Article Figures & Data
Figures
Tables
Supplemental data
Files in this Data Supplement:
- supplemental material - Supplemental Figure 1. Calibration of [Ca2+] signals from 2 synthetic indicators in response to different trains of action potentials allows quantitation of indicator saturation and reliable measurement of ?[Ca2+]. (A) Determination of the degree of saturation (?) using trains of action potentials at different frequencies (30, 50 , and 70 Hz) with the same temporal duration. This method allows determination of the saturation during each experiment, which is critical for translation of fluorescence into [Ca2+]. (B) Model of the change in fluorescence for an unsaturated Ca2+ indicator using parameters derived from experiments in primary apical dendrites. (C-D) Plot of the normalized peak fluorescence ratio (Qnorm) of X-Rhod-5F (C) and Fluo4-FF (D) versus the normalized ratio of AP train frequencies (Vnorm). Error bars are � SEM. Black line indicates the same relationship for the ideal unsaturated (B; ? = 0; Qnorm = Vnorm). Percentages (red and green text) represent saturation (?) for the given AP train frequency (black text).
- supplemental material - Supplemental Figure 2. Buffering of influxing [Ca2+] only slightly perturbs amplitude and timecourse of GECI fluorescence. (A) Recordings made from cells during loading with X-Rhod-5F were monitored using single AP transients (red traces; bottom row; time after break-in indicated). The decay of the amplitude of the transients (from 60% to 15%) and the increase in decay time (from 880 msec to 2200 msec) reveals added buffering as cell fills, but single AP and AP train GECI responses (green traces; middle and top rows; time after break-in indicated) were only slightly perturbed in amplitude. There were also slight differences in the rise and fall of the GECI signal depending on buffering, but that were consistent with ?[Ca2+] as buffer was added.
- supplemental material - Supplemental Figure 3. AP trains delivered at 20, 30, 50, and 70 Hz reveal different responses between and GECIs and synthetic indicators and each other. Trains of action potentials delivered at 20, 30, 50 & 70 Hz (columns 1-4, all rows) yielded GECI responses much different from those observed with synthetic indicators. The black trace is the mean response across cells, and the light colored traces are the mean � SEM. Analysis of SNR is given in the main text, Figure 3. (A-E) The red (X-Rhod-5F; KD = 1.9 ?M; N = 8 cells) and green (Fluo4-FF; KD = 10.4 ?M; N = 8 cells) synthetic calcium indicators respond to all stimuli, including single APs, while the GECIs show weaker responses to low frequency stimuli, making it difficult to generate fluorescence above background.
- supplemental material - Supplemental Figure 4. Determination of synthetic calcium indicator biophysical parameters in situ. (A) Models of fluorescence saturation, ?, versus [Ca2+] with different Hill coefficients (n). When n = 1, the model is identical to a hyperbolic binding model. As n increases, the transition from low to high ? sharpens. (B) The Hill coefficient versus the log interval change in calcium between 10% and 90% fluorescence saturation, ?(log10([Ca2+])10%?90%. The larger the n, the narrower the [Ca2+] range over which the transition occurs. (C) Determination of the fluorescence saturation,?, of X-Rhod-5F as a function of [Ca2+] measured with Fluo4-FF. The 50% saturation point is the effective KD of X-Rhod-5F and this value agrees well with values determined for X-Rhod-5F in vitro (KD in situ = 1.32 ?M; KD in vitro = 1.9 ?M). (D) Hill plot of data from (C) allows the determination of n (slope of least-square fit line; n = 1.11 � .21). The value obtained is close to the expected value of 1 (since X-Rhod-5F is a stochiometric calcium buffer with hyperbolic saturation).
- supplemental material - Supplemental Figure 5. Calcium buffering by GECIs perturbs X-Rhod-5F decay like Fluo4-FF, but does not alter the peak response substantially. Single cell X-Rhod-5F responses (thin red lines) and means (thick black lines) to a train of 20 APs delivered at 30 Hz in different conditions. Responses were filtered using a 50 msec averaging window and scaled to have a maximum of 1. (A) GFP transfected cells (? = .76 sec) show similar X-Rhod-5F responses as un-transfected cells. Thus, transfection of neurons did not disturb their calcium handling, but rather expression of an endogenous calcium buffer (GECI) did. (B) Cells filled with Fluo4-FF and X-Rhod-5F (? = 1.92 sec) showed slower decay than cells expressing GFP [A] or filled with the Alexa 488. (C) Cells filled with Alexa-488 (? = .45 sec), which does not bind Ca2+, show similar decay times as GFP transfected cells. (D-F) Expression of GECIs slows decay of X-Rhod-5F signal like Fluo-4FF. (GCaMP ? = 1.18 sec; Camgaroo2 ? = 1.23 sec; Inverse Pericam ? = 1.33 sec). (G) Mean responses from [A-F] comparing decay times.
- supplemental material - Supplemental Figure 6. Model of [Ca2+] accumulation and GCaMP fluorescence saturation with different KDs Model of primary apical dendrite [Ca2+] accumulation (black trace) and GCaMP fluorescence saturation assuming different dissociation constants (green traces). The lower KD (.24 ?M) corresponds to the previously reported value (Nakai et al., 2001), while the higher one (1.7 ?M) was measured here. Our data and models of [Ca2+] in large apical dendrites, as shown here, are incompatible with a KD for GCaMP less than 1 ?M.
- supplemental material - Supplemental Materials and Methods