- Email: [email protected]
Spontaneous neural activity in the primary visual cortex of retinal degenerated rats
Neuroscience Letters 623 (2016) 42–46
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Research paper
Spontaneous neural activity in the primary visual cortex of retinal degenerated rats Yi Wang a , Ke Chen c , Ping Xu a , Tsz Kin Ng d , Leanne Lai Hang Chan a,b,∗ a
Department of Electronic Engineering, City University of Hong Kong, Hong Kong Center for Biosystems, Neuroscience, and Nanotechnology, City University of Hong Kong, Hong Kong Key Laboratory for Neuroinformation of Ministry of Education, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China d Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong b c
h i g h l i g h t s • Investigated the spontaneous neural activities in the primary visual cortex of RD model. • Neurons in the primary visual cortex of RD model were hyperactive comparing to control group. • Complexity of ISI sequence of spontaneous activities in the primary visual cortex of RD model decreased.
a r t i c l e
i n f o
Article history: Received 26 June 2015 Received in revised form 16 January 2016 Accepted 27 April 2016 Available online 27 April 2016 Keywords: Retinal degeneration Primary visual cortex Spontaneous activity Firing rate LZ complexity
a b s t r a c t Retinal degeneration (RD) models have been widely used to study retinal degenerative diseases for a long time. The biological and electrophysiological presentations of changes in the retina during degeneration progress have been well investigated; thus, the present study is aimed at investigating the electrophysiological effects of RD in the primary visual cortex. We extracellularly recorded the spontaneous neural activities in the primary visual cortex of RD rats. The firing rate, interspike interval (ISI) and Lempel-Ziv (LZ) complexity of spontaneous neural activities were subsequently analyzed. When compared to the control group, it was found that the neurons in primary visual cortex of the RD model fired more frequently. In addition, there was a decrease in LZ complexity of spontaneous neural firing in the RD model. These results suggest that the progress of RD may not only affect the retina itself but also the primary visual cortex, which may result in an unbalanced inhibition-excitation system as well as the decreased arising rate of new patterns of spontaneous activities. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Retinal degeneration is the deterioration of the retina with the consequent death of retinal cells, especially photoreceptors, which results in progressive impairment or total loss of visual function. In order to better understand this disease, various animal models have been produced, such as mouse models rd1, rd2, as well as rat models RCS, P23H, and S334ter. Since mutations in the rhodopsin (Rho) gene are very common in RD and contribute to the majority of known genetic forms of autosomal dominant (ad) retinitis pigmentosa (RP), substantial efforts have been made to develop models
∗ Corresponding author at: Department of Electronic Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon Tong, Kowloon, Hong Kong. E-mail address: [email protected] (L.L.H. Chan). http://dx.doi.org/10.1016/j.neulet.2016.04.062 0304-3940/© 2016 Elsevier Ireland Ltd. All rights reserved.
that express rhodopsin mutation [1]. Among all the rat models of RD, P23H and S334ter express rhodopsin aggregation defect and rhodopsin inactivation defect, respectively [1]. Specifically, S334ter rhodopsin transgenic rats, as a model of several rhodopsin truncation mutations in human RP patients, express rhodopsin gene with an early termination codon at residue 334, resulting in the expression of a rhodopsin protein without 15C-terminal amino acids that are involved in rhodopsin trafficking to the photoreceptor outer segments and in the inactivation of rhodopsin protein after light absorption [2,3]. Although there are several lines of S334ter rat with different rates of photoreceptor degeneration, there is a similar course of photoreceptor degeneration followed by secondary modifications [4,5]. Among the research on RD models, many efforts have been made in the scope of electrophysiology, as electrical activities play an essential role in early development of the nervous system, as well
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
as at later stages in refinement of connections [6]. In particular, spontaneous activities have been considered to be important for the refinement of neural projections and maintenance of topographic maps (neuroanatomy) in the brain [7,8]. There has been substantial evidence showing that altering the pattern of spontaneous activities disrupts this refinement [9]. The spontaneous activities, therefore, have caught major attention in the research community. Notably, an increase in spontaneous hyperactivity of retinal ganglion cells (RGCs) occurs during the progression of retinal degeneration [10–13], and this spontaneous hyperactivity is sustained well into adulthood for weeks after photoreceptors have disappeared [11]. It has been reported that an increase in glutamate concentration and a decrease in photoreceptors partially contributes to the increase in the spontaneous firing rate of ganglion cells in RD models [10,12]. The research on RD models, however, represents a bias towards physiology and electrophysiology of the retina, or cortical organization in the macular degeneration subject by analyzing the fMRI data [14–16]. The data on spontaneous activities in the visual cortex of RD models is still insufficient. Therefore, in the present study, the spontaneous activities, particularly the spiking in primary visual cortex of RD rats, have been investigated, and we expect to know the changes in inherent neural activities in the primary visual cortex of RD model in terms of electrophysiological properties. 2. Material and methods 2.1. Animals and surgical procedures The animals were obtained from Animal Center of Chinese University of Hong Kong, and housed in animal facilities at City University of Hong Kong. All experiments were conducted in accordance with protocols approved by Animal Research Ethics Sub-Committee in City University of Hong Kong and Department of Health, HKSAR. Long Evans and S334ter rats (P80–P90, n = 5 for each group), were used as the control (wild-type, WT) and RD groups, respectively. For surgery, the procedures were similar to those in previous work [17]. The animal was anesthetized with intraperitoneal injection of Ketamine- Xylazine combination (Ketamine: 70 mg/kg, Xylazine: 7 mg/kg) initially, then isoflurane (2%) was applied during the recording. After craniotomy, a bone screw was fixed in the skull and a glasspipette tungsten electrode was placed on the surface of primary visual cortex (AP: −6 ∼ −9 mm; ML: 2 ∼ 4 mm; DV: −0.1 ∼ −1 mm), as the ground and recording electrode, respectively. The exposed cortical surface was covered with 2% agarose to avoid drying and reduce body oscillation caused by breath, then a shield was used to attenuate the environmental interference. The animal was maintained at 38 ◦ C by a warm pad, and an ocular solution was regularly applied to keep the eyes moist during the surgery and recording.
