- Email: [email protected]
The local anesthetic properties and toxicity of saxitonin homologues for rat sciatic nerve block in vivo
T h e Local A n e s t h e t i c P r o p e r t i e s a n d Toxicity of S a x i t o n i n H o m o l o g u e s for Rat Sciatic N e r v e B l o c k In Vivo Daniel S. Kohane, M.D., Ph.D., Nu T. Lu, B.A., Arman C. G6kg6l-Kline, Maria Shubina, M.S., Yu Kuang, Sherwood Hall, Ph.D., Gary R. Strichartz, Ph.D., and Charles B. Berde, M.D., Ph.D. Background and Objectives. Saxitoxin and its homologues are naturally occurring compounds that block the sodium channel with high potency. They have the potential for providing prolonged duration local anesthesia w h e n coinjected with vasoconstrictors or conventional local anesthetics and are devoid of local neurotoxicity. Here, we compare sciatic nerve block with saxitoxin to those with neosaxitoxin, decarbamoyl saxitoxin, and tetrodotoxin (Trx), in a search for even safer compounds. Methods. Rats received percutaneous sciatic nerve block with toxins. The compounds were compared in terms of lethality, onset and duration of action for thermal analgesia (hot-plate testing), and motor block (weight-bearing). Data were expressed as medians with 25th and 75th percentiles, and median effective concentrations were determined. Results. The median concentrations at which analgesia of 60 minutes duration was achieved were neosaxitoxin, 34 ± 2 Rmol/L; saxitoxin, 58 -+ 3 lamol/L; TTX, 92 -+ 5 lamol/L; and decarbamoyl saxitoxin, 268 z 8 lamol/L. Similar trends were observed for other measures of effectiveness (block duration of 90 minutes, maximal block), and for lethality so that the therapeutic indices were similar. No toxin had a marked predominance of sensory or motor block. The potency of T r x was intermediate between those of the saxitoxins, and its therapeutic index was slightly better. No difference was observed in time to onset of nerve blockade among the toxins. Conclusions. Substitutions on the saxitoxin nucleus result in large differences in incidence and duration of block, and toxicity. The therapeutic indices of the saxitoxins are similar; that of TrX is slightly better. Reg Anesth Pain Bled 2000;25:52-59.
Key W o r d s :
EDs0, LD5o, Local anesthetic, Peripheral nerve, Saxitoxin, Sciatic, Tetrodotoxin.
F
o r m a n y years, t h e r e h a s b e e n a n i n t e r e s t in t h e l o c a l a n e s t h e t i c a c t i v i t y of a g r o u p of h i g h l y potent naturally occurring toxins that bind the o u t e r o p e n i n g of s o d i u m c h a n n e l s at a l o c a t i o n t e r m e d "site 1. "~ T h e s e a g e n t s (Fig 1), i n c l u d i n g s a x i t o x i n (STX) a n d its d e r i v a t i v e s a n d t e t r o d o t o x i n (TrX), b i n d to c h a n n e l s w i t h h i g h affinity a n d r a p i d reversibility. 2-4 S a x i t o x i n , t h e p a r e n t c o m p o u n d of t h e f a m i l y of t o x i n s t h a t c a u s e s p a r a l y t i c shellfish
p o i s o n i n g , is p r o d u c e d b y m a r i n e d i n o f l a g e l l a t e s of t h e g e n u s Alexandrium. S a x i t o x i n a n d its n a t u rally occurring derivatives have a common nucleus, b u t differ in t h e s u b s t i t u e n t s at 4 sites (Fig 1). T h e 3 s a x i t o x i n s w e h a v e e x a m i n e d in this s t u d y a r e STX itselL n e o s a x i t o x i n (neoSTX), a n d d e c a r b a m o y l STX (dcSTX). N e o s a x i t o x i n differs f r o m STX b y t h e a d d i t i o n of 1 o x y g e n a t o m : t h e - H at N~ is a n - O H in n e o s a x i t o x i n . I n dcSTX, t h e c a r b a m o y l
From the Department of Anesthesia, Children's Hospital (D.S.K., N.T.L., A.C.G.-K., C.B.B.), Boston, Massachusetts; Harvard School of Public Health (M.S.), Boston, Massachusetts; the Department of Chemical Engineering, Massachusetts Institute of Technology (D.S.K., Y.K.), Cambridge, Massachusetts; Washington Seafood Laboratory Branch, Office of Seafood, U.S. Food and Drug Administration (S.H.), Washington, DC; the Department of Anesthesia and Pharmacology, Brigham and Women's Hospital and Harvard Medical School (G.R.S.), Boston, Massachusetts; the Departments of Anesthesia (C.B.B.) and Pediatrics (D.S.K., C.B.B), Harvard Medical School, Boston, Massachusetts; and the Department of Pediatrics, Massachusetts General Hospital, Boston, Massachusetts (D.S.K.).
Accepted for publication June 24, 1999. All experiments were performed at Children's Hospital, Boston, MA. Supported by grants from the CHMC Anesthesia Foundation and the Anesthesia Pain Research Endowment Fund, and National Institutes of Health Grant No. GM 35647. Reprint requests: Charles B. Berde, M.D., Ph.D., Department of Anesthesia, Children's Hospital, 333 Longwood Ave, Room 555, Boston, MA 02115. © 2000 by the American Society of Regional Anesthesia and Pain Medicine. 1098-7339/00/2501-001055.00/0
52
Regional Anesthesia and Pain Medicine, Vo125, No 1 (January-February), 2000: pp 52-59
Saxitoxin Homologuesfor Sciatic Nerve Block • Kohaneet al. 53 Saxitoxins
R4/0~/H R1 -- N ( ~ 6 ~ ,
. . - ° N ",
13 Ji2 9N~+--"~NHz H2N ""~+"~--~ ./ /0 12~i"-15I-I
.I "OH
R3~,R#OH
necessarily correspond to observations made in vitro.7.s The relative lack of central nervous system and cardiac toxicities of site 1 sodium channel blockers and their recently described lack of local neurotoxicity 9 have stimulated renewed interest in their therapeutic potential as local anesthetics. ~° In this study, we apply an in vivo neurobehavioral paradigm l°Al to test the hypothesis that functional group substitutions on the STX nucleus result in differences in incidence or duration of thermal nociception and motor strength. Materials and Methods
RI
R2
R3
Saxitoxin
-H
-H
-H -CONH2
Stock Solutions
Neosaxitoxin
-OH
-H
-H -CONH2
Decarbamoyl Saxitoxin
-H
-H
-H
Saxitoxin, neoSTX, and dcSTX (Washington Seafood Laboratories, Washington, DC), and TTX (Sigma Chemical Co., St. Louis, MO) were reconstituted in 20 mol/L citrate buffer pH 4.5 to 4.6, then diluted in saline as needed. The 3 STX derivatives were each at least 90% pure and contained less than 5% of the other derivatives. Coming from natural sources the preparations were presumed to be enantiomerically pure. The optical isomer has no ( < 1 % ) biologic activity. ° Bupivacaine hydrochloride (Astra USA, Westborough, MA) was obtained in 0.5% solution and then diluted to the desired concentration.
Tetrodotoxin
R4
-H
o O,H)\
' o OH HzN~N~
l_______~ I
\
A n i m a l Care Fig 1. Molecular structures of the saxitoxins and of tetrodotoxin. Tetrodotoxin is shown as the hydroxy lactone, one of the major species in solution.
