- Email: [email protected]
Biliverdin-induced brainstem auditory evoked potential abnormalities in the jaundiced Gunn rat
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s
Research Report
Biliverdin-induced brainstem auditory evoked potential abnormalities in the jaundiced Gunn rat Ann C. Rice ⁎, Steven M. Shapiro Department of Neurology, Box 980599, Virginia Commonwealth University, Richmond, VA 23298-0599, USA
A R T I C LE I N FO
AB S T R A C T
Article history:
Brainstem auditory evoked potential (BAEP) abnormalities occur in jaundiced Gunn rats
Accepted 2 June 2006
given sulfadimethoxine to displace bilirubin bound to serum albumin, releasing it into the
Available online 7 July 2006
tissues. One problem with the model is that after displacement, plasma bilirubin levels drop and do not correlate with neurological dysfunction. In this report, we administered
Keywords:
biliverdin, the immediate precursor of bilirubin, in 15- to 17-day-old Gunn rat pups to create
Kernicterus
an improved model of bilirubin-induced neurological dysfunction. Total plasma bilirubin
Bilirubin
(TB) levels were measured with a Leica bilirubinometer. Biliverdin (40 mg/kg) or phosphate-
Neurotoxicity
buffered saline (PBS) was administered either once and BAEPs recorded 8 h later or twice, 12 h apart, and BAEPs recorded 24 h after the initial injection. A single biliverdin injection
Abbreviations:
produced a significantly decreased amplitude of BAEP wave III, 1.21 ± 0.25 vs. 0.49 ± 0.27 μV
ABRs, auditory brainstem responses
(control vs. biliverdin). The two-injection paradigm resulted in a significantly elevated TB
BAEPs, brainstem auditory evoked
(9.9 ± 1.2 vs. 14.9 ± 3.1 mg/dl; control vs. biliverdin), significant increases in I–II (1.15 ± 0.08 vs.
potentials
1.42 ± 0.09 ms) and I–III (2.17 ± 0.08 vs. 2.5 ± 0.13 ms) interwave intervals and a decrease in the
BIND, bilirubin-induced neurological
amplitude of wave III (1.36 ± 0.30 vs. 0.38 ± 0.26 μV). Additionally, there were significant
dysfunction
correlations between TB and the amplitude of wave III (r2 = 0.74) and TB and the I–III
BV, biliverdin
interwave interval (r2 = 0.51). In summary, biliverdin administration in jaundiced Gunn rat
sulfa, sulfadimethoxine
pups produces BAEP abnormalities consistent with those observed in the sulfadimethoxine
TB, total plasma bilirubin
model and human newborn hyperbilirubinemia and resulted in increased plasma bilirubin levels that correlate with the degree of neurological dysfunction. © 2006 Elsevier B.V. All rights reserved.
1.
Introduction
The acute and chronic manifestations of excessive neonatal hyperbilirubinemia, acute bilirubin encephalopathy in newborns and chronic bilirubin encephalopathy also known as kernicterus and/or bilirubin-induced neurological dysfunction (BIND) in older infants, children and adults are on the rise in the United States in part due to early discharge of newborns from the hospital (for reviews, see Maisels and Newman, 1995; Maisels and Watchko, 2000; Johnson and
Bhutani, 1998; Dennery et al., 2001; Shapiro, 2003; Stevenson et al., 2004; Kaplan and Hammerman, 2004, 2005). Total bilirubin (TB) levels in plasma increase normally 3–5 days after birth due to immaturity of the enzyme uridine diphosphate (UDP) glucuronysyl transferase, which functions to add glucuronic acid residues to bilirubin rendering it more water soluble and promoting its excretion into bile (Maisels and Watchko, 2000; Shapiro, 2003; Kaplan and Hammerman, 2004; Stevenson et al., 2004). This increase in TB is generally self-correcting by upregulation of the transferase enzyme
⁎ Corresponding author. Fax: +1 804 828 5654. E-mail address: [email protected] (A.C. Rice). 0006-8993/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2006.06.005
216
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –22 1
(Stevenson et al., 2004). However, in cases where the TB levels rise excessively, signs of acute damage can appear, with subsequent development of a range of neurological sequelae, which include auditory dysfunction, impaired upward gaze, dystonia, athetosis and other cerebral palsy-like movement disorders (Shapiro, 2003; Kaplan and Hammerman, 2004; Stevenson et al., 2004; Shapiro, 2005). One major difficulty with the diagnosis of BIND is that the elevated bilirubin levels occur shortly after birth, but irreversible neurological dysfunction may take months or years to recognize. The current practice is that if a high TB is detected in the newborn, treatments such as phototherapy and, in extreme cases, exchange transfusion are initiated to attempt to prevent the irreversible neurological damage. Tests such as brainstem auditory evoked potentials (BAEPs), also known as auditory brainstem responses (ABRs), can be performed to assess auditory function during and after hyperbilirubinemia to determine if there is likely to be permanent damage (Bratlid, 1991; Kaplan and Hammerman, 2004; Shapiro, 2005). BAEP recordings detect the efficacy of neural transmission between brainstem auditory nuclei, one of the brain areas damaged by high bilirubin levels. Another difficulty with the diagnosis of BIND stems from identifying the level of plasma bilirubin responsible for the range of neurological dysfunctions because often there are additional factors involved that influence brain concentrations of bilirubin (Bratlid, 1991; Maisels and Watchko, 2000; Stevenson et al., 2004; Shapiro, 2005; Wennberg et al., 2006). Furthermore, it is not known how high levels of plasma bilirubin cause permanent devastating neurological dysfunction. The Gunn strain of Wistar rat (Gunn, 1938) is presently the best animal model of BIND. These animals have a spontaneous mutation of the UDP-glucuronysyl transferase enzyme (Strebel and Odell, 1971), and the homozygous recessive (jj) animals exhibit significantly elevated TB levels (Johnson et al., 1959, 1961; Schutta and Johnson, 1969). Bilirubin levels peak around 15–17 days of age and slowly drop over the lifetime of the animal (Johnson et al., 1961; Sawasaki et al., 1976; personal observation). In the circulation, bilirubin is tightly bound to serum albumin (Brodersen, 1978; Rose and Wisniewski, 1979; Shimabuku et al., 1983; Shapiro, 1988). When the albumin binding capacity is exceeded, the free or unbound bilirubin leaves the circulation and enters the tissues including the brain. The untreated jj animals often exhibit mild ataxia due to the consistently observed cerebellar hypoplasia (Schutta and Johnson, 1967; Sawasaki et al., 1976; Rose and Wisniewski, 1979; Yamamura and Takagishi, 1993; Conlee and Shapiro, 1997). However, more severe neurological sequelae can be induced by administration of sulfadimethoxine (sulfa), which displaces the bilirubin bound to albumin allowing even higher levels to enter the tissues (Blanc and Johnson, 1959; Diamond and Schmid, 1966; Schutta and Johnson, 1967; Rose and Wisniewski, 1979; Shapiro, 1988, 2002; Rashid et al., 1998). Dramatic BAEP abnormalities are observed within hours following sulfa administration to jj pups (Shapiro, 1988, 1993, 2002; Shapiro and Hecox, 1988). Additionally, sulfa-treated jj pups typically develop severe dystonia after a day or two, thereby exhibiting many of the same characteristic neurological sequelae observed in humans.
In the Gunn rat sulfa model of bilirubin-induced neurotoxicity, plasma bilirubin levels drop to near zero, making it difficult to correlate the plasma level of bilirubin, pre- and post-injection of sulfonamide, with the neurological sequelae. Biliverdin (BV) is the immediate precursor to bilirubin in the catabolism of hemoglobin and is more soluble and less neurotoxic than bilirubin. In this report, we assess whether BV administration raises plasma bilirubin to levels that produce neurological dysfunction, as an improvement in the Gunn rat model to enable the study of varying degrees of BIND with the advantage of detectable increased plasma bilirubin levels.
2.
Results
2.1.
Total plasma bilirubin (TB) levels
A time course response of BV was assessed on 15-day-old jj rat pups to determine if TB levels could be significantly elevated by biliverdin administration. Fig. 1 depicts the time course of TB levels following a single injection of 40 mg/kg BV, phosphate-buffered saline (PBS, BV-vehicle) and 200 mg/kg sulfa or saline (sulfa-vehicle). A pre-injection blood sample was drawn, the injection administered and a post-injection blood sample drawn at a specified time for each animal. Time course curves were compared using a split-plot analysis of variance (ANOVA). Plasma bilirubin levels were significantly elevated by a single injection of BV and decreased by sulfadimethoxine, whereas the two vehicle groups were significantly different from each drug group, but not from each other (F3,99 = 570). In the BV group receiving a single injection of BV, the plasma bilirubin level peaked 4 h after the injection, and by 24 h plasma bilirubin levels recovered to near pre-injection levels.