43
ous penetrations and depths in order to reach different layers of the primary visual cortex. In total, three to four locations per penetration, and two to three penetrations were recorded for each animal. For every location, the recording lasted over 5 min. At the end of recording, the animal was euthanized by overdose of Dorminal (300 mg/kg). 2.3. Histology After the recording, eyeballs of both WT and RD animals were enucleated and fixed in 10% formalin in 0.1 M phosphate buffer (pH 7.4) at 4 ◦ C for overnight. Samples were then dehydrated with a graded series of ethanol and xylene and subsequently embedded in paraffin wax. Retinal sections (5 m) with pupil-optic nerve position were stained with hematoxylin and eosin (H&E). Retinal sections were imaged with a light microscrope to assess the status of different layers of retina. 2.4. Data analysis Spike sorting was initially conducted to distinguish between single units. A MATLAB (R2012b, MathWorks, US) package named Wave Clus [18] was used for this purpose, and the units were classified based on the features of their waveforms. After spike sorting, the firing rate of spontaneous activities was calculated. To compare the distribution of firing rates of different groups, we first performed the Shapiro-Wilk test to evaluate the normality of the data distribution. If the data satisfied a normal distribution, a t-test was conducted; otherwise, the Mann-Whitney test was used. The ISI was subsequently obtained. For every single unit, the mean and LZ complexity of ISI sequence were determined for comparing the ISI distribution and complexity of all the units from both WT and RD groups. Specifically, for calculating the LZ complexity, an algorithm based on Abraham Lempel and Jacob Ziv’s work was applied [19]. The statistical tests were as same as above. 3. Results 3.1. Degeneration in RD retina The H/E staining of both WT and RD rat retinas showed that the thickness of degenerated retinas of RD rats at P90 is less than that of the age matched control group (Fig. 1). Specifically, the outer plexiform layer (OPL), outer nuclear layer (ONL) as well as photoreceptor inner and outer segments (IS/OS) had almost disappeared. This validated the condition of retinal degeneration in RD model, which also agrees with previous studies [2,4,5,20]. Moreover, it has been reported that the inner nuclear layer (INL), inner plexiform layer (IPL) and ganglion cell layer (GCL) appear intact even at the age of P226 [21]. In the present study, however, we found that these layers in the RD retinas were also notably atrophied at the age of P90 when compared to the WT retinas.
2.2. Recording 3.2. Spontaneous firing rate The A-M Systems 3600 (A-M Systems, US) and CED Micro 14013 (Cambridge Electronic Design, UK) were used as the amplifier and data acquisition system, respectively. By using a micromanipulator, the recording electrode was lowered into the brain. The reading was recorded as zero reference when the electrode just passed through the cortical surface. As spikes appeared, the electrode was suspended froI m advancing and was allowed to settle for approximately 10 min before recording. The room was kept dark during the extracellular recording of spontaneous activities. The signal was filtered from 300 to 5 kHz and sampled at 25 kHz. The electrodes were placed in vari-
In the present study, 96 single units from 5 WT rats and 89 single units from 5 RD rats were isolated through spike sorting. Then, the representative waveforms of spontaneous activities were plotted on top of the averaged curve (Fig. 2). We have investigated the spontaneous activity of single neurons in both WT and RD groups. First, we determined the firing rates for each unit in both groups. In the WT group, the firing rates displayed a range from 0.11 Hz to 17.49 Hz with the mean ± SD of 3.04 ± 3.77 Hz. In the RD group, the firing rates showed a range from 0.11 Hz to 24.52 Hz with the mean ± SD of 4.05 ± 4.17 Hz.
44
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
Fig. 1. Light microscopy with H&E staining from a normal retina and a degenerated retina. (A) Ten laminar retinal layers are clearly observed in WT (3 months) normal retina. (B) Retinal thickness is reduced in the F1(3 months) retina with the absence of photoreceptor layers. (RPE, retinal pigment epithelium; CC, choriocapillaris; S, sclera).
Fig. 2. Waveforms of spontaneous firing in primary visual cortex of rats. (A) Spontaneous firing in the primary visual cortex of WT rats. (B) Spontaneous firing in the primary visual cortex of RD rats. (C) Two clusters have been sorted out by using principle component analysis (PCA). (D) The corresponding waveforms and their averaged waveforms of the isolated units shown in C.
By performing the Shapiro-Wilk test [22], we found that all the firing rates from both groups were not from the normal distribution (WT, p = 1.4e-10; RD, p = 5.9e-10), which is consistent with the hypothesis that the firing rates probably have a skewed (typically lognormal) distribution [23]. We will further elaborate on this in the discussion section. The Mann-Whitney test was subsequently conducted. The results suggested that the firing rates of spontaneous activity in the primary visual cortex of RD rats were significantly higher (p = 0.0005) than that of WT rats (Fig. 3).
3.3. Interspike interval distribution As an important property of spontaneous activities, the ISI was also examined. To compare the ISI distribution of all the units from both WT and RD groups, the mean ISI was calculated. In the WT group, the mean ISI has an extent from 0.06 s to 9.61 s with the mean ± SD of 1.18 ± 1.58 s. In contrast, for the RD group, the mean ISI has decreased, with the range from 0.04 s to 7.84 s with the mean ± SD of 0.69 ± 1.18 s. The results of Shapiro-Wilk test indicated that the mean ISI from both groups did not satisfy the normal distribution (WT, p = 4.5e-12; RD, p = 2.8e-14). The Mann-Whitney test was then performed. As shown in Fig. 4, although both of the
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
45
Fig. 3. Comparison of the firing rate of spontaneous activity in the primary visual cortex. The results indicate that the V1 cells of the RD model exhibit a higher firing rate than that of the control group. Mann-Whitney test, p < 0.001.
