear primarily to set in motion the inductive processes critical to the origin of the malformation. The source of Shh is the prechordal mesenchyme so that its proximate field of effect must be the extreme ventral midline of the neural tube, corresponding to the rostral and ventral boundary region of the diencephalic–telelencephalic anlagen (Shimamura et al., 1995; Weed et al., 1997; Wallis and Muenke, 1999; Patten and Placzek, 2000). This zone would become the region of transition between suprachiasmatic diencephalon and telencephalic lamina terminalis, that is, the TT segment. In each of these brains the topologic abnormality is maximum in this zone. However, in all it extends well beyond this zone, throughout much if not all the TT segment. This corresponds sequentially from rostroventral to dorsal to zones where FGF8 and BMP2/4 rather than Shh, are the primary morphogenetic agents. Whereas experimental evidence has not yet directly implicated these signaling systems in the origin of the HPE malformation, it is pertinent that in the BMPr1a KO there is failure of specification of the septal limb of the choroid plexus at the telencephalic hemisphere margin (Hebert et al., 2002). This is associated with inchoate overgrowth of the affected zone, reproducing a structure not unlike the seam encountered in these brains. Normally the action of BMP morphogen in this zone induces a cessation of proliferation and specification of choroid plexus, both of which are postulated to have been disrupted in this series of brains with HPE. To the extent that the model is valid there is the implication that activations within the dorsal BMP pathways and perhaps also that of FGF8 are downstream contingencies of a normal activation sequence initiated by the Shh signaling system (Hebert et al., 2002).

Whereas the seven brains in the present series are uniform in their topology they are greatly variable respecting the extent and degrees of each of the several patterns of topologic abnormality. Of note is the observation that there are quite substantial direct correlations between the telencephalic segment lengths and forebrain volume deficits implying that the conduct of the inductive process as it leads to specification is also profoundly determinant with respect to cell production and growth. In the normal brain, by contrast, there is essentially no correlation between the topologic segment measures and the variance of actual versus predicted size. Whereas the basis for this dramatic difference is not known, there is at least the suggestion that below some normal threshold of inductive effect there is a dependence of forebrain volume upon the level of inductive effect. Where subthreshold and to a variable degree in the HPE brains, brains are not only small but coefficients of variation are high. In the normal brain with this threshold exceeded, growth is modulated more dominantly by postindeuctive histogenetic interactions.

For the present, the gradations of anomaly in this series can not be matched to a molecular basis where a mutation in the Shh or other currently recognized ‘HPE loci’ has been identified in the series. It is perhaps pertinent in this regard that the HPE anomaly as expressed in these brains is both mild and topologically uniform. With respect to gravity of malformation, a minimum estimate places the incidence of HPE at conception at 1/250 (Matsunaga and Shiota, 1977), with <1/20 000 surviving to term (Bullen et al., 2001), a large proportion of which succumb in early postnatal life (Cohen, 1989a; Bullen et al., 2001). With respect to topological uniformity, the seven brains in this series contrast greatly with the larger set of brains from which they were drawn in our own study series of brains with HPE, and that series is only a limited sampling of the topologic range encountered in the literature. In any event, we infer from both the mildness and uniformity of the series that the points of mutation in the series are not only remotely downstream from the primary Shh inductive signal but also reflect mutation at a convergent point in the multiple pathways that subtend the inductive process. A test of this hypothesis must await further mutation discovery.

This work was supported by grants from the Carter Foundation, NIH Grant DA 09467, by Human Brain Project Grant NS34189, by USPHS grant NS12005, and by grants from the Fairway Trust and the Giovanni Armenise Harvard Foundation for Advanced Scientific Research.

Figure 1. Normal growth curves. This plot demonstrates the volume of the total forebrain as a function of age. The normative subjects (diamonds) demonstrate logarithmic growth as a function of age (curve fit and parameters indicated). For any given HPE subject (squares), the deviation from the normative curve for the subjects age can be calculated.

Figure 1. Normal growth curves. This plot demonstrates the volume of the total forebrain as a function of age. The normative subjects (diamonds) demonstrate logarithmic growth as a function of age (curve fit and parameters indicated). For any given HPE subject (squares), the deviation from the normative curve for the subjects age can be calculated.

Figure 2. Schematic diagram of topographic segments pertaining to developmental staging as superimposed upon a three dimensional reconstruction of the midsagittal aspect of the normal brain. DD is the diencephalic segment extending from the subthalamus to the suprachiasmatic junction with telencephalon. TT is the telencephalic segment extending from the DD segment through the septal limb of the prosencephalic ventricles to the hippocampal commissure. DT is the diencephalic-telencephalic segment which diverges laterally from the midline to follow the temporal limb of the choroid fissure to the amygdala.

Figure 2. Schematic diagram of topographic segments pertaining to developmental staging as superimposed upon a three dimensional reconstruction of the midsagittal aspect of the normal brain. DD is the diencephalic segment extending from the subthalamus to the suprachiasmatic junction with telencephalon. TT is the telencephalic segment extending from the DD segment through the septal limb of the prosencephalic ventricles to the hippocampal commissure. DT is the diencephalic-telencephalic segment which diverges laterally from the midline to follow the temporal limb of the choroid fissure to the amygdala.

Figure 3. Surface reconstructions of control and HPE brains as seen from frontal, superior, inferior and left lateral aspects. A transparent veil is superposed over the rostral region of conjunction of left and right cortex with no intervening interhemispheric fissure. The two HPE brains present the minimal and maximal extremes of midline cortical conjunction from the series. In the minimal example, the zone without interhemispheric fissure extends from the suprachiasmatic through the frontopolar region and the zone of partial fissure into the posterior frontal region. In the maximal example the corresponding zones without fissure and with partial fissure continue further caudally into the frontoparietal region.

