, 2009 and Rheinlaender and Schaffer, 2009; see also below). Therefore, in practice, we could not break the presynaptic cell membrane and obtain whole-cell patch-clamp recordings when using the original scanning nanopipettes. To overcome this limitation, we optimized a method to widen the ultra-fine pipette tip after the completion of the high-resolution 3D topography scan by breaking it against the glass coverslip (Böhle and Benndorf, 1994), using programmable feedback control of the HPICM scanner controller. The nanopipette tip-breaking procedure consisted JQ1 of three steps (Figure 3A). First, the pipette was navigated to a previously identified area of the coverslip free

of neuronal

processes. Second, the fall rate (the rate at which the pipette repeatedly approaches the surface during “hopping”) was increased from the standby rate (typically 60 nm/ms) by approximately one order of magnitude (to ∼500 nm/ms). At this fall rate, the noncontact mode of HPICM could no longer be preserved because of the inherent latency of the z axis piezo feedback control. As a result, the pipette repeatedly crashed into the coverslip, breaking its tip and increasing its diameter because of the conical shape of the pipette. Pipette Alectinib mouse tip breaking resulted in stepwise increases of the pipette current as its resistance dropped (red arrows in Figure 3B). The breaking was automatically stopped by returning all the fall rate to baseline (60 nm/ms)

once the pipette current reached a desired level. This process could be repeated to fine-tune the desired pipette tip diameter in steps as small as 10% by varying the stop criteria for current increase, duration, and “breaking” fall rate (Figure 3B). To characterize the properties of widened nanopipettes, we obtained scanning electron microscopy (SEM) images of intact and modified pipette tips (Figures 3C and 3D; see Experimental Procedures for details). Importantly the controlled breaking procedure did not change the overall shape of the pipette tip but reliably allowed the inner tip diameter to be increased approximately 4-fold: from 107 ± 16 nm (mean ± SD, n = 4) to 417 ± 48 nm (mean ± SD, n = 8). The experimentally determined relationship between pipette resistance and inner pipette tip diameter for both intact and widened pipettes was in close agreement with theoretical predictions based on the tip geometry (Figure 3G). On average the resistance of the widened pipettes was decreased ∼2.4-fold (from 92.2 ± 8.9 MΩ to 38.7 ± 4.0 MΩ, mean ± SD, n = 17), thus making the modified pipettes more suitable for whole-cell patch-clamp recordings. Importantly, because the pipette was held vertically at all times, the x, y coordinates of the pipette tip did not change (Figures S4A–S4C).