Hearing Research, 46 (1990) 53-62

Elsevier

53

HEARES 01372

Blood flow changes in chicken brain stem auditory nuclei following cochlea removal

Brent E. Richardson ’ and Dianne Durham ‘v2 Hearing Development Laboratories, RL-30 Departments oJ ’ Otolatyngology and ’ Biological Structure, University oJ Washington,

Seattle, Washington, U.S.A.

(Received 5 May 1989; accepted 30 December 1989)

Cochlea removal results in rapid and persistent metabolic and morphological changes in avian brain stem auditory nuclei. Because

such changes in the central nervous system are often associated with changes in local blood flow, we examined blood flow in

second-order auditory nucleus magnocellularis (NM) and third-order nucleus laminaris (NL). The diffusible tracer [‘4C]-iodoan-

tipyrine was infused intravenously into 20- to 26-day-old chickens either 30 min or 6 h after unilateral cochlea removal. This tracer

rapidly equilibrates between blood and tissue in proportion to local blood flow. Unoperated animals served as controls. Thirty

seconds after tracer infusion, brains were removed and frozen. Cryostat sections were prepared for quantitative film autoradiography.

Blood flow in normal and deafferented areas within NM and NL was compared.

Nucleus magnocellularis receives its only excitatory input from the ipsilateral cochlea via the eighth nerve. Axons from NM

bifurcate and project to the ipsilateral dorsal dendritic region of NL (NLd) and the contralateral ventral dendritic region of NL

(NLv). Thirty minutes after cochlea removal, blood flow in ipsilateral NM decreases by 30%. This decrease persists at 6 hours. Blood

flow in NL does not change in accordance with the pattern of afferent input from NM. Rather, blood flow in NLd and NLv

ipsilateraf to cochlea removal is significantly decreased 6 h post lesion. These results are in contrast to the pattern of morphological

and metabolic changes observed in NL after cochlea removal.

Deafferentation; Nucleus laminaris; Nucleus magnocellularis; fodoantipyrine

Introduction

Afferent activity is important for the normal development and maintenance of the central nervous system (Cowan, 1970; Guillery, 1974; Globus, 1975; Smith, 1977). The chicken auditory system has been the subject of a number of studies examining the metabolic and morphological ef- fects of deafferentation on brain stem auditory nuclei. Ablation of the otocyst in embryos (Levi- Montalcini, 1949; Parks, 1979) or cochlea removal

in hatchlings (Born and Rubel, 1985) decreases neuron number and soma size in ipsilateral sec- ond-order nucleus magnocellularis (NM) within

two days of deafferentation *. In NM of young hatchlings, cochlea removal also causes an im- mediate and complete cessation of electrical activ- ity (Born and Rubel, 1984) a rapid and persistent decrease in glucose uptake (Lippe et al., 1980; Heil and Scheich, 1986) a dramatic decrease in protein synthesis within 30 minutes postlesion (Steward and Rubel, 1985), and biphasic changes

Correspondence to: Dianne Durham, Hearing Development Laboratories, RL-30 University of Washington, Seattle, WA

98195. U.S.A.

l The term ‘deafferentation’ usually refers to a situation in which some portion of a neuron’s afferent input has degen-

erated. In chick brain stem, afferents to ipsilateral NM are

degenerating 2 days after cochlea removal (Parks and Rubel,

1978); in NL, evidence for secondary degeneration of NM

axons is not seen until 4 days after cochlea removal. (Rubel, Smith and Steward, unpublished observations). For lack of a

better term, however, we will refer to NM ipsilateral to

cochlea removal and areas in NL receiving input from that

NM as ‘deafferented’ beginning at the time of cochlea

removal.

0378-5955/90/$03.50 0 1990 Elsevier Science Publishers B.V. (Biomedical Division)

54

in activities of the mitochondrial enzymes suc-

cinate dehydrogenase (SDH) and cytochrome oxidase (CO) (Durham and Rubel, 1985; Hyde and Durham, 1989). Neurons in third-order nucleus laminaris (NL) receive spatially segre- gated binaural input from NM; the ipsilateral NM projects to dorsal NL dendrites (NLd) and the contralateral NM projects to ventral NL dendrites (NLv) (See top Fig. 5). Following unilateral cochlea removal, glucose uptake in deafferented

areas of NL declines immediately and to a greater extent than in NM (Lippe et al., 1980; Heil and

Scheich, 1985). The spatial pattern of decreased glucose uptake matches the pattern of input from NM. Levels of SDH and CO in deafferented den-

dritic regions of NL are persistently decreased after three days postlesion (Durham and Rubel, 1985; Hyde and Durham, 1989).