Fig. 4. Comparison of the ISI mean of spontaneous activity in the primary visual cortex. The results show that the V1 cells of the RD model show a decrease in ISI mean compared to that of the control group. Mann-Whitney test, p < 0.001.
mean ISI were skewed, the difference between the two groups was significant (p = 0.0006). 3.4. LZ complexity of ISI sequence Furthermore, with the purpose of providing a better understanding about spontaneous firing pattern, the complexity of ISI sequence was also investigated (Fig. 5). In present study, LZ complexity has been employed for the complexity analysis. In the WT group, LZ complexity of ISI sequence was ranged from 0.36 to 1.28 with the mean ± SD of 0.89 ± 0.16, while for the RD group, LZ complexity displayed an extent from 0.34 to 1.14 with the mean ± SD of 0.79 ± 0.17. By carrying out the ShapiroWilk test, we found that the LZ complexities of ISI sequence of WT and RD groups come from a skewed distribution (p = 0.0437) and normal distribution (p = 0.2798), respectively. The Mann-Whitney test was subsequently performed, and the results indicated that the overall LZ complexity of ISI sequence of the RD group was significantly lower than that of the WT group (p = 2.4822e-05).
Fig. 5. Comparison of the LZ complexity of the ISI sequence. The LZ complexity of the ISI sequence of RD group exhibits an evident decrease compared to that of the WT group. Mann-Whitney test, p < 0.001.
4. Discussion Consistent with the findings in retina [10,11,24], the spontaneous activities in primary visual cortex of RD model exhibited an increased firing rate, which suggests a thorough refinement presumably occurred along the visual pathway. In addition, the ISI sequence did not only have a smaller mean value but also represented a decrease in complexity, implying that the cortical circuitry
of visual system has impaired capability for generating new patterns of inherent activities. Therefore, the changes in spontaneous activities in primary visual cortex of RD model may provide better understanding of the progress of retinal degenerative diseases. Given that the spontaneous activities in primary visual cortex of RD rats exhibited higher firing rate than that of WT rats, it can be confirmed that the visual system could be affected by the progress
46
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
of degeneration not only in early phase but also in later stages in terms of both morphology and function, such as circuitry reformation and retinotopy reorganization, due to the neural plasticity. According to the views of Hubel and Wiesel, the abnormality in the eyes could cause morphological changes in geniculate and defect in striate cortex [25]. Moreover, our results are also consistent with the finding that a significant degeneration in the cortical processing of visual information before development has been properly completed [26]. In addition, we found that neurons in primary visual cortex of RD model possess smaller ISI than that of control group, which agrees with our preceding results about firing rate. Besides, we also investigated the complexity of ISI sequence. In general, the ISI sequence of spontaneous activities could display nonrandom internal structure, and the degree of this structural nonrandomness could be quantified by studying its complexity [27]. The concept of complexity provides a promising tool for clinical study that may benefit the evaluation of the disease progress or treatment outcomes [28]. LZ complexity has been extensively applied to measure the generation rate of new patterns along a digital sequence and is also closely related to important information-theoretic properties such as compression ratio, redundancy and coding efficiency [29,30]. In recent years, LZ complexity has been widely used in the context of biomedical signal analysis, for instance, to characterize the spike trains with estimation of entropy, complexity curve and the number of neural source states [29], or to study mental disorders [31]. In general, lower complexity indicates higher regularity, which is consistent with previous observation that neurons that fire more rapidly fire more regularly [32]. Combined with our results of LZ complexity, it could be implied that the cortical visual system of RD model has attenuated capability to generate new patterns of neural activity, which is related to instant responses to stimuli or lasting changes of circuitry that profoundly depend on neural plasticity. On the other hand, since only the pattern of spontaneous activities was studied in present study, the question regarding responsive activities remains unanswered. Nevertheless, we did record responsive activities in primary visual cortex under light stimuli (data not shown), and we found that even for the rats at late stage of retinal degeneration, weak visual responses can still be observed. Further studies are required to address this question. In summary, our results suggest that the inhibition-excitation system in the primary visual cortex of RD model probably has been disrupted, resulting in an increase in firing rate of spontaneous activities. Moreover, the decreased LZ complexity of ISI sequence indicated a lower arising rate of new patterns of spontaneous activities in RD model. These findings therefore implied that the degeneration progress in RD model does not only affect the retina itself but also the primary visual cortex. Finally, to validate these implications, other approaches, such as molecular and cellular signaling, are needed. Acknowledgement This work was supported by the Research Grants Council of the Hong Kong Special Administrative Region, China [grant number CityU 123412, CityU 111013 and CityU 7004433]; the National Science Foundations of China [grant number 31300912]; the Guangdong Innovative and Entrepreneurial Research Team Program [grant number 2013S046]; and the Shenzhen Peacock Plan. References [1] R.E. Marc, B.W. Jones, C.B. Watt, E. Strettoi, Neural remodeling in retinal degeneration, Prog. Retin. Eye Res. 22 (2003) 607–655.