group, -CONH2 at O18, has been replaced by a hydrogen atom. We also compare the actions of the saxitoxins to TrX, a c o m p o u n d that has been isolated from m a n y species of animal, most notoriously the puffer fish Fugu of the family Tetraodontidae. ~ In vitro studies have shown that alterations in the R-groups result in differences in the potencies and binding characteristics of the saxitoxins. The EC~0 for inhibition of c o m p o u n d action potentials for neoSTX is 0.22 times that of STX, 4 whereas the EC~0 for dcSTX is 2.2 times that of STX. 6 The EDso for inhibition of c o m p o u n d action potentials from TrX is 3 times that of STX.4 Based on these data, the relative potencies of these compounds should be neoSTX > STX > dcSTX > TrX w h e n injected perineurally. However, these differences in potency are relatively small; all 4 are encompassed by approximately 1 order of magnitude. Furthermore, the in vivo performance characteristics of compounds with local anesthetic properties do not
Young adult male Sprague-Dawley rats weighing 310 to 420 g were used. Animals were cared for in compliance with protocols approved by the Animal Care and Use Committee at Children's Hospital. Sprague-Dawley rats were obtained from Charles River Laboratories (Wilmington, MA). They were housed in groups and kept in a 6 AM tO 6 PM light-dark cycle. Sciatic Block Technique Before the nerve block injections, rats were anesthetized briefly (< 1 minute) with halothane by face mask (2% to 4% inspired concentration in 100% oxygen). A 23-gauge needle was introduced posteromedial to the greater trochanter, pointing in all anteromedial direction. Once bone was contacted, the needle was withdrawn 1 m m and 0.1 mL of drug-containing solution was injected. The left leg was always used for blocks; the fight served as control. The entire block procedure took about 1 minute, inducting anesthesia. Assessment of Nerve Block The effectiveness of block was measured, applying the methods of T h a l h a m m e r et al.,It or modifica-
54
Regional Anesthesia and Pain Medicine Vol. 25 No. 1 January-February 2000
tions thereof. ~° M e a s u r e m e n t s were m a d e every 2 minutes to d e t e r m i n e time to onset of block and every 30 minutes to determine duration of block. Functional deficits were also m e a s u r e d in the uninjected (contralateral) leg to provide a control for systemic effects (residual h a l o t h a n e anesthesia and systematically distributed toxin). ~o.~x.,3 All n e u r o b e havioral testing was done by a blinded observer. Nociceptive block was assessed by a modified hotplate test.14 Hindpaws were exposed in sequence (left t h e n right) to a hot plate at 56°C (Model 39D Hot Plate Analgesia Meter, IITC Inc., Woodland Hills, CA), and the time (latency) until p a w withdrawal was m e a s u r e d by a stopwatch. If the p a w r e m a i n e d in contact for 12 seconds, it was r e m o v e d by the e x p e r i m e n t e r to avoid injury to the animal or the d e v e l o p m e n t of hyperalgesia. This test was repeated 3 times for each rat at every time point. M o t o r strength was assessed by holding the rat with its posterior placed above a digital balance and allowing it to bear its o w n weight on 1 h i n d p a w at a time. The m a x i m u m weight that the rat could bear w i t h o u t its ankle touching the balance was quantified. This m e a s u r e is referred to as extensor postural thrust.l' Rats that b e c a m e apneic for 20 seconds following toxin injection were euthanized with intraperitoneal pentobarbital (100 mg/kg). Death was ascertained by clinical criteria.
Data Processing The data for nociceptive block are reported in terms of t h e r m a l latency, onset of block, and duration of block. The time to onset is the time required for t h e r m a l latency to rise to a value of 7 seconds (which is the midpoint b e t w e e n the m a x i m u m latency [12 seconds] and the baseline latency of a p p r o x i m a t e l y 2 seconds). Similarly, the duration of block is the time for latency to return to 7 seconds f r o m a higher value. Animals that did not survive w e r e not included in the calculation of block duration. All other animals were included in the calculations of m e d i a n block durations, including those in which the block was unsuccessful (defined as injections that did not result in a t h e r m a l latency of at least 7 seconds or a suppression in m o t o r strength of at least 50%). Block duration was considered zero for those injections. These zero-duration blocks probably result f r o m real pharmacologic differences b e t w e e n drugs rather t h a n operator error, since the rate of successful block is greater t h a n 99% w h e n we h a v e used 0.5% b u p i v a c a i n e w i t h the s a m e v o l u m e and m e t h o d of block.
Because some experinaental groups had large n u m b e r s of zero-duration blocks (i.e., the frequency of various block durations was not normally distributed), we have used the median as the m e a s u r e of central tendency t h r o u g h o u t this report, with 25th and 75th percentiles. A median or 75th percentile value of zero does not necessarily m e a n that all blocks were unsuccessful. Statistical Analysis Statistical inferences on medians were made using the Kruskal-Wallis test a n d / o r the M a n n W h i t n e y U test. Dose-response analyses and the determination of median effective concentrations or doses were conducted using logistic regression. The variances were calculated using the delta method, and the 95% confidence intervals were derived from those. These analyses were conducted using Stata statistical software (Stata Corp, College Station, TX).
Results D u r a t i o n of Block Rats were injected with 0.1 mL of solution containing increasing concentrations of the various toxins, and the resulting durations of block were calculated. Block durations increased progressively with increasing doses of all the toxins. Figure 2 shows the results for thermal nociception. For each c o m p o u n d , low concentrations were associated with a high incidence of unsuccessful blocks. At higher concentrations, there was a c o n c e n t r a t i o n - d e p e n d e n t increase in the duration of block. For each toxin, we calculated the median effective concentrations required to achieve t h e r m a l nociceptive blocks of 60 and 90 minutes (the ECs0.dur6o and ECs0.durgo). These values are presented in Table 1. Based on these data (P < .0001 for all comparisons), the order of in vivo potencies was neoSTX > STX > TTX > dcSTX. F u n c t i o n a l Specificity M o t o r and sensory block were c o m p a r e d in terms of the concentrations (ECso-maxS) at which 50% of rats developed m a x i m a l t h e r m a l nociceptive block (latency of 12 seconds) and m o t o r block (extensor postural thrust -<40 g). The potencies in terms of ECso.max (Table 2) w e r e in the order neoSTX > STX > TTX > dcSTX for both t h e r m a l nociception and m o t o r function (P < .0001 for all comparisons). No difference was observed b e t w e e n the EC50.maxs for t h e r m a l nociception and m o t o r strength for STX and dcSTX. There was a small (16%) t e n d e n c y to sensory selectivity for neoSTX (P = .025). Tetrodotoxin s h o w e d no statistically significant difference
Saxitoxin Homologues
for Sciatic Nerve Block
•
K o h a n e et al.
55
150 + neoSTX ---<>-- STX -----B--- TIN ---Am dcSTX
.~ 125
100 >
Fig 2. Dose-response curve for duration of thermal nodceptive block for saxitoxins and tetrodotoxin. Data points are medians.
O
"~ 75
r~ O
~
50
e-. o
25 /
0--c 10
=
.~.~-
Lethality a n d T h e r a p e u t i c Indices of the Toxins To calculate the therapeutic index for these compounds, we determined the LD~0 (i.e., median dose [in nmol/kg]) resulting in terminal a p n e a in rats injected with toxin at the sciatic nerve (Table 3). As with the ECsos, the order of LDs0s was neoSTX < STX < "Iq?X < dcSTX (P < .0001 for comparisons b e t w e e n toxins). The LDs0 and ECs0 values (from Tables 1 and 2) were converted into pg/kg, and the therapeutic indices (ratio LDs0/EDso) were obtained with respect to both EC~0-max (which reflects the incidence and intensity of block) and EC~0-dur60 (which reflects incidence and duration). W h e n the ECs0.rnax was used to derive the denominator, no statistically significant differences were observed Table 1. Median Concentration Resulting in Thermal
Nociceptive Block Lasting 60 and 90 Minutes ECso-durS0 (pmol/L)
STX neoSTX dcSTX
58 +- 3 34 _ 2 268 _+ 8 92 __+5
TTX
ECso-durgo (pmol/L) 70 40 334 111
._-- .
,~ . . . ,.'.
.-.
50 100 Toxin concentration (tdVl)
b e t w e e n the ECso.maxS for thermal nociceptive block and m o t o r block.
Compound
i :,
+ + -
6 3 55 15
NOTE. ECso-durSOand ECso-du~9oare the concentrations resulting in median durations of block of 60 and 90 minutes, respectively. Values are shown with 95% confidence intervals. Pairwise statistical comparisons of the toxins yielded P < .0001 in all cases, n, number of rats studied to derive the ECsos. STX, Saxitoxin; neoSTX, neosaxitonin; dcSTX, decarbamoyl saxitoxin; -I-rx, tetrodotoxin.
1000
b e t w e e n the therapeutic indices for the various toxins, except for a weakly significant difference b e t w e e n TI"X and the toxin with the worst therapeutic index, STX. W h e n the therapeutic indices w e r e calculated based on the EC~0-dur60, dcSTX had the lowest therapeutic index; both T r x and neoSTX had significantly higher values. In all cases, the differences w e r e of relatively small m a g n i t u d e (the highest and lowest therapeutic indices differed by app r o x i m a t e l y 2 5 %). Onset of N e r v e B l o c k Rats were injected with selected concentrations of the various toxins that had b e e n s h o w n by the above experiments to uniformly result in successful blocks. (The different toxins were c o m p a r e d at different concentrations because of the differences
Table 2. Median Effective Concentration (ECs0.max) Required for Maximal Thermal Nociceptive and Motor Block
n 72 127 64 89
500
ECso.max (pmol/L) Compound
Thermal Nociception
Motor Block
STX neoSTX dcSTX
51 _ 4
53 -.+ 8 36 -.+ 5
TTX
30 - 1 202 +_ 22
202 -+ 20
76 - 11
73 - 11
P .66
.025 .98 .71
NOTE. Values for ECsos are given with 95% confidence intervals. P values compare thermal nociceptive block and motor block. The number of rats studied is the same as in Table 2. STX, saxitonin; neoSTX, neosaxitonin; dcSTX, decarbamoyl, saxitonin; F I X , tetrodotoxin.
