Fast versus slow: different saccadic behavior in ...

Ann. N.Y. Acad. Sci. ISSN 0077-8923

A N N A L S O F T H E N E W Y O R K A C A D E M Y O F SC I E N C E S Issue: Basic and Clinical Ocular Motor and Vestibular Research

Fast versus slow: different saccadic behavior in cerebellar ataxias Alessandra Rufa and Pamela Federighi Eye Tracking and Visual Application Laboratory, Department of Neurological, Neurosurgical, and Behavioral Science, University of Siena, Siena, Italy Address for correspondence: Alessandra Rufa, M.D., Ph.D., Eye Tracking and Visual Application Lab, Department of Neurological, Neurosurgical, and Behavioral Science, University of Siena, Siena, Italy. [email protected]

Spinocerebellar ataxia type 2 (SCA2) is a genetic neurodegenerative disorder primarily characterized by involvement of the brainstem and cerebellum, basal ganglia, spinal cord, cerebral cortex, but white matter is also involved. In lateonset cerebellar ataxia (LOCA), the cerebellum is mainly involved, as demonstrated by clinical and neuroradiological findings. These neurodegenerative diseases are often associated with progressive abnormalities in eye movement control, particularly saccadic changes. We recorded saccadic eye movements in eight SCA2 patients and 10 LOCA patients. Here, we suggest that abnormalities in saccadic parameters differ in the two groups of patients according to specific anatomical substrates. The different saccadic behavior observed in these two clinically distinct degenerative cerebellar diseases offers the opportunity to simplify some general mechanisms of saccadic motor control. Like his mentor Fred Plum, John Leigh strongly encouraged younger neuroscientists to tackle neurological problems by investigating “pathological physiology.” With this teaching in mind, we studied patients with rare neurometabolic and neurodegenerative diseases. Keywords: saccade accuracy; saccade velocity; spinocerebellar ataxia type 2; late-onset cerebellar ataxia Preferred citation: Rufa, A. & P. Federighi. 2011. Fast versus slow: different saccadic behavior in cerebellar ataxias. In Basic and Clinical Ocular Motor and Vestibular Research. Janet Rucker & David S. Zee, Eds. Ann. N.Y. Acad. Sci. 1233: 148–154.

Saccadic control Saccades are rapid eye movements shifting foveation from point to point in visual space. The goal of a saccade is to be simultaneously fast and accurate, particularly when the stimulus is unexpected and potentially alerting.1,2 Since the speed of saccades is usually faster than that of visual processing, any correction by visual feedback is denied; thus, saccade accuracy is ensured by an efferent copy of the ongoing motor command (forward model) that allows motor commands themselves to be controlled by predicting the sensory effects of the current movement and modifying movement trajectory if necessary.3–5 In daily living, however, speed and accuracy need not necessarily be maximized for each saccade. In some conditions, saccades may last longer, making visual information available for directing the eye to its goal. Moreover, since the visual space normally explored by humans in daily living is progressively contracting, it is ar-

guable that a visually driven saccade system could be favored by evolution. A useful method for understanding how the saccade system monitors dynamic parameters in various conditions is the application of optimal control theory frameworks to saccadic motor control.6,7 This theory suggests that the saccade system is affected by intrinsic costs that tend to reduce its efficiency. The first cost is the poor vision for target eccentricity; the second is saccade endpoint variability due to signal-dependent noise. In this model, faster saccades are generated by higher motor commands, which are associated with greater noise that in turn produce higher endpoint variability; on the contrary, slow movements are more accurate at the expense of speed.8 To minimize these costs, the saccade system modulates motor commands by controlling the trade-off between speed and accuracy.9,10 Further implementation of this theory exploits neuroeconomic constructs based on the assumption that the relationship between saccade duration and velocity may be doi: 10.1111/j.1749-6632.2011.06126.x

148

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154 

Rufa & Federighi

Fast versus slow: cerebellar ataxias

Functional anatomy of the saccade motor control system

Figure 1. Recording of two 18◦ horizontal saccades made by a SCA2 patient. Saccades of this patient were very slow. The broken line shows target position, and the black line indicates eye position. Positive values of eye position represent rightward saccades, negative values leftward saccades.