[2] V.M. Shinde, O.S. Sizova, J.H. Lin, M.M. LaVail, M.S. Gorbatyuk, Er stress in retinal degeneration in s334ter Rho rats, PLoS One 7 (2012) e33266. [3] J. Chen, C.L. Makino, N.S. Peachey, D.A. Baylor, M.I. Simon, Mechanisms of rhodopsin inactivation in vivo as revealed by a cooh-terminal truncation mutant, Science 267 (1995) 374–377. [4] G. Martinez-Navarrete, M.J. Seiler, R.B. Aramant, L. Fernandez-Sanchez, I. Pinilla, N. Cuenca, Retinal degeneration in two lines of transgenic S334ter rats, Exp. Eye Res. 92 (2011) 227–237. [5] A. Ray, G.J. Sun, L. Chan, N.M. Grzywacz, J. Weiland, E.-J. Lee, Morphological alterations in retinal neurons in the S334ter-line3 transgenic rat, Cell Tissue Res. 339 (2010) 481–491. [6] N.C. Spitzer, Electrical activity in early neuronal development, Nature 444 (2006) 707–712. [7] M.G. Hanson, L.T. Landmesser, Increasing the frequency of spontaneous rhythmic activity disrupts pool-specific axon fasciculation and pathfinding of embryonic spinal motoneurons, J. Neurosci. 26 (2006) 12769–12780. [8] N.X. Tritsch, E. Yi, J.E. Gale, E. Glowatzki, D.E. Bergles, The origin of spontaneous activity in the developing auditory system, Nature 450 (2007) 50–55. [9] L.A. Kirkby, G.S. Sack, A. Firl, M.B. Feller, A role for correlated spontaneous activity in the assembly of neural circuits, Neuron 80 (2013) 1129–1144. [10] E.L. Nylen, Quantitative analysis of retinal function in early stages of retinal degeneration in the rd1 mouse, (2010) http://ir.uiowa.edu/etd/565/. [11] S.F. Stasheff, Emergence of sustained spontaneous hyperactivity and temporary preservation of off responses in ganglion cells of the retinal degeneration (rd1) mouse, J. Neurophysiol. 99 (2008) 1408–1421. [12] D.J. Margolis, P.B. Detwiler, Cellular origin of spontaneous ganglion cell spike activity in animal models of retinitis pigmentosa, J. Ophthalmol. 2011 (2011) 1–6. [13] S.F. Stasheff, M. Shankar, M.P. Andrews, Developmental time course distinguishes changes in spontaneous and light-evoked retinal ganglion cell activity in rd1 and rd10 mice, J. Neurophysiol. 105 (2011) 3002–3009. [14] H.A. Baseler, A. Gouws, K.V. Haak, C. Racey, M.D. Crossland, A. Tufail, et al., Large-scale remapping of visual cortex is absent in adult humans with macular degeneration, Nat. Neurosci. 14 (2011) 649–655. [15] C.I. Baker, Reorganization of visual processing in macular degeneration, J. Neurosci. 25 (2005) 614–618. [16] Y. Shao, G.A. Keliris, A. Papanikolaou, M.D. Fischer, D. Zobor, H. Jägle, et al., Visual cortex organisation in a macaque monkey with macular degeneration, Eur. J. Neurosci. 38 (2013) 3456–3464. [17] C.M. Niell, M.P. Stryker, Highly selective receptive fields in mouse visual cortex, J. Neurosci. 28 (2008) 7520–7536. [18] R. Quiroga, Z. Nadasdy, Y. Ben-Shaul, Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering, Neural Comput. 16 (2004) 1661–1687. [19] A. Lempel, J. Ziv, Complexity of finite sequences, IEEE Trans. Inf. Theory 22 (1976) 75–81. [20] B.W. Jones, R.E. Marc, Retinal remodeling during retinal degeneration, Exp. Eye Res. 81 (2005) 123–137. [21] D.J. Margolis, G. Newkirk, T. Euler, P.B. Detwiler, Functional stability of retinal ganglion cells after degeneration-induced changes in synaptic input, J. Neurosci. 28 (2008) 6526–6536. [22] N.M. Razali, Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling tests, J. Stat. Model. Anal. 2 (2011) 21–33. [23] G. Buzsáki, K. Mizuseki, The log-dynamic brain: how skewed distributions affect network operations, Nat. Rev. Neurosci. 15 (2014) 264–278. [24] C. Sekirnjak, L.H. Jepson, P. Hottowy, A. Sher, W. Dabrowski, A.M. Litke, et al., Changes in physiological properties of rat ganglion cells during retinal degeneration, J. Neurophysiol. 105 (2011) 2560–2571. [25] D.H. Hubel, T.N. Wiesel, Effects of monocular deprivation in kittens, Naunyn Schmiedebergs Arch. Pharmacol. 248 (1964) 492–497. [26] C. Gias, A. Vugler, J. Lawrence, A.J. Carr, L.L. Chen, A. Ahmado, et al., Degeneration of cortical function in the Royal College of Surgeons rat, Vision Res. 51 (2011) 2176–2185. [27] P.E. Rapp, I.D. Zimmerman, E.P. Vining, N. Cohen, A.M. Albano, M.A. Jimenez-Montano, The algorithmic complexity of neural spike trains increases during focal seizures, J. Neurosci. 14 (1994) 4731–4739. [28] A.C. Yang, S.-J. Tsai, Complexity of mental illness: a new research dimension, Prog. Neuropsychopharmacol. Biol. Psychiatry 45 (2013) 251–252. ´ [29] J. Szczepanski, J.M. Amigó, E. Wajnryb, M.V. Sanchez-Vives, Characterizing spike trains with Lempel–Ziv complexity, Neurocomputing 58–60 (2004) 79–84. [30] M. Aboy, R. Hornero, D. Abasolo, D. Alvarez, Interpretation of the Lempel-Ziv complexity measure in the context of biomedical signal analysis, IEEE Trans. Biomed. Eng. 53 (2006) 2282–2288. [31] C. Gómez, R. Hornero, A. Fernández, D. Abásolo, M. López, Decreased Lempel-Ziv complexity in Alzheimer’s disease patients’ magnetoencephalograms, Conf. Proc. IEEE Eng. Med. Biol. Soc. (2005) 4514–4517. [32] B.D. Bennett, C.J. Wilson, Spontaneous activity of neostriatal cholinergic interneurons in vitro, J. Neurosci. 19 (1999) 5586–5596.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Research paper
Spontaneous neural activity in the primary visual cortex of retinal degenerated rats Yi Wang a , Ke Chen c , Ping Xu a , Tsz Kin Ng d , Leanne Lai Hang Chan a,b,∗ a
Department of Electronic Engineering, City University of Hong Kong, Hong Kong Center for Biosystems, Neuroscience, and Nanotechnology, City University of Hong Kong, Hong Kong Key Laboratory for Neuroinformation of Ministry of Education, School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, China d Department of Ophthalmology and Visual Sciences, The Chinese University of Hong Kong, Hong Kong b c
h i g h l i g h t s • Investigated the spontaneous neural activities in the primary visual cortex of RD model. • Neurons in the primary visual cortex of RD model were hyperactive comparing to control group. • Complexity of ISI sequence of spontaneous activities in the primary visual cortex of RD model decreased.