56
Regional Anesthesia and Pain Medicine Vol. 25 No. 1 January-February 2000
Table 3. Median Lethal Doses (LDso) and Therapeutic Indices Derived From ECs0-max
Derived From EC5o-dur6o
Compound STX neoSTX dcSTX "I-FX
LDso (nmol/kg) 23 14 94 41
+_ 1 -+ 1 + 5 ___2
Therapeutic Index 1.3 1.42 1.2 1.51
P Values for Therapeutic Indices neoSTX
dcSTX
TTX
0.44
0.53 0.01
0.27 0.8 0.01
Therapeutic Index 1.49 1.61 1.59 1.82
P Values for Therapeutic Indices neoSTX
dcSTX
"l-lX
0.23
0.35 0.9
0.04 0.18 0.18
NOTE. LDsos for compounds injected at the sciatic nerve are given with 95% confidence intervals. All of the LDsos are statistically significantly different from each other (P < .0001 ). Therapeutic index = LDso/EDso.The Pvalues shown compare the therapeutic indices of the compounds listed in the first column with other toxins. The number of rats studied for each LDs0 were as follows. STX: 108; neoSTX: 150; dcSTX: 90;-I-I'X: 101. STX, saxitonin; neoSTX, neosaxitonin; dcSTX, docarbamoyl saxitonin; TTX, tetmdotoxin.
in their individual potencies.) The time of onset of t h e r m a l nociceptive block for each toxin is s h o w n in Table 4. There was no statistically significant difference in onset time b e t w e e n the three saxitoxins and TrX (P > .05 for Kruskal-Wallis comparing all 4 c o m p o u n d s ) . These onsets were all slower than that of 15.4 mol/L (0.5%) bupivacaine, a conventional a m i n o - a m i d e local anesthetic (P = .02 in comparison of TTX and bupivacaine). Rats injected with saline after general anesthesia n e v e r achieved our criteria for t h e r m a l analgesia (latency ->7 seconds). M o t o r block was detectable within 2 minutes (the first time point tested) for all groups, including saline-injected controls. Motor block p r o m p t l y resolved (by 4 minutes postinjection) in the salineinjected rats, but persisted in those injected with active compounds. Thus, we could not determine the exact time of onset of m o t o r block (since it occurred while general anesthesia was wearing off). This brief initial m o t o r block was c o m m o n l y seen in the contralateral leg and, therefore, was probably caused by w e a k n e s s f r o m residual general anesthesia. Discussion
Chemical substitutions on the saxitoxin molecule h a v e m a r k e d effects on p o t e n c y in vitro. In this study, we found large differences b e t w e e n the toxins in t e r m s of incidence and duration of block and
Table 4. Time to Onset of Thermal Nociceptive Block Compound and Concentration
Time to Onset of Nociceptive Block (rain)
n
S'IX (60 pmol/L) neoSTX (36 pmol/L) dcSTX (270 pmol/L) TTX (125 pmol/L) Bupivacaine (15.4 mol/L) Saline (0.9%)
6.2 (4.5-7) 8.2 (7-9.5) 6.2 (3.1-6.9) 5.2 (4.3-7) 2.3 (2.4-2.9) Never achieved
11 12 11 12 12 6
NOTE. Values are medians with 25th and 75th percentiles in parentheses. STX, saxitonin; neoS'l-X, neosaxitonin; dcSTX, decarl0amoyl saxitonin; TTX, tetrodotoxin.
LDso in vivo. HoweveL although the relative potencies of the STXs in rat sciatic nerve block mirrored their relative potencies in vitro, there was not an exact proportionality b e t w e e n in vitro and in vivo data. In Fig 3, we plotted the thermal nociceptive ECso.dur6o and ECso-max values from this study for each toxin against published values for 2 in vitro measures of potency: the relative ECs0s for depressing the c o m p o u n d action potentiaP .6 (Fig 3A) and the equilibrium dissociation constant KD determ i n e d from functional block of open single channels Is (Fig 3B). Both in vitro measures found similar proportionalities b e t w e e n the potencies of the saxitoxins (ECso or KD of neoSTX ~ 0.25 times that of STX, dcSTX ~ 2.3 times STX). The corresponding proportionalities from our study were neoSTX ~ 0.6 times STX, dcSTX = 4.3 times STX. There was close a g r e e m e n t b e t w e e n the relative potencies of the toxins as determined by the ECso-maxand ECs0.dur6 0. The ECso-maxof TrX in vivo was greater t h a n that of STX, but m u c h lower t h a n that of dcSTX. Previously published in vitro data had suggested that the potencies of dcSTX 6 and TrX 4 relative to STX w e r e similar or that dcSTX was slightly m o r e potent t h a n TI'X. This disparity b e t w e e n in vivo and in vitro data could be a result of T r x having some property that enables it to cross biologic m e m b r a n e s m o r e readily. The difference b e t w e e n the absolute values of in vivo potencies described here and those previously reported in vitro 4,6 is consistent with the established view that these barriers (such as the epi- and p e r i n e u r i u m ) are a real obstacle to the action of these toxins. The in vivo EthoS for the toxins w e r e a p p r o x i m a t e l y 1,000 to 10,000 times the in vitro KD (Fig 3). Greater ease in penetrating such barriers has b e e n used to explain the higher ratio of in vivo to in vitro potency~° for bupivacaine (approximately 100) c o m p a r e d with T r x (approximately 1,000). Tetrodotoxin is quite similar to the STXs in most physicochemical characteristics that affect p h a r m a c o k i n e t -
Saxitoxin Homologues for Sciatic Nerve Block
A 300
i
[
•
i
25O
dcSTX
I
o
~ 150
8 8
TTX a ~STX 50 I
]
neoSTX
I
~
I
0.5 1 1.5 2 2.5 EC58 for CAP inhibition normalized to STX
B 300
I Ir
250 dcSTX
i
~ 150
8 100
STX
5o
TTX
:
~ neoSTX
0 0
10 15 20 Equilibrium constant Kv (nM)
25
30
Fig 3. Comparison of in vivo and in vitro effectiveness of STX homologues and Trx. The principal measures of in vivo effectiveness'were the ECso-maxfor thermal nociceptive block (open circles), and the concentration yielding a duration of 60 minutes (ECs0-dur6O,dark squares). These data are compared with the in vitro ratio of the ECso for compound action potential (CAP) inhibition to the value for STX (A), and the equilibrium dissociation constant (Ko; B) for each toxin. The value of Ko is given by the ratio of the rate constants for dissociation and binding from the sodium channel, and are the primary determinants of the potencies of toxins (see Hall et ai.24).
ics. None of these c o m p o u n d s h a v e appreciable hydrophobicity, and they are all of a p p r o x i m a t e l y the same molecular weight (300 to 400 da). Saxitoxin and dcSTX are divalent cations at physiologic pH (with positive charges associated with the guanidinium groups [Na and Na]). Neosaxitoxin has the same two positive charges, but also has a partial negative charge (pKb = 6.75) at physiologic pH. Tetrodotoxin is a m o n o v a l e n t cation (because of its single g u a n i d i n i u m group). The lesser charge m i g h t provide an advantage in crossing lipid-rich perineural m e m b r a n e barriers, thus m a k i n g it m o r e perm e a n t and effectively m o r e p o t e n t t h a n dcSTX in
•
Kohane et al.