Metabolic and mo~holo~cal changes in the central nervous system have been associated with changes in blood flow. Roy and Sherrington (1890) first proposed that local blood flow in the brain is coupled to functional activity through the action of cerebral metabolites on its vascuiature. Changes in arterial partial pressures of oxygen and carbon dioxide, perfusion pressure, or activity in the peri- vascular nerve plexus also affect blood flow (Las- sen, 1982). Deafferentation has been associated

with changes in blood flow. For example, uni- lateral enucleation causes a 10% decrease in the number of perfused capillaries in rat superior col- liculus (Collins et al., 1987). Capillary recruitment and increased local cerebral blood flow have been demonstrated in humans during increased neural activity induced by somatosenso~ stimulation (Raichle et al., 1987a,b).

Given the metabolic changes that occur in chick brain stem auditory nuclei as a consequence of cochlea removal, we wished to determine whether blood flow in these areas might also change after this manipulation. In addition to the time course of such blood flow changes, the spatial resolution of blood flow regulation was of interest. Cochlea removal is known to eliminate neuronal activity and cause metabolic changes in specific, well-de- fined regions of the brain stem auditory nuclei, especially in NL. These nuclei seemed therefore to

be an excellent system in which to examine whether the spatial resolution of blood flow regulation

matches that of metabolic changes such as glucose uptake.

We present evidence that local blood flow de- creases within the second- and third-order brain stem auditory nuclei following unilateral cochlea removal, and that these changes are not always spatially correlated with the metabolic and mor- phological changes previously described.

Materials and Methods

Chicken eggs (White leghorn, H and N strain)

were obtained from a local supplier and hatched in a forced draft incubator. Twenty-four 20- to

26day-old chickens were the subjects of these experiments. Fifteen chickens were anesthetized with ketamine (80 mg/kg, intramuscular adminis- tration) and pentobarbital (19 mg/kg, intraperi- toneal ad~nistration) and underwent unilateral cochlea removal as previously described by Born and Rubel (1985). Briefly, the tympanic mem- brane was cut away and the columella, the chicken’s single ossicle, was removed. The cochlea was extracted through the oval window with fine forceps and examined to verify complete removal. Seven birds were only anesthetized and served as unoperated controls. All birds were kept anesthe-

tized throughout the remainder of the experiment. Blood flow was determined using the autora-

diographic method of Sakurada and colleagues (1978). In the wing contralateral to cochlea re- moval, both the bra&al artery at the junction of the radial and ulnar arteries and the cutaneous vein of the elbow were exposed. At 30 min or 6 h following cochlea removal in experimental animals, or immediately following wing vessel preparation in control animals, a catheter 100 mm in length (PE 10, ID 0.28 mm x OD 0.61 mm, Clay Adams) was inserted into the exposed artery, flushed with heparin (1000 U/ml) and tied in place. Ten PCi of 4-Iodo [ N-methyl-‘4C] antipyrine ([i4C]IAP, Amersham Corporation, specific activity 50-56 mCi/mmol) was infused via the exposed vein. Tracer was dissolved in normal saline and admin- istered as a ramp injection over 30 s using a 1 cc

insulin syringe with a 28G needle (U-100, Becton Dickinson). During the [‘4C]IAP infusion, birds were exposed to ambient laboratory sound condi-

55

tions. Sound levels measured with a sound level meter were approximately 70 dBC. Blood samples from the free-flowing arterial catheter were col- lected at two-second intervals. At 30 s from the beginning of [14C]IAP infusion, the birds were decapitated; the brains were removed within 2.5 min and rapidly frozen in heptane cooled with dry ice (- 65 o C). The brains were stored at - 80 o C until sectioning.

Sectioning and Autoradiography Coronal 20 pm sections through the brain stem

auditory nuclei were cut on a cryostat at - 25 *C. One in every three sections in the area of interest was thaw-mounted onto a gelatin and chrome alum-coated glass coverslip and dried on a hot plate at 60 o C. Coverslips were mounted on card- board and exposed to film (Kodak XTL) along with calibrated [‘4C]-methylmethacrylate stan- dards (Amersh~~. After development of the films, the cryostat sections were stained with thionin, dehydrated and mounted on glass slides with DPX.