Figure 3. Surface reconstructions of control and HPE brains as seen from frontal, superior, inferior and left lateral aspects. A transparent veil is superposed over the rostral region of conjunction of left and right cortex with no intervening interhemispheric fissure. The two HPE brains present the minimal and maximal extremes of midline cortical conjunction from the series. In the minimal example, the zone without interhemispheric fissure extends from the suprachiasmatic through the frontopolar region and the zone of partial fissure into the posterior frontal region. In the maximal example the corresponding zones without fissure and with partial fissure continue further caudally into the frontoparietal region.

Figure 4. Coronal images of control and HPE brains, referenced to cerebral exteriors, with minimal and maximal degrees of malformation. The midforebrain images (B–D) of the control brain illustrate discontinuity of neocortical faces of interhemispheric fissure in superior relation to corpus callosum. The fornices are separated from thalamus by closed choroid fissure and choroid plexus (B–D). Rostrally in the two brains with HPE left and right cortex are continuous at the fundus of a partial depth interhemispheric fissure (A, B). There are no evident septal nuclei, fornix or septal choroid plexus and the ventricular system is continuous through the open septal limbs of the choroid fissures (B, C). Caudally the interhemispheric fissure becomes full depth and left and right cortical faces are bridged at the fundus by a seam of gray matter (C) (masks). The seam may be traced rostrally to where it becomes continuous with the conjoined heads of the caudate encapsulating the anterior tip of a probe shaped tip of the forebrain ventricle. In caudal relation to the corpus callosum, left and right cortex are normally discontinuous in the fundus of the full depth interhemispheric fissure. Occipital and temporal structures are topologically normal (D). In contrast to the abnormal structures of the septal limb, the temporal limb of choroid fissure and hippocampus extending from amygdale to hippocampal commissure (B–D) as well as the topology of internal capsules, thalamus (B), subthalamus, peduncles (C) and lower brain stem and cerebellum (D) are normal.

Figure 4. Coronal images of control and HPE brains, referenced to cerebral exteriors, with minimal and maximal degrees of malformation. The midforebrain images (B–D) of the control brain illustrate discontinuity of neocortical faces of interhemispheric fissure in superior relation to corpus callosum. The fornices are separated from thalamus by closed choroid fissure and choroid plexus (B–D). Rostrally in the two brains with HPE left and right cortex are continuous at the fundus of a partial depth interhemispheric fissure (A, B). There are no evident septal nuclei, fornix or septal choroid plexus and the ventricular system is continuous through the open septal limbs of the choroid fissures (B, C). Caudally the interhemispheric fissure becomes full depth and left and right cortical faces are bridged at the fundus by a seam of gray matter (C) (masks). The seam may be traced rostrally to where it becomes continuous with the conjoined heads of the caudate encapsulating the anterior tip of a probe shaped tip of the forebrain ventricle. In caudal relation to the corpus callosum, left and right cortex are normally discontinuous in the fundus of the full depth interhemispheric fissure. Occipital and temporal structures are topologically normal (D). In contrast to the abnormal structures of the septal limb, the temporal limb of choroid fissure and hippocampus extending from amygdale to hippocampal commissure (B–D) as well as the topology of internal capsules, thalamus (B), subthalamus, peduncles (C) and lower brain stem and cerebellum (D) are normal.

Figure 5. Minimal and maximal degrees of malformation in the present series as viewed in midsagittal and axial planes. The cortical contribution to the seam is covered by masks in the sagittal images. The extent of corpus callosum is greatly different. At its rostral edge it abuts the caudal limit of the seam. Continuity of claustrum across the midline (covered by mask) is evident, in the axial image from the maximally involved. The sagittal views of the HPE brains confirm the normal contours of diencephalon, brain stem and cerebellum.

Figure 5. Minimal and maximal degrees of malformation in the present series as viewed in midsagittal and axial planes. The cortical contribution to the seam is covered by masks in the sagittal images. The extent of corpus callosum is greatly different. At its rostral edge it abuts the caudal limit of the seam. Continuity of claustrum across the midline (covered by mask) is evident, in the axial image from the maximally involved. The sagittal views of the HPE brains confirm the normal contours of diencephalon, brain stem and cerebellum.

Figure 6. Three-dimensional reconstruction of frontal (a, b) and interhemispheric (c, d) perspectives of brain from the T1 series of HPE (b, d) and a normal brain (a, c). The cerebral exterior is projected as a transparency in (c, d). Normally paired caudates (red) are conjoined over the ventricular (purple) anterior tip in HPE. The rostral infolded cortex and cortical contribution to the seam (yellow) are continuous with surface and interhemispheric cortex and are part of a gray matter continuum with caudate heads rostrally. The gray matter continuum terminates caudally at the callosal decussation (not reconstructed). The thalamus (blue) lies deep and medial to the temporal to septal arc of the ventricular system and choroid fissure.

Figure 6. Three-dimensional reconstruction of frontal (a, b) and interhemispheric (c, d) perspectives of brain from the T1 series of HPE (b, d) and a normal brain (a, c). The cerebral exterior is projected as a transparency in (c, d). Normally paired caudates (red) are conjoined over the ventricular (purple) anterior tip in HPE. The rostral infolded cortex and cortical contribution to the seam (yellow) are continuous with surface and interhemispheric cortex and are part of a gray matter continuum with caudate heads rostrally. The gray matter continuum terminates caudally at the callosal decussation (not reconstructed). The thalamus (blue) lies deep and medial to the temporal to septal arc of the ventricular system and choroid fissure.

Figure 7. The three grades of anomaly ranging from no interhemispheric fissure, to partial and full depth fissure with seam are referenced to the appearance in coronal images. Topology is normal caudally in dorsal relation to the corpus callosum.

Figure 7. The three grades of anomaly ranging from no interhemispheric fissure, to partial and full depth fissure with seam are referenced to the appearance in coronal images. Topology is normal caudally in dorsal relation to the corpus callosum.