57
vivo. However, the fact that the onset of TTX block is not shorter than that of the STXs argues against this explanation. Bupivacaine, which is amphiphilic and has an intermediate degree of hydrophobicity 16 at physiologic pH, has m o r e rapid onset than TTX and the STXs. No correlation was found between in vitro or in vivo potency and the time to onset of sciatic nerve block. A likely explanation for this finding is that the pharmacokinetic barriers to the penetration of the hydrophilic toxins into or within the nerve bundle are the rate-limiting step to the d e v e l o p m e n t of nerve block. Differences b e t w e e n in vitro and in vivo effectiveness have also been noted with conventional local anesthetics. Mepivacaine is m o r e potent in vivo than prilocaine and lidocaine in vivo, but less potent t h a n t h e m in vitro. 7 The reduced performance in the nerve bath has been attributed to differential ionization of the local anesthetics. Differential effects of local anesthetics on the vasculature of the n e r v e could also have a m a r k e d effect on potency in vivo. The ratios b e t w e e n the ECso.maxvalues for sensory and m o t o r block for the saxitoxins and tetrodotoxin were similar. These findings are s o m e w h a t surprising because one might have expected m o t o r pred o m i n a n c e in t o x i n - d e p e n d e n t block based on current research on the differential actions of site 1 sodium channel blockers on nerve fiber subpopulations. A-alpha and A-delta peaks of c o m p o u n d action potentials of frog sciatic nerves have b e e n s h o w n to be abolished by application of 1 iamol/L T r x , whereas the C p e a k was unaffected. 17 Other investigators 18 using rat L4-L5 dorsal root ganglia have also s h o w n action potentials in A-alpha/beta cells to be blocked by 0.1 to 1 pmol/L TTX, while a proportion of A-delta and all C cells had impulses that w e r e T r x insensitive, generating the concept that TrX-insensitive sodium channels are critical for impulse activation in p r i m a r y nociceptors. H o w ever, similar discrepancies h a v e b e e n f o u n d bet w e e n electrophysiologic data and in vivo observations of sensorimotor selectivity with veratridine s and ropivacaine.~2.~9 The transient bilateral m o t o r block seen in all e x p e r i m e n t a l groups (including the saline-injected controls) was ascribable to the initial inhalational anesthetic. It was of very brief duration (it was only observed during experiments to determine onset of n e r v e block), and consequently it is unlikely that it contributed to the m u c h longer n e r v e deficits resulting f r o m toxin injection. One caution in interpreting the differences bet w e e n sensory and m o t o r block in o u r e x p e r i m e n t a l paradigm is that in b o t h cases w e are dealing w i t h
58
Regional Anesthesia and Pain Medicine Vol. 25 No. 1 January-February 2000
"censored" data with s o m e w h a t arbitrarily established p e r f o r m a n c e limits. This is of particular importance in the determination of the ECsos of m a x i m a l block. In the case of sensory data, "maximal" latency is set at 12 seconds, beyond which there is a risk of injury or sensitization to the rat's foot. With m o t o r data, "maximal" block could not simply be defined as the lowest weight b o r n e by a n y given rat, because by that m e a s u r e every rat m u s t achieve m a x i m a l block. Consequently, an arbitrary cut-off of less t h a n 40 g was adopted. Therefore, "maximal" sensory and m o t o r block as defined here m a y not reflect exactly the same degrees of block of the respective functions. Nonetheless, both measures indicate a m a j o r deficit in the respective functions.S. 10-J3 The toxicity of site 1 sodium channel blockers has been described extensively, m.2°.21 Unlike conventional local anesthetics, w h e r e the m a j o r toxicity is neurologic (seizures and coma) and cardiovascular (dysrhythmias and cardiac depression), morbidity f r o m these toxins results primarily from respiratory failure from diaphragmatic paralysis. The latter is believed to r e s u l t f r o m toxin action on the diap h r a g m and phrenic nerves, rather t h a n in the central n e r v o u s system. 2° Large differences were observed in the LDs0s of the saxitoxins that also mirrored the potencies predicted by in vitro data. However, the differences in LDs0s b e t w e e n individual toxins w e r e paralleled by differences in the ECs0s, w h i c h resulted in similar therapeutic indices. Surprisingly, the therapeutic index of TTX was generally higher (i.e., safer) t h a n that of the saxitoxins, although the difference was probably too small to be of clinical significance. A priori, one might h a v e expected the opposite, because T r x tends to cause substantial h y p o t e n s i o n (via direct action on s m o o t h muscle cells and s y m p a t h e t i c nerves) while this is m u c h less of a feature of STX. 2~ The LDso value for TTX in this study was very similar to that f o u n d in a previous study ~° (41 vs 40 m m o l / L / k g ) . However, because the v o l u m e of injectate was 0.3 m L in the earlier article, and only 0.1 m L in this article, the relative toxicities of a n y given T r x concentration w e r e m a r k e d l y different. For example, 100 lamol/L T r x in the earlier article (corresponding to 10 lag or 89 n m o l / L / k g ) was u n i f o r m l y fatal, w h e r e a s h e r e it resulted in no deaths (as it corresponded to 3.3 lag or 30 nmol/kg). These findings further confirm the usefulness of total dose (rather t h a n concentration), w h e n comp a n g the systemic toxicities of different preparations, The ECs0-max reported here is a p p r o x i m a t e l y twice t h a t in the earlier article, for similar reasons. BeCause a ~ v e n concentration in this article repre-
sents one third tile a m o u n t of drug in the previous one, one would expect a lesser local effect. Although it is important to understand the perform a n c e characteristics of the individual toxins alone, we have proposed that they are likely to be used clinically in combination with either conventional local anesthetics or vasoconstrictors. The latter compounds greatly increase the duration of block from site sodium channel blockers, io.22.23 and the vasoconstrictors greatly improve their therapeutic index. 1° Bupivacaine also increases the LDso of TrX to a small degree, possibly by a vasoconstrictive m e c h a nism. It seems reasonable to expect that the h o m o logues of STX would be potentiated in a similar manner. However, it is impossible to predict w h e t h e r the rank order of their potencies in combination with a second drug will be the same. In s u m m a r y , chemical substitutions on the STX molecule result in m a r k e d changes in potency and toxicity in vivo. In vitro measures of effectiveness or potency are not perfect predictors of in vivo potency, as evidenced by the greater potency of T r x in relation to the STXs in vivo, and the differences in the relative potencies of STXs in vitro and in vivo. The differences in intrinsic potency b e t w e e n site 1 sodium channel blockers only result in small differences in therapeutic indices because the LDs0s are affected in a proportional manner.
References 1. Catterall WA. Neurotoxins that act on voltagesensitive sodium channels in excitable membranes. Annu Rev PhaTw,acol Toxicol 1980;20:15-43. 2. Hille B. The receptor for tetrodotoxin and saxitoxin. A structural hypothesis. Biophys J 1975; 15:615-619. 3. Narahashi T. Mechanism of action of tetrodotoxin and saxitoxin on excitable membranes. Fed Proc 1972; 31:1124-1132. 4. Strichartz G, Rando T, Hall S, Gitschier J, Hall L, Magnani B, Bay CH. On the mechanism by which saxitoxin binds to and blocks sodium channels. Ann N YAcad Sci 1986;479:96-112. 5. Fuhrman FA. Tetrodotoxin. Sci Am 1967;217:60-71. 6. Strichartz GR, Hall S, Magnani B, Hong CY, Kishi Y, Debin JA. The potencies of synthetic analogues of saxitoxin and the absolute stereoselectivity of decarbamoyl saxitoxin. Toxicon 1995;33:723-737. 7. Pateromichelakis S, Prokopiou AA. Local anesthesia efficacy: Discrepancies between in vitro and in vivo studies. Acta Anesthesiol Scand 1988;32:672-675. 8. Vladimirov M, Thalhammer JG, Hunt N, Feldman HS, Strichartz G. Neurophysiologic actions and neurological consequences of veratridine on the rat sciatic nerve. Anesthesiology 1997;86:945-956. 9. Sakura S, Bollen AW, Ciriales R, Drasner K. Local anesthetic neurotoxicity does not result from blockade of voltage-gated sodium channels. Anesth An alg 1995;81:338-346.
Saxitoxin Homologues for Sciatic Nerve Block 10. Kohane DS, Yieh J, Lu NT, Langer R, Strichartz G, Berde CB. A reexamination o[ tetrodotoxin for prolonged anesthesia. Anesthesiology 1998;89:119-131. 11. Thalhammer JG, Vladimirova M, Bershadsky B, Strichartz GR. Neurologic evaluation of the rat during sciatic nerve block with lidocaine. Anesthesiology 1995; 82:1013-1025. 12. Kohane DS, Sankar W, Hu D, Shubina M, Refai N, Berde CB. Sciatic nerve blockade in infant, adolescent, and adult rats: A comparison of ropivacaine and bupivacaine. Anesthesiology 1998;89:1199-1208. 13. Kohane DS, Kuang Y, Lu NT, Langer R, Strichartz GR, Berde CB. Vanilloid receptor agonists potentiate the in vivo local anesthetic activity of percutaneously injected site 1 sodium channel blockers. Anesthesiology 1999;90:524-534. 14. Masters DB, Berde CB, Dutta SK, Griggs CT, Hu D, Kupsky W, Langer R. Prolonged regional nerve blockade by controlled release of local anesthetic from a biodegradable polymer matrix. Anesthesiology 1993;79: 100-107. 15. Guo X, Uehara A, Ravindran A, Bryan SH, Hall S, Moczydlowski E. Kinetic basis for insensitivity to tetrodotoxin and saxitoxin in sodium channels of canine heart and denervated rat skeletal muscle. Biochemistry 1987;26:7546-7556. 16. Bernards CM, Hill HE Physical and chemical properties of drug molecules governing their diffusion through the spinal meninges. Anesthesiology 1992; 77:750-756.
•
Kohane et al.