Film autoradiographs were analyzed using the Bioquant IV computerized image analysis system (R and M Biometrics). For each brain, three to six sections around a point located 50% of the ante- rior to posterior extent of the nucleus were selected for m~urement. Since nucleus ma~~llula~s lies more caudal than nucleus huninaris, the sec- tions used for measurement were different for each nucleus. Nuclear boundaries and section edges were traced over the image of each thionin- stained section and this overlay image was stored in the computer. Using the section edges on the overlay, the overlay and the image of the corre- sponding film were aligned, and the average opti- cal density (OD) of each outlined nucleus was determined For each OD measurement, a tissue concentration of tracer (nCi/g) was computed by comparison with methylmethacrylate standards. Each tissue concentration was converted to a value for blood flow by solving the Kety integral (Sakurada et al., 1978). Blood flow values were averaged for all measured sections in a given nucleus.

Calculation of local blood ji’ow Local cerebral blood flow was calculated using

an iterative solution to the Kety equation:

where Ci(2’) is the tissue concentration of tracer at time p, C, is the arterial concentration of tracer at a given time; X is the tissue to blood partition coefficient; and K is a constant incorpo- rating the rate of blood flow in the tissue. The constant is defined as:

1y = mF/Wh

where F/W is the blood flow/unit mass of tissue; and m is a constant whose value is assumed to be 1 in the absence of arteriovenous shunts and diffu- sion abnormalities.

The value of X was determined empirically in 2 birds according to the method of Sakurada and colleagues (1978) with the exception that the renal and hepatic vascular supplies were not ligated. All liquid scintillation counting for determining X and the values of C, was done in a Packard Minimax Tricarb 4000 liquid scintillation counter using 226Ra external standardization and calibration with [14C]-toluene quenched standards (Amersham/ Searle).

Absolute values for blood flow were found to vary widely among animals in our experiment (see Discussion). Thus, to allow comparison between individual birds and between groups of birds, vari- ous ratios of the average blood flow in a given area to the average blood flow in another area in the same brain were computed for each animal. The average ratios for animals in a given group were compared using both parametric and non- parametric statistical analyses *. In addition, cor- relations between pairs of average ratios were ex- amined.

* Non-parametric analyses were done on the data since, with small sample sizes, ratios probably are not distributed nor- malty. Additional parametric analyses were also carried out on the data, including one-way ANOVA on the ratios them- selves and on the arc sine square root and log transforma- tion of the ratios. All statistical analyses produced the same results; primarily the non-parametric analyses will be re- ported exphcity.

Results

Nucleus mugnocelluluris

Fig. 1 shows a Nissl-stained section and its corresponding autoradiograph from an animal in which tracer was infused 6 h after cochlea re- moval. Note the relatively high blood flow in the

dorsolateral portion of the brain stem, which con- tains the auditory and vestibular nuclei. This labeling pattern is similar to that seen in experi-

Fig. 1. Low magnification photomierograph of a Nissl-stained brain stem section (top) and its corresponding autoradiograph (bottom). This animal was infused with tracer 6 hours after w&lea removal. Notch in the right ventral hwin stem marks the deafferented side. Arrows mark NM. Tracer uptake in the dorsdaceral brain stem containing auditory and vestibular nuclei is much greater than that of surrounding tissue. Tracer uptake in deafferented NM is less than that in normal NM. This pattern of blood flow is similar IO the pattern of &cose

uptake in NM. (sccrle bar = 1 mm).

Fig. 2. Quantitative analysis of blood flow changes in NM. Bars represent the mean ratio of blood flow tn NM ipsiiateral to cochlea removal to that on the contralateral side, f S.E.M. Asterisks denote values which are significantly different from control values (Mann-Whitney U. P c. 0.05). Blood flow in NM lpsilateral to cochlea removal decreases significantly by 3Q mm posllesion and remains dweascd at 6 h. There is no

slgmlicant diflerence between the two experimental groups.

ments in which glucose uptake is examined using radiolabeled 2-deoxyglucose (e.g. Heil and Scheich,