Figure 8. Three dimensional frontal reconstruction of cerebrum in T1 HPE brain. The conjoined polar region of the cerebrum and the central white matter are computationally removed. As noted earlier by Yakovlev (Yakovlev, 1959), there is apparent continuity of insular cortex across the midline at expense of frontal association regions. The rostral infolded cortex and its continuity with the gray matter seam (yellow), continuing into the caudate heads (red) is visible through the rostral window of the reconstruction. The inner surface of posterior convexity cortex visually flanking the deep gray structures is rendered in blue in order to heighten the apparent depth projection of the central gray matter structures. The temporal tips of the ventricular system (purple) are visible through the windowed anterior temporal lobes.

Figure 8. Three dimensional frontal reconstruction of cerebrum in T1 HPE brain. The conjoined polar region of the cerebrum and the central white matter are computationally removed. As noted earlier by Yakovlev (Yakovlev, 1959), there is apparent continuity of insular cortex across the midline at expense of frontal association regions. The rostral infolded cortex and its continuity with the gray matter seam (yellow), continuing into the caudate heads (red) is visible through the rostral window of the reconstruction. The inner surface of posterior convexity cortex visually flanking the deep gray structures is rendered in blue in order to heighten the apparent depth projection of the central gray matter structures. The temporal tips of the ventricular system (purple) are visible through the windowed anterior temporal lobes.

Figure 9. Three dimensional reconstruction of seam and adjacent structures as seen from the midsagittal aspect in HPE with maximal malformation in the series. The locations of diencephalic-diencephalic (DD), telencephalic-telencephalic (TT) and diencephalic-telencephalic (DT) segments are projected upon the reconstruction.

Figure 9. Three dimensional reconstruction of seam and adjacent structures as seen from the midsagittal aspect in HPE with maximal malformation in the series. The locations of diencephalic-diencephalic (DD), telencephalic-telencephalic (TT) and diencephalic-telencephalic (DT) segments are projected upon the reconstruction.

Figure 10. Linearized schema of diencephalon and rim of telencephalic vesicles in the control and the T1 series of HPE brains. From left to right, the normal sequence is diencephalic-diencephalic (DD), telencephalic-telencephalic (TT) and diencephalic-telencephalic (DT) segments. In the HPE series, the DD and DT segments are topologically normal. The grades of topological anomaly in the TT segment includes ‘no interhemispheric fissure’, followed by ‘partial’ and then ‘full depth interhemispheric fissure’ where there is cortical continuity across midline in the first two grades and seam as continuity for the third. The absolute lengths correspond to the number of inclusive coronal planes multiplied by the coronal plane thickness. These lengths are also represented a percentiles normalized to the total length of the full sequence for each brain.

Figure 10. Linearized schema of diencephalon and rim of telencephalic vesicles in the control and the T1 series of HPE brains. From left to right, the normal sequence is diencephalic-diencephalic (DD), telencephalic-telencephalic (TT) and diencephalic-telencephalic (DT) segments. In the HPE series, the DD and DT segments are topologically normal. The grades of topological anomaly in the TT segment includes ‘no interhemispheric fissure’, followed by ‘partial’ and then ‘full depth interhemispheric fissure’ where there is cortical continuity across midline in the first two grades and seam as continuity for the third. The absolute lengths correspond to the number of inclusive coronal planes multiplied by the coronal plane thickness. These lengths are also represented a percentiles normalized to the total length of the full sequence for each brain.

Figure 11. Schematic representations of model for the projection of the postulated grades of induction failure reconstructed for a stage 14 human embryo. (A, B) The cerebral hemispheres emerge as bilaterally paired evaginations from the single telencephalic vesicle of the terminal neural tube (Muller and O’Rahilly, 1985, 1986, 1987, 1988a,b,c; O’Rahilly and Muller, 1989, 1994) (Figure 8). The anterior arcs of evagination of right and left vesicles are paramedian and parallel within the telencephalic vesicle. The posterior arcs of evagination of right and left vesicles, by contrast, diverge laterally from the midline to follow the right and left telencephalic–diencephalic borders. That is, the telencephalic rims of the two hemispheres arise from the midline but the telencephalic–diencephalic rims of the two hemispheres emerge from opposite sides of the neural tube (C, D) In the present series of HPE brains the topologic anomaly is expressed only within the telencephalic regions of evagination. We have described three grades of anomaly which we suggest correspond to the relative degrees of failure of specification (darkly shaded at midline) and differential growth (lightly shaded paramedian). Thus, the rostral zone where there is no interhemispheric fissure represents the maximum degree of failure of differential growth and failure of specification, the zone of partial fissure represents partial failure of differential growth and complete failure of specification and the zone of cortex to seam continuity at the fundus of a fissure of full depth represents normal differential growth and partial failure of specification

Figure 11. Schematic representations of model for the projection of the postulated grades of induction failure reconstructed for a stage 14 human embryo. (A, B) The cerebral hemispheres emerge as bilaterally paired evaginations from the single telencephalic vesicle of the terminal neural tube (Muller and O’Rahilly, 1985, 1986, 1987, 1988a,b,c; O’Rahilly and Muller, 1989, 1994) (Figure 8). The anterior arcs of evagination of right and left vesicles are paramedian and parallel within the telencephalic vesicle. The posterior arcs of evagination of right and left vesicles, by contrast, diverge laterally from the midline to follow the right and left telencephalic–diencephalic borders. That is, the telencephalic rims of the two hemispheres arise from the midline but the telencephalic–diencephalic rims of the two hemispheres emerge from opposite sides of the neural tube (C, D) In the present series of HPE brains the topologic anomaly is expressed only within the telencephalic regions of evagination. We have described three grades of anomaly which we suggest correspond to the relative degrees of failure of specification (darkly shaded at midline) and differential growth (lightly shaded paramedian). Thus, the rostral zone where there is no interhemispheric fissure represents the maximum degree of failure of differential growth and failure of specification, the zone of partial fissure represents partial failure of differential growth and complete failure of specification and the zone of cortex to seam continuity at the fundus of a fissure of full depth represents normal differential growth and partial failure of specification