Fig. 1 – Plasma bilirubin levels. Plasma bilirubin levels following a single injection of biliverdin (40 mg/kg), sulfa (200 mg/kg), PBS (BV-vehicle) or saline (sulfa-vehicle). Blood was drawn twice from each animal at 0 h and a designated time post-injection. Values are mean ± SD; n = 5 at each time point.
217
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
2.2.
Brainstem auditory evoked potentials (BAEPs)
To validate BV treatment as a model of BIND, we performed BAEPs to assess auditory dysfunction in another set of animals. BAEPs are computer averaged surface recorded electrical potentials in response to an auditory stimulus. The resultant waveforms reflect neural transmission in the afferent auditory pathways between the auditory nerve and auditory brainstem nuclei.
2.2.1.
Eight-hour single injection experiment
A baseline BAEP was recorded on 15-day-old jj pups, then a single injection of BV (40 mg/kg) or PBS (equivalent volume) was injected and 8 h later a second BAEP was recorded. The 8-h BAEP latencies and amplitudes were compared between groups (Table 1). There was a small but significant decrease in the amplitude of wave III after BV injection, whereas no other parameters were significantly different between the two groups. Due to the small change in BAEPs following a single injection of BV, we examined the effects of two BV injections on BAEPs in another group of animals.
2.2.2.
Twenty-four-hour double injection experiment
For TB level determination, an initial blood sample was drawn prior to injection and a second one 24 h after the initial BV injection following the BAEP recording. BV (40 mg/ kg) was administered twice, 12 h apart, in 16-day-old jj pups and BAEPs recorded 24 h after the first injection. The physiological parameters are displayed in Table 2. Initial mean body weights did not differ significantly between groups; however, 24 h after the first injection the BV group gained significantly less weight than the PBS group. Weight loss is consistently observed following sulfa administration in auditory impaired animals (Conlee and Shapiro, 1991). The pre-BV injection TB levels were not different between groups;
Table 2 – Physiological parameters
jj-BV jj-PBS P
N
Initial Weight (g)
Weight Gain (g)
8 7
26.9 ± 2.6 26.8 ± 3.0 NS
2.0 ± 1.0 3.2 ± 0.6 0.0189**
Initial bilirubin mg/dl 10.1 ± 1.1 10.8 ± 1.2 NS
24 h bilirubin mg/dl 14.9 ± 3.1 9.9 ± 1.2 0.0014**
Physiological parameters in jaundiced rat pups before and 24 h after the first of 2 injections of biliverdin (jj-BV) or vehicle (jj-PBS) 12 h apart. These animals had BAEPs recorded at 24 h after 2 injections and the final blood sample drawn after the BAEP recording. Values are mean ± SD. Significance is at P = 0.05. NS = not significant.
however, at 24 h the TB levels were significantly increased by approximately 50% in the BV group. Fig. 2 depicts representative waveforms observed following two injections of either PBS or BV (40 mg/kg) in jj pups. Vertical dashed lines from the middle of each peak in the PBS-treated animal to the waveforms in the BV-treated animal demonstrate the increase in latency of waves II and III. BAEP latencies, interwave intervals and amplitudes in jj pups given two injections of either BV (40 mg/kg) or PBS (equivalent volume) are given in Table 3. There were statistically significant increases in BAEP I–II and I–III interwave intervals and decreases in the amplitude of wave III in the BV-treated group compared to PBS-treated animals. The TB levels obtained at the time of the BAEP from the double injection experiment were compared to the significantly changed BAEP parameters. BAEP wave III amplitude
Table 1 – BAEP data following a single injection of biliverdin Group
N
jj-BV jj-PBS P
10 5
Group
N
jj-BV jj-PBS P
10 5
Latency (ms) I
I–II
I–III
1.46 ± 0.18 1.35 ± 0.15 NS
1.29 ± 0.16 1.22 ± 0.05 NS
2.48 ± 0.21 2.26 ± 0.07 0.046
Amplitude (μV) I
II
III
1.17 ± 0.25 1.05 ± 0.23 NS
0.82 ± 0.48 1.10 ± 0.4 NS
0.49 ± 0.27 1.21 ± 0.25 0.00027**
BAEP changes following a single injection of biliverdin. Biliverdin (40 mg/kg) or PBS was injected into jaundiced Gunn rat pups (jj-BV and jj-PBS, respectively). Eight hours later, BAEPs were recorded. The latency of wave I and the interwave interval between waves I to II and I to III are shown (upper half). The amplitudes of waves I, II and III are reported (lower half). All values are mean ± SD, significant P values are at 0.0083 denoted by **. NS = not significant.
Fig. 2 – BAEP waves following 2 injections of Biliverdin (40 mg/kg) or PBS. Representative BAEP waves recorded 24 h after the first injection of 2 injections 12 h apart in an animal given PBS (upper) and an animal given BV (lower). Waves I, II and III are identified. A dashed line drawn vertically from the peak of the waves in PBS-treated animal to the waveform of the BV-treated animal shows the changes in latencies.
218
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –22 1
Table 3 – BAEP data following two injections of biliverdin Group
N
jj-BV jj-PBS P
8 7
Group
N
jj-BV jj-PBS P
8 7
Latency (ms) I
I–II
I–III
1.19 ± 0.06 1.27 ± 0.13 NS
1.42 ± 0.09 1.15 ± 0.08 0.00003**
2.50 ± 0.13 2.17 ± 0.08 0.00006**
Amplitude (μV) I
II
III
1.19 ± 0.35 1.23 ± 0.26 NS
1.08 ± 0.3 1.36 ± 0.48 NS
0.38 ± 0.26 1.36 ± 0.30 0.000012**
BAEP data following two injections of biliverdin. Biliverdin (40 mg/ kg) or PBS was injected twice into jaundiced Gunn rat pups (jj-BV and jj-PBS, respectively) 12 h apart. BAEPs were recorded 24 h after the first injection. The latency of wave I and the interwave interval between waves I to II and I to III are shown (upper half). The amplitudes of waves I, II and III are reported (lower half). All values are mean ± SD, significant p values are at 0.0083 denoted by **. NS = not significant.
was inversely correlated to the TB. Linear regression shows a significant inverse correlation between the amplitude of wave III and the total plasma bilirubin level (r2 = 0.7111, P = 7.84 × 10−5); however, there was a better fit of the data with a power function (r2 = 0.7417; Fig. 3). Additionally, there was a significant linear correlation between TB and the I–III interwave interval (r2 = 0.5056, P = 0.00297; data not shown). These correlations support our hypothesis that the TB level can potentially indicate the amount of auditory dysfunction in this model in contrast to the sulfa model where they do not.
3.
The BAEP is a very sensitive tool for studying bilirubininduced auditory dysfunction. In rats (Huang, 1980), and generalizing from studies in other mammals (Buchwald and Huang, 1975; Huang, 1980; Huang and Buchwald, 1977, 1978; Melcher and Kiang, 1996; Melcher et al., 1996a,b) and humans (Mρller et al., 1981) and correlation of abnormalities of BAEP waves with neuroanatomical changes in our rat model (Conlee and Shapiro, 1991; Shapiro and Conlee, 1991), waves I, II and III in the rat correspond to waves I, III and V in human and reflect neural transmission between the auditory nerve (wave I) and the cochlear nucleus (wave II in rat, III in human) and midbrain structures (wave III in rat, V in human). In previous studies, we have recorded normal cochlear microphonic responses in animals with abnormal BAEPs in this model (Shapiro and TeSelle, 1994) and shown histologically normal cochlea with abnormalities of spiral ganglion and auditory nerve (Shaia et al., 2005) supporting the fact that the waveforms characterized represent brainstem and not cochlear damage. With the single BV injection experiment, the only abnormality observed was with the amplitude of wave III, probably reflecting relatively minor brainstem dysfunction, whereas the double injection experiment, with higher plasma bilirubin levels, decreased amplitudes of waves II and III and increased interwave intervals between waves I–II and waves I– III, displayed a greater degree of brainstem dysfunction. Both BV experiments, however, exhibited less auditory dysfunction than generally observed with the sulfa model, which typically displays more severe latency increases and amplitude decreases in waves II and III, often leading to disappearance of these waves, as well as abnormalities and disappearance of wave I, causing total absence of BAEP (Shapiro, 2002). Therefore, not only is BV administration an additional relevant
Discussion
In summary, we have demonstrated that biliverdin administration to jaundiced Gunn rat pups significantly elevates plasma bilirubin levels. Furthermore, these animals exhibit brainstem auditory dysfunction as assessed with BAEP recordings. The pattern of BAEP abnormalities (increased latency of waves II and III and decreased amplitude of waves II and III) is similar to that seen with sulfonamide injection in jj Gunn rat pups (Shapiro, 1988, 1993). Thus, the BV model closely mimics the Gunn rat sulfa model and the human disorder and provides us with another model to assess how hyperbilirubinemia in the newborn produces such devastating neurological dysfunction. This new model has the distinct advantage over the sulfa model in that there is a direct correlation of plasma bilirubin level with neurophysiological function. Therefore, we can begin to understand in what situations high plasma bilirubin levels produce neurological dysfunction. The disadvantage of our model however is that due to the limited solubility of BV, a large volume of fluid and at least two injections have to be administered. It is possible that with a shorter interval between injections an overall higher plasma bilirubin level could be obtained. Additional studies will further characterize BV-induced neurological dysfunction compared with the wellknown sulfa model.