1986; Lippe et al. 1980). The right NM, ipsilateral to cochlea removal, appears less densely labeled than NM on the contralateral side of the brain

stem, suggesting decreased blood flow. Similar decreases in labeling in NM were seen 30 mm

after cochlea removal. Results of quantitative analysis of blood flow

changes in NM are shown in Fig. 2. The ratio of blood flow in NM ipsilateral to cochlea removal to that in NM contralateral to cochlea removal was computed for each animal. Average ratios for groups of animals are plotted against survival time. As expected, the ratio of average blood flow in NM for control animals is not different from unity (95% confidence limits for the control animals include 1.0). A between group Kruskal- Walks analysis of variance shows a reliable effect

of Group [H = 5.96, P < 0.05]. Post hoc pairwise comparisons (Mann-Whitney U) show that blood flow ratios at both 30 min and 6 h are lower than those of controls (P -C 0.05), but that blood flow ratios in these two experimental groups are not reliably different from one another (P < O,Ol>. Thus, there is a decrease in blood flow in ipsi- Iateral NM which is evident by 30 min postlesion with no apparent change in magnitude 6 h after deafferentation.

Nucleus laminaris We examined NL both ipsilateral and con-

tralateral to cochlea removal. On each side of the brain, the ratio of blood flow in the dendritic region in NL receiving input from deafferented NM to that in the region of NL receiving input from normally innervated NM was examined. If afferent activity can locally regulate blood flow with the same spatial resolution as other metabolic parameters, then deafferented and innervated re- gions of NL should demonstrate a difference in labeling, and the ratio of blood flow in these areas should be different from that in control animals. The results of this analysis are shown in Fig. 3a. In control animals, the ratio of blood flow in dorsal to ventral NL neuropil regions is nearly one, suggesting that blood flow is equal in dorsal and ventral dendritic regions. Following cochlea removal, the ratio of blood flow in the deaffer- ented dendritic region divided by the blood flow in the control dendritic region for each side of the brain stem is not significantly different from that in control animals at either 30 min or 6 h follow- ing cochlea removal (H = 1.42, P > 0.10 ipsi- lateral; H = 0.62, P > 0.10 contralateral). For all ratios in Fig. 3a, average ratios are not different from unity (95% confidence limits include 1.0). Thus, blood flow in the two dendritic domains of each NL remain constant with respect to one another following cochlea removal.

However, when we compared blood flow in NL dendritic regions on opposite sides of the brain we observed reliable differences (Fig. 3b). We calcu- lated the ratio of blood flow in either dorsal or ventral NL ipsilateral to cochlea removal to that in the corresponding dendritic region on the con- tralateral side of the brain. As before, in control animals these ratios are nearly unity, indicating no side-to-side differences in blood flow in NL den- dritic regions (95% confidence limits include 1.0). Following cochlea removal, however, these ratios are reliably less than those in control animals (H = 6.49 P < 0.01 dorsal; H = 6.25, P < 0.05, ventral). Post-hoc pairwise comparisons showed that ratios are not reliably different from controls 30 min postlesion but are reliably different 6 h after cochlea removal (Mann-Whitney U, P < 0.05).

Measurements in NL were usually more varia-

a.

hoperated 30 mm 6 hrs

Survival time after cochlea removal

b.

I Unoperated 30 min 6 hrs

Survival time after cochlea removal

Fig. 3. Quantitative analysis of blood flow changes in NL. Bars represent mean ratiofS.E.M. (a) The ratio of the blood flow in the deafferented region of NL to that of the control region of NL for each side of the brain (ipsilateral or contralateral to cochlea removal) is plotted against survival time. There are no significant differences between groups when compared in this manner. (b) The ratio of blood flow in either dorsal or ventral ipsilateral dendritic region of NL divided by that of the corre- sponding contralateral dendritic region of NL is plotted against survival time after cochlea removal. An asterisk indicates that a value is significantly different from that of control (P < 0.05). Both NLd and NLv show a statistically significant decrease in blood flow on the side ipsilateral to cochlea removal at

6 h post-lesion.

ble than those in NM, probably because the smaller neuropil areas in NL were harder to out- line. To further examine changes in NL blood flow, we calculated the ratios of blood flow in dorsal or ventral ipsilateral NL neuropil to that in contralateral NM. Since blood flow in con- tralateral NM presumably does not change, any difference in this ratio among groups would reflect differences in NL blood flow. Kruskal-Wallis analysis of variance showed that these ratios were reliably different for both dorsal (H = 7.2, P <

58

0.01) and ventral (II = 6.49, P < 0.01) neuropil regions. Post-hoc pairwise comparisons showed that for dorsal neuropil, both 30 min and 6 h groups were different from the control group. Only the 6 h group was different for ventral neuropil (Mann-W~tney I-I, P < 0.05).