Table 1


 Comparative measures of brain volumes (cc)

  Total  Forebrain  Telencephalon  Diencephalon 
Class Age Volume Brain Deviation  Volume Deviation  Volume Deviation  Volume Deviation 
Control-1 2.2 1086.7 1.05   963.7 1.05   943.1 1.05  20.6 1.01 
Control-2 1.3 1012.8 1.05   891.8 1.05   872.9 1.05  18.8 0.98 
Control-3 5.3 1261.2 1.08  1109.5 1.08  1085.3 1.08  24.3 1.10 
Control-4 6.0 1106.8 0.94   988.2 0.95   966.6 0.95  21.6 0.97 
Control-5 3.8 1033.8 0.93   898.0 0.91   878.1 0.91  19.8 0.93 
Control-6 0.3  723.4 0.96   639.7 0.96   623.2 0.96  16.6 1.02 
Mean 3.2 1037.5 1.00   915.2 1.00   894.9 1.00  20.3 1.00 
SD 2.3  177.0 0.07   156.3 0.07   153.8 0.07   2.6 0.06 
CV     0.17 0.07     0.17 0.07     0.17 0.07   0.13 0.06 
             
T1-1 4.2  348.7 0.31   300.6 0.30   288.1 0.29  12.6 0.58 
T1-2 1.6  761.2 0.77   618.3 0.71   603.8 0.71  14.5 0.74 
T1-3 3.0  931.0 0.86   788.9 0.82   770.4 0.82  18.6 0.89 
T1-4 1.6  490.4 0.49   409.1 0.47   397.3 0.46  11.8 0.60 
T1-5 0.9  298.7 0.33   232.8 0.29   223.5 0.28   9.3 0.50 
T1-6 0.1  328.7 0.52   277.0 0.49   268.0 0.49   9.0 0.61 
T1-7 6.1  668.2 0.56   562.0 0.54   543.3 0.53  18.7 0.84 
Mean 2.5  546.7 0.55   455.5 0.52   442.0 0.51  13.5 0.68 
SD 2.1  244.9 0.21   206.8 0.20   203.3 0.20   4.0 0.14 
CV     0.45 0.38     0.45 0.38     0.46 0.39   0.30 0.21 
  Total  Forebrain  Telencephalon  Diencephalon 
Class Age Volume Brain Deviation  Volume Deviation  Volume Deviation  Volume Deviation 
Control-1 2.2 1086.7 1.05   963.7 1.05   943.1 1.05  20.6 1.01 
Control-2 1.3 1012.8 1.05   891.8 1.05   872.9 1.05  18.8 0.98 
Control-3 5.3 1261.2 1.08  1109.5 1.08  1085.3 1.08  24.3 1.10 
Control-4 6.0 1106.8 0.94   988.2 0.95   966.6 0.95  21.6 0.97 
Control-5 3.8 1033.8 0.93   898.0 0.91   878.1 0.91  19.8 0.93 
Control-6 0.3  723.4 0.96   639.7 0.96   623.2 0.96  16.6 1.02 
Mean 3.2 1037.5 1.00   915.2 1.00   894.9 1.00  20.3 1.00 
SD 2.3  177.0 0.07   156.3 0.07   153.8 0.07   2.6 0.06 
CV     0.17 0.07     0.17 0.07     0.17 0.07   0.13 0.06 
             
T1-1 4.2  348.7 0.31   300.6 0.30   288.1 0.29  12.6 0.58 
T1-2 1.6  761.2 0.77   618.3 0.71   603.8 0.71  14.5 0.74 
T1-3 3.0  931.0 0.86   788.9 0.82   770.4 0.82  18.6 0.89 
T1-4 1.6  490.4 0.49   409.1 0.47   397.3 0.46  11.8 0.60 
T1-5 0.9  298.7 0.33   232.8 0.29   223.5 0.28   9.3 0.50 
T1-6 0.1  328.7 0.52   277.0 0.49   268.0 0.49   9.0 0.61 
T1-7 6.1  668.2 0.56   562.0 0.54   543.3 0.53  18.7 0.84 
Mean 2.5  546.7 0.55   455.5 0.52   442.0 0.51  13.5 0.68 
SD 2.1  244.9 0.21   206.8 0.20   203.3 0.20   4.0 0.14 
CV     0.45 0.38     0.45 0.38     0.46 0.39   0.30 0.21 

Deviation of whole brain, forebrain, telencephalon and diencephalon are with reference to the estimated normative means (see Materials and Methods). For these measures whole brain includes forebrain as well as brain stem and cerebellum. Telencephalon and diencephalon are classical separate, complementary partitions of forebrain so that the sum of the respective volumes under telencephalon and diencephalon equal that for forebrain. All volumes are in cc. CV is STD/Average. Ages are given in decimal fractions of years.