a r t i c l e
i n f o
Article history: Received 26 June 2015 Received in revised form 16 January 2016 Accepted 27 April 2016 Available online 27 April 2016 Keywords: Retinal degeneration Primary visual cortex Spontaneous activity Firing rate LZ complexity
a b s t r a c t Retinal degeneration (RD) models have been widely used to study retinal degenerative diseases for a long time. The biological and electrophysiological presentations of changes in the retina during degeneration progress have been well investigated; thus, the present study is aimed at investigating the electrophysiological effects of RD in the primary visual cortex. We extracellularly recorded the spontaneous neural activities in the primary visual cortex of RD rats. The firing rate, interspike interval (ISI) and Lempel-Ziv (LZ) complexity of spontaneous neural activities were subsequently analyzed. When compared to the control group, it was found that the neurons in primary visual cortex of the RD model fired more frequently. In addition, there was a decrease in LZ complexity of spontaneous neural firing in the RD model. These results suggest that the progress of RD may not only affect the retina itself but also the primary visual cortex, which may result in an unbalanced inhibition-excitation system as well as the decreased arising rate of new patterns of spontaneous activities. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Retinal degeneration is the deterioration of the retina with the consequent death of retinal cells, especially photoreceptors, which results in progressive impairment or total loss of visual function. In order to better understand this disease, various animal models have been produced, such as mouse models rd1, rd2, as well as rat models RCS, P23H, and S334ter. Since mutations in the rhodopsin (Rho) gene are very common in RD and contribute to the majority of known genetic forms of autosomal dominant (ad) retinitis pigmentosa (RP), substantial efforts have been made to develop models
∗ Corresponding author at: Department of Electronic Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon Tong, Kowloon, Hong Kong. E-mail address: [email protected] (L.L.H. Chan). http://dx.doi.org/10.1016/j.neulet.2016.04.062 0304-3940/© 2016 Elsevier Ireland Ltd. All rights reserved.
that express rhodopsin mutation [1]. Among all the rat models of RD, P23H and S334ter express rhodopsin aggregation defect and rhodopsin inactivation defect, respectively [1]. Specifically, S334ter rhodopsin transgenic rats, as a model of several rhodopsin truncation mutations in human RP patients, express rhodopsin gene with an early termination codon at residue 334, resulting in the expression of a rhodopsin protein without 15C-terminal amino acids that are involved in rhodopsin trafficking to the photoreceptor outer segments and in the inactivation of rhodopsin protein after light absorption [2,3]. Although there are several lines of S334ter rat with different rates of photoreceptor degeneration, there is a similar course of photoreceptor degeneration followed by secondary modifications [4,5]. Among the research on RD models, many efforts have been made in the scope of electrophysiology, as electrical activities play an essential role in early development of the nervous system, as well
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
as at later stages in refinement of connections [6]. In particular, spontaneous activities have been considered to be important for the refinement of neural projections and maintenance of topographic maps (neuroanatomy) in the brain [7,8]. There has been substantial evidence showing that altering the pattern of spontaneous activities disrupts this refinement [9]. The spontaneous activities, therefore, have caught major attention in the research community. Notably, an increase in spontaneous hyperactivity of retinal ganglion cells (RGCs) occurs during the progression of retinal degeneration [10–13], and this spontaneous hyperactivity is sustained well into adulthood for weeks after photoreceptors have disappeared [11]. It has been reported that an increase in glutamate concentration and a decrease in photoreceptors partially contributes to the increase in the spontaneous firing rate of ganglion cells in RD models [10,12]. The research on RD models, however, represents a bias towards physiology and electrophysiology of the retina, or cortical organization in the macular degeneration subject by analyzing the fMRI data [14–16]. The data on spontaneous activities in the visual cortex of RD models is still insufficient. Therefore, in the present study, the spontaneous activities, particularly the spiking in primary visual cortex of RD rats, have been investigated, and we expect to know the changes in inherent neural activities in the primary visual cortex of RD model in terms of electrophysiological properties. 2. Material and methods 2.1. Animals and surgical procedures The animals were obtained from Animal Center of Chinese University of Hong Kong, and housed in animal facilities at City University of Hong Kong. All experiments were conducted in accordance with protocols approved by Animal Research Ethics Sub-Committee in City University of Hong Kong and Department of Health, HKSAR. Long Evans and S334ter rats (P80–P90, n = 5 for each group), were used as the control (wild-type, WT) and RD groups, respectively. For surgery, the procedures were similar to those in previous work [17]. The animal was anesthetized with intraperitoneal injection of Ketamine- Xylazine combination (Ketamine: 70 mg/kg, Xylazine: 7 mg/kg) initially, then isoflurane (2%) was applied during the recording. After craniotomy, a bone screw was fixed in the skull and a glasspipette tungsten electrode was placed on the surface of primary visual cortex (AP: −6 ∼ −9 mm; ML: 2 ∼ 4 mm; DV: −0.1 ∼ −1 mm), as the ground and recording electrode, respectively. The exposed cortical surface was covered with 2% agarose to avoid drying and reduce body oscillation caused by breath, then a shield was used to attenuate the environmental interference. The animal was maintained at 38 ◦ C by a warm pad, and an ocular solution was regularly applied to keep the eyes moist during the surgery and recording.