Thus, there is a decrease in blood flow in both the dorsal dendritic region of NL and the ventral

dendritic region of NL ipsilateral to cochlea re- moval. This decrease is most reliably seen 6 h after

cochlea removal. The decreases in dorsal and ventral ipsilateral neuropil appear to be equiv- alent, since the ipsilateral dorsal to ventral ratio is not significantly different from unity (see above).

The relationship between changes in blood flow in NM and NL was also examined. The ratios for NLd and NLv from Fig. 3b were compared with NM ipsilateral/contralateral ratios for each animal. Blood flow changes in NLd correlate well with those in NM (Spearman rank correlation r = 0.77), suggesting that blood flow in NL and

NM may be coregulated. Blood flow changes in NLv correlate less well with those in NM (r =

0.60), suggesting that coregulation may be better coupled in NLd than NLv. Data from control animals suggests that NL blood flow is greater than that of NM. The ratio of NLd or NLv blood flow divided by NM blood flow is greater than

Fig. 4. Photo~cr~aph of blood vessels supplying brain stem

auditory nuclei. Broken black lines mark the outline of NM

and cell body region of NL. A Go&-stained impregnated

vessel (marked with white dots) branches to supply NM and

NL. The branch to NL forms an arcade coursing from NLd

through the single layer of cell bodies to NLv. Scale bar = 100 pm.

unity for both dorsal neuropil (1.35 _t 0.14 SEM) and ventral neuropil (1.44 4 0.13).

The spatial pattern of blood flow changes and the correlation between changes in NM and NL prompted us to investigate the pattern of blood vessels supplying the brain stem auditory nuclei. We examined Golgi-stained brain stem sections

containing impregnated blood vessels from previ-

ously prepared animals (Deitch and Rubel, 1984). One such section (Fig. 4) demonstrates a vessel branching to supply both NM and NL. The NL branch forms an arcade coursing from dorsal neu- ropil, through perikarya, to ventral neuropil. A rigorous tree-dimensional reconstruction of the vessels, which we have not done, would be neces- sary to determine whether the pattern shown pro- vides a possible explanation of the data.

In summary, there is a decrease in blood flow in NL on the side ipsilateral to cochlea removal which is of equivalent magnitude in NLd and NLv and which may be correlated with the decrease in

blood flow in NM.

Discussion

We found decreased blood flow in second- and third-order brain stem auditory nuclei ipsilateral to cochlea removal. Blood flow in NM decreases within 30 min of deafferentation and remains de- creased for at least 6 h. In NL, blood flow is decreased in both dorsal and ventral ipsilateral neuropil regions, a pattern which does not coin- cide with that of afferent innervation. The results also suggest blood flow may be coregulated to some degree in NM and NL. We will first discuss

some meth~olo~cal considerations relevant to our results, and then discuss the relations~p be- tween blood flow and metabolic activity.

Methodological considerations Three aspects of the methods we employed are

known to alter absolute blood flow, but probably do not affect the conclusions of this study. First, Jay and colleagues (1988) have shown recently that corrections for lag time and catheter washout can change the computed value of local blood flow by as much as 30% when working with small animals. Our experimental design meets the criteria for reducing these errors to insignificance, so we

59

have not included these corrections. Second, ketamine, used as an anesthetic in our experi- ments, has been shown to increase rates of cerebral blood flow and metabolism (Takeshita et al., 1972; Cavazzuti et al., 1987). Third, pentobarbital is known to cause hypotension and respiratory de- pression, which could affect systemic blood flow via changes in perfusion pressure and arterial carbon dioxide tension. Although variations in these factors may explain in part the ten-fold range in absolute blood flow we observed among individual animals, we compared blood flow as ratios between the two sides of the brain of the same animal, which prevents systemic errors from confounding the results.

Variability in the sound environment in which the blood flow measurements were taken may have caused differences in both the absolute blood flow and ratios of blood flow values from animal to animal. Injection of the isotope for blood flow measurements took place under ambient noise conditions. Thus the level of activity on the ‘con- trol’ side may have varied in different subjects. Without knowing the quantitative relationship be- tween blood flow and electrophysiological activity, it is not possible to determine to what extent variability in sound exposure might have in- fluenced absolute blood flow levels.

