First, calcium elevations in astrocytes in all of these studies were monitored using synthetic dyes, loaded into cells using the membrane permeant AM ester form, and by identifying astrocytes using either genetic markers (Zhuo et al., 1997) or sulforhodamine 101 (Nimmerjahn et al., 2004). However, the dye is taken up by all cells, and

even when using counterstains (Figures 3C and 5C), signal separation can become difficult (Göbel and CH5424802 mw Helmchen, 2007 and Grewe and Helmchen, 2009). In addition, the time course of calcium responses in neurons and astrocytes is influenced by the properties of the indicator as well as the endogenous calcium buffer capacity (Helmchen et al., 1996 and Neher and Augustine, 1992), although onset kinetics are probably not affected significantly. A second scenario is that calcium mTOR inhibitor changes in astrocytes do occur earlier than functional hyperemia but that they are not picked up by the indicator. This may be either because the affinity of typical bulk-loaded indicators is too low to detect very subtle calcium changes or because the indicators tend to accumulate in somata

and larger processes, leaving out the extensive network of smaller astrocytic processes and their even finer ramifications. There is clear evidence for differences in calcium signals recorded in astrocyte somata and fine processes (Reeves et al., 2011). Perhaps progress can be made if astrocytes can be selectively labeled with calcium indicators, especially with genetically encoded indicators such as GCaMPs (Nakai et al., 2001, Shigetomi et al., 2010 and Tian et al., 2009), troponin-based probes (Mank et al., 2006 and Mank et al., 2008), and chameleons (Atkin et al., during 2009, Miyawaki et al., 1997 and Truong et al., 2007). A third scenario is that calcium changes in astrocytes indeed appear later than functional hyperemia. For example, it is possible that nonastrocytic mechanisms—e.g., neuronal NO or dedicated interneurons—trigger the initial rise of functional hyperemia,

but that astrocytic pathways are necessary to maintain the response. Moreover, signaling steps between astrocytic activation and calcium increase, such as diacylglycerol production, may also be vasoactive. It is also feasible that calcium represents just one of many different vasoactive astrocytic messengers, such as sodium (Bernardinelli et al., 2004), protons (Amato et al., 1994 and Chesler and Kraig, 1987), cAMP (Moldrich et al., 2002), ATP (Cotrina et al., 2000 and Pascual et al., 2005), or lactate (Gordon et al., 2008). Future experiments may benefit from monitoring changes in these parameters within astrocytes in vivo. In addition to monitoring calcium rises, it is also important to be able to perturb calcium levels within astrocytes at will.