represented by a reward temporal discount function.11 In this scenario, the expected value of each saccade is related to the possible reward associated with its goal, depending on the value assigned to the stimulus. The temporal discounting of reward is the cost paid by the system for the duration of a movement that influences its dynamics.11 The oculomotor vermis (OMV) and associated fastigial nucleus (cFN) are thought to be where this modulation occurs.12–14 Another efficient mechanism that normal saccades adopt for maintaining good speed at the expense of accuracy is to undershoot the target when its displacement is >10◦ .15 This is unnecessary for displacements of the target below that value as central vision discriminates it, even if the saccade is less accurate. When saccades are as slow as 100 ms, vision may be used to get them on target. It has already been noticed that normal saccades are open loops, because they are too fast to be modified in flight by visual feedback. Actually, in the case of slow saccades, the visual reference of target position may be continuously available for monitoring eye displacement and if necessary adjusting it via closed-loop (visual feedback).16,17 Recent research on saccade dynamics in patients with two distinct degenerative cerebellar diseases, spinocerebellar ataxia type 2 (SCA2) and late-onset cerebellar ataxia (LOCA), has showed that at least two different solutions are used by the brain to make the saccade subsystem efficient. The first uses a low-speed, visually accurate system compatible with brainstem burst generator failure, as observed in spinocerebellar ataxia type 2 (Fig. 1); the second adopts a less-accurate, high-speed saccade system compatible with cerebellar OMV-cFN changes.18

Significant advances have recently been made in understanding the neural pathways and physiological mechanisms underlying the saccade motor program and its execution at cortical and subcortical levels. Several lines of research have demonstrated that the superior colliculus, brainstem, and cerebellum take part in a circuitry necessary for online correction of saccade amplitude and its rapid adaptation in different behavioral conditions.19–22 Converging evidence from anatomical, pathological, and physiological cell recording studies have demonstrated that the signal generating a saccade of specific amplitude and direction originates in the superior colliculus (SC). The SC contains a dorsal portion that is visual in terms of functions and connections. It also contains a retinotopic representation of space with the fovea represented anteriorly and the retinal periphery posteriorly. The medial layer of the SC contains the motor map in polar coordinates. Its stimulation produces saccades in the opposite direction. The localization of neurons in the topographic map, rather than their spike frequency, codes specific information regarding saccade direction and amplitude following isoamplitude or isodirectional vectors. In this map, rostral neurons discharge before saccades of small amplitude, whereas caudal neurons discharge for saccades of larger amplitudes.23 Electrical stimulation of the medial layer produces a saccade-generating signal that simultaneously propagates without discontinuity to the dorsal layer. This suggests that the sensory effect (visual consequence) of a motor displacement is continuously monitored at that level.24 Saccade motor commands, such as desired displacement vector, encoded by the SC, are sent to the pontine burst generator (BG) by a direct pathway and indirectly though the cerebellar OMV, which receives mossy fibers from the nucleus reticularis tegmenti pontis, conveying motor commands for saccades of any given vector, and ascending fibers from the inferior olives, carrying an error signal.12,25–27 Ascending fibers contact Purkinje cells (PCs) directly, whereas mossy fibers synapse with cell elements in the granular layer before contacting several PCs via parallel fibers.28 Cerebellar OMV output goes to the caudal fastigial nucleus cFN and from there to the contralateral pontine burst

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154 

149

Fast versus slow: cerebellar ataxias

Rufa & Federighi

generator;29 another indirect contingent of fibers reach cortical areas via basal ganglia and thalamus (all these structures may in turn modulate SC activity). The direct pathway is thought to be crucial for controlling the short-term accuracy of saccades, whereas the indirect pathway presumably participates in a more enduring resiliency of trajectories due to target remapping. According to this scheme, the saccade-related cerebellum controls accuracy, probably monitoring motor command by a forward model,3,30,31 in which internal representation of motor command corrects for anticipated errors by rapid modifications of saccade duration.32 In other words, it is involved in predicting the sensory consequences of motor command.33 Although the size and direction of position error are coded elsewhere and sent to the cerebellum as an efference copy,34–36 the OMV is necessary for error correction and rapid adaptation.12,13,37–44 SCA2 and LOCA SCA2 is a rare neurodegenerative disorder with autosomal dominant inheritance due to CAG trinucleotide repeat expansion on chromosome 12q23–24.1 of the gene encoding ataxin-2, a protein involved in RNA splicing and that may confer resistance to degeneration.45 Clinically, SCA2 presents with progressive stance, limb, and truncal ataxia; dysarthria; and a variable combination of other symptoms, including oculomotor disorders, somatosensory deficits, pyramidal, and extrapyramidal dysfunctions, swallowing problems, as well as peripheral neuropathy.46 The neurodegeneration in SCA2 mainly affects the brainstem; the cerebellum is also affected with a region-specific pattern of atrophy including Crus I, lobules VI and VII, and cerebral cortex, particularly frontotemporal areas. Oculomotor disorders in these patients typically include a substantial reduction in saccade velocity. Previous reports indicate that slowing of saccades runs parallel to disease progression and CAG trinucleotide repeat expansion, besides being observed in presymptomatic subjects.47 In this respect, postmortem studies have documented a dramatic reduction in pontine excitatory burst neurons, accounting for the severe slowing of horizontal saccades observed in these patients.48 Anatomical studies in large case series have clarified that the brainstem is affected early and diffusely in SCA2 and that all precerebellar nuclei are targets of neurodegeneration.49