59
17. Buchanan S, Harper AA, Elliott dR. Differential effects of tetrodotoxin (TTX) and high external K+ on A and C fibre compound action potential peaks in frog sciatic nerve. Neurosci Lett 1996;219:131 - 134. 18. Villiere V, McLachlan EM. Electrophysiological properties of neurons in intact rat dorsal root ganglia classified by conduction velocity and action potential d u ration, d Neurophysiol 1996;76:1924-1941. 19. Rosenberg PI-I, I-Ieinonen E. Differential sensitivity of A and C fibers to long-acting amide local anesthetics. Br J Anaesth 1983;55:163-167. 20. Kao CY. Tetrodotoxin, saxitoxin, and their significance in the study of excitation phenomena. Pharmacol Rev 1966; 18:997-1049. 21. Kao CY. Pharmacology of tetrodotoxin and saxitoxin. Fed Proc 1972;31:1117-1123. 22. Adams HJ, Blair MR, Jr., Takman BH. The local anesthetic activity of tetrodotoxin alone and in combination with vasoconstrictors and local anesthetics. Anesth Analg 1976;55:568-573. 23. Adams HJ, Blair MR, Jr., Takman BH. The local anesthetic activity of saxitoxin alone and with vasoconstrictor and local anesthetic agents. Arch Int Pharmacodyn 1976;224:275-282. 24. Hall S, Strichartz G, Moczydlowski E, Ravindran A, Reichardt PB. The saxitoxins. Sources, chemistry, and pharmacology. In: Hall S, Strichartz G, eds. Marine Toxins. Origin, Structure, and Molecular Pharmacology. Washington, DC: American Chemical Society, 1990;29-65.
Key W o r d s :
EDs0, LD5o, Local anesthetic, Peripheral nerve, Saxitoxin, Sciatic, Tetrodotoxin.
F
o r m a n y years, t h e r e h a s b e e n a n i n t e r e s t in t h e l o c a l a n e s t h e t i c a c t i v i t y of a g r o u p of h i g h l y potent naturally occurring toxins that bind the o u t e r o p e n i n g of s o d i u m c h a n n e l s at a l o c a t i o n t e r m e d "site 1. "~ T h e s e a g e n t s (Fig 1), i n c l u d i n g s a x i t o x i n (STX) a n d its d e r i v a t i v e s a n d t e t r o d o t o x i n (TrX), b i n d to c h a n n e l s w i t h h i g h affinity a n d r a p i d reversibility. 2-4 S a x i t o x i n , t h e p a r e n t c o m p o u n d of t h e f a m i l y of t o x i n s t h a t c a u s e s p a r a l y t i c shellfish
p o i s o n i n g , is p r o d u c e d b y m a r i n e d i n o f l a g e l l a t e s of t h e g e n u s Alexandrium. S a x i t o x i n a n d its n a t u rally occurring derivatives have a common nucleus, b u t differ in t h e s u b s t i t u e n t s at 4 sites (Fig 1). T h e 3 s a x i t o x i n s w e h a v e e x a m i n e d in this s t u d y a r e STX itselL n e o s a x i t o x i n (neoSTX), a n d d e c a r b a m o y l STX (dcSTX). N e o s a x i t o x i n differs f r o m STX b y t h e a d d i t i o n of 1 o x y g e n a t o m : t h e - H at N~ is a n - O H in n e o s a x i t o x i n . I n dcSTX, t h e c a r b a m o y l
From the Department of Anesthesia, Children's Hospital (D.S.K., N.T.L., A.C.G.-K., C.B.B.), Boston, Massachusetts; Harvard School of Public Health (M.S.), Boston, Massachusetts; the Department of Chemical Engineering, Massachusetts Institute of Technology (D.S.K., Y.K.), Cambridge, Massachusetts; Washington Seafood Laboratory Branch, Office of Seafood, U.S. Food and Drug Administration (S.H.), Washington, DC; the Department of Anesthesia and Pharmacology, Brigham and Women's Hospital and Harvard Medical School (G.R.S.), Boston, Massachusetts; the Departments of Anesthesia (C.B.B.) and Pediatrics (D.S.K., C.B.B), Harvard Medical School, Boston, Massachusetts; and the Department of Pediatrics, Massachusetts General Hospital, Boston, Massachusetts (D.S.K.).
Accepted for publication June 24, 1999. All experiments were performed at Children's Hospital, Boston, MA. Supported by grants from the CHMC Anesthesia Foundation and the Anesthesia Pain Research Endowment Fund, and National Institutes of Health Grant No. GM 35647. Reprint requests: Charles B. Berde, M.D., Ph.D., Department of Anesthesia, Children's Hospital, 333 Longwood Ave, Room 555, Boston, MA 02115. © 2000 by the American Society of Regional Anesthesia and Pain Medicine. 1098-7339/00/2501-001055.00/0
52
Regional Anesthesia and Pain Medicine, Vo125, No 1 (January-February), 2000: pp 52-59
Saxitoxin Homologuesfor Sciatic Nerve Block • Kohaneet al. 53 Saxitoxins
R4/0~/H R1 -- N ( ~ 6 ~ ,
. . - ° N ",
13 Ji2 9N~+--"~NHz H2N ""~+"~--~ ./ /0 12~i"-15I-I
.I "OH
R3~,R#OH
necessarily correspond to observations made in vitro.7.s The relative lack of central nervous system and cardiac toxicities of site 1 sodium channel blockers and their recently described lack of local neurotoxicity 9 have stimulated renewed interest in their therapeutic potential as local anesthetics. ~° In this study, we apply an in vivo neurobehavioral paradigm l°Al to test the hypothesis that functional group substitutions on the STX nucleus result in differences in incidence or duration of thermal nociception and motor strength. Materials and Methods
RI
R2
R3
Saxitoxin
-H
-H
-H -CONH2
Stock Solutions
Neosaxitoxin
-OH
-H
-H -CONH2
Decarbamoyl Saxitoxin
-H
-H
-H
Saxitoxin, neoSTX, and dcSTX (Washington Seafood Laboratories, Washington, DC), and TTX (Sigma Chemical Co., St. Louis, MO) were reconstituted in 20 mol/L citrate buffer pH 4.5 to 4.6, then diluted in saline as needed. The 3 STX derivatives were each at least 90% pure and contained less than 5% of the other derivatives. Coming from natural sources the preparations were presumed to be enantiomerically pure. The optical isomer has no ( < 1 % ) biologic activity. ° Bupivacaine hydrochloride (Astra USA, Westborough, MA) was obtained in 0.5% solution and then diluted to the desired concentration.
Tetrodotoxin
R4
-H
o O,H)\
' o OH HzN~N~
l_______~ I
\
A n i m a l Care Fig 1. Molecular structures of the saxitoxins and of tetrodotoxin. Tetrodotoxin is shown as the hydroxy lactone, one of the major species in solution.
group, -CONH2 at O18, has been replaced by a hydrogen atom. We also compare the actions of the saxitoxins to TrX, a c o m p o u n d that has been isolated from m a n y species of animal, most notoriously the puffer fish Fugu of the family Tetraodontidae. ~ In vitro studies have shown that alterations in the R-groups result in differences in the potencies and binding characteristics of the saxitoxins. The EC~0 for inhibition of c o m p o u n d action potentials for neoSTX is 0.22 times that of STX, 4 whereas the EC~0 for dcSTX is 2.2 times that of STX. 6 The EDso for inhibition of c o m p o u n d action potentials from TrX is 3 times that of STX.4 Based on these data, the relative potencies of these compounds should be neoSTX > STX > dcSTX > TrX w h e n injected perineurally. However, these differences in potency are relatively small; all 4 are encompassed by approximately 1 order of magnitude. Furthermore, the in vivo performance characteristics of compounds with local anesthetic properties do not
Young adult male Sprague-Dawley rats weighing 310 to 420 g were used. Animals were cared for in compliance with protocols approved by the Animal Care and Use Committee at Children's Hospital. Sprague-Dawley rats were obtained from Charles River Laboratories (Wilmington, MA). They were housed in groups and kept in a 6 AM tO 6 PM light-dark cycle. Sciatic Block Technique Before the nerve block injections, rats were anesthetized briefly (< 1 minute) with halothane by face mask (2% to 4% inspired concentration in 100% oxygen). A 23-gauge needle was introduced posteromedial to the greater trochanter, pointing in all anteromedial direction. Once bone was contacted, the needle was withdrawn 1 m m and 0.1 mL of drug-containing solution was injected. The left leg was always used for blocks; the fight served as control. The entire block procedure took about 1 minute, inducting anesthesia. Assessment of Nerve Block The effectiveness of block was measured, applying the methods of T h a l h a m m e r et al.,It or modifica-
54
Regional Anesthesia and Pain Medicine Vol. 25 No. 1 January-February 2000
tions thereof. ~° M e a s u r e m e n t s were m a d e every 2 minutes to d e t e r m i n e time to onset of block and every 30 minutes to determine duration of block. Functional deficits were also m e a s u r e d in the uninjected (contralateral) leg to provide a control for systemic effects (residual h a l o t h a n e anesthesia and systematically distributed toxin). ~o.~x.,3 All n e u r o b e havioral testing was done by a blinded observer. Nociceptive block was assessed by a modified hotplate test.14 Hindpaws were exposed in sequence (left t h e n right) to a hot plate at 56°C (Model 39D Hot Plate Analgesia Meter, IITC Inc., Woodland Hills, CA), and the time (latency) until p a w withdrawal was m e a s u r e d by a stopwatch. If the p a w r e m a i n e d in contact for 12 seconds, it was r e m o v e d by the e x p e r i m e n t e r to avoid injury to the animal or the d e v e l o p m e n t of hyperalgesia. This test was repeated 3 times for each rat at every time point. M o t o r strength was assessed by holding the rat with its posterior placed above a digital balance and allowing it to bear its o w n weight on 1 h i n d p a w at a time. The m a x i m u m weight that the rat could bear w i t h o u t its ankle touching the balance was quantified. This m e a s u r e is referred to as extensor postural thrust.l' Rats that b e c a m e apneic for 20 seconds following toxin injection were euthanized with intraperitoneal pentobarbital (100 mg/kg). Death was ascertained by clinical criteria.