Fig. 3 – Correlation between plasma bilirubin level and amplitude of BAEP wave III. The correlation compares total plasma bilirubin 24 h after the first injection of 2 BV or PBS injections, 12 h apart in both BV (gray circles)- and PBS (open circles)-treated animals. The correlation best fits a power function (y = 693.99x −2.8162; r 2 = 0.7417); n = 8 BV-treated animals; n = 7 PBS-treated animals.
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
model of BIND, but we can study milder abnormalities with greater consistency and reproducibility. The strong correlations between TB and the conduction time between the auditory nerve and midbrain (I–III interwave interval) and the amplitude of wave III demonstrates that Gunn rats given BV may be a useful model to begin to identify specific neurological sequela likely to develop at defined TB levels. Unbound (free) bilirubin in uninjected jj pups (Ahlfors and Shapiro, 1997) and humans (Ahlfors, 2000) is a better indicator of BAEP dysfunction than TB alone. Total bilirubin and unbound bilirubin can be compared during acute bilirubin encephalopathy in this model, whereas with displacers such as sulfa, both total and free bilirubin drop to near zero and their relative predictive values pre- and post-injection would be expected to be lost. The BV model of BIND characterized thus far produces a milder form of neurological impairment than the sulfa model. As apparent from Fig. 1, some animals had peak plasma bilirubin levels in the upper teens; however, animals that survived 48 h from other experiments not presented here did not display the characteristic dystonia observed with the sulfa model. This indicates that at least in our animal model, the auditory system may be more sensitive to bilirubin toxicity than the basal ganglia. Alternatively, the basal ganglia may require significantly more damage to occur before abnormalities appear, and/or that our classic behavioral assessment is not as sensitive in assessing basal ganglia function as BAEPs are in assessing auditory function. In conclusion, we are developing and characterizing a model of BIND that allows us to study milder neurological sequelae in the whole animal. This is relevant to the human cases of learning disabilities that ultimately may be solely the product of milder hearing impairment or a central auditory processing disorder for which there is presently no rodent model.
4.
Experimental procedures
All procedures were approved by the Institutional Animal Care and Use Committee. All efforts were made to minimize pain and stress to the animals and to reduce the number of animals used. The spontaneously jaundiced Gunn strain of Wistar rats used for all experiments was obtained from our in-house breeding colony. Experiments were performed on 15- to 17day-old rat pups when endogenous bilirubin levels are significantly elevated. These studies used only the homozygous recessive (jaundiced, jj) pups. Biliverdin hydrochloride was purchased from Frontier Scientific (Logan, UT). BV was sonicated until it dissolved in phosphate-buffered saline (PBS) at 0.4 mg/ml and injected at 40 mg/kg, i.p. An equivalent volume of PBS was injected into control animals. Due to the limited solubility of BV, the injection volume is 10% of the animal's body weight. Sulfadimethoxine (sulfa) diluted in saline was injected at 200 mg/kg in a volume of 1% of the animal's body weight.
4.1.
Total plasma bilirubin (TB) levels
Blood samples were collected from each animal via a cheek puncture into a heparinized hematocrit tube and centrifuged.
219
Two samples were obtained from each animal (pre- and posttreatment for both time course and double BV injection BAEP experiments). Animals were sacrificed after the second blood sample was drawn. Plasma bilirubin levels were assessed using a Leica Unistat Bilirubinometer. This instrument requires 20 μl of plasma for analysis. Generally, 50–75 μl of blood were collected for each sample. If excessive bleeding occurred, the cheek was pinched for a few seconds to control the bleeding.
4.1.1.
BV time course curve
A total of 103 animals were used for the time course experiment. Animals were randomly assigned to each sampling time point prior to the initial blood sample being collected. Immediately following the initial blood sampling, the animal was injected with either a single 40 mg/kg BV dose or equivalent volume of PBS. Total plasma bilirubin values were compared by a split-plot ANOVA with a Tukey post hoc test using SSPS software for the time course curve experiment, or paired or unpaired t test using Microsoft Excel for the BAEP experiments.
4.2.
Brainstem auditory evoked potentials (BAEPs)
Thirty jj rat pups were used for these experiments. Fifteen were used for the single injection 8-h experiment and fifteen were used for the 24-h double injection experiment. Weightmatched paired littermates from at least 11 different litters were used to account for inter-litter variation. Pups were anesthetized with an intramuscular injection of 60 mg/kg ketamine and 6 mg/kg acepromazine. BAEPs were recorded using a Nicolet Spirit 2000 evoked potential system (Biosys, Inc.). The left ear was occluded with petrolatum, and BAEPs were obtained to monaural 100-μs duration rarefaction clicks delivered at 31.7/s to the right ear through a Sony Walkman 4LIS headphone speaker, a small speaker we have used previously with response characteristics relatively flat from about 250 to 18,000 Hz (Shapiro and Hecox, 1989). The sound intensity was nominally set at 70 dB, which we calibrated biologically and corresponded to a level of about 62 dB above a normal Gunn rat pup BAEP threshold level as described below. Surface electrical activity was recorded from 13 mm long subcutaneous platinum needle electrodes inserted on the scalp over the vertex and behind the left and right mastoid bullae with a ground electrode in the flank. Rectal temperature was controlled at 37.0 ± 0.1 °C using a heating blanket and red heat lamp and stabilized for a minimum of 5 min before and during the recording. Two channel BAEP recordings were obtained from the contralateral to the ipsilateral mastoid (horizontal) and the vertex to the ipsilateral mastoid (vertical) electrode pairs, filtered from 30 to 3000 Hz. As there was not a qualitative difference between the two orientations, only the horizontal data are presented in this report. The vertical channel was used to help identify uncertain peaks. Each individual BAEP was the averaged response to 2000 stimuli, and three or more replicated responses were obtained for each animal. The individual BAEP replications were summed, and the peak latencies and peak-to-trough amplitudes of BAEP waves I, II and III were scored using a cursor. Latencies were subtracted to obtain interwave intervals. Eight-hour single injection experiment—A baseline BAEP was recorded, the animal was injected with either 40 mg/kg BV or PBS
220
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –22 1
(equivalent volume) and 8 h later a second BAEP was recorded. Twenty-four-hour double injection experiment—A pre-injection blood sample was drawn from the left cheek to ensure there was no damage to the ear canal on the side being subjected to BAEP testing. Then either 40 mg/kg BV or PBS (equivalent volume) was injected. Twelve hours later a second injection was administered and 12 h after that (24 h after original injection) a BAEP was recorded followed by a second blood sample being drawn. For biological calibration of the system, BAEP threshold levels were recorded in six 16-day-old, phenotypically normal heterozygous non-jaundiced Gunn rat pups from different litters at successively lower intensities. In 5 of 6 animals, replicable responses were obtained at 8 dB and no replicable response at 7 dB, and in 1 animal there were replicable responses at 7 dB and no response at 6 dB. The criteria for no replicable responses was defined as no waves ≥ 0.02 μV present in any of 3 recordings, 8000 sweeps each, with no or minimal muscle activity on the raw EEG, and with absent BAEP responses appearing essentially flat without significant artifact. For threshold determination, a rate of 91.7/s was used; anesthesia and recording conditions were otherwise identical to those in the experiments. All recordings were done in a relatively quite room, but machine noise from the Nicolet Spirit placed about 2 m from the rat likely caused some masking near threshold. We have previously shown that I–II and I–III interwave intervals and the amplitudes of waves II and III are the important BAEP variables affected by bilirubin toxicity in the sulfa model (Shapiro, 1988, 1993). For statistical analysis, we chose to examine 6 relatively independent BAEP-dependent variables a priori: the latency of wave I, I–II and I–III interwave intervals and the amplitudes of waves I, II and III in the mastoid to mastoid electrode montage. To correct for multiple independent observations, we divided an overall experimentwise p value of 0.05 by 6 to obtain an individual criterion p value of 0.0083 (Keppel, 1982; Lagakos, 2006).
Acknowledgments This work was supported by NIH R01 grants DC00369 and NS47151 to SMS. We would like to thank Dr. Charles E. Ahlfors who initially suggested using BV in our model. Additionally, we would like to thank Drs. Claudio Tiribelli and J. Donald Ostrow for helpful discussions in planning the project.
