between the pattern of Fos immunoreactivity evoked by the visceral stimulus 
and that observed in response to hindpaw formalin stimulation (figure 4). First, 
the densest staining was in the caudal thoracic rather than the lumbar cord, as 
was seen following formalin. Second, the staining following visceral stimulation 
was largely restricted to laminae I and V of the dorsal horn and to the area 
around the central canal. This different topographic distribution further argues 
that the induction of Fos immunoreactivity in spinal neurons is not secondary 
to stress but is related directly to the particular stimulus. Third, there was a 
far more extensive rostral caudal distribution of the Fos immunoreactivity, 
particularly in lamina I, where staining from the upper cervical to the sacral cord 
was found. This is consistent with the known anatomy of the primary afferent 
fibers that carry noxious visceral information; these fibers terminate over many 
segments rostral and caudal to their major termination point in the cord (Sugiura 
et al. 1989). The absence of Fos immunoreactivity in lamina II is also 
remarkably consistent with the central termination pattern of visceral afferents 
(Cervero 1985; Sugiura et al. 1989). 
FOS EXPRESSION AFTER NONNOXIOUS STIMULATION 
To further validate the use of Fos immunocytochemistry to monitor activity in 
pain pathways in the CNS, it was important to show that the pattern of Fos 
immunoreactivity in the cord following a nonnoxious stimulus was different from 
that seen following a noxious stimulus. Initial work by Hunt and colleagues 
(1 987) suggested that this would be difficult to document. In their study, 
nonnoxious stimuli produced only minimal increases in Fos immunoreactivity, 
albeit in appropriate regions of the dorsal horn. However, Gogas and 
colleagues (1990) found that allowing an animal to walk on a slow-turning 
rotarod device for 1 hour evokes widespread Fos immunoreactivity in the 
cord. Fos expression was notably lacking in laminae I and No of the cervical 
and lumbar segments. However, the inner portion of lamina II as well as 
laminae III and IV, areas that contain neurons that respond exclusively to 
nonnoxious stimulation (Bennett et al. 1980; Light et al. 1979), contained many 
labeled cells. Additional staining was found in the medial portion of laminae V 
and VI, an area that receives inputs from joint afferents (Brown and Fyffe 1978, 
1979), and in lamina IX, where many Fos-immunoreactive MNs were found. 
These results are important since they provide considerable support for the 
hypothesis that Fos protein expression is stimulus specific and can, therefore, 
be used to study those populations of neurons in the cord that are activated by 
noxious stimulation. Furthermore, the modulation of noxious stimulus-evoked 
Fos expression by pain-relieving drugs such as opioids can also be studied, 
which can provide information about the mechanisms by which these 
compounds produce analgesia. 
141 