43
ous penetrations and depths in order to reach different layers of the primary visual cortex. In total, three to four locations per penetration, and two to three penetrations were recorded for each animal. For every location, the recording lasted over 5 min. At the end of recording, the animal was euthanized by overdose of Dorminal (300 mg/kg). 2.3. Histology After the recording, eyeballs of both WT and RD animals were enucleated and fixed in 10% formalin in 0.1 M phosphate buffer (pH 7.4) at 4 ◦ C for overnight. Samples were then dehydrated with a graded series of ethanol and xylene and subsequently embedded in paraffin wax. Retinal sections (5 m) with pupil-optic nerve position were stained with hematoxylin and eosin (H&E). Retinal sections were imaged with a light microscrope to assess the status of different layers of retina. 2.4. Data analysis Spike sorting was initially conducted to distinguish between single units. A MATLAB (R2012b, MathWorks, US) package named Wave Clus [18] was used for this purpose, and the units were classified based on the features of their waveforms. After spike sorting, the firing rate of spontaneous activities was calculated. To compare the distribution of firing rates of different groups, we first performed the Shapiro-Wilk test to evaluate the normality of the data distribution. If the data satisfied a normal distribution, a t-test was conducted; otherwise, the Mann-Whitney test was used. The ISI was subsequently obtained. For every single unit, the mean and LZ complexity of ISI sequence were determined for comparing the ISI distribution and complexity of all the units from both WT and RD groups. Specifically, for calculating the LZ complexity, an algorithm based on Abraham Lempel and Jacob Ziv’s work was applied [19]. The statistical tests were as same as above. 3. Results 3.1. Degeneration in RD retina The H/E staining of both WT and RD rat retinas showed that the thickness of degenerated retinas of RD rats at P90 is less than that of the age matched control group (Fig. 1). Specifically, the outer plexiform layer (OPL), outer nuclear layer (ONL) as well as photoreceptor inner and outer segments (IS/OS) had almost disappeared. This validated the condition of retinal degeneration in RD model, which also agrees with previous studies [2,4,5,20]. Moreover, it has been reported that the inner nuclear layer (INL), inner plexiform layer (IPL) and ganglion cell layer (GCL) appear intact even at the age of P226 [21]. In the present study, however, we found that these layers in the RD retinas were also notably atrophied at the age of P90 when compared to the WT retinas.
2.2. Recording 3.2. Spontaneous firing rate The A-M Systems 3600 (A-M Systems, US) and CED Micro 14013 (Cambridge Electronic Design, UK) were used as the amplifier and data acquisition system, respectively. By using a micromanipulator, the recording electrode was lowered into the brain. The reading was recorded as zero reference when the electrode just passed through the cortical surface. As spikes appeared, the electrode was suspended froI m advancing and was allowed to settle for approximately 10 min before recording. The room was kept dark during the extracellular recording of spontaneous activities. The signal was filtered from 300 to 5 kHz and sampled at 25 kHz. The electrodes were placed in vari-
In the present study, 96 single units from 5 WT rats and 89 single units from 5 RD rats were isolated through spike sorting. Then, the representative waveforms of spontaneous activities were plotted on top of the averaged curve (Fig. 2). We have investigated the spontaneous activity of single neurons in both WT and RD groups. First, we determined the firing rates for each unit in both groups. In the WT group, the firing rates displayed a range from 0.11 Hz to 17.49 Hz with the mean ± SD of 3.04 ± 3.77 Hz. In the RD group, the firing rates showed a range from 0.11 Hz to 24.52 Hz with the mean ± SD of 4.05 ± 4.17 Hz.
44
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
Fig. 1. Light microscopy with H&E staining from a normal retina and a degenerated retina. (A) Ten laminar retinal layers are clearly observed in WT (3 months) normal retina. (B) Retinal thickness is reduced in the F1(3 months) retina with the absence of photoreceptor layers. (RPE, retinal pigment epithelium; CC, choriocapillaris; S, sclera).
Fig. 2. Waveforms of spontaneous firing in primary visual cortex of rats. (A) Spontaneous firing in the primary visual cortex of WT rats. (B) Spontaneous firing in the primary visual cortex of RD rats. (C) Two clusters have been sorted out by using principle component analysis (PCA). (D) The corresponding waveforms and their averaged waveforms of the isolated units shown in C.
By performing the Shapiro-Wilk test [22], we found that all the firing rates from both groups were not from the normal distribution (WT, p = 1.4e-10; RD, p = 5.9e-10), which is consistent with the hypothesis that the firing rates probably have a skewed (typically lognormal) distribution [23]. We will further elaborate on this in the discussion section. The Mann-Whitney test was subsequently conducted. The results suggested that the firing rates of spontaneous activity in the primary visual cortex of RD rats were significantly higher (p = 0.0005) than that of WT rats (Fig. 3).
3.3. Interspike interval distribution As an important property of spontaneous activities, the ISI was also examined. To compare the ISI distribution of all the units from both WT and RD groups, the mean ISI was calculated. In the WT group, the mean ISI has an extent from 0.06 s to 9.61 s with the mean ± SD of 1.18 ± 1.58 s. In contrast, for the RD group, the mean ISI has decreased, with the range from 0.04 s to 7.84 s with the mean ± SD of 0.69 ± 1.18 s. The results of Shapiro-Wilk test indicated that the mean ISI from both groups did not satisfy the normal distribution (WT, p = 4.5e-12; RD, p = 2.8e-14). The Mann-Whitney test was then performed. As shown in Fig. 4, although both of the
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
45
Fig. 3. Comparison of the firing rate of spontaneous activity in the primary visual cortex. The results indicate that the V1 cells of the RD model exhibit a higher firing rate than that of the control group. Mann-Whitney test, p < 0.001.