150

LOCA is a group of cerebellar ataxias with undefined genetic mutations but clinical and MRI evidence of isolated cerebellar atrophy, including midline structures.50 Although pathophysiologically and genetically heterogeneous, this group of diseases is clinically homogeneous, being characterized by pure cerebellar ataxia, dysarthria, and dysmetria, including eye movement inaccuracy and often nystagmus. In the study reported here, we evaluated differences in saccade dynamics between eight SCA2 (five male and three female, mean age 46.8 years, range 28–65 years) and 10 LOCA (six male and four female, mean age 51.5, range 39– 63 years) patients, showing that subjects with slow saccades maintained accuracy due to visual feedback, which was active in directing the eyes to their target, whereas subjects with LOCA maintained fast saccades at the price of inaccuracy. Here, we suggest that abnormalities in saccadic parameters differ in the two groups of patients according to specific anatomical substrate. The different saccadic behavior observed in these two clinically distinct degenerative cerebellar diseases offers the opportunity to simplify some general mechanisms of saccadic motor control. Methods Saccades were recorded in a dark room from one eye using high-resolution video-oculography. Stimuli were generated by a microcomputer-controlled LCD stimulator. An interactive program was used for eye calibration. Horizontal and vertical gaze were recorded during the experiment. Data were sampled at a frequency of 240 Hz and stored in a personal computer. The visual stimulus was a red dot with a diameter subtending a visual angle of 0.4◦ presented on a black background (luminance 2.5 cd/m2 ). Each subject’s head was immobilized by a bite bar and chinrest. After the central fixation point was switched off, the visual target appeared to the right or left at four possible horizontal locations (10 and 18◦ of amplitude) in an unpredictable manner without “gap” periods. Subjects were instructed to make a saccade as fast and precisely as possible toward the target. The velocity threshold for identifying the start and end of a saccade was fixed at 10◦ /sec. The dynamic parameters calculated for each subject were speed, duration, amplitude, latency, gain, Q value, absolute error, and variance of saccade amplitude. Comparisons between the two

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154 

Rufa & Federighi

Fast versus slow: cerebellar ataxias

Figure 2. Summary of saccade accuracy of SCA2 and LOCA patients. (A1–2) Absolute error of the initial saccade for 10◦ prosaccade tests. (B1–2) Absolute error of the initial saccade for 18◦ prosaccade tests. Each data point is a subject’s mean. Error bars represent standard deviation.

groups were conducted for each variable using the means and sample variances of all saccadic parameters. Details of signal processing and data analysis have already been reported.18 Results Peak saccade velocity was significantly lower in SCA2 than LOCA patients (10◦ : P < 0.001, 18◦ : P < 0.001). Peak velocity was 159 ± 79◦ /s versus 380 ± 82◦ /s for 10◦ , and 213 ± 124◦ /s versus 453 ± 104◦ /s for 18◦ in SCA2 and LOCA patients, respectively. The mean saccade velocity was also significantly lower in SCA2 than LOCA patients (10◦ : P < 0.001, 18◦ : P < 0.001). Mean velocity was 77 ± 34◦ /s versus 201 ± 42◦ /s for 10◦ and 108 ± 60◦ /s versus 228 ± 50◦ /s for 18◦ in SCA2 and LOCA patients, respectively. Saccade duration was significantly longer in SCA2 than LOCA patents (10◦ : P < 0.05, 18◦ : P < 0.05). The duration was 158 ± 85 ms versus 46 ± 6 ms for 10◦ and 238 ± 162 ms versus 70 ± 18◦ /s for 18◦ in SCA2 and LOCA patients, respectively. Significant differences in saccade accuracy were also evident between the two groups (10◦ : P = 0.001; 18◦ : P < 0.05). Absolute error of initial saccades was higher in LOCA (10◦ : 2.50 ± 0.97◦ ; 18◦ : 4.16 ± 2.51◦ ) than SCA2 patients (10◦ : 1.49 ± 0.40◦ ; 18◦ : 2.24 ± 0.81◦ ). Figure 2 shows the accuracy