Data Processing The data for nociceptive block are reported in terms of t h e r m a l latency, onset of block, and duration of block. The time to onset is the time required for t h e r m a l latency to rise to a value of 7 seconds (which is the midpoint b e t w e e n the m a x i m u m latency [12 seconds] and the baseline latency of a p p r o x i m a t e l y 2 seconds). Similarly, the duration of block is the time for latency to return to 7 seconds f r o m a higher value. Animals that did not survive w e r e not included in the calculation of block duration. All other animals were included in the calculations of m e d i a n block durations, including those in which the block was unsuccessful (defined as injections that did not result in a t h e r m a l latency of at least 7 seconds or a suppression in m o t o r strength of at least 50%). Block duration was considered zero for those injections. These zero-duration blocks probably result f r o m real pharmacologic differences b e t w e e n drugs rather t h a n operator error, since the rate of successful block is greater t h a n 99% w h e n we h a v e used 0.5% b u p i v a c a i n e w i t h the s a m e v o l u m e and m e t h o d of block.
Because some experinaental groups had large n u m b e r s of zero-duration blocks (i.e., the frequency of various block durations was not normally distributed), we have used the median as the m e a s u r e of central tendency t h r o u g h o u t this report, with 25th and 75th percentiles. A median or 75th percentile value of zero does not necessarily m e a n that all blocks were unsuccessful. Statistical Analysis Statistical inferences on medians were made using the Kruskal-Wallis test a n d / o r the M a n n W h i t n e y U test. Dose-response analyses and the determination of median effective concentrations or doses were conducted using logistic regression. The variances were calculated using the delta method, and the 95% confidence intervals were derived from those. These analyses were conducted using Stata statistical software (Stata Corp, College Station, TX).
Results D u r a t i o n of Block Rats were injected with 0.1 mL of solution containing increasing concentrations of the various toxins, and the resulting durations of block were calculated. Block durations increased progressively with increasing doses of all the toxins. Figure 2 shows the results for thermal nociception. For each c o m p o u n d , low concentrations were associated with a high incidence of unsuccessful blocks. At higher concentrations, there was a c o n c e n t r a t i o n - d e p e n d e n t increase in the duration of block. For each toxin, we calculated the median effective concentrations required to achieve t h e r m a l nociceptive blocks of 60 and 90 minutes (the ECs0.dur6o and ECs0.durgo). These values are presented in Table 1. Based on these data (P < .0001 for all comparisons), the order of in vivo potencies was neoSTX > STX > TTX > dcSTX. F u n c t i o n a l Specificity M o t o r and sensory block were c o m p a r e d in terms of the concentrations (ECso-maxS) at which 50% of rats developed m a x i m a l t h e r m a l nociceptive block (latency of 12 seconds) and m o t o r block (extensor postural thrust -<40 g). The potencies in terms of ECso.max (Table 2) w e r e in the order neoSTX > STX > TTX > dcSTX for both t h e r m a l nociception and m o t o r function (P < .0001 for all comparisons). No difference was observed b e t w e e n the EC50.maxs for t h e r m a l nociception and m o t o r strength for STX and dcSTX. There was a small (16%) t e n d e n c y to sensory selectivity for neoSTX (P = .025). Tetrodotoxin s h o w e d no statistically significant difference
Saxitoxin Homologues
for Sciatic Nerve Block
•
K o h a n e et al.
55
150 + neoSTX ---<>-- STX -----B--- TIN ---Am dcSTX
.~ 125
100 >
Fig 2. Dose-response curve for duration of thermal nodceptive block for saxitoxins and tetrodotoxin. Data points are medians.
O
"~ 75
r~ O
~
50
e-. o
25 /
0--c 10
=
.~.~-
Lethality a n d T h e r a p e u t i c Indices of the Toxins To calculate the therapeutic index for these compounds, we determined the LD~0 (i.e., median dose [in nmol/kg]) resulting in terminal a p n e a in rats injected with toxin at the sciatic nerve (Table 3). As with the ECsos, the order of LDs0s was neoSTX < STX < "Iq?X < dcSTX (P < .0001 for comparisons b e t w e e n toxins). The LDs0 and ECs0 values (from Tables 1 and 2) were converted into pg/kg, and the therapeutic indices (ratio LDs0/EDso) were obtained with respect to both EC~0-max (which reflects the incidence and intensity of block) and EC~0-dur60 (which reflects incidence and duration). W h e n the ECs0.rnax was used to derive the denominator, no statistically significant differences were observed Table 1. Median Concentration Resulting in Thermal
Nociceptive Block Lasting 60 and 90 Minutes ECso-durS0 (pmol/L)
STX neoSTX dcSTX
58 +- 3 34 _ 2 268 _+ 8 92 __+5
TTX
ECso-durgo (pmol/L) 70 40 334 111
._-- .
,~ . . . ,.'.
.-.
50 100 Toxin concentration (tdVl)
b e t w e e n the ECso.maxS for thermal nociceptive block and m o t o r block.
Compound
i :,
+ + -
6 3 55 15
NOTE. ECso-durSOand ECso-du~9oare the concentrations resulting in median durations of block of 60 and 90 minutes, respectively. Values are shown with 95% confidence intervals. Pairwise statistical comparisons of the toxins yielded P < .0001 in all cases, n, number of rats studied to derive the ECsos. STX, Saxitoxin; neoSTX, neosaxitonin; dcSTX, decarbamoyl saxitoxin; -I-rx, tetrodotoxin.
1000
b e t w e e n the therapeutic indices for the various toxins, except for a weakly significant difference b e t w e e n TI"X and the toxin with the worst therapeutic index, STX. W h e n the therapeutic indices w e r e calculated based on the EC~0-dur60, dcSTX had the lowest therapeutic index; both T r x and neoSTX had significantly higher values. In all cases, the differences w e r e of relatively small m a g n i t u d e (the highest and lowest therapeutic indices differed by app r o x i m a t e l y 2 5 %). Onset of N e r v e B l o c k Rats were injected with selected concentrations of the various toxins that had b e e n s h o w n by the above experiments to uniformly result in successful blocks. (The different toxins were c o m p a r e d at different concentrations because of the differences
Table 2. Median Effective Concentration (ECs0.max) Required for Maximal Thermal Nociceptive and Motor Block
n 72 127 64 89
500
ECso.max (pmol/L) Compound
Thermal Nociception
Motor Block
STX neoSTX dcSTX
51 _ 4
53 -.+ 8 36 -.+ 5
TTX
30 - 1 202 +_ 22
202 -+ 20
76 - 11
73 - 11
P .66
.025 .98 .71
NOTE. Values for ECsos are given with 95% confidence intervals. P values compare thermal nociceptive block and motor block. The number of rats studied is the same as in Table 2. STX, saxitonin; neoSTX, neosaxitonin; dcSTX, decarbamoyl, saxitonin; F I X , tetrodotoxin.