Fig. 4. Comparison of the ISI mean of spontaneous activity in the primary visual cortex. The results show that the V1 cells of the RD model show a decrease in ISI mean compared to that of the control group. Mann-Whitney test, p < 0.001.
mean ISI were skewed, the difference between the two groups was significant (p = 0.0006). 3.4. LZ complexity of ISI sequence Furthermore, with the purpose of providing a better understanding about spontaneous firing pattern, the complexity of ISI sequence was also investigated (Fig. 5). In present study, LZ complexity has been employed for the complexity analysis. In the WT group, LZ complexity of ISI sequence was ranged from 0.36 to 1.28 with the mean ± SD of 0.89 ± 0.16, while for the RD group, LZ complexity displayed an extent from 0.34 to 1.14 with the mean ± SD of 0.79 ± 0.17. By carrying out the ShapiroWilk test, we found that the LZ complexities of ISI sequence of WT and RD groups come from a skewed distribution (p = 0.0437) and normal distribution (p = 0.2798), respectively. The Mann-Whitney test was subsequently performed, and the results indicated that the overall LZ complexity of ISI sequence of the RD group was significantly lower than that of the WT group (p = 2.4822e-05).
Fig. 5. Comparison of the LZ complexity of the ISI sequence. The LZ complexity of the ISI sequence of RD group exhibits an evident decrease compared to that of the WT group. Mann-Whitney test, p < 0.001.
4. Discussion Consistent with the findings in retina [10,11,24], the spontaneous activities in primary visual cortex of RD model exhibited an increased firing rate, which suggests a thorough refinement presumably occurred along the visual pathway. In addition, the ISI sequence did not only have a smaller mean value but also represented a decrease in complexity, implying that the cortical circuitry
of visual system has impaired capability for generating new patterns of inherent activities. Therefore, the changes in spontaneous activities in primary visual cortex of RD model may provide better understanding of the progress of retinal degenerative diseases. Given that the spontaneous activities in primary visual cortex of RD rats exhibited higher firing rate than that of WT rats, it can be confirmed that the visual system could be affected by the progress
46
Y. Wang et al. / Neuroscience Letters 623 (2016) 42–46
of degeneration not only in early phase but also in later stages in terms of both morphology and function, such as circuitry reformation and retinotopy reorganization, due to the neural plasticity. According to the views of Hubel and Wiesel, the abnormality in the eyes could cause morphological changes in geniculate and defect in striate cortex [25]. Moreover, our results are also consistent with the finding that a significant degeneration in the cortical processing of visual information before development has been properly completed [26]. In addition, we found that neurons in primary visual cortex of RD model possess smaller ISI than that of control group, which agrees with our preceding results about firing rate. Besides, we also investigated the complexity of ISI sequence. In general, the ISI sequence of spontaneous activities could display nonrandom internal structure, and the degree of this structural nonrandomness could be quantified by studying its complexity [27]. The concept of complexity provides a promising tool for clinical study that may benefit the evaluation of the disease progress or treatment outcomes [28]. LZ complexity has been extensively applied to measure the generation rate of new patterns along a digital sequence and is also closely related to important information-theoretic properties such as compression ratio, redundancy and coding efficiency [29,30]. In recent years, LZ complexity has been widely used in the context of biomedical signal analysis, for instance, to characterize the spike trains with estimation of entropy, complexity curve and the number of neural source states [29], or to study mental disorders [31]. In general, lower complexity indicates higher regularity, which is consistent with previous observation that neurons that fire more rapidly fire more regularly [32]. Combined with our results of LZ complexity, it could be implied that the cortical visual system of RD model has attenuated capability to generate new patterns of neural activity, which is related to instant responses to stimuli or lasting changes of circuitry that profoundly depend on neural plasticity. On the other hand, since only the pattern of spontaneous activities was studied in present study, the question regarding responsive activities remains unanswered. Nevertheless, we did record responsive activities in primary visual cortex under light stimuli (data not shown), and we found that even for the rats at late stage of retinal degeneration, weak visual responses can still be observed. Further studies are required to address this question. In summary, our results suggest that the inhibition-excitation system in the primary visual cortex of RD model probably has been disrupted, resulting in an increase in firing rate of spontaneous activities. Moreover, the decreased LZ complexity of ISI sequence indicated a lower arising rate of new patterns of spontaneous activities in RD model. These findings therefore implied that the degeneration progress in RD model does not only affect the retina itself but also the primary visual cortex. Finally, to validate these implications, other approaches, such as molecular and cellular signaling, are needed. Acknowledgement This work was supported by the Research Grants Council of the Hong Kong Special Administrative Region, China [grant number CityU 123412, CityU 111013 and CityU 7004433]; the National Science Foundations of China [grant number 31300912]; the Guangdong Innovative and Entrepreneurial Research Team Program [grant number 2013S046]; and the Shenzhen Peacock Plan. References [1] R.E. Marc, B.W. Jones, C.B. Watt, E. Strettoi, Neural remodeling in retinal degeneration, Prog. Retin. Eye Res. 22 (2003) 607–655.