REFERENCES
Ahlfors, C.E., 2000. Measurement of plasma unbound unconjugated bilirubin. Anal. Biochem. 279, 130–135. Ahlfors, C.E., Shapiro, S.M., 1997. Brainstem auditory evoked potentials (BAEPs) and bilirubin-albumin binding in jaundiced (jj) Gunn rat pups. Pediatr. Res. 41, 35A. Blanc, W.A., Johnson, L., 1959. Studies on kernicterus: relationship with sulfonamide intoxication, report on kernicterus in rats with glucuronyl transferase deficiency and review of pathogenesis. J. Neuropathol. Exp. Neurol. 18, 165–189. Bratlid, D., 1991. Bilirubin toxicity: pathophysiology and assessment of risk factors. N.Y. State J. Med. 91, 489–492.
Brodersen, R., 1978. Determination of the vacant amount of high-affinity bilirubin binding sites on serum albumin. Acta Pharmacol. Toxicol. 42, 153–158. Buchwald, J.S., Huang, C.-M., 1975. Far-field acoustic response: origins in the cat. Science 189, 382–384. Conlee, J.W., Shapiro, S.M., 1991. Morphological changes in the cochlear nucleus and nucleus of the trapezoid body in Gunn rat pups. Hear. Res. 57, 23–30. Conlee, J.W., Shapiro, S.M., 1997. Development of cerebellar hypoplasia in jaundiced Gunn rats: a quantitative light microscopic analysis. Acta Neuropathol. 93, 450–460. Dennery, P.A., Seidmann, D.S., Stevenson, D.K., 2001. Neonatal hyperbilirubinemia. N. Engl. J. Med. 344, 581–590. Diamond, I., Schmid, R., 1966. Experimental bilirubin encephalopathy. The mode of entry of bilirubin-14C into he central nervous system. J. Clin. Invest. 45, 678–689. Gunn, C.K., 1938. Hereditary acholuric jaundice in a new mutant strain of rats. J. Heredity 29, 137–139. Huang, C.-M., 1980. A comparative study of the brain stem auditory response in mammals. Brain Res. 184, 215–219. Huang, C.-M., Buchwald, J.S., 1977. Interpretation of the vertex short-latency acoustic response: a study of single neurons in the brain stem. Brain Res. 137, 291–303. Huang, C.-M., Buchwald, J.S., 1978. Factors that affect the amplitudes and latencies of the vertex short latency acoustic responses in the cat. Electroencephalogr. Clin. Neurophysiol. 44, 179–186. Johnson, L., Bhutani, V.K., 1998. Guidelines for management of the jaundiced term and near-term infant. Curr. Controv. Perinatal Care III. 25, 555–574. Johnson, L., Sarmiento, F., Blanc, W.A., Day, R., 1959. Kernicterus in rats with an inherited deficiency of glucuronyl transferase. Am. J. Dis. Child 97, 591–608. Johnson, L., Garcia, M.L., Figueroa, E., Sarmiento, F., 1961. Kernicterus in rats lacking glucuronyl transferase. Am. J. Dis. Child. 101, 322–349. Kaplan, M., Hammerman, C., 2004. Understanding and preventing severe neonatal hyperbilirubinemia: is bilirubin neurotoxicity really a concern in the developed world? Clin. Perinatol. 31, 555–575 (Review). Kaplan, M., Hammerman, C., 2005. Understanding and preventing kernicterus: adjuncts in the interpretation of neonatal serum bilirubin. Clin. Chim. Acta 356, 9–21 (Review). Keppel, G., 1982. Design and Analysis: A Researcher's Handbook. Prentice-Hall, Inc., Englewood Cliffs, NJ. Lagakos, S.W., 2006. The challenge of subgroup analyses—Reporting without distorting. N. Engl. J. Med. 354, 1667–1669. Maisels, M.J., Newman, T.B., 1995. Kernicterus in otherwise healthy, breast-fed term newborns. Pediatrics 96, 730–733. Maisels, M.J., Watchko, J.F., 2000. Neonatal Jaundice. Harwood Academic Publishers, Amsterdam. Melcher, J.R., Kiang, N.Y., 1996. Generators of the brainstem auditory evoked potential in cat: III. Identified cell populations. Hear. Res. 93, 52–71. Melcher, J.R., Knudson, I.M., Fullerton, B.C., Guinan Jr., J.J., Norris, B.E., Kiang, N.Y., 1996a. Generators of the brainstem auditory evoked potential in cat: I. An experimental approach to their identification. Hear. Res. 93, 1–27. Melcher, J.R., Guinan Jr., J.J., Knudson, I.M., Kiang, N.Y., 1996b. Generators of the brainstem auditory evoked potential in cat: II. Correlating lesion sites with waveform changes. Hear. Res. 93, 28–51. Mρller, A.R., Jannetta, P.J., Mρller, M.B., 1981. Neural generators of brainstem evoked potentials: results from human intracranial recordings. Ann. Oto. Rhiol. Laryngol. 90, 591–596. Rashid, H., Muzammil, S., Tayyab, S., 1998. Comparison of the bilirubin binding and other molecular properties of the serum albumin of several mammalian species. Biochem. Mol. Biol. Int. 44, 165–173.
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
Rose, A.L., Wisniewski, H.K., 1979. Acute bilirubin encephalopathy induced with sulfadimethoxine in Gunn rats. J. Neuropathol. Exp. Neurol. 38, 152–164. Sawasaki, Y., Yamada, N., Nakajima, H., 1976. Developmental features of cerebellar hypoplasia and brain bilirubin levels in a mutant (Gunn) rat with hereditary hyperbilirubinaemia. J. Neurochem. 27, 577–583. Schutta, H.S., Johnson, L., 1967. Bilirubin encephalopathy in the Gunn rat: a fine structure study of the cerebellar cortex. J. Neuropathol. Exp. Neurol. 26, 377–396. Schutta, H.S., Johnson, L., 1969. Clinical signs and morphologic abnormalities in Gunn rats treated with sulfadimethoxine. J. Pediatr. 75, 1070–1079. Shaia, W.T., Shapiro, S.M., Spencer, R.F., 2005. The jaundiced Gunn rat model of auditory neuropathy/dys-synchrony. Laryngoscope 115, 2167–2173. Shapiro, S.M., 1988. Acute brainstem auditory evoked potential abnormalities in jaundiced Gunn rats given sulfonamide. Pediatr. Res. 23, 306–310. Shapiro, S.M., 1993. Reversible brainstem auditory evoked potential abnormalities in jaundiced Gunn rats given sulfonamide. Pediatr. Res. 34, 629–633. Shapiro, S.M., 2002. Somatosensory and brainstem auditory evoked potentials in the Gunn rat model of acute bilirubin neurotoxicity. Pediatr. Res. 52, 844–849. Shapiro, S.M., 2003. Bilirubin toxicity in the developing nervous system. Pediatr. Neurol. 29, 410–421. Shapiro, S.M., 2005. Definition of the clinical spectrum of
221
kernicterus and bilirubin-induced neurologic dysfunction (BIND). J. Perinatol. 25, 54–59 (Review). Shapiro, S.M., Hecox, K.E., 1988. Development of brainstem auditory evoked potentials in heterozygous and homozygous jaundiced Gunn rats. Dev. Brain Res. 41, 147–157. Shapiro, S.M., Hecox, K.E., 1989. Brain stem auditory evoked potentials in jaundiced Gunn rats. Ann. Otol. Rhinol. Laryngol. 98, 308–317. Shapiro, S.M., Conlee, J.W., 1991. Brainstem auditory evoked potentials correlate with morphological changes in Gunn rat pups. Hear. Res. 57, 16–22. Shapiro, S.M., TeSelle, M.E., 1994. Cochlear microphonics in the jaundiced Gunn rat. Am. J. Otolaryngol. 15, 129–137. Shimabuku, R., Nakamura, H., Matsuo, T., 1983. Effect of sulfisoxazole on bilirubin–albumin binding in Gunn rats. Acta Paediatr. Jpn. 25, 304–308. Stevenson, D.K., Wong, R.J., DeSandre, G.H., Vreman, H.J., 2004. A primer on neonatal jaundice. Adv. Pediatr. 51, 263–288 (Review). Strebel, L., Odell, G.B., 1971. Bilirubin uridine diphosphoglucuronyltransferase in rat liver microsomes: genetic variation and maturation. Pediatr. Res. 5, 548–559. Wennberg, R.B., Ahlfors, C.E., Bhutani, V.K., Johnson, L.H., Shapiro, S.M., 2006. Toward understanding kernicterus: a challenge to improve the management of jaundiced newborns. Pediatrics 117, 474–485. Yamamura, H., Takagishi, Y., 1993. Cerebellar hypoplasia in the hyperbilirubinemic Gunn rat: morphological aspects. Nagoya J. Med. Sci. 55, 11–21.