Table 2


 Comparative measures of topological segments (mm)

    DD  TT  DT  
Class Age Total length Deviation Length %of total  Deviation  Length %of total  Deviation  Length %of Total  Deviation  
Control-1 2.20 132 1.18 18.0 0.14  1.44  97.5 0.74  1.22  16.5 0.13  0.85  
Control-2 1.30 112.5 1.00  9.0 0.08  0.72  82.5 0.73  1.03  21.0 0.19  1.08  
Control-3 5.30 123 1.10 18.0 0.15  1.44  90.0 0.73  1.13  15.0 0.12  0.77  
Control-4 6.00  91.5 0.82 13.5 0.15  1.08  60.0 0.66  0.75  18.0 0.20  0.92  
Control-5 3.80 126 1.13  9.0 0.07  072  93.0 0.74  1.16  24.0 0.19  1.23  
Control-6 0.30  87 0.78  7.5 0.09  0.6  57.0 0.66  0.71  22.5 0.26  1.15  
Mean  112.0 1.00 12.5 0.11  1.00  80.0 0.71  1.00  19.5 0.18  1.00  
SD   18.8 0.17  4.7 0.04  0.38  17.4 0.04  0.22   3.5 0.05  0.18  
CV    0.17 0.17  0.38 0.32  0.38   0.22 0.06  0.22   0.18 0.28  0.18  
                   
T1-1 4.23  64.5 0.58  9.0 0.14  0.72  45.0 0.70  0.56  10.5 0.16  0.54  
T1-2 1.57 120.4 1.08 16.8 0.14  1.34  85.4 0.71  1.07  18.2 0.15  0.93  
T1-3 3.03 130.5 1.17 16.5 0.13  1.32  96.0 0.74  1.20  18.0 0.14  0.92  
T1-4 1.58 105 0.94  4.5 0.04  0.36  78.0 0.74  0.98  22.5 0.21  1.15  
T1-5 0.94  81.2 0.73  7.0 0.09  0.56  60.2 0.74  0.75  14.0 0.17  0.72  
T1-6 0.13  76.8 0.69  7.2 0.09  0.58  57.6 0.75  0.72  12.0 0.16  0.62  
T1-7 6.10 111 0.99 13.5 0.12  1.08  75.0 0.68  0.94  22.5 0.20  1.15  
Mean   98.5 0.88 10.6 0.11  0.85  71.0 0.72  0.89  16.8 0.17  0.86  
SD   24.6 0.22  4.9 0.04  0.39  17.7 0.03  0.22   4.8 0.03  0.25  
CV    0.25 0.25  0.46 0.33  0.46   0.25 0.04  0.25   0.29 0.16  0.29  
                   
Total abnormality   Grade 1  Grade 2  Grade 3  TT normal 
Class Length % of TT Length % of TT Deviation  Length  % of TT Deviation  Length  % of TT Deviation  Length Deviation 
T1-1  22.5  0.50  9.0 0.20 0.40   6.0  0.13 0.27   7.5  0.17 0.33  22.5 0.50 
T1-2 64.4 0.75  7.0 0.08 0.11  19.6  0.23 0.30  37.8  0.44 0.59  21.0 0.25 
T1-3 73.5 0.77 18.0 0.19 0.24   4.5  0.05 0.06  51  0.53 0.69  22.5 0.23 
T1-4 42.0 0.54 18.0 0.23 0.43   1.5  0.02 0.04  22.5  0.29 0.54  36.0 0.46 
T1-5 21.0 0.35  8.4 0.14 0.40   2.8  0.05 0.13   9.8  0.16 0.47  39.2 0.65 
T1-6 49.2 0.85 14.4 0.25 0.29  21.6  0.38 0.44  13.2  0.23 0.27   8.4 0.15 
T1-7 40.5 0.54 16.5 0.22 0.41   6.0  0.08 0.15  18  0.24 0.44  34.5 0.46 
Mean 44.73 0.61 13.04 0.19 0.33   8.86  0.13 0.20  22.83  0.29 0.48  26.30 0.39 
SD 19.65 0.18  4.78 0.06 0.12   8.20  0.13 0.14  16.01  0.14 0.15  10.84 0.18 
CV  0.44 0.29  0.37 0.31 0.36   0.93  0.96 0.73   0.70  0.4« 0.31   0.41 0.47 
    DD  TT  DT  
Class Age Total length Deviation Length %of total  Deviation  Length %of total  Deviation  Length %of Total  Deviation  
Control-1 2.20 132 1.18 18.0 0.14  1.44  97.5 0.74  1.22  16.5 0.13  0.85  
Control-2 1.30 112.5 1.00  9.0 0.08  0.72  82.5 0.73  1.03  21.0 0.19  1.08  
Control-3 5.30 123 1.10 18.0 0.15  1.44  90.0 0.73  1.13  15.0 0.12  0.77  
Control-4 6.00  91.5 0.82 13.5 0.15  1.08  60.0 0.66  0.75  18.0 0.20  0.92  
Control-5 3.80 126 1.13  9.0 0.07  072  93.0 0.74  1.16  24.0 0.19  1.23  
Control-6 0.30  87 0.78  7.5 0.09  0.6  57.0 0.66  0.71  22.5 0.26  1.15  
Mean  112.0 1.00 12.5 0.11  1.00  80.0 0.71  1.00  19.5 0.18  1.00  
SD   18.8 0.17  4.7 0.04  0.38  17.4 0.04  0.22   3.5 0.05  0.18  
CV    0.17 0.17  0.38 0.32  0.38   0.22 0.06  0.22   0.18 0.28  0.18  
                   
T1-1 4.23  64.5 0.58  9.0 0.14  0.72  45.0 0.70  0.56  10.5 0.16  0.54  
T1-2 1.57 120.4 1.08 16.8 0.14  1.34  85.4 0.71  1.07  18.2 0.15  0.93  
T1-3 3.03 130.5 1.17 16.5 0.13  1.32  96.0 0.74  1.20  18.0 0.14  0.92  
T1-4 1.58 105 0.94  4.5 0.04  0.36  78.0 0.74  0.98  22.5 0.21  1.15  
T1-5 0.94  81.2 0.73  7.0 0.09  0.56  60.2 0.74  0.75  14.0 0.17  0.72  
T1-6 0.13  76.8 0.69  7.2 0.09  0.58  57.6 0.75  0.72  12.0 0.16  0.62  
T1-7 6.10 111 0.99 13.5 0.12  1.08  75.0 0.68  0.94  22.5 0.20  1.15  
Mean   98.5 0.88 10.6 0.11  0.85  71.0 0.72  0.89  16.8 0.17  0.86  
SD   24.6 0.22  4.9 0.04  0.39  17.7 0.03  0.22   4.8 0.03  0.25  
CV    0.25 0.25  0.46 0.33  0.46   0.25 0.04  0.25   0.29 0.16  0.29  
                   