[2] V.M. Shinde, O.S. Sizova, J.H. Lin, M.M. LaVail, M.S. Gorbatyuk, Er stress in retinal degeneration in s334ter Rho rats, PLoS One 7 (2012) e33266. [3] J. Chen, C.L. Makino, N.S. Peachey, D.A. Baylor, M.I. Simon, Mechanisms of rhodopsin inactivation in vivo as revealed by a cooh-terminal truncation mutant, Science 267 (1995) 374–377. [4] G. Martinez-Navarrete, M.J. Seiler, R.B. Aramant, L. Fernandez-Sanchez, I. Pinilla, N. Cuenca, Retinal degeneration in two lines of transgenic S334ter rats, Exp. Eye Res. 92 (2011) 227–237. [5] A. Ray, G.J. Sun, L. Chan, N.M. Grzywacz, J. Weiland, E.-J. Lee, Morphological alterations in retinal neurons in the S334ter-line3 transgenic rat, Cell Tissue Res. 339 (2010) 481–491. [6] N.C. Spitzer, Electrical activity in early neuronal development, Nature 444 (2006) 707–712. [7] M.G. Hanson, L.T. Landmesser, Increasing the frequency of spontaneous rhythmic activity disrupts pool-specific axon fasciculation and pathfinding of embryonic spinal motoneurons, J. Neurosci. 26 (2006) 12769–12780. [8] N.X. Tritsch, E. Yi, J.E. Gale, E. Glowatzki, D.E. Bergles, The origin of spontaneous activity in the developing auditory system, Nature 450 (2007) 50–55. [9] L.A. Kirkby, G.S. Sack, A. Firl, M.B. Feller, A role for correlated spontaneous activity in the assembly of neural circuits, Neuron 80 (2013) 1129–1144. [10] E.L. Nylen, Quantitative analysis of retinal function in early stages of retinal degeneration in the rd1 mouse, (2010) http://ir.uiowa.edu/etd/565/. [11] S.F. Stasheff, Emergence of sustained spontaneous hyperactivity and temporary preservation of off responses in ganglion cells of the retinal degeneration (rd1) mouse, J. Neurophysiol. 99 (2008) 1408–1421. [12] D.J. Margolis, P.B. Detwiler, Cellular origin of spontaneous ganglion cell spike activity in animal models of retinitis pigmentosa, J. Ophthalmol. 2011 (2011) 1–6. [13] S.F. Stasheff, M. Shankar, M.P. Andrews, Developmental time course distinguishes changes in spontaneous and light-evoked retinal ganglion cell activity in rd1 and rd10 mice, J. Neurophysiol. 105 (2011) 3002–3009. [14] H.A. Baseler, A. Gouws, K.V. Haak, C. Racey, M.D. Crossland, A. Tufail, et al., Large-scale remapping of visual cortex is absent in adult humans with macular degeneration, Nat. Neurosci. 14 (2011) 649–655. [15] C.I. Baker, Reorganization of visual processing in macular degeneration, J. Neurosci. 25 (2005) 614–618. [16] Y. Shao, G.A. Keliris, A. Papanikolaou, M.D. Fischer, D. Zobor, H. Jägle, et al., Visual cortex organisation in a macaque monkey with macular degeneration, Eur. J. Neurosci. 38 (2013) 3456–3464. [17] C.M. Niell, M.P. Stryker, Highly selective receptive fields in mouse visual cortex, J. Neurosci. 28 (2008) 7520–7536. [18] R. Quiroga, Z. Nadasdy, Y. Ben-Shaul, Unsupervised spike detection and sorting with wavelets and superparamagnetic clustering, Neural Comput. 16 (2004) 1661–1687. [19] A. Lempel, J. Ziv, Complexity of finite sequences, IEEE Trans. Inf. Theory 22 (1976) 75–81. [20] B.W. Jones, R.E. Marc, Retinal remodeling during retinal degeneration, Exp. Eye Res. 81 (2005) 123–137. [21] D.J. Margolis, G. Newkirk, T. Euler, P.B. Detwiler, Functional stability of retinal ganglion cells after degeneration-induced changes in synaptic input, J. Neurosci. 28 (2008) 6526–6536. [22] N.M. Razali, Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling tests, J. Stat. Model. Anal. 2 (2011) 21–33. [23] G. Buzsáki, K. Mizuseki, The log-dynamic brain: how skewed distributions affect network operations, Nat. Rev. Neurosci. 15 (2014) 264–278. [24] C. Sekirnjak, L.H. Jepson, P. Hottowy, A. Sher, W. Dabrowski, A.M. Litke, et al., Changes in physiological properties of rat ganglion cells during retinal degeneration, J. Neurophysiol. 105 (2011) 2560–2571. [25] D.H. Hubel, T.N. Wiesel, Effects of monocular deprivation in kittens, Naunyn Schmiedebergs Arch. Pharmacol. 248 (1964) 492–497. [26] C. Gias, A. Vugler, J. Lawrence, A.J. Carr, L.L. Chen, A. Ahmado, et al., Degeneration of cortical function in the Royal College of Surgeons rat, Vision Res. 51 (2011) 2176–2185. [27] P.E. Rapp, I.D. Zimmerman, E.P. Vining, N. Cohen, A.M. Albano, M.A. Jimenez-Montano, The algorithmic complexity of neural spike trains increases during focal seizures, J. Neurosci. 14 (1994) 4731–4739. [28] A.C. Yang, S.-J. Tsai, Complexity of mental illness: a new research dimension, Prog. Neuropsychopharmacol. Biol. Psychiatry 45 (2013) 251–252. ´ [29] J. Szczepanski, J.M. Amigó, E. Wajnryb, M.V. Sanchez-Vives, Characterizing spike trains with Lempel–Ziv complexity, Neurocomputing 58–60 (2004) 79–84. [30] M. Aboy, R. Hornero, D. Abasolo, D. Alvarez, Interpretation of the Lempel-Ziv complexity measure in the context of biomedical signal analysis, IEEE Trans. Biomed. Eng. 53 (2006) 2282–2288. [31] C. Gómez, R. Hornero, A. Fernández, D. Abásolo, M. López, Decreased Lempel-Ziv complexity in Alzheimer’s disease patients’ magnetoencephalograms, Conf. Proc. IEEE Eng. Med. Biol. Soc. (2005) 4514–4517. [32] B.D. Bennett, C.J. Wilson, Spontaneous activity of neostriatal cholinergic interneurons in vitro, J. Neurosci. 19 (1999) 5586–5596.