performance of all patients for 10◦ and 18◦ target distances. Mean and variance results of saccade amplitude confirmed the accuracy results. The amplitude of the initial saccade was lower in LOCA than SCA2 patients for 10◦ and 18◦ , even though amplitude was not significantly different in two groups of patients. Variance of saccade amplitude was quite high in both LOCA and SCA2 patients. The variance of saccade amplitude was greater in LOCA than SCA2 patients (18◦ : P < 0.05), but was only significant for 18◦ . Main sequence relationships of peak velocity and duration versus amplitude of saccades from SCA2 and LOCA patients are shown in Figure 3A and B. The 95% prediction bounds for controls are also shown.18 Figure 3 shows that saccades of SCA2 patients were outside the prediction bounds for controls. The plot of saccade peak velocity multiplied by duration versus amplitude is shown in Figure 3C. We found a correlation between saccade amplitude and peak velocity in LOCA (P < 0.001) and SCA2 patients (P < 0.001) when we studied the relationship between peak velocity and amplitude; however, LOCA patients were well represented by a linear or an exponential model, whereas SCA2 patients were poorly represented by both.

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154 

151

Fast versus slow: cerebellar ataxias

Rufa & Federighi

Figure 3. Plots of main sequence relationships. Data points are saccades of SCA2 patients (green) and LOCA patients (red). Broken curves indicate 95% prediction bounds for controls. (A) Plot of peak velocity versus amplitude of saccades. (B) Plot of duration versus amplitude of saccades. (C) Plot of peak velocity duration versus amplitude of saccades. The slope of the curve indicates the ratio of peak saccade velocity to mean saccade velocity.

152

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154 

Rufa & Federighi

Fast versus slow: cerebellar ataxias

Conclusions

References

Our study demonstrates for the first time that these two neurodegenerative diseases illustrate different strategies adopted by the brain to maintain the saccade system efficient for vision. According to the main neuropathological findings in these two groups of patients, prevalent degeneration of the OMV-cFN region accounts for fast but inaccurate saccades, such as those observed in LOCA, whereas changes in the pontine BG network determine slow but accurate saccades, such as those of SCA2 patients. These differences are in line with neuroeconomic theories postulating that movements are programmed to minimize costs and maximize rewards, and assuming that temporal discounting of rewards may influence the velocity and duration of saccades.11 Patients with slow saccades also offer a unique opportunity to study the influence of visual feedback on saccade dynamics. It was previously suggested that saccades are ballistic because they are too fast to be modified in flight by visual feedback. In fact, the duration of saccades is usually shorter than the visual sensory input being processed. Actually, saccades lasting longer than a 100 ms, such as those of patients with SCA2, make visual information available for directing the eyes onto the target (closedloop). This implies that the saccade system could be continuously online with the sensory system at subcortical level, acquiring visual references of target position, which may be used in monitoring and adjusting current eye displacement. This mechanism becomes available when saccades are slow enough to be modified in flight by visual feedback and could be commonly used by the visual system. In conclusion, patients with slow saccades, such as SCA2 patients, offer the opportunity to study how the visual system influences saccadic motor control. More generally, the study of these rare neurodegenerative diseases makes it possible to test new hypotheses on mechanisms of brain function.