a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m
w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s
Research Report
Biliverdin-induced brainstem auditory evoked potential abnormalities in the jaundiced Gunn rat Ann C. Rice ⁎, Steven M. Shapiro Department of Neurology, Box 980599, Virginia Commonwealth University, Richmond, VA 23298-0599, USA
A R T I C LE I N FO
AB S T R A C T
Article history:
Brainstem auditory evoked potential (BAEP) abnormalities occur in jaundiced Gunn rats
Accepted 2 June 2006
given sulfadimethoxine to displace bilirubin bound to serum albumin, releasing it into the
Available online 7 July 2006
tissues. One problem with the model is that after displacement, plasma bilirubin levels drop and do not correlate with neurological dysfunction. In this report, we administered
Keywords:
biliverdin, the immediate precursor of bilirubin, in 15- to 17-day-old Gunn rat pups to create
Kernicterus
an improved model of bilirubin-induced neurological dysfunction. Total plasma bilirubin
Bilirubin
(TB) levels were measured with a Leica bilirubinometer. Biliverdin (40 mg/kg) or phosphate-
Neurotoxicity
buffered saline (PBS) was administered either once and BAEPs recorded 8 h later or twice, 12 h apart, and BAEPs recorded 24 h after the initial injection. A single biliverdin injection
Abbreviations:
produced a significantly decreased amplitude of BAEP wave III, 1.21 ± 0.25 vs. 0.49 ± 0.27 μV
ABRs, auditory brainstem responses
(control vs. biliverdin). The two-injection paradigm resulted in a significantly elevated TB
BAEPs, brainstem auditory evoked
(9.9 ± 1.2 vs. 14.9 ± 3.1 mg/dl; control vs. biliverdin), significant increases in I–II (1.15 ± 0.08 vs.
potentials
1.42 ± 0.09 ms) and I–III (2.17 ± 0.08 vs. 2.5 ± 0.13 ms) interwave intervals and a decrease in the
BIND, bilirubin-induced neurological
amplitude of wave III (1.36 ± 0.30 vs. 0.38 ± 0.26 μV). Additionally, there were significant
dysfunction
correlations between TB and the amplitude of wave III (r2 = 0.74) and TB and the I–III
BV, biliverdin
interwave interval (r2 = 0.51). In summary, biliverdin administration in jaundiced Gunn rat
sulfa, sulfadimethoxine
pups produces BAEP abnormalities consistent with those observed in the sulfadimethoxine
TB, total plasma bilirubin
model and human newborn hyperbilirubinemia and resulted in increased plasma bilirubin levels that correlate with the degree of neurological dysfunction. © 2006 Elsevier B.V. All rights reserved.
1.
Introduction
The acute and chronic manifestations of excessive neonatal hyperbilirubinemia, acute bilirubin encephalopathy in newborns and chronic bilirubin encephalopathy also known as kernicterus and/or bilirubin-induced neurological dysfunction (BIND) in older infants, children and adults are on the rise in the United States in part due to early discharge of newborns from the hospital (for reviews, see Maisels and Newman, 1995; Maisels and Watchko, 2000; Johnson and
Bhutani, 1998; Dennery et al., 2001; Shapiro, 2003; Stevenson et al., 2004; Kaplan and Hammerman, 2004, 2005). Total bilirubin (TB) levels in plasma increase normally 3–5 days after birth due to immaturity of the enzyme uridine diphosphate (UDP) glucuronysyl transferase, which functions to add glucuronic acid residues to bilirubin rendering it more water soluble and promoting its excretion into bile (Maisels and Watchko, 2000; Shapiro, 2003; Kaplan and Hammerman, 2004; Stevenson et al., 2004). This increase in TB is generally self-correcting by upregulation of the transferase enzyme
⁎ Corresponding author. Fax: +1 804 828 5654. E-mail address: [email protected] (A.C. Rice). 0006-8993/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2006.06.005
216
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –22 1
(Stevenson et al., 2004). However, in cases where the TB levels rise excessively, signs of acute damage can appear, with subsequent development of a range of neurological sequelae, which include auditory dysfunction, impaired upward gaze, dystonia, athetosis and other cerebral palsy-like movement disorders (Shapiro, 2003; Kaplan and Hammerman, 2004; Stevenson et al., 2004; Shapiro, 2005). One major difficulty with the diagnosis of BIND is that the elevated bilirubin levels occur shortly after birth, but irreversible neurological dysfunction may take months or years to recognize. The current practice is that if a high TB is detected in the newborn, treatments such as phototherapy and, in extreme cases, exchange transfusion are initiated to attempt to prevent the irreversible neurological damage. Tests such as brainstem auditory evoked potentials (BAEPs), also known as auditory brainstem responses (ABRs), can be performed to assess auditory function during and after hyperbilirubinemia to determine if there is likely to be permanent damage (Bratlid, 1991; Kaplan and Hammerman, 2004; Shapiro, 2005). BAEP recordings detect the efficacy of neural transmission between brainstem auditory nuclei, one of the brain areas damaged by high bilirubin levels. Another difficulty with the diagnosis of BIND stems from identifying the level of plasma bilirubin responsible for the range of neurological dysfunctions because often there are additional factors involved that influence brain concentrations of bilirubin (Bratlid, 1991; Maisels and Watchko, 2000; Stevenson et al., 2004; Shapiro, 2005; Wennberg et al., 2006). Furthermore, it is not known how high levels of plasma bilirubin cause permanent devastating neurological dysfunction. The Gunn strain of Wistar rat (Gunn, 1938) is presently the best animal model of BIND. These animals have a spontaneous mutation of the UDP-glucuronysyl transferase enzyme (Strebel and Odell, 1971), and the homozygous recessive (jj) animals exhibit significantly elevated TB levels (Johnson et al., 1959, 1961; Schutta and Johnson, 1969). Bilirubin levels peak around 15–17 days of age and slowly drop over the lifetime of the animal (Johnson et al., 1961; Sawasaki et al., 1976; personal observation). In the circulation, bilirubin is tightly bound to serum albumin (Brodersen, 1978; Rose and Wisniewski, 1979; Shimabuku et al., 1983; Shapiro, 1988). When the albumin binding capacity is exceeded, the free or unbound bilirubin leaves the circulation and enters the tissues including the brain. The untreated jj animals often exhibit mild ataxia due to the consistently observed cerebellar hypoplasia (Schutta and Johnson, 1967; Sawasaki et al., 1976; Rose and Wisniewski, 1979; Yamamura and Takagishi, 1993; Conlee and Shapiro, 1997). However, more severe neurological sequelae can be induced by administration of sulfadimethoxine (sulfa), which displaces the bilirubin bound to albumin allowing even higher levels to enter the tissues (Blanc and Johnson, 1959; Diamond and Schmid, 1966; Schutta and Johnson, 1967; Rose and Wisniewski, 1979; Shapiro, 1988, 2002; Rashid et al., 1998). Dramatic BAEP abnormalities are observed within hours following sulfa administration to jj pups (Shapiro, 1988, 1993, 2002; Shapiro and Hecox, 1988). Additionally, sulfa-treated jj pups typically develop severe dystonia after a day or two, thereby exhibiting many of the same characteristic neurological sequelae observed in humans.
In the Gunn rat sulfa model of bilirubin-induced neurotoxicity, plasma bilirubin levels drop to near zero, making it difficult to correlate the plasma level of bilirubin, pre- and post-injection of sulfonamide, with the neurological sequelae. Biliverdin (BV) is the immediate precursor to bilirubin in the catabolism of hemoglobin and is more soluble and less neurotoxic than bilirubin. In this report, we assess whether BV administration raises plasma bilirubin to levels that produce neurological dysfunction, as an improvement in the Gunn rat model to enable the study of varying degrees of BIND with the advantage of detectable increased plasma bilirubin levels.
2.
Results
2.1.
Total plasma bilirubin (TB) levels
A time course response of BV was assessed on 15-day-old jj rat pups to determine if TB levels could be significantly elevated by biliverdin administration. Fig. 1 depicts the time course of TB levels following a single injection of 40 mg/kg BV, phosphate-buffered saline (PBS, BV-vehicle) and 200 mg/kg sulfa or saline (sulfa-vehicle). A pre-injection blood sample was drawn, the injection administered and a post-injection blood sample drawn at a specified time for each animal. Time course curves were compared using a split-plot analysis of variance (ANOVA). Plasma bilirubin levels were significantly elevated by a single injection of BV and decreased by sulfadimethoxine, whereas the two vehicle groups were significantly different from each drug group, but not from each other (F3,99 = 570). In the BV group receiving a single injection of BV, the plasma bilirubin level peaked 4 h after the injection, and by 24 h plasma bilirubin levels recovered to near pre-injection levels.