Total abnormality   Grade 1  Grade 2  Grade 3  TT normal 
Class Length % of TT Length % of TT Deviation  Length  % of TT Deviation  Length  % of TT Deviation  Length Deviation 
T1-1  22.5  0.50  9.0 0.20 0.40   6.0  0.13 0.27   7.5  0.17 0.33  22.5 0.50 
T1-2 64.4 0.75  7.0 0.08 0.11  19.6  0.23 0.30  37.8  0.44 0.59  21.0 0.25 
T1-3 73.5 0.77 18.0 0.19 0.24   4.5  0.05 0.06  51  0.53 0.69  22.5 0.23 
T1-4 42.0 0.54 18.0 0.23 0.43   1.5  0.02 0.04  22.5  0.29 0.54  36.0 0.46 
T1-5 21.0 0.35  8.4 0.14 0.40   2.8  0.05 0.13   9.8  0.16 0.47  39.2 0.65 
T1-6 49.2 0.85 14.4 0.25 0.29  21.6  0.38 0.44  13.2  0.23 0.27   8.4 0.15 
T1-7 40.5 0.54 16.5 0.22 0.41   6.0  0.08 0.15  18  0.24 0.44  34.5 0.46 
Mean 44.73 0.61 13.04 0.19 0.33   8.86  0.13 0.20  22.83  0.29 0.48  26.30 0.39 
SD 19.65 0.18  4.78 0.06 0.12   8.20  0.13 0.14  16.01  0.14 0.15  10.84 0.18 
CV  0.44 0.29  0.37 0.31 0.36   0.93  0.96 0.73   0.70  0.4« 0.31   0.41 0.47 

The lengths of linearized DD, TT and DT topological segments and their sums (total length, upper panel, column 3) are given in mm. Values from the controls are given in upper panel and for the T1 series of HPE in the middle panel. For each segment in the control and HPE series, ‘deviations’ refers to the observed segment length as a fraction of the length predicted for age from the growth curve of the normalized subjects. The lower panel provides measures of the lengths of grades 1–3 abnormality within the total TT segment for the HPE brains. Total abnormality refers to the summated lengths of grades 1–3 for each HPE brain in column 2 with the length of the abnormal segment as fraction of the total TT segment length in column 3. Note that there is for each TT segment a normal component lying superior to the corpus callosum. The columns given to grades 1–3 provide corresponding analysis for the respective subsegments of the TT segment. The final set, TT Normal provides corresponding analysis for the normal, that is supracallosal, TT subsegment.

Table 3


 Topologic segment to volume correlations

  T1  Control 
  r2 P  r2 P 
Telenceph. volume Total length 0.927 0.003  0.317 0.244 
 DD length 0.853 0.015  0.631 0.059 
 TT length 0.805 0.006  0.297 0.263 
 DT length 0.328 0.179  0.554 0.090 
Dienceph. volume Total length 0.561 0.053  0.198 0.377 
 DD length 0.626 0.034  0.673 0.046 
 TT length 0.440 0.104  0.175 0.408 
 DT length 0.339 0.170  0.619 0.063 
Total abnorm. L Telenceph. deviation 0.980 0.000  n.a. n.a. 
 Dienceph. deviation 0.575 0.048  n.a. n.a. 
Grade 1 L Telenceph. deviation 0.137 0.414  n.a. n.a. 
 Dienceph. deviation 0.230 0.277  n.a. n.a. 
Grade 2 L Telenceph. deviation 0.080 0.539  n.a. n.a. 
 Dienceph. deviation 0.003 0.901  n.a. n.a. 
Grade 3 L Telenceph. deviation 0.899 0.001  n.a. n.a. 
 Dienceph. deviation 0.574 0.049  n.a. n.a. 
  T1  Control 
  r2 P  r2 P 
Telenceph. volume Total length 0.927 0.003  0.317 0.244 
 DD length 0.853 0.015  0.631 0.059 
 TT length 0.805 0.006  0.297 0.263 
 DT length 0.328 0.179  0.554 0.090 
Dienceph. volume Total length 0.561 0.053  0.198 0.377 
 DD length 0.626 0.034  0.673 0.046 
 TT length 0.440 0.104  0.175 0.408 
 DT length 0.339 0.170  0.619 0.063 
Total abnorm. L Telenceph. deviation 0.980 0.000  n.a. n.a. 
 Dienceph. deviation 0.575 0.048  n.a. n.a. 
Grade 1 L Telenceph. deviation 0.137 0.414  n.a. n.a. 
 Dienceph. deviation 0.230 0.277  n.a. n.a. 
Grade 2 L Telenceph. deviation 0.080 0.539  n.a. n.a. 
 Dienceph. deviation 0.003 0.901  n.a. n.a. 
Grade 3 L Telenceph. deviation 0.899 0.001  n.a. n.a. 
 Dienceph. deviation 0.574 0.049  n.a. n.a. 

The collective values for telencephalic or diencephalic volumes of topologic segment lengths are correlated with measures of telencephalic or diencephalic volumes for the T1 HPE and Normative or Control brain sets. The upper panel provides measures for total topologic segment length for the HPE and Control sets. The lower panel provides measures for total and grades 1–3 subsegment lengths. These measures are not applicable for the control set.