1. Leigh, R.J. & D.S. Zee. 2006. The Neurology of Eye Movements. 4th ed. Oxford University Press. New York. 2. Ramat, S., R.J. Leigh, D.S. Zee, et al. 2007. What clinical disorders tell us about the neural control of saccadic eye movements. Brain 130: 10–35. 3. Robinson, D.A. 1975. Oculomotor control signals. In Basic Mechanisms of Ocular Motility and Their Clinical Implications. F. Lennerstrand & P. Bach-y-Rita, Eds.: 337–374. Pergamon Press. Oxford. 4. Optican, L.M. 2005. Sensorimotor transformation for visually guided saccades. Ann. N.Y. Acad. Sci. 1039: 132–148. 5. Quaia, C., M. Par´e, R.H. Wurtz, et al. 2000. Extent of compensation for variations in monkey saccadic eye movements. Exp. Brain Res. 132: 39–51. 6. Todorov, E. & M.I. Jordan. 2002. Optimal feedback control as a theory of motor coordination. Nat. Neurosci. 5: 1226– 1235. 7. Shadmehr, R. & J.W. Krakauer. 2008. A computational neuroanatomy for motor control. Exp. Brain Res. 185: 359–381. 8. Harris, C.M. & D.M. Wolpert. 1998. Signal-dependent noise determines motor planning. Nature 394: 780–784. 9. Harris, C.M. & D.M. Wolpert. 2006. The main sequence of saccades optimizes speed-accuracy trade-off. Biol. Cybern. 95: 21–29. 10. van Beers, R. 2008. Saccadic eye movements minimize the consequence of motor noise. PLoS One 3: e2070: 1–8. 11. Shadmehr, R. 2010. Control of movements and temporal discounting of reward. Curr. Opin. Neurobiol. 20: 726–730. 12. Hopp, J.J. & A.F. Fuchs. 2004. The characteristics and neuronal substrate of saccadic eye movement plasticity. Prog. Neurobiol. 72: 27–53. 13. Catz, N., P.W. Dicke & P. Thier. 2005. Cerebellar complex spike firing is suitable to induce as well as to stabilize motor learning. Curr. Biol. 15: 2179–2189. 14. Kojima, Y., R. Soetedjo & A.F. Fuchs. 2010. Changes in simple spike activity of some Purkinje cells in the oculomotor vermis during saccade adaptation are appropriate to participate in motor learning. J. Neurosci. 30: 3715–3727. 15. Henson, D.B. 1978. Corrective saccades: effects of altering visual feedback. Vis. Res. 18: 63–67. 16. Zee, D.S., R.D. Yee, D.G. Cogan, et al. 1976. Ocular motor abnormalities in hereditary cerebellar ataxia. Brain 99: 207– 234. 17. MacAskill, M.R., T.J. Anderson & R.D. Jones. 2000. Suppression of displacement in severely slowed saccades. Vis. Res. 40: 3405–3413. 18. Federighi, P., G. Cevenini, M.T. Dotti, et al. 2011. Differences in saccade dynamics between spinocerebellar ataxia 2 and late-onset cerebellar ataxias. Brain 134: 879–891. 19. Robinson, F.R., A. Straube & A.F. Fuchs. 1993. Role of the caudal fastigial nucleus in saccade generation. II. Effects of muscimol inactivation. J. Neurophysiol. 70: 1741–1758. 20. Takagi, M., D.S. Zee & R.J. Tamargo. 1998. Effects of lesions of the oculomotor vermis on eye movements in primate: saccades. J. Neurophysiol. 80: 1911–1931. 21. Iwamoto, Y. & K. Yoshida. 2002. Saccadic dysmetria following inactivation of the primate fastigial oculomotor region. Neurosci. Lett. 325: 211–215.

Acknowledgments We thank the patients and their families, who willingly gave their time, and continue to support our research. Conflicts of interest The authors declare no conflicts of interest.