Blood flow and metabolic activity We expected that the spatial pattern of changes

in blood flow in NM and NL would correspond to the pattern of afferent innervation. Although this is the case in NM, blood flow changes in NL occur unilaterally rather than in accordance with the distribution of afferent activity. This disparity, illustrated in Fig. 5, argues for spatial regulation of blood flow in a different pattern than that in which neuronal activity is regulated.

The association between metabolic parameters and blood flow forms the basis of the Roy and Sherrington (1980) metabolic homeostasis hy- pothesis of blood flow regulation. While in many cases altered metabolic rate is accompanied by changes in blood flow (Lou et al., 1987; Fox and Raichle, 1986; Sokoloff, 1981; Cremer et al., 1983; Pasternak and Groothuis, 1984), several studies have demonstrated cases in which various mea- sures of metabolic rate and local blood flow do

Afferent Input

NM NM

NLd

NLv

NLd

NLv

Blood Flow

Fig. 5. Schematic diagram of afferent input (top) and blood flow (bottom) after cochlea removal. In the top panel, nuclei in which afferent activity decreases following right cochlea re- moval are shaded. In the bottom panel, nuclei with decreased blood flow following right cochlea removal are shaded. Blood flow in NLv decreases ipsilateral to cochlea removal, but not contralaterally. which is the inverse of the afferent

activity pattern.

not correlate well (Lou et al., 1987; Sundermann et al., 1985; Fox and Raichle, 1986; Raichle et al., 1987b; Powers et al., 1987). Raichle and col- leagues (1987b) showed that increases in local cerebral blood flow were 2 to 3 times the magni- tude of changes in local cerebral glucose consump- tion in response to somatosensory stimulation. Increased local blood flow does not appear to be necessary for normal neuronal activity induced by somatosensory stimulation (Powers et al., 1987). Knowledge of the quantitative relationship be- tween blood flow and various metabolic parame- ters may help resolve these discrepancies.

The spatial pattern of blood flow changes, stat- istical analyses suggesting correlations between blood flow changes in NM and NL, and the pattern of blood vessels supplying NM and NL all suggest that blood flow may be co-regulated in these two nuclei. Regional regulation may explain why both ipsilateral NLd and NLv show corre- lated, decreased blood flow. It is not clear why the correlation between NM and NL was weaker for ventral than dorsal NL. Coupling between NM and NL may explain the observation that blood flow in contralateral, denervated ventral NL does not decrease following cochlea removal. However,

60

the conclusion that blood flow to NM and NL is coregulated is compromised by the fact that while equal changes in blood flow were observed at 30 min and 6 h in NM, decreases could only be

demonstrated reliably at 6 h in NL. Because of their small size, NL dendritic regions are relatively

more difficult to measure than are other areas of

the brain stem, resulting in data which are more

variable. Analysis of ratios in which ipsilateral NL was compared to contralateral NM suggested that

at least for dorsal NL significant changes may be occurring at 30 min.

We examined the temporal and spatial char- acteristics of blood flow changes in part to see whether blood flow might be involved in the metabolic cascade of events following cochlea re- moval in the auditory system. For NL, the dif- ferences between the spatial pattern of blood flow and other cellular changes makes such a relation- ship unlikely. For NM, both the spatial pattern of blood flow changes and the rapidity with which

they occur suggest that some relationship is possi- ble. Although only experiments in which blood

flow is manipulated independently of cochlea re- moval could more appropriately address this ques- tion, several conclusions can be drawn from the present experiments. First, glucose uptake seems unlikely to be regulated by blood flow, since changes in glucose use in NM and NL are not followed by similar changes in blood flow. Sec- ond, protein synthesis and blood flow in ipsi- lateral NM both decrease within 30 min after cochlea removal (Steward and Rubel, 1985). How-

ever, activity-dependent changes in protein synthesis also are observed in vitro where blood

flow changes can not contribute to the effect (Hyson and Rubel, 1989). In addition, there is no difference in the amount of protein synthesis in glial cell bodies between sides after cochlea re- moval, even though blood flow differs on the two sides of the brain (Steward and Rubel, 1985).

In summary, the time course of blood flow changes in brain stem auditory nuclei is tempor- ally correlated with the earliest metabolic and morphological changes, but the spatial pattern of blood flow is not consistent with these other changes. Many factors may influence blood flow, including perfusion pressure, capillary recruit- ment, and arterial 0, and CO2 tensions (Phelps et

al., 1981; Tyson et al., 1987; Collins et al., 1987; Raichle et al., 1987a; Schmidt et al., 1945; Lassen, 1959). Further experiments will be needed to de- termine which of them may be most important for local regulation of blood flow.