Fig. 1 – Plasma bilirubin levels. Plasma bilirubin levels following a single injection of biliverdin (40 mg/kg), sulfa (200 mg/kg), PBS (BV-vehicle) or saline (sulfa-vehicle). Blood was drawn twice from each animal at 0 h and a designated time post-injection. Values are mean ± SD; n = 5 at each time point.
217
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
2.2.
Brainstem auditory evoked potentials (BAEPs)
To validate BV treatment as a model of BIND, we performed BAEPs to assess auditory dysfunction in another set of animals. BAEPs are computer averaged surface recorded electrical potentials in response to an auditory stimulus. The resultant waveforms reflect neural transmission in the afferent auditory pathways between the auditory nerve and auditory brainstem nuclei.
2.2.1.
Eight-hour single injection experiment
A baseline BAEP was recorded on 15-day-old jj pups, then a single injection of BV (40 mg/kg) or PBS (equivalent volume) was injected and 8 h later a second BAEP was recorded. The 8-h BAEP latencies and amplitudes were compared between groups (Table 1). There was a small but significant decrease in the amplitude of wave III after BV injection, whereas no other parameters were significantly different between the two groups. Due to the small change in BAEPs following a single injection of BV, we examined the effects of two BV injections on BAEPs in another group of animals.
2.2.2.
Twenty-four-hour double injection experiment
For TB level determination, an initial blood sample was drawn prior to injection and a second one 24 h after the initial BV injection following the BAEP recording. BV (40 mg/ kg) was administered twice, 12 h apart, in 16-day-old jj pups and BAEPs recorded 24 h after the first injection. The physiological parameters are displayed in Table 2. Initial mean body weights did not differ significantly between groups; however, 24 h after the first injection the BV group gained significantly less weight than the PBS group. Weight loss is consistently observed following sulfa administration in auditory impaired animals (Conlee and Shapiro, 1991). The pre-BV injection TB levels were not different between groups;
Table 2 – Physiological parameters
jj-BV jj-PBS P
N
Initial Weight (g)
Weight Gain (g)
8 7
26.9 ± 2.6 26.8 ± 3.0 NS
2.0 ± 1.0 3.2 ± 0.6 0.0189**
Initial bilirubin mg/dl 10.1 ± 1.1 10.8 ± 1.2 NS
24 h bilirubin mg/dl 14.9 ± 3.1 9.9 ± 1.2 0.0014**
Physiological parameters in jaundiced rat pups before and 24 h after the first of 2 injections of biliverdin (jj-BV) or vehicle (jj-PBS) 12 h apart. These animals had BAEPs recorded at 24 h after 2 injections and the final blood sample drawn after the BAEP recording. Values are mean ± SD. Significance is at P = 0.05. NS = not significant.
however, at 24 h the TB levels were significantly increased by approximately 50% in the BV group. Fig. 2 depicts representative waveforms observed following two injections of either PBS or BV (40 mg/kg) in jj pups. Vertical dashed lines from the middle of each peak in the PBS-treated animal to the waveforms in the BV-treated animal demonstrate the increase in latency of waves II and III. BAEP latencies, interwave intervals and amplitudes in jj pups given two injections of either BV (40 mg/kg) or PBS (equivalent volume) are given in Table 3. There were statistically significant increases in BAEP I–II and I–III interwave intervals and decreases in the amplitude of wave III in the BV-treated group compared to PBS-treated animals. The TB levels obtained at the time of the BAEP from the double injection experiment were compared to the significantly changed BAEP parameters. BAEP wave III amplitude
Table 1 – BAEP data following a single injection of biliverdin Group
N
jj-BV jj-PBS P
10 5
Group
N
jj-BV jj-PBS P
10 5
Latency (ms) I
I–II
I–III
1.46 ± 0.18 1.35 ± 0.15 NS
1.29 ± 0.16 1.22 ± 0.05 NS
2.48 ± 0.21 2.26 ± 0.07 0.046
Amplitude (μV) I
II
III
1.17 ± 0.25 1.05 ± 0.23 NS
0.82 ± 0.48 1.10 ± 0.4 NS
0.49 ± 0.27 1.21 ± 0.25 0.00027**
BAEP changes following a single injection of biliverdin. Biliverdin (40 mg/kg) or PBS was injected into jaundiced Gunn rat pups (jj-BV and jj-PBS, respectively). Eight hours later, BAEPs were recorded. The latency of wave I and the interwave interval between waves I to II and I to III are shown (upper half). The amplitudes of waves I, II and III are reported (lower half). All values are mean ± SD, significant P values are at 0.0083 denoted by **. NS = not significant.
Fig. 2 – BAEP waves following 2 injections of Biliverdin (40 mg/kg) or PBS. Representative BAEP waves recorded 24 h after the first injection of 2 injections 12 h apart in an animal given PBS (upper) and an animal given BV (lower). Waves I, II and III are identified. A dashed line drawn vertically from the peak of the waves in PBS-treated animal to the waveform of the BV-treated animal shows the changes in latencies.
218
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –22 1
Table 3 – BAEP data following two injections of biliverdin Group
N
jj-BV jj-PBS P
8 7
Group
N
jj-BV jj-PBS P
8 7
Latency (ms) I
I–II
I–III
1.19 ± 0.06 1.27 ± 0.13 NS
1.42 ± 0.09 1.15 ± 0.08 0.00003**
2.50 ± 0.13 2.17 ± 0.08 0.00006**
Amplitude (μV) I
II
III
1.19 ± 0.35 1.23 ± 0.26 NS
1.08 ± 0.3 1.36 ± 0.48 NS
0.38 ± 0.26 1.36 ± 0.30 0.000012**
BAEP data following two injections of biliverdin. Biliverdin (40 mg/ kg) or PBS was injected twice into jaundiced Gunn rat pups (jj-BV and jj-PBS, respectively) 12 h apart. BAEPs were recorded 24 h after the first injection. The latency of wave I and the interwave interval between waves I to II and I to III are shown (upper half). The amplitudes of waves I, II and III are reported (lower half). All values are mean ± SD, significant p values are at 0.0083 denoted by **. NS = not significant.
was inversely correlated to the TB. Linear regression shows a significant inverse correlation between the amplitude of wave III and the total plasma bilirubin level (r2 = 0.7111, P = 7.84 × 10−5); however, there was a better fit of the data with a power function (r2 = 0.7417; Fig. 3). Additionally, there was a significant linear correlation between TB and the I–III interwave interval (r2 = 0.5056, P = 0.00297; data not shown). These correlations support our hypothesis that the TB level can potentially indicate the amount of auditory dysfunction in this model in contrast to the sulfa model where they do not.
3.
The BAEP is a very sensitive tool for studying bilirubininduced auditory dysfunction. In rats (Huang, 1980), and generalizing from studies in other mammals (Buchwald and Huang, 1975; Huang, 1980; Huang and Buchwald, 1977, 1978; Melcher and Kiang, 1996; Melcher et al., 1996a,b) and humans (Mρller et al., 1981) and correlation of abnormalities of BAEP waves with neuroanatomical changes in our rat model (Conlee and Shapiro, 1991; Shapiro and Conlee, 1991), waves I, II and III in the rat correspond to waves I, III and V in human and reflect neural transmission between the auditory nerve (wave I) and the cochlear nucleus (wave II in rat, III in human) and midbrain structures (wave III in rat, V in human). In previous studies, we have recorded normal cochlear microphonic responses in animals with abnormal BAEPs in this model (Shapiro and TeSelle, 1994) and shown histologically normal cochlea with abnormalities of spiral ganglion and auditory nerve (Shaia et al., 2005) supporting the fact that the waveforms characterized represent brainstem and not cochlear damage. With the single BV injection experiment, the only abnormality observed was with the amplitude of wave III, probably reflecting relatively minor brainstem dysfunction, whereas the double injection experiment, with higher plasma bilirubin levels, decreased amplitudes of waves II and III and increased interwave intervals between waves I–II and waves I– III, displayed a greater degree of brainstem dysfunction. Both BV experiments, however, exhibited less auditory dysfunction than generally observed with the sulfa model, which typically displays more severe latency increases and amplitude decreases in waves II and III, often leading to disappearance of these waves, as well as abnormalities and disappearance of wave I, causing total absence of BAEP (Shapiro, 2002). Therefore, not only is BV administration an additional relevant
Discussion
In summary, we have demonstrated that biliverdin administration to jaundiced Gunn rat pups significantly elevates plasma bilirubin levels. Furthermore, these animals exhibit brainstem auditory dysfunction as assessed with BAEP recordings. The pattern of BAEP abnormalities (increased latency of waves II and III and decreased amplitude of waves II and III) is similar to that seen with sulfonamide injection in jj Gunn rat pups (Shapiro, 1988, 1993). Thus, the BV model closely mimics the Gunn rat sulfa model and the human disorder and provides us with another model to assess how hyperbilirubinemia in the newborn produces such devastating neurological dysfunction. This new model has the distinct advantage over the sulfa model in that there is a direct correlation of plasma bilirubin level with neurophysiological function. Therefore, we can begin to understand in what situations high plasma bilirubin levels produce neurological dysfunction. The disadvantage of our model however is that due to the limited solubility of BV, a large volume of fluid and at least two injections have to be administered. It is possible that with a shorter interval between injections an overall higher plasma bilirubin level could be obtained. Additional studies will further characterize BV-induced neurological dysfunction compared with the wellknown sulfa model.