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154 

153

Fast versus slow: cerebellar ataxias

Rufa & Federighi

22. Robinson, F.R., A.F. Fuchs & C.T. Noto. 2002. Cerebellar influences on saccade plasticity. Ann. N.Y. Acad. Sci. 956: 155–163. 23. Sparks, D.L. 2002. The brainstem control of saccadic eye movements. Nat. Rev. Neurosci. 3: 952–964. 24. Munoz, D.P. & R.H. Wurtz. 1995. Saccade-related activity in monkey superior colliculus. I. Characteristics of burst and buildup cells. J. Neurophysiol. 73: 2313–2333. 25. B¨uttner-Ennever, J.A. 2008. Mapping the oculomotor system. Prog. Brain Res. 171: 3–11. 26. Iwamoto, Y. & Y. Kaku. 2010. Saccade adaptation as a model of learning in voluntary movements. Exp. Brain Res. 204: 145–162. 27. Azizi, S.A. 2007. And the olive said to the cerebellum: organization and functional significance of the olivo-cerebellar system. Neuroscientist 13: 616–625. 28. Ito, M. 2006. Cerebellar circuitry as a neuronal machine. Prog. Neurobiol. 78: 272–303. 29. Ohtsuka, K. & H. Noda. 1991. The effect of microstimulation of the oculomotor vermis on discharges of fastigial neurons and visually-directed saccades in manques. Neurosci. Res. 10: 290–295. 30. Wolpert, D.M., Z. Ghahramani & M.I. Jordan. 1995. An internal model for sensorimotor integration. Science 269: 1880–1882. 31. Quaia, C., P. Lef`evre & L.M. Optican. 1999. Model of the control of saccades by superior colliculus and cerebellum. J. Neurophysiol. 82: 999 –1018. 32. Chen-Harris, H., W.M. Joiner, V. Ethier , et al. 2008. Adaptive control of saccades via internal feedback. J. Neurosci. 28: 2804–2813. 33. Blakemore, S.J., C.D. Frith & D.M. Wolpert. 2001. The cerebellum is involved in predicting the sensory consequences of action. Neuroreport 12: 1879–1884. 34. Takeichi, N., C.R. Kaneko & A.F. Fuchs. 2007. Activity changes in monkey superior colliculus during saccade adaptation. J. Neurophysiol. 97: 4096–4107. 35. Soetedjo, R., Y. Kojima & A.F. Fuchs. 2008. Complex spike activity in the oculomotor vermis of the cerebellum: a vectorial error signal for saccade motor learning? J. Neurophysiol. 100: 1949–1466. 36. Soetedjo, R., A.F. Fuchs & Y. Kojima. 2009. Subthreshold activation of the superior colliculus drives saccade motor learning. J. Neurosci. 29: 15213–15222.

154

37. Golla, H., K. Tziridis, T. Haarmeier, et al. 2008. Reduced saccadic resilience and impaired saccadic adaptation due to cerebellar disease. Eur. J. Neurosci. 27: 132–144. 38. Tian, J., V. Ethier, R. Shadmehr, et al. 2009. Some perspectives on saccade adaptation. Ann. N.Y. Acad. Sci. 1164: 166– 172. 39. Marr, D. 1969. A theory of cerebellar cortex. J. Physiol. 202: 437–470. 40. Albus, J.S. 1971. A theory of cerebellar function. Math. Biosci. 10: 25–61. 41. Noda, H. & T. Fujikado. 1987. Involvement of Purkinje cells in evoking saccadic eye movements by microstimulation of the posterior cerebellar vermis of monkeys. J. Neurophysiol. 57: 1247–1261. 42. Fuchs, A.F., F.R. Robinson & A. Straube. 1993. Role of the caudal fastigial nucleus in saccade generation. I. Neuronal discharge pattern. J. Neurophysiol. 70: 1723–1740. 43. Scudder, C.A. 2002. Role of the fastigial nucleus in controlling horizontal saccades during adaptation. Ann. N. Y. Acad. Sci. 978: 63–78. 44. Catz, N., P.W. Dicke & P. Thier. 2008. Cerebellar-dependent motor learning is based on pruning a Purkinje cell population response. Proc. Natl. Acad. Sci. U.S.A. 105: 7309– 7314. 45. Gispert, S., R. Twells, G. Orozco, et al. 1993. Chromosomal assignment of the second locus for autosomal dominant cerebellar ataxia (SCA2) to chromosome 12q23–24.1. Nat. Genet. 4: 295–299. 46. Sch¨ols, L., P. Bauer, T. Schmidt, et al. 2004. Autosomal dominant cerebellar ataxias: clinical features, genetics and pathogenesis. Lancet Neurol. 3: 291–304. 47. Vel´azquez-P´erez, L., C. Seifried, M. Abele, et al. 2009. Saccade velocity is reduced in presymptomatic spinocerebellar ataxia Type 2. Clin. Neurophysiol. 120: 632–635. 48. Geiner, S., A.K. Horn, N.H. Wadia, et al. 2008. The neuroanatomical basis of slow saccades in spinocerebellar ataxia type 2 (Wadia-subtype). Prog. Brain Res. 171: 575– 581. 49. R¨ub, U., K. B¨urk, L. Sch¨ols, et al. 2004. Damage to the reticulotegmental nucleus of the pons in spinocerebellar ataxia type 1, 2, and 3. Neurology 63: 1258–1263. 50. Kerber, K.A., J.C. Jen, S. Perlman, et al. 2005. Late-onset pure cerebellar ataxia: differentiating those with and without identifiable mutations. J. Neurol. Sci. 238: 41–45.

c 2011 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1233 (2011) 148–154