Acknowledgments

We wish to acknowledge Katie Swanson for producing the program to evaluate the Kety in- tegral; Rick Hyson for comments on the manu- script, statistical advice, and assistance and gui- dance in the use of the Bioquant image analysis system; Judy Debel for technical assistance; Geo- rge Hashisaki for help on some of the early pilot studies; Shannon Wood for help with preparation of the manuscript; and Greg Hyde, Nell Cant, and Edwin Rubel for comments on the manuscript. This work was supported by NIH grants NS24518 and NS26521. B.E.R. was supported by the Medi- cal Student Research Traning Program at the Uni- versity of Washington.

References

Born, D.E. and Rubel. E.W (1984) Cochlea removal eliminates

physiological activity in brain stem auditory nuclei of the

chicken. Sot. Neurosci. Abstr. 10, 843.

Born, D.E. and Rubel, E.W (1985) Afferent influences on

brain stem auditory nuclei of the chicken: neuron number

and size following cochlea removal. J. Comp. Neurol. 231,

435-445.

Cavazzuti, M., Porro, CA., Biral, G.P. Benassi, C. and Barbieri,

G.C. (1987) Ketamine effects on local cerebral blood flow

and metabolism in the rat. J. Cerebral Blood Flow Meta-

bol. 7, 806-811.

Collins, R.C., Wagman. I.L.. Lymer, L. and Matter, J.M.

(1987) Distribution and recruitment of capillaries in rat

brain. J. Cerebral Blood Flow Metabol. 7. S336.

Cowan, W.M. (1970) Anterograde and retrograde transneuro-

nal degeneration in the central and peripheral nervous

system. In: W.J.H. Nauta and S.O.E. Ebbesson (Eds.),

Contemporary Research Methods in Neuroanatomy.

Springer, Berlin, pp. X7-249.

Cremer, J.E., Cunningham, V.J. and Seville, M.P. (1983) Rela-

tionship between extraction and metabohsm of glucose,

blood flow, and tissue blood volume in regions of rat brain.

J. Cerebral Blood Flow and Metabolism 3, 291-302.

Deitch. J.A. and Rubel, E.W (1984) Afferent influences on

brain stem auditory nuclei of the chicken: Time course and

specificity of dendritic atrophy following deafferentation. J.

Comp. Neurol., 229, 66-79.

61

Durham, D. and Rubel, E.W (1985) Afferent influences on

brain stem auditory nuclei of the chicken: Changes in

succinate dehydrogenase activity following cochlea re-

moval. J. Comp. Neurol. 231, 446-456.

Fox, P.T. and Raichle, M.E. (1986) Focal physiological uncou-

pling of cerebral blood flow and oxidative metabolism

during ~mato~so~ st~ulation in human subjects. Proc.

Natl. Acad. Sci. 83, 1140-1144.

Globus, A. (1975) Brain morphology as a function of presyn-

aptic morphology and activity. In: A.H. Riesen (Ed.), De-

velopmental Neuropsychology of Sensory Deprivation,

Academic Press, New York, pp. 9-91.

Guillery, R.W. (1974) On structural changes that can be pro-

duced experimentally in the mammalian visual pathways.

In: R. Bellairs and E.G. Gray (Eds) Essays on the Nervous

System. Clarendon Press, Oxford, pp. 299-326.

Heil, P. and Scbeich, H. (1986) Effects of unilateral and

bilateral cochlea removal on 2-deoxyglucose patterns in the

chick auditory system. J. Comp. Neurol. 252, 279-301.

Hyde, G.E. and Durham, D. (1989) Afferent influences on

brain stem auditory nuclei of the chicken: Changes in

cytochrome oxidase activity following cochlea removal. As-

soc. Res. Otolaryngol., St. Petersburg Beach, Florida.

Hyson, R.L. and Rubel, E.W (1989) Transneuronal regulation

of protein synthesis in the brain-stem auditory system of

the chick requires synaptic activation. J. Neurosci. 9, 2835-

2845.

Jay. T.M., Lucignani, G., Crane, A.M., Jehle, J. and Sokoloff,

L. (1988) Measurement of local cerebral blood flow with

[‘4C]10doantipyrine in the mouse. J. Cerebral Blood Flow

Metabol. 8, 121-129.