Fig. 3 – Correlation between plasma bilirubin level and amplitude of BAEP wave III. The correlation compares total plasma bilirubin 24 h after the first injection of 2 BV or PBS injections, 12 h apart in both BV (gray circles)- and PBS (open circles)-treated animals. The correlation best fits a power function (y = 693.99x −2.8162; r 2 = 0.7417); n = 8 BV-treated animals; n = 7 PBS-treated animals.
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
model of BIND, but we can study milder abnormalities with greater consistency and reproducibility. The strong correlations between TB and the conduction time between the auditory nerve and midbrain (I–III interwave interval) and the amplitude of wave III demonstrates that Gunn rats given BV may be a useful model to begin to identify specific neurological sequela likely to develop at defined TB levels. Unbound (free) bilirubin in uninjected jj pups (Ahlfors and Shapiro, 1997) and humans (Ahlfors, 2000) is a better indicator of BAEP dysfunction than TB alone. Total bilirubin and unbound bilirubin can be compared during acute bilirubin encephalopathy in this model, whereas with displacers such as sulfa, both total and free bilirubin drop to near zero and their relative predictive values pre- and post-injection would be expected to be lost. The BV model of BIND characterized thus far produces a milder form of neurological impairment than the sulfa model. As apparent from Fig. 1, some animals had peak plasma bilirubin levels in the upper teens; however, animals that survived 48 h from other experiments not presented here did not display the characteristic dystonia observed with the sulfa model. This indicates that at least in our animal model, the auditory system may be more sensitive to bilirubin toxicity than the basal ganglia. Alternatively, the basal ganglia may require significantly more damage to occur before abnormalities appear, and/or that our classic behavioral assessment is not as sensitive in assessing basal ganglia function as BAEPs are in assessing auditory function. In conclusion, we are developing and characterizing a model of BIND that allows us to study milder neurological sequelae in the whole animal. This is relevant to the human cases of learning disabilities that ultimately may be solely the product of milder hearing impairment or a central auditory processing disorder for which there is presently no rodent model.
4.
Experimental procedures
All procedures were approved by the Institutional Animal Care and Use Committee. All efforts were made to minimize pain and stress to the animals and to reduce the number of animals used. The spontaneously jaundiced Gunn strain of Wistar rats used for all experiments was obtained from our in-house breeding colony. Experiments were performed on 15- to 17day-old rat pups when endogenous bilirubin levels are significantly elevated. These studies used only the homozygous recessive (jaundiced, jj) pups. Biliverdin hydrochloride was purchased from Frontier Scientific (Logan, UT). BV was sonicated until it dissolved in phosphate-buffered saline (PBS) at 0.4 mg/ml and injected at 40 mg/kg, i.p. An equivalent volume of PBS was injected into control animals. Due to the limited solubility of BV, the injection volume is 10% of the animal's body weight. Sulfadimethoxine (sulfa) diluted in saline was injected at 200 mg/kg in a volume of 1% of the animal's body weight.
4.1.
Total plasma bilirubin (TB) levels
Blood samples were collected from each animal via a cheek puncture into a heparinized hematocrit tube and centrifuged.
219
Two samples were obtained from each animal (pre- and posttreatment for both time course and double BV injection BAEP experiments). Animals were sacrificed after the second blood sample was drawn. Plasma bilirubin levels were assessed using a Leica Unistat Bilirubinometer. This instrument requires 20 μl of plasma for analysis. Generally, 50–75 μl of blood were collected for each sample. If excessive bleeding occurred, the cheek was pinched for a few seconds to control the bleeding.
4.1.1.
BV time course curve
A total of 103 animals were used for the time course experiment. Animals were randomly assigned to each sampling time point prior to the initial blood sample being collected. Immediately following the initial blood sampling, the animal was injected with either a single 40 mg/kg BV dose or equivalent volume of PBS. Total plasma bilirubin values were compared by a split-plot ANOVA with a Tukey post hoc test using SSPS software for the time course curve experiment, or paired or unpaired t test using Microsoft Excel for the BAEP experiments.
4.2.
Brainstem auditory evoked potentials (BAEPs)
Thirty jj rat pups were used for these experiments. Fifteen were used for the single injection 8-h experiment and fifteen were used for the 24-h double injection experiment. Weightmatched paired littermates from at least 11 different litters were used to account for inter-litter variation. Pups were anesthetized with an intramuscular injection of 60 mg/kg ketamine and 6 mg/kg acepromazine. BAEPs were recorded using a Nicolet Spirit 2000 evoked potential system (Biosys, Inc.). The left ear was occluded with petrolatum, and BAEPs were obtained to monaural 100-μs duration rarefaction clicks delivered at 31.7/s to the right ear through a Sony Walkman 4LIS headphone speaker, a small speaker we have used previously with response characteristics relatively flat from about 250 to 18,000 Hz (Shapiro and Hecox, 1989). The sound intensity was nominally set at 70 dB, which we calibrated biologically and corresponded to a level of about 62 dB above a normal Gunn rat pup BAEP threshold level as described below. Surface electrical activity was recorded from 13 mm long subcutaneous platinum needle electrodes inserted on the scalp over the vertex and behind the left and right mastoid bullae with a ground electrode in the flank. Rectal temperature was controlled at 37.0 ± 0.1 °C using a heating blanket and red heat lamp and stabilized for a minimum of 5 min before and during the recording. Two channel BAEP recordings were obtained from the contralateral to the ipsilateral mastoid (horizontal) and the vertex to the ipsilateral mastoid (vertical) electrode pairs, filtered from 30 to 3000 Hz. As there was not a qualitative difference between the two orientations, only the horizontal data are presented in this report. The vertical channel was used to help identify uncertain peaks. Each individual BAEP was the averaged response to 2000 stimuli, and three or more replicated responses were obtained for each animal. The individual BAEP replications were summed, and the peak latencies and peak-to-trough amplitudes of BAEP waves I, II and III were scored using a cursor. Latencies were subtracted to obtain interwave intervals. Eight-hour single injection experiment—A baseline BAEP was recorded, the animal was injected with either 40 mg/kg BV or PBS
220
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –22 1
(equivalent volume) and 8 h later a second BAEP was recorded. Twenty-four-hour double injection experiment—A pre-injection blood sample was drawn from the left cheek to ensure there was no damage to the ear canal on the side being subjected to BAEP testing. Then either 40 mg/kg BV or PBS (equivalent volume) was injected. Twelve hours later a second injection was administered and 12 h after that (24 h after original injection) a BAEP was recorded followed by a second blood sample being drawn. For biological calibration of the system, BAEP threshold levels were recorded in six 16-day-old, phenotypically normal heterozygous non-jaundiced Gunn rat pups from different litters at successively lower intensities. In 5 of 6 animals, replicable responses were obtained at 8 dB and no replicable response at 7 dB, and in 1 animal there were replicable responses at 7 dB and no response at 6 dB. The criteria for no replicable responses was defined as no waves ≥ 0.02 μV present in any of 3 recordings, 8000 sweeps each, with no or minimal muscle activity on the raw EEG, and with absent BAEP responses appearing essentially flat without significant artifact. For threshold determination, a rate of 91.7/s was used; anesthesia and recording conditions were otherwise identical to those in the experiments. All recordings were done in a relatively quite room, but machine noise from the Nicolet Spirit placed about 2 m from the rat likely caused some masking near threshold. We have previously shown that I–II and I–III interwave intervals and the amplitudes of waves II and III are the important BAEP variables affected by bilirubin toxicity in the sulfa model (Shapiro, 1988, 1993). For statistical analysis, we chose to examine 6 relatively independent BAEP-dependent variables a priori: the latency of wave I, I–II and I–III interwave intervals and the amplitudes of waves I, II and III in the mastoid to mastoid electrode montage. To correct for multiple independent observations, we divided an overall experimentwise p value of 0.05 by 6 to obtain an individual criterion p value of 0.0083 (Keppel, 1982; Lagakos, 2006).
Acknowledgments This work was supported by NIH R01 grants DC00369 and NS47151 to SMS. We would like to thank Dr. Charles E. Ahlfors who initially suggested using BV in our model. Additionally, we would like to thank Drs. Claudio Tiribelli and J. Donald Ostrow for helpful discussions in planning the project.