References

Briscoe J, Ericson J (
1999
) The specification of neuronal identity by graded Sonic Hedgehog signalling.
Semin Cell Dev Biol
 
10
:
353
–362.
Brocklehurst G (
1973
) Diencephalic cysts.
J Neurosurg
 
38
:
47
–51.
Bullen PJ, Rankin JM, Robson SC (
2001
) Investigation of the epidemiology and prenatal diagnosis of holoprosencephaly in the North of England.
Am J Obstet Gynecol
 
184
:
1256
–1262.
Caviness V, Takahashi T, Nowakowski R (
1995
) Numbers, time and neocortical neuronogenesis: A general developmental and evolutionary model.
Trends Neurosci
 
18
:
379
–383.
Caviness V, Kennedy DN, Richelme C, Rademacher R, Filipek P (
1996
) The human brain age
7
–11 years: a volumetric analysis based upon magnetic resonance images.
Cereb Cortex
 
6
:
726
–736.
Cepko CL (
2001
) Tackling the brain’s genetic complexity.
Nat Neurosci
  4 Suppl:
1159
–1160.
Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H, Beachy PA (
1996
) Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function.
Nature
 
383
:
407
–413.
Cohen MM Jr (
1989
) Perspectives on holoprosencephaly. Part I. Epidemiology, genetics, and syndromology.
Teratology
 
40
:
211
–235.
Cohen MM Jr (
1989
) Perspectives on holoprosencephaly. Part III. Spectra, distinctions, continuities, and discontinuities.
Am J Med Genet
 
34
:
271
–288.
Cohen MM Jr (
2001
) Problems in the definition of holoprosencephaly.
Am J Med Genet
 
103
:
183
–187.
Cohen MM Jr, Sulik KK (
1992
) Perspectives on holoprosencephaly: Part II. Central nervous system, craniofacial anatomy, syndrome commentary, diagnostic approach, and experimental studies.
J Craniofac Genet Dev Biol
 
12
:
196
–244.
Cohen, MM Jr, Shiota, K (
2002
) Teratogenesis of holoprosencephaly.
Am J Med Genet
 
109
:
1
–15.
Cooper MK, Porter JA, Young KE, Beachy PA (
1998
) Teratogen-mediated inhibition of target tissue response to Shh signaling.
Science
 
280
:
1603
–1607.
DeMyer W (
1977
) Holoprosencephaly (cyclopia–arhinencephaly). In: Handbook of clinical neurology: congenital malformations of the brain and skull, Part I (Vinken P, Bruyn G, eds), pp.
431
–478. Amsterdam: North Holland.
DeMyer W, Zeman W (
1963
) Alobar holoprosencephaly (arhinencephaly) with median cleft lip and palate.
Conf Neurol
 
23
:
1
–36.
Edlund T, Jessell TM (
1999
) Progression from extrinsic to intrinsic signaling in cell fate specification: a view from the nervous system.
Cell
 
96
:
211
–224.
Filipek PA, Richelme C, Kennedy DN, Caviness V (
1994
) The young adult human brain: an MRI-based morphometric analysis.
Cereb Cortex
 
4
:
344
–360.
Gofflot F, Gaoua W, Bourguignon L, Roux C, Picard JJ (
2001
) Expression of Sonic Hedgehog downstream genes is modified in rat embryos exposed in utero to a distal inhibitor of cholesterol biosynthesis.
Dev Dyn
 
220
:
99
–111.
Golden JA (
1998
) Holoprosencephaly: a defect in brain patterning.
J Neuropathol Exp Neurol
 
57
:
991
–999.
Golden JA (
1999
) Towards a greater understanding of the pathogenesis of holoprosencephaly.
Brain Dev
 
21
:
513
–521.
Gong Q, Shipley, MT (
1995
) Evidence that pioneer olfactory axons regulate telencephalon cell cycle kinetics to induce the formation of the olfactory bulb.
Neuron
 .
Hebert J, Mishina Y, McConnell S (
2002
) BMP Signaling is required locally to pattern the dorsal telencephalic midline.
Neuron
 
35
:
1029
–1041.
Kelley RI (
2000
) Inborn errors of cholesterol biosynthesis.
Adv Pediatr
 
47
:
1
–53.
Kennedy DN, Filipek P, Caviness V (
1989
) Anatomic segmentation and volumetric analysis in nuclear magnetic resonance imaging.
IEEE Trans Med Imag
 
7
:
1
–7.
Kobori JA, Herrick MK, Urich H (
1987
) Arhinencephaly. The spectrum of associated malformations.
Brain
 
110
:
237
–260.
Kretschmann H, Schleicher A, Wingert F, Zilles K (
1976
) Morphometrical growth analysis. Verh Anat Ges 70 Pt
2
:
1035
–1041.
Kretschmann HJ, Kammradt G, Krauthausen I, Sauer B, Wingert F (
1986
) Brain growth in man. In: Brain growth (Kretschmann HJ, ed.), pp. 1–26. Basel: Karger.
Kundrat H (
1882
) Arhinencepyhalie als typische Art von Missbildung. Graz: von Leuschner and Lubensky.
Leech RW, Shuman RM (
1986
) Holoprosencephaly and related midline cerebral anomalies: a review.
J Child Neurol
 
1
:
3
–18.
Lemire RJ, Loeser JD, Leech RW, Alvord EC Jr (
1975
) Normal and abnormal development of the human nervous system. Hagerstown, MD: Harper & Row.
Livesey FJ, Cepko CL (
2001
) Vertebrate neural cell-fate determination: lessons from the retina.
Nat Rev Neurosci
 
2
:
109
–118.
Matsunaga E, Shiota K (
1977
) Holoprosencephaly in human embryos: epidemiologic studies of 150 cases.
Teratology
 
16
:
261
–272.
Ming JE, Roessler E, Muenke M (
1998
) Human developmental disorders and the Sonic Hedgehog pathway.
Mol Med Today
 
4
:
343
–349.
Ming JE, Kaupas ME, Roessler E, Brunner HG, Golabi M, Tekin M, Stratton RF, Sujansky E, Bale SJ, Muenke M (
2002
) Mutations in PATCHED-1, the receptor for SONIC HEDGEHOG, are associated with holoprosencephaly.
Hum Genet
 