56
Regional Anesthesia and Pain Medicine Vol. 25 No. 1 January-February 2000
Table 3. Median Lethal Doses (LDso) and Therapeutic Indices Derived From ECs0-max
Derived From EC5o-dur6o
Compound STX neoSTX dcSTX "I-FX
LDso (nmol/kg) 23 14 94 41
+_ 1 -+ 1 + 5 ___2
Therapeutic Index 1.3 1.42 1.2 1.51
P Values for Therapeutic Indices neoSTX
dcSTX
TTX
0.44
0.53 0.01
0.27 0.8 0.01
Therapeutic Index 1.49 1.61 1.59 1.82
P Values for Therapeutic Indices neoSTX
dcSTX
"l-lX
0.23
0.35 0.9
0.04 0.18 0.18
NOTE. LDsos for compounds injected at the sciatic nerve are given with 95% confidence intervals. All of the LDsos are statistically significantly different from each other (P < .0001 ). Therapeutic index = LDso/EDso.The Pvalues shown compare the therapeutic indices of the compounds listed in the first column with other toxins. The number of rats studied for each LDs0 were as follows. STX: 108; neoSTX: 150; dcSTX: 90;-I-I'X: 101. STX, saxitonin; neoSTX, neosaxitonin; dcSTX, docarbamoyl saxitonin; TTX, tetmdotoxin.
in their individual potencies.) The time of onset of t h e r m a l nociceptive block for each toxin is s h o w n in Table 4. There was no statistically significant difference in onset time b e t w e e n the three saxitoxins and TrX (P > .05 for Kruskal-Wallis comparing all 4 c o m p o u n d s ) . These onsets were all slower than that of 15.4 mol/L (0.5%) bupivacaine, a conventional a m i n o - a m i d e local anesthetic (P = .02 in comparison of TTX and bupivacaine). Rats injected with saline after general anesthesia n e v e r achieved our criteria for t h e r m a l analgesia (latency ->7 seconds). M o t o r block was detectable within 2 minutes (the first time point tested) for all groups, including saline-injected controls. Motor block p r o m p t l y resolved (by 4 minutes postinjection) in the salineinjected rats, but persisted in those injected with active compounds. Thus, we could not determine the exact time of onset of m o t o r block (since it occurred while general anesthesia was wearing off). This brief initial m o t o r block was c o m m o n l y seen in the contralateral leg and, therefore, was probably caused by w e a k n e s s f r o m residual general anesthesia. Discussion
Chemical substitutions on the saxitoxin molecule h a v e m a r k e d effects on p o t e n c y in vitro. In this study, we found large differences b e t w e e n the toxins in t e r m s of incidence and duration of block and
Table 4. Time to Onset of Thermal Nociceptive Block Compound and Concentration
Time to Onset of Nociceptive Block (rain)
n
S'IX (60 pmol/L) neoSTX (36 pmol/L) dcSTX (270 pmol/L) TTX (125 pmol/L) Bupivacaine (15.4 mol/L) Saline (0.9%)
6.2 (4.5-7) 8.2 (7-9.5) 6.2 (3.1-6.9) 5.2 (4.3-7) 2.3 (2.4-2.9) Never achieved
11 12 11 12 12 6
NOTE. Values are medians with 25th and 75th percentiles in parentheses. STX, saxitonin; neoS'l-X, neosaxitonin; dcSTX, decarl0amoyl saxitonin; TTX, tetrodotoxin.
LDso in vivo. HoweveL although the relative potencies of the STXs in rat sciatic nerve block mirrored their relative potencies in vitro, there was not an exact proportionality b e t w e e n in vitro and in vivo data. In Fig 3, we plotted the thermal nociceptive ECso.dur6o and ECso-max values from this study for each toxin against published values for 2 in vitro measures of potency: the relative ECs0s for depressing the c o m p o u n d action potentiaP .6 (Fig 3A) and the equilibrium dissociation constant KD determ i n e d from functional block of open single channels Is (Fig 3B). Both in vitro measures found similar proportionalities b e t w e e n the potencies of the saxitoxins (ECso or KD of neoSTX ~ 0.25 times that of STX, dcSTX ~ 2.3 times STX). The corresponding proportionalities from our study were neoSTX ~ 0.6 times STX, dcSTX = 4.3 times STX. There was close a g r e e m e n t b e t w e e n the relative potencies of the toxins as determined by the ECso-maxand ECs0.dur6 0. The ECso-maxof TrX in vivo was greater t h a n that of STX, but m u c h lower t h a n that of dcSTX. Previously published in vitro data had suggested that the potencies of dcSTX 6 and TrX 4 relative to STX w e r e similar or that dcSTX was slightly m o r e potent t h a n TI'X. This disparity b e t w e e n in vivo and in vitro data could be a result of T r x having some property that enables it to cross biologic m e m b r a n e s m o r e readily. The difference b e t w e e n the absolute values of in vivo potencies described here and those previously reported in vitro 4,6 is consistent with the established view that these barriers (such as the epi- and p e r i n e u r i u m ) are a real obstacle to the action of these toxins. The in vivo EthoS for the toxins w e r e a p p r o x i m a t e l y 1,000 to 10,000 times the in vitro KD (Fig 3). Greater ease in penetrating such barriers has b e e n used to explain the higher ratio of in vivo to in vitro potency~° for bupivacaine (approximately 100) c o m p a r e d with T r x (approximately 1,000). Tetrodotoxin is quite similar to the STXs in most physicochemical characteristics that affect p h a r m a c o k i n e t -
Saxitoxin Homologues for Sciatic Nerve Block
A 300
i
[
•
i
25O
dcSTX
I
o
~ 150
8 8
TTX a ~STX 50 I
]
neoSTX
I
~
I
0.5 1 1.5 2 2.5 EC58 for CAP inhibition normalized to STX
B 300
I Ir
250 dcSTX
i
~ 150
8 100
STX
5o
TTX
:
~ neoSTX
0 0
10 15 20 Equilibrium constant Kv (nM)
25
30
Fig 3. Comparison of in vivo and in vitro effectiveness of STX homologues and Trx. The principal measures of in vivo effectiveness'were the ECso-maxfor thermal nociceptive block (open circles), and the concentration yielding a duration of 60 minutes (ECs0-dur6O,dark squares). These data are compared with the in vitro ratio of the ECso for compound action potential (CAP) inhibition to the value for STX (A), and the equilibrium dissociation constant (Ko; B) for each toxin. The value of Ko is given by the ratio of the rate constants for dissociation and binding from the sodium channel, and are the primary determinants of the potencies of toxins (see Hall et ai.24).
ics. None of these c o m p o u n d s h a v e appreciable hydrophobicity, and they are all of a p p r o x i m a t e l y the same molecular weight (300 to 400 da). Saxitoxin and dcSTX are divalent cations at physiologic pH (with positive charges associated with the guanidinium groups [Na and Na]). Neosaxitoxin has the same two positive charges, but also has a partial negative charge (pKb = 6.75) at physiologic pH. Tetrodotoxin is a m o n o v a l e n t cation (because of its single g u a n i d i n i u m group). The lesser charge m i g h t provide an advantage in crossing lipid-rich perineural m e m b r a n e barriers, thus m a k i n g it m o r e perm e a n t and effectively m o r e p o t e n t t h a n dcSTX in
•
Kohane et al.
57
vivo. However, the fact that the onset of TTX block is not shorter than that of the STXs argues against this explanation. Bupivacaine, which is amphiphilic and has an intermediate degree of hydrophobicity 16 at physiologic pH, has m o r e rapid onset than TTX and the STXs. No correlation was found between in vitro or in vivo potency and the time to onset of sciatic nerve block. A likely explanation for this finding is that the pharmacokinetic barriers to the penetration of the hydrophilic toxins into or within the nerve bundle are the rate-limiting step to the d e v e l o p m e n t of nerve block. Differences b e t w e e n in vitro and in vivo effectiveness have also been noted with conventional local anesthetics. Mepivacaine is m o r e potent in vivo than prilocaine and lidocaine in vivo, but less potent t h a n t h e m in vitro. 7 The reduced performance in the nerve bath has been attributed to differential ionization of the local anesthetics. Differential effects of local anesthetics on the vasculature of the n e r v e could also have a m a r k e d effect on potency in vivo. The ratios b e t w e e n the ECso.maxvalues for sensory and m o t o r block for the saxitoxins and tetrodotoxin were similar. These findings are s o m e w h a t surprising because one might have expected m o t o r pred o m i n a n c e in t o x i n - d e p e n d e n t block based on current research on the differential actions of site 1 sodium channel blockers on nerve fiber subpopulations. A-alpha and A-delta peaks of c o m p o u n d action potentials of frog sciatic nerves have b e e n s h o w n to be abolished by application of 1 iamol/L T r x , whereas the C p e a k was unaffected. 17 Other investigators 18 using rat L4-L5 dorsal root ganglia have also s h o w n action potentials in A-alpha/beta cells to be blocked by 0.1 to 1 pmol/L TTX, while a proportion of A-delta and all C cells had impulses that w e r e T r x insensitive, generating the concept that TrX-insensitive sodium channels are critical for impulse activation in p r i m a r y nociceptors. H o w ever, similar discrepancies h a v e b e e n f o u n d bet w e e n electrophysiologic data and in vivo observations of sensorimotor selectivity with veratridine s and ropivacaine.~2.~9 The transient bilateral m o t o r block seen in all e x p e r i m e n t a l groups (including the saline-injected controls) was ascribable to the initial inhalational anesthetic. It was of very brief duration (it was only observed during experiments to determine onset of n e r v e block), and consequently it is unlikely that it contributed to the m u c h longer n e r v e deficits resulting f r o m toxin injection. One caution in interpreting the differences bet w e e n sensory and m o t o r block in o u r e x p e r i m e n t a l paradigm is that in b o t h cases w e are dealing w i t h
