Neamine and 2-deoxystreptamine neomycin derivatives exhibit antinociceptive activity in rat models of phasic, incision and neuropathic pain

Article abstract of DOI:10.1111/jphp.12480

Objectives To assess the antinociceptive activity of the neomycin derivatives neamine and 2-deoxystreptamine following intraspinal administration in rats. Methods We used the tail-flick test and measured the threshold to mechanical stimulation in models of incisional and neuropathic pain. Key findings The derivatives produced antinociception in the tail-flick test and reduced mechanical allodynia in models of incisional and neuropathic pain. The approximate ED₅₀ in milligrams (confidence limits in parenthesis) in these tests were 1.35 mg (0.61; 2.95), 0.20 mg (0.14; 0.27) and 0.28 mg (0.12; 0.63) for neamine, and 1.05 mg (0.68; 1.60), 0.78 mg (0.776; 0.783) and 0.79 mg (0.46; 1.34) for 2-deoxystreptamine, respectively. Neamine was more potent than 2-deoxystreptamine in the incisional and neuropathic pain models, but they had similar potency in the tail-flick test. Tetra-azidoneamine, a neamine derivative in which free amino groups are replaced with azido groups, did not change the incisional mechanical allodynia. The reduction of incisional allodynia by neamine and 2-deoxystreptamine was transitorily antagonized by intrathecal administration of calcium chloride. Conclusions The intraspinal administration of neamine and 2-deoxystreptamine is antinociceptive in rats. The presence of amino groups in the structure of these derivatives is fundamental to their antinociceptive effect, which may be due to a calcium antagonist activity.

Full text of DOI:10.1111/jphp.12480

Research Paper
Journal of Pharmacy
And Pharmacology
Neamine and 2-deoxystreptamine neomycin derivatives
exhibit antinociceptive activity in rat models of phasic,
incision and neuropathic pain
a
a
b
b
Wiliam A. Prado , Ana C. Rossaneis , Ivone Carvalho , Luis Otávio B. Zamoner and
a
Alexandre P. Corrado
a
b
Department of Pharmacology, Ribeirão Preto Medical School and Department of Pharmaceutical Sciences, Faculty of Pharmaceutical Sciences of
Ribeirão Preto, University of São Paulo, Ribeirão Preto, Brazil
Keywords
Abstract
2
-deoxystreptamine; aminoglycoside
antibiotic; incision pain; neamine; neuropathic
pain
Objectives To assess the antinociceptive activity of the neomycin derivatives
neamine and 2-deoxystreptamine following intraspinal administration in rats.
Methods We used the tail-flick test and measured the threshold to mechanical
stimulation in models of incisional and neuropathic pain.
Key findings The derivatives produced antinociception in the tail-flick test and
reduced mechanical allodynia in models of incisional and neuropathic pain. The
approximate ED50 in milligrams (confidence limits in parenthesis) in these tests
were 1.35 mg (0.61; 2.95), 0.20 mg (0.14; 0.27) and 0.28 mg (0.12; 0.63) for
neamine, and 1.05 mg (0.68; 1.60), 0.78 mg (0.776; 0.783) and 0.79 mg (0.46;
Correspondence
Wiliam A. Prado, Department of
Pharmacology, Ribeirão Preto Medical School,
Bandeirantes, Avenue 3900, CEP 14049-900,
Ribeirão Preto, SP, Brazil.
E-mail: wadprado@fmrp.usp.br
Received November 10, 2014
Accepted July 19, 2015
1
2
.34) for 2-deoxystreptamine, respectively. Neamine was more potent than
-deoxystreptamine in the incisional and neuropathic pain models, but they had
similar potency in the tail-flick test. Tetra-azidoneamine, a neamine derivative in
which free amino groups are replaced with azido groups, did not change the
incisional mechanical allodynia. The reduction of incisional allodynia by neamine
and 2-deoxystreptamine was transitorily antagonized by intrathecal administra-
tion of calcium chloride.
doi: 10.1111/jphp.12480
Conclusions The intraspinal administration of neamine and 2-deoxystreptamine
is antinociceptive in rats. The presence of amino groups in the structure of these
derivatives is fundamental to their antinociceptive effect, which may be due to a
calcium antagonist activity.
Introduction
Aminoglycoside antibiotics, such as gentamicin, streptomy-
cin, neomycin and kanamycin, produce antinociception in
rodent models of experimental pain following intraperito-
antibiotic produces a severe decline in blood pressure
accompanied by neuromuscular blockade, both of which
are competitively antagonized by calcium.
[9,10]
[
1]
[2,3]
neal, intracerebroventricular or intraspinal administra-
Streptidine, an aminocyclitol derivative from streptomy-
cin, produces a neuromuscular blockade with the same
[
3]
tion. The antinociceptive effect of intraspinal gentamicin
was antagonized by intraspinal administration of calcium
[
11]
characteristics as the entire molecule of the antibiotic,
including a competitive interaction with calcium.
[
3]
[12]
chloride, thus leading to the idea that the effect of gen-
tamicin depends on its calcium-antagonist property. Gen-
tamicin and other aminoglycoside antibiotics act as
More recently, it was shown that neamine, but not
2-deoxystreptamine (aminoglycoside-aminocyclitol and
aminocyclitol derivatives of neomycin, respectively)
(Figure 1), retains the antibacterial and ototoxic properties
[
4,5]
antagonists of N-type calcium channels
and compete
[6]
with calcium in several biological processes. However, the
chronic utilization of these compounds is frequently fol-
lowed by oto- and nephrotoxicity. Moreover, the fast
increase of the plasmatic concentration of aminoglycoside
[13]
of the entire antibiotic molecule.
[
7]
[8]
Discovery of new pain relieving drugs is necessary since
many patients still suffer from pain mainly because the
1
696
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704

Wiliam A. Prado et al.
Neomycin derivatives and antinociception.
NH2
‘
6
NH2
O
HO
HO
2‘
O
NH2
3
5
HO
HO
I
NH2
O
O
NH2
H2N
O
HO
NH2
H2N
OH
OH
O
HO
NH2
OH
2”
III
H2N
O
OH
O
OH
neomycin
neamine
N3
O
HO
HO
H2N
N3
HO
HO
NH2
N3
O
HO
N3
OH
-deoxystreptamine
Figure 1 Chemical structures of neomycin and its derivatives neamine, 2-deoxystreptamine and tetra-azidoneamine.
OH
2
tetra-azidoneamine
[
14]
available drugs are effective but limited by side effects.
lines of the Committee for Research and Ethical Issues of
the International Association for the Study of Pain.
[18]
Here, we focused on the potential of neomycin derivatives
as new analgesic drugs. It is currently accepted that estab-
lishing efficacy across a variety of models reduces transla-
tional risk associated with the target mechanism, and
variability in potency or maximal effects can help inform
both clinical trial design and patient population selec-
Tail-flick test
Each animal was placed in a ventilated tube with the ventral
surface of the tail between 4 and 6 cm from the tip laid
across a wire coil maintained at room temperature
[
15]
tion. By this reason, this study examines comparatively
whether neamine and 2-deoxystreptamine exhibit anti-
