Surgical Techniques for Chronic Implantation of Microwire Arrays in Rodents and Primates

The study of electrophysiology started with the work of Luigi Galvani (1737–1798), who was the first to provide evidence for the electrical nature of “the mysterious fluid” (at the time referred to as “animal spirits”). Galvani’s nephew, Giovanni Aldini (1762–1834), continued this line of inquiry in 1803, using Galvani’s and Alessandro Volta’s (bimetallic electricity) principles together, despite the fact that Volta did not believe in animal electricity. Carlo Mateucci (1818–1868) in Bologna and Emil Du Bois-Reymond (1818–1896) in Berlin described the phenomenon called “negative variation” when a galvanometer showed an unexpected decrease in current intensity during muscle contraction. The study of the electrophysiology of the nervous system began when Julius Berstein (1839–1917) proposed his theory of the nerve impulse as a wave of negativity (membrane theory of the nerve tissue). Later, using a galvanometer with one electrode in the gray matter and one on the skull surface (or electrodes in different points of the external surface of the brain), Richard Caton (1842–1926) recorded a feeble current in the brain. In 1870, for the first time, Gustav Fritsch (1838–1927) and Eduard Hitzig (1838–1907) inserted an electrode in the dura of a dog brain and stimulated the motor area, generating movement in the contralateral side of the animal’s body (Niedermeyer 1993; Piccolino 1998). The work of these and many other scientists marked the beginning of the study of the electrophysiology of the nervous system, opening doors to the possibility of stimulating different areas of the brain through electrical current and subsequently recording the brain electrical activity. Improvements in electrode manufacturing, the advent of modern acquisition equipment, and better surgical and asepsis techniques have provided us the ability to chronically implant multiple electrodes simultaneously in several areas of the brain in the same animal (Nicolelis, Baccala et al., 1995) and to study the interactions of populations of neurons (Nicolelis, Fanselow et al., 1997; Ghazanfar, Stambaugh et al., 2000). Upon the animal’s recovery from surgery, we have been able to record simultaneously from different brain areas of mice (Costa, Cohen et al., 2004), rats (Faggin, Nguyen et al., 1997; Ghazanfar and Nicolelis 1997; Nicolelis, Ghazanfar et al., 1997) and nonhuman primate brains (Nicolelis, Stambaugh et al., 1999; Nicolelis, Dimitrov et al., 2003) for long periods of time, from a couple of months in rodents (Ghazanfar and Nicolelis 1997; Nicolelis, Ghazanfar et al., 1997) to up to years in non-human primates, such as owl monkeys (Nicolelis, Ghazanfar et al., 1998) and Rhesus monkeys (Nicolelis, Dimitrov et al., 2003). These recordings are carried out under several different experimental conditions and behavior tasks (Kralik, Dimitrov et al., 2001; Nicolelis and Ribeiro 2002). With chronically implanted multiple electrodes, it is also possible to record different layers of the same area of the brain (Chapin and Lin 1984) and study spatiotemporal response of many neurons (Nicolelis and Chapin 1994). Microcannulae can be attached to the electrode arrays and are used to inject drugs in the areas of the implant during chronic experimental recordings (Shuler, Krupa et al., 2002). Chronically implanted electrodes offer unparalleled advantages for correlating neuronal activity and animal behavior. In our lab, these techniques were developed in rodents and later adapted to primates. Over the last several years, there have been significant strides in making rodent implantations more reliable, faster, and easier. We have identified and resolved many of the issues that now permit larger neuronal yields that last longer. As a consequence of continuous improvement in techniques, the length of time required for surgery has been reduced. At the same time, over the last 14 years, we have developed a surgical technique adapted to the unique features of primates. This has made primate implantations routine and reproducible. Here, we will describe detailed technical aspects of the current surgical implantation approach used in our laboratory at the Duke University Center for Neuroengineering (DUCN). Such a surgical protocol has evolved and benefited from almost two decades of accumulated experience on chronic multielectrode neural recordings (Nicolelis, Stambaugh et al., 1999; Nicolelis, Dimitrov et al., 2003, Kralik, Dimitrov et al., 2001; Nicolelis and Ribeiro 2002).