58
Regional Anesthesia and Pain Medicine Vol. 25 No. 1 January-February 2000
"censored" data with s o m e w h a t arbitrarily established p e r f o r m a n c e limits. This is of particular importance in the determination of the ECsos of m a x i m a l block. In the case of sensory data, "maximal" latency is set at 12 seconds, beyond which there is a risk of injury or sensitization to the rat's foot. With m o t o r data, "maximal" block could not simply be defined as the lowest weight b o r n e by a n y given rat, because by that m e a s u r e every rat m u s t achieve m a x i m a l block. Consequently, an arbitrary cut-off of less t h a n 40 g was adopted. Therefore, "maximal" sensory and m o t o r block as defined here m a y not reflect exactly the same degrees of block of the respective functions. Nonetheless, both measures indicate a m a j o r deficit in the respective functions.S. 10-J3 The toxicity of site 1 sodium channel blockers has been described extensively, m.2°.21 Unlike conventional local anesthetics, w h e r e the m a j o r toxicity is neurologic (seizures and coma) and cardiovascular (dysrhythmias and cardiac depression), morbidity f r o m these toxins results primarily from respiratory failure from diaphragmatic paralysis. The latter is believed to r e s u l t f r o m toxin action on the diap h r a g m and phrenic nerves, rather t h a n in the central n e r v o u s system. 2° Large differences were observed in the LDs0s of the saxitoxins that also mirrored the potencies predicted by in vitro data. However, the differences in LDs0s b e t w e e n individual toxins w e r e paralleled by differences in the ECs0s, w h i c h resulted in similar therapeutic indices. Surprisingly, the therapeutic index of TTX was generally higher (i.e., safer) t h a n that of the saxitoxins, although the difference was probably too small to be of clinical significance. A priori, one might h a v e expected the opposite, because T r x tends to cause substantial h y p o t e n s i o n (via direct action on s m o o t h muscle cells and s y m p a t h e t i c nerves) while this is m u c h less of a feature of STX. 2~ The LDso value for TTX in this study was very similar to that f o u n d in a previous study ~° (41 vs 40 m m o l / L / k g ) . However, because the v o l u m e of injectate was 0.3 m L in the earlier article, and only 0.1 m L in this article, the relative toxicities of a n y given T r x concentration w e r e m a r k e d l y different. For example, 100 lamol/L T r x in the earlier article (corresponding to 10 lag or 89 n m o l / L / k g ) was u n i f o r m l y fatal, w h e r e a s h e r e it resulted in no deaths (as it corresponded to 3.3 lag or 30 nmol/kg). These findings further confirm the usefulness of total dose (rather t h a n concentration), w h e n comp a n g the systemic toxicities of different preparations, The ECs0-max reported here is a p p r o x i m a t e l y twice t h a t in the earlier article, for similar reasons. BeCause a ~ v e n concentration in this article repre-
sents one third tile a m o u n t of drug in the previous one, one would expect a lesser local effect. Although it is important to understand the perform a n c e characteristics of the individual toxins alone, we have proposed that they are likely to be used clinically in combination with either conventional local anesthetics or vasoconstrictors. The latter compounds greatly increase the duration of block from site sodium channel blockers, io.22.23 and the vasoconstrictors greatly improve their therapeutic index. 1° Bupivacaine also increases the LDso of TrX to a small degree, possibly by a vasoconstrictive m e c h a nism. It seems reasonable to expect that the h o m o logues of STX would be potentiated in a similar manner. However, it is impossible to predict w h e t h e r the rank order of their potencies in combination with a second drug will be the same. In s u m m a r y , chemical substitutions on the STX molecule result in m a r k e d changes in potency and toxicity in vivo. In vitro measures of effectiveness or potency are not perfect predictors of in vivo potency, as evidenced by the greater potency of T r x in relation to the STXs in vivo, and the differences in the relative potencies of STXs in vitro and in vivo. The differences in intrinsic potency b e t w e e n site 1 sodium channel blockers only result in small differences in therapeutic indices because the LDs0s are affected in a proportional manner.
References 1. Catterall WA. Neurotoxins that act on voltagesensitive sodium channels in excitable membranes. Annu Rev PhaTw,acol Toxicol 1980;20:15-43. 2. Hille B. The receptor for tetrodotoxin and saxitoxin. A structural hypothesis. Biophys J 1975; 15:615-619. 3. Narahashi T. Mechanism of action of tetrodotoxin and saxitoxin on excitable membranes. Fed Proc 1972; 31:1124-1132. 4. Strichartz G, Rando T, Hall S, Gitschier J, Hall L, Magnani B, Bay CH. On the mechanism by which saxitoxin binds to and blocks sodium channels. Ann N YAcad Sci 1986;479:96-112. 5. Fuhrman FA. Tetrodotoxin. Sci Am 1967;217:60-71. 6. Strichartz GR, Hall S, Magnani B, Hong CY, Kishi Y, Debin JA. The potencies of synthetic analogues of saxitoxin and the absolute stereoselectivity of decarbamoyl saxitoxin. Toxicon 1995;33:723-737. 7. Pateromichelakis S, Prokopiou AA. Local anesthesia efficacy: Discrepancies between in vitro and in vivo studies. Acta Anesthesiol Scand 1988;32:672-675. 8. Vladimirov M, Thalhammer JG, Hunt N, Feldman HS, Strichartz G. Neurophysiologic actions and neurological consequences of veratridine on the rat sciatic nerve. Anesthesiology 1997;86:945-956. 9. Sakura S, Bollen AW, Ciriales R, Drasner K. Local anesthetic neurotoxicity does not result from blockade of voltage-gated sodium channels. Anesth An alg 1995;81:338-346.
Saxitoxin Homologues for Sciatic Nerve Block 10. Kohane DS, Yieh J, Lu NT, Langer R, Strichartz G, Berde CB. A reexamination o[ tetrodotoxin for prolonged anesthesia. Anesthesiology 1998;89:119-131. 11. Thalhammer JG, Vladimirova M, Bershadsky B, Strichartz GR. Neurologic evaluation of the rat during sciatic nerve block with lidocaine. Anesthesiology 1995; 82:1013-1025. 12. Kohane DS, Sankar W, Hu D, Shubina M, Refai N, Berde CB. Sciatic nerve blockade in infant, adolescent, and adult rats: A comparison of ropivacaine and bupivacaine. Anesthesiology 1998;89:1199-1208. 13. Kohane DS, Kuang Y, Lu NT, Langer R, Strichartz GR, Berde CB. Vanilloid receptor agonists potentiate the in vivo local anesthetic activity of percutaneously injected site 1 sodium channel blockers. Anesthesiology 1999;90:524-534. 14. Masters DB, Berde CB, Dutta SK, Griggs CT, Hu D, Kupsky W, Langer R. Prolonged regional nerve blockade by controlled release of local anesthetic from a biodegradable polymer matrix. Anesthesiology 1993;79: 100-107. 15. Guo X, Uehara A, Ravindran A, Bryan SH, Hall S, Moczydlowski E. Kinetic basis for insensitivity to tetrodotoxin and saxitoxin in sodium channels of canine heart and denervated rat skeletal muscle. Biochemistry 1987;26:7546-7556. 16. Bernards CM, Hill HE Physical and chemical properties of drug molecules governing their diffusion through the spinal meninges. Anesthesiology 1992; 77:750-756.
•
Kohane et al.
59
17. Buchanan S, Harper AA, Elliott dR. Differential effects of tetrodotoxin (TTX) and high external K+ on A and C fibre compound action potential peaks in frog sciatic nerve. Neurosci Lett 1996;219:131 - 134. 18. Villiere V, McLachlan EM. Electrophysiological properties of neurons in intact rat dorsal root ganglia classified by conduction velocity and action potential d u ration, d Neurophysiol 1996;76:1924-1941. 19. Rosenberg PI-I, I-Ieinonen E. Differential sensitivity of A and C fibers to long-acting amide local anesthetics. Br J Anaesth 1983;55:163-167. 20. Kao CY. Tetrodotoxin, saxitoxin, and their significance in the study of excitation phenomena. Pharmacol Rev 1966; 18:997-1049. 21. Kao CY. Pharmacology of tetrodotoxin and saxitoxin. Fed Proc 1972;31:1117-1123. 22. Adams HJ, Blair MR, Jr., Takman BH. The local anesthetic activity of tetrodotoxin alone and in combination with vasoconstrictors and local anesthetics. Anesth Analg 1976;55:568-573. 23. Adams HJ, Blair MR, Jr., Takman BH. The local anesthetic activity of saxitoxin alone and with vasoconstrictor and local anesthetic agents. Arch Int Pharmacodyn 1976;224:275-282. 24. Hall S, Strichartz G, Moczydlowski E, Ravindran A, Reichardt PB. The saxitoxins. Sources, chemistry, and pharmacology. In: Hall S, Strichartz G, eds. Marine Toxins. Origin, Structure, and Molecular Pharmacology. Washington, DC: American Chemical Society, 1990;29-65.