REFERENCES
Ahlfors, C.E., 2000. Measurement of plasma unbound unconjugated bilirubin. Anal. Biochem. 279, 130–135. Ahlfors, C.E., Shapiro, S.M., 1997. Brainstem auditory evoked potentials (BAEPs) and bilirubin-albumin binding in jaundiced (jj) Gunn rat pups. Pediatr. Res. 41, 35A. Blanc, W.A., Johnson, L., 1959. Studies on kernicterus: relationship with sulfonamide intoxication, report on kernicterus in rats with glucuronyl transferase deficiency and review of pathogenesis. J. Neuropathol. Exp. Neurol. 18, 165–189. Bratlid, D., 1991. Bilirubin toxicity: pathophysiology and assessment of risk factors. N.Y. State J. Med. 91, 489–492.
Brodersen, R., 1978. Determination of the vacant amount of high-affinity bilirubin binding sites on serum albumin. Acta Pharmacol. Toxicol. 42, 153–158. Buchwald, J.S., Huang, C.-M., 1975. Far-field acoustic response: origins in the cat. Science 189, 382–384. Conlee, J.W., Shapiro, S.M., 1991. Morphological changes in the cochlear nucleus and nucleus of the trapezoid body in Gunn rat pups. Hear. Res. 57, 23–30. Conlee, J.W., Shapiro, S.M., 1997. Development of cerebellar hypoplasia in jaundiced Gunn rats: a quantitative light microscopic analysis. Acta Neuropathol. 93, 450–460. Dennery, P.A., Seidmann, D.S., Stevenson, D.K., 2001. Neonatal hyperbilirubinemia. N. Engl. J. Med. 344, 581–590. Diamond, I., Schmid, R., 1966. Experimental bilirubin encephalopathy. The mode of entry of bilirubin-14C into he central nervous system. J. Clin. Invest. 45, 678–689. Gunn, C.K., 1938. Hereditary acholuric jaundice in a new mutant strain of rats. J. Heredity 29, 137–139. Huang, C.-M., 1980. A comparative study of the brain stem auditory response in mammals. Brain Res. 184, 215–219. Huang, C.-M., Buchwald, J.S., 1977. Interpretation of the vertex short-latency acoustic response: a study of single neurons in the brain stem. Brain Res. 137, 291–303. Huang, C.-M., Buchwald, J.S., 1978. Factors that affect the amplitudes and latencies of the vertex short latency acoustic responses in the cat. Electroencephalogr. Clin. Neurophysiol. 44, 179–186. Johnson, L., Bhutani, V.K., 1998. Guidelines for management of the jaundiced term and near-term infant. Curr. Controv. Perinatal Care III. 25, 555–574. Johnson, L., Sarmiento, F., Blanc, W.A., Day, R., 1959. Kernicterus in rats with an inherited deficiency of glucuronyl transferase. Am. J. Dis. Child 97, 591–608. Johnson, L., Garcia, M.L., Figueroa, E., Sarmiento, F., 1961. Kernicterus in rats lacking glucuronyl transferase. Am. J. Dis. Child. 101, 322–349. Kaplan, M., Hammerman, C., 2004. Understanding and preventing severe neonatal hyperbilirubinemia: is bilirubin neurotoxicity really a concern in the developed world? Clin. Perinatol. 31, 555–575 (Review). Kaplan, M., Hammerman, C., 2005. Understanding and preventing kernicterus: adjuncts in the interpretation of neonatal serum bilirubin. Clin. Chim. Acta 356, 9–21 (Review). Keppel, G., 1982. Design and Analysis: A Researcher's Handbook. Prentice-Hall, Inc., Englewood Cliffs, NJ. Lagakos, S.W., 2006. The challenge of subgroup analyses—Reporting without distorting. N. Engl. J. Med. 354, 1667–1669. Maisels, M.J., Newman, T.B., 1995. Kernicterus in otherwise healthy, breast-fed term newborns. Pediatrics 96, 730–733. Maisels, M.J., Watchko, J.F., 2000. Neonatal Jaundice. Harwood Academic Publishers, Amsterdam. Melcher, J.R., Kiang, N.Y., 1996. Generators of the brainstem auditory evoked potential in cat: III. Identified cell populations. Hear. Res. 93, 52–71. Melcher, J.R., Knudson, I.M., Fullerton, B.C., Guinan Jr., J.J., Norris, B.E., Kiang, N.Y., 1996a. Generators of the brainstem auditory evoked potential in cat: I. An experimental approach to their identification. Hear. Res. 93, 1–27. Melcher, J.R., Guinan Jr., J.J., Knudson, I.M., Kiang, N.Y., 1996b. Generators of the brainstem auditory evoked potential in cat: II. Correlating lesion sites with waveform changes. Hear. Res. 93, 28–51. Mρller, A.R., Jannetta, P.J., Mρller, M.B., 1981. Neural generators of brainstem evoked potentials: results from human intracranial recordings. Ann. Oto. Rhiol. Laryngol. 90, 591–596. Rashid, H., Muzammil, S., Tayyab, S., 1998. Comparison of the bilirubin binding and other molecular properties of the serum albumin of several mammalian species. Biochem. Mol. Biol. Int. 44, 165–173.
BR A I N R ES E A RC H 1 1 0 7 ( 2 00 6 ) 2 1 5 –2 21
Rose, A.L., Wisniewski, H.K., 1979. Acute bilirubin encephalopathy induced with sulfadimethoxine in Gunn rats. J. Neuropathol. Exp. Neurol. 38, 152–164. Sawasaki, Y., Yamada, N., Nakajima, H., 1976. Developmental features of cerebellar hypoplasia and brain bilirubin levels in a mutant (Gunn) rat with hereditary hyperbilirubinaemia. J. Neurochem. 27, 577–583. Schutta, H.S., Johnson, L., 1967. Bilirubin encephalopathy in the Gunn rat: a fine structure study of the cerebellar cortex. J. Neuropathol. Exp. Neurol. 26, 377–396. Schutta, H.S., Johnson, L., 1969. Clinical signs and morphologic abnormalities in Gunn rats treated with sulfadimethoxine. J. Pediatr. 75, 1070–1079. Shaia, W.T., Shapiro, S.M., Spencer, R.F., 2005. The jaundiced Gunn rat model of auditory neuropathy/dys-synchrony. Laryngoscope 115, 2167–2173. Shapiro, S.M., 1988. Acute brainstem auditory evoked potential abnormalities in jaundiced Gunn rats given sulfonamide. Pediatr. Res. 23, 306–310. Shapiro, S.M., 1993. Reversible brainstem auditory evoked potential abnormalities in jaundiced Gunn rats given sulfonamide. Pediatr. Res. 34, 629–633. Shapiro, S.M., 2002. Somatosensory and brainstem auditory evoked potentials in the Gunn rat model of acute bilirubin neurotoxicity. Pediatr. Res. 52, 844–849. Shapiro, S.M., 2003. Bilirubin toxicity in the developing nervous system. Pediatr. Neurol. 29, 410–421. Shapiro, S.M., 2005. Definition of the clinical spectrum of
221
kernicterus and bilirubin-induced neurologic dysfunction (BIND). J. Perinatol. 25, 54–59 (Review). Shapiro, S.M., Hecox, K.E., 1988. Development of brainstem auditory evoked potentials in heterozygous and homozygous jaundiced Gunn rats. Dev. Brain Res. 41, 147–157. Shapiro, S.M., Hecox, K.E., 1989. Brain stem auditory evoked potentials in jaundiced Gunn rats. Ann. Otol. Rhinol. Laryngol. 98, 308–317. Shapiro, S.M., Conlee, J.W., 1991. Brainstem auditory evoked potentials correlate with morphological changes in Gunn rat pups. Hear. Res. 57, 16–22. Shapiro, S.M., TeSelle, M.E., 1994. Cochlear microphonics in the jaundiced Gunn rat. Am. J. Otolaryngol. 15, 129–137. Shimabuku, R., Nakamura, H., Matsuo, T., 1983. Effect of sulfisoxazole on bilirubin–albumin binding in Gunn rats. Acta Paediatr. Jpn. 25, 304–308. Stevenson, D.K., Wong, R.J., DeSandre, G.H., Vreman, H.J., 2004. A primer on neonatal jaundice. Adv. Pediatr. 51, 263–288 (Review). Strebel, L., Odell, G.B., 1971. Bilirubin uridine diphosphoglucuronyltransferase in rat liver microsomes: genetic variation and maturation. Pediatr. Res. 5, 548–559. Wennberg, R.B., Ahlfors, C.E., Bhutani, V.K., Johnson, L.H., Shapiro, S.M., 2006. Toward understanding kernicterus: a challenge to improve the management of jaundiced newborns. Pediatrics 117, 474–485. Yamamura, H., Takagishi, Y., 1993. Cerebellar hypoplasia in the hyperbilirubinemic Gunn rat: morphological aspects. Nagoya J. Med. Sci. 55, 11–21.