The 2-deoxyglucose (2-DG) technique was developed originally by Louis Sokoloff and coworkers to measure real values of glucose utilization in the brain (Kennedy et al., 1975). Shortly thereafter, the technique was used to great advantage to show that different odorants activate distinct spatial patterns of activity in the olfactory bulb (Stewart et al., 1979). We perform a modification of this technique that gives a semi-quantitative measure of relative neural activity in the olfactory bulb (Royet et al., 1987; Johnson et al., 1999). In this method, a tracer amount of [14C ]2-deoxyglucose is injected into a rat immediately prior to placing the animal in a glass Mason jar. Odorized vapor then is introduced into the jar via a hole that has been drilled into the glass top. The radiolabeled 2-DG is taken into cells through the glucose transporter in proportion to the cells' demand for glucose. In more active areas, more 2-DG is accumulated. The 2-DG is phosphorylated, but cannot be broken down in most neurons. For the majority of odorant exposures illustrated on this website, rats were exposed to the odorant for 45 minutes, which is intended to result in clearance of the radiolabel from blood vessels and to enhance the contrast of the odorant-evoked radioactivity by avoiding registering radiolabel that is associated with unphosphorylated 2-DG. It should be noted that we have obtained very similar patterns for a number of odorants using exposure times as short as 2 to 5 minutes (Woo et al., manuscript in preparation). 

The 2-DG method has a number of benefits and a few disadvantages relative to other techniques that have been used recently in the study of glomerular layer activity patterns. The first advantage is that the 2-DG method involves sectioning bulb tissue, thereby allowing total access to the entire structure from dorsal to ventral, rostral to caudal, and lateral to medial. Optical imaging techniques that have been applied to monitor reflectance, calcium levels, and voltages in the glomerular layer of the rodent olfactory bulb are commonly limited to the extreme dorsal surface of the bulb (Rubin and Katz, 1999; Meister and Bonhoeffer, 2001; Wachowiak and Cohen, 2001, 2002; Spors and Grinvald, 2002). However, most odorants do not activate dorsal glomeruli as well as they activate glomeruli in other parts of the bulb, as is evident from the activity patterns in this website. In the worst case, limiting one's experiments to a small part of the dorsal bulb may result in exclusively studying and interpreting responses to minor odorant contaminants (Johnson et al., 2004). One laboratory has successfully applied optical imaging techniques to the lateral bulbar surface in addition to the dorsal surface, and as might be expected, their results tend to agree with those obtained using the 2-DG method (Uchida et al., 2000; Takahashi et al., 2004). Also, similar optical recording techniques have proven extremely effective on simpler olfactory systems such as zebrafish (Friedrich and Korsching, 1997; Fuss and Korsching, 2001), salamanders (Cinelli et al., 1995), and honeybees (Galizia et al., 1999; Sachse et al., 1999), where a more complete complement of the olfactory glomeruli can be imaged simultaneously. 

A second advantage of the 2-DG technique is that it involves the study of animals that are awake and not anesthetized. These rats can inhale odorants naturally, and their neural responses are unaffected by drugs. Most other procedures for monitoring the spatial distributions of responses to odorants, including the optical imaging methods, fMRI (Xu et al., 2000), and electrophysiology (Mori et al., 1992; Imamura et al., 1992) are performed under anesthesia. Anesthesia suppresses sniffing and appears to change the activity of bulbar neurons (Motokizawa et al., 1996; Rinberg et al., 2004). A third advantage of the 2-DG method is that the label is internal and that radioactivity standards can be exposed along with every film. Thus, one can relate any signal to a standardized value that does not involve arbitrary thresholding or scaling of the observed response. Furthermore, the 2-DG technique is free of the section-to-section sources of variability in backgrounds, antibody penetration, or efficacy of washing that plague immunohistochemical or in situ hydribization techniques for the detection of activity-dependent expression of genes such as c-fos (Guthrie and Gall, 1995; Johnson et al., 1995). A fourth advantage of the 2-DG method is that the standardization allows a quantitative approach to bulbar activity patterns across groups of individuals, including the use of statistics to characterize the variance in the patterns across different animals exposed under the same conditions. A fifth advantage is that the 2-DG method has the spatial resolution to detect the activation of a single glomerulus without thresholding the response (Johnson et al., 1998; 1999; 2005b).  

Odorant-evoked 2-DG uptake visualized as a pseudocolor image with an overlying image of an adjacent section stained using cresyl violet to locate neuronal cell bodies. The radiolabeled metabolic marker is concentrated in foci of uptake corresponding in position to olfactory glomeruli.

A major disadvantage of the 2-DG technique as we perform it is that it is only possible to obtain a single odorant-evoked pattern in each animal. Given the biological and experimental variation in the location of activated glomeruli between different animals, we are unable to tell whether the very same glomeruli are stimulated by two different odorants. Other methods, such as optical imaging and electrophysiology, are able to study the response of the same glomerulus or neuron to multiple different odorants. However, it also is possible that a previous exposure to one odorant might change the response to a subsequently presented odorant as measured by these methods (Fletcher and Wilson, 2003). 

A second perceived disadvantage of the 2-DG technique is that it cannot detect the responses of the olfactory system during the very early time points (fractions of seconds) during which odor perception actually occurs, nor can it follow any possible changes in the spatial pattern of activity that might be occurring in the early seconds or even minutes of a continuous odorant exposure (Spors and Grinvald, 2002). However, others have shown that rats trained to sniff an odorant intermittently for only a few seconds at a time during a 45-minute exposure period have activity patterns very similar to those seen in rats exposed to odorant continuously throughout the 45-minute period (Slotnick et al., 1989). Moreover, others have shown that glomerular responses have been shown to remain constant over many minutes (Guthrie and Gall, 1995).