Lassen, N.A. (1959) Cerebral blood flow and oxygen consump-

tion in man. Physiol. Rev. 39, 183-238.

Lassen, N.A. (1982) Regulation of cerebral blood flow: Com-

ments on a few specific problems. Develop. Neurosci. 14,

535.

Levi-Montalcini, R. (1949) Tbe development of the acoustico-

vestibular centers in the chick embryo in the absence of the

afferent root fibers and of descending fiber tracts. J. Comp.

Neurol. 91, 209-242.

Lippe, W.R., Steward, 0. and Rubel, E.W. (1980) The effect of

unilateral basilar papilla removal upon nuclei laminaris and

ma~~llul~s of the chick examined with [3H]2-deoxy-

D-glucose autoradiography. Brain Res. 196, 43-58.

Lou, H.C., Edvinsson, L. and MacKenzie, E.T. (198’7) The

concept of coupling blood flow to brain function: Revision

Required? AM. Neurol. 22, 289-297.

Parks, T.N. (1979) Afferent influences on the development of

the brain stem auditory nuclei of the chicken: otocyst

ablation. J. Comp. Neurol. 183, 665-678.

Parks, T.N. and Rubel, E.W (1978) Organization and develop-

ment of the brain stem auditory nuclei of the chicken:

primary afferent projections. J. Comp. Neurol. 180, 439-

448.

Pastemak, J.F., Groothuis, D.R. (1984) Regional variability of

blood flow and glucose utilization within the subependymal

germinal matrix. Brain Res. 299, 281-288.

Phelps, ME., Huang, S.-C., Hoffman, E.J., Selin, C. and Kuhl,

D.E. (1981) Cerebral extraction of N-13 ammonia: its

dependence on cerebral blood flow and capillary permea-

bility - surface area product. Stroke 12, 607-619.

Powers, W.J., Fox, P.T., Snyder, A.Z. and Raichle, M.E. (1987)

Normal neuronal function does not require normal physio-

logic increase in CBF. J. Cerebral Blood Flow Metabol. 7,

s310.

Raicble, M.E., Fox, P.T., Herscovitch, P. and Mintun, M.A.

(1987a) Capillary recruitment occurs during focal physio-

logical activation in the human brain. J. Cerebral Blood

Flow Metabol. 7, S338.

Raichle, M.E., Fox, P.T., Mintun, M.A. and Dense, C. (1987b)

Cerebral blood flow and oxidative glycolysis are uncoupled

by neuronal activity. .I. Cerebral Blood Flow Metabol. 7,

s300.

Roy, C.S. and Sherrington, C.S. (1980) On the regulation of the

blood-supply of the brain. J. Physiol. (London) 11, 85-108.

Sakurada, O., Kennedy, C., Jehle, J., Brown, J.D., Carbin, G.L.

and Sokoloff, L. (1978) Measurement of local cerebral

blood flow with iodo[‘4C]antipyrine. Am. J. Physiol. 3,

H59-H66.

Schmidt, C.F., Kety, S.S. and Pennes, H.H. (1945) The gaseous

metabolism of the brain of the monkey. Am. J. Phys. 143,

33-52.

Smith, D.E. (1977) The effect of deafferentation on the devel- opment of brain and spinal nuclei. Prog. Neurobiol. 8,

349-367.

Sokoloff, L. (1981) Relationships among local functional activ-

ity, energy metabolism, and blood flow in the central

nervous system. FASEB 40, 2312-2316.

Steward, 0. and Rubel, E.W (1985) Afferent influences on

brain stem auditory nuclei of the chicken: cessation of

amino acid incorporation as an antecedent to age-depen-

dent transneuronal degeneration. J. Comp. Neurol. 231,

385-395.

Sundermann. R., Gitelman, D.. Toga, A.W. and Collins, R.C.

(198s) There is a loose relationship between cerebral blood

flow and metabolism during visual stimulation. Sot. Neuro-

sci. Abstr. 11, 1125.

Takeshita, H., Okuda, Y. and Sari, A. (1972) The effects of

ketamine on cerebral circulation and metabolism in man.

Anesthesiology 36, 69-75.

Tyson, G., Gross, P., Tajima, T., Nakata, H., Pettigrew, K.,

Patlak, C. and Fenstermacher, J. (1987) Local cerebral

blood flow, size of capillary beds, and velocity of capillary

blood flow in seven brain structures, J. Cerebral Blood

Flow Metaboi. 7, S341.