110
:
297
–301.
Muenke M, Cohen MM Jr (
2000
) Genetic approaches to understanding brain development: holoprosencephaly as a model.
Ment Retard Dev Disabil Res Rev
 
6
:
15
–21.
Muller F, O’Rahilly R (
1985
) The first appearance of the neural tube and optic primordium in the human embryo at stage 10.
Anat Embryol
 
172
:
157
–169.
Muller F, O’Rahilly R (
1986
) The development of the human brain and the closure of the rostral neuropore at stage 11.
Anat Embryol
 
175
:
205
–222.
Muller F, O’Rahilly R (
1987
) The development of the human brain, the closure of the caudal neuropore, and the beginning of secondary neurulation at stage 12.
Anat Embryol
 
176
:
413
–430.
Muller F, O’Rahilly R (
1988
) The first appearance of the future cerebral hemispheres in the human embryo at stage 14.
Anat Embryol
 
177
:
495
–511.
Muller F, O’Rahilly R (
1988
) The development of the human brain from a closed neural tube at stage 13.
Anat Embryol
 
177
:
203
–224.
Muller F, O’Rahilly R (
1988
) The development of the human brain, including the longitudinal zoning in the diencephalon at stage 15.
Anat Embryol
 
179
:
55
–71.
Muller F, O’Rahilly R (
1989
) Mediobasal prosencephalic defects, including holoprosencephaly and cyclopia, in relation to the development of the human forebrain.
Am J Anat
 
185
:
391
–414.
Nanni L, Ming JE, Bocian M, Steinhaus K, Bianchi DW, Die-Smulders C, Giannotti A, Imaizumi K, Jones KL, Campo MD, Martin RA, Meinecke P, Pierpont ME, Robin NH, Young ID, Roessler E, Muenke M (
1999
) The mutational spectrum of the sonic hedgehog gene in holoprosencephaly: SHH mutations cause a significant proportion of autosomal dominant holoprosencephaly.
Hum Mol Genet
 
8
:
2479
–2488.
Nanni L, Croen LA, Lammer EJ, Muenke M (
2000
) Holoprosencephaly: molecular study of a California population.
Am J Med Genet
 
90
:
315
–319.
O’Rahilly R, Muller F (
1986
) The meninges in human development.
J Neuropathol Exp Neurol
 
45
:
588
–608.
O’Rahilly R, Muller F (
1989
) Interpretation of some median anomalies as illustrated by cyclopia and symmelia.
Teratology
 
40
:
409
–421.
O’Rahilly R, Muller F (
1990
) Ventricular system and choroid plexuses of the human brain during the embryonic period proper.
Am J Anat
 
189
:
285
–302.
O’Rahilly R, Muller F (
1994
) Neurulation in the normal human embryo.
Ciba Found Symp
 
181
:
70
–82.
Patten I, Placzek M (
2000
) The role of Sonic hedgehog in neural tube patterning.
Cell Mol Life Sci
 
57
:
1695
–1708.
Pfefferbaum A, Mathalon DH, Sullivan EV, Rawles JM, Zipursky RB, Lim KO (
1994
) A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood.
Arch Neurol
 
51
:
874
–887.
Probst F (
1979
) The porencephalies: neuroradiological appearances and differential diagnsosis. Berlin: Springer Verlag.
Roessler E, Muenke M (
1998
) Holoprosencephaly: a paradigm for the complex genetics of brain development.
J Inherit Metab Dis Dis
 
21
:
481
–497.
Roux C, Wolf C, Mulliez N, Gaoua W, Cormier V, Chevy F, Citadelle D (
2000
) Role of cholesterol in embryonic development.
Am J Clin Nutr
 
71
:
1270S
–1279S.
Shimamura K, Hartigan D, Martinez S, Puelles L, Rubenstein J (
1995
) Longitudinal organization of the anterior neural plate and neural tube.
Development
 
121
:
3923
–3933.
Sidman RL, Rakic P (
1982
) Development of the human central nervous system. In: Histology and histopathology of the nervous system (Haymaker W, Adams RD, eds), pp. 3–145. Springfield, IL: Charles C Thomas.
Siebert J, Cohen MM Jr, Sulik KK, Shaw C-M, Lemire R (
1990
) Holoprosencephaly: an overview and atlas of cases. New York: Wiley-Liss.
Simon EM, Hevner RF, Pinter J, Clegg NJ, Delgado M, Kinsman SL, Hahn JS, Barkovich AJ (
2001
) The dorsal cyst in holoprosencephaly and the role of the thalamus in its formation.
Neuroradiology
 
43
:
787
–791.
Wallis DE, Muenke M (
1999
) Molecular mechanisms of holoprosencephaly.
Mol Genet Metab
 
68
:
126
–138.
Wallis DE, Roessler E, Hehr U, Nanni L, Wiltshire T, Richieri-Costa A, Gillessen-Kaesbach G, Zackai EH, Rommens J, Muenke M (
1999
) Mutations in the homeodomain of the human SIX3 gene cause holoprosencephaly.
Nature Genet
 
22
:
196
–198.
Walsh C (
1999
) Genetic malformations of the human cerebral cortex.
Neuron
 
23
:
19
–29.
Walterhouse DO, Yoon JW, Iannaccone PM (
1999
) Developmental pathways: Sonic Hedgehog-patched-GLI.
Environ Health Perspect
 
107
:
167
–171.
Weed M, Mundlos S, Olsen BR (
1997
) The role of sonic hedgehog in vertebrate development.
Matrix Biol
 
16
:
53
–58.
Yakovlev PI (
1959
) Pathoarchitectonic studies of cerebral malformations. III. Arrhinencephalies (holotelencephalies).
J Neuropathol Exp Neurol
 
18
:
22
–55.
Yokota A, Oota T, Matsukado Y (
1984
) Dorsal cyst malformations. Part I. Clinical study and critical review on the definition of holoprosencephaly.
Childs Brain
 
11
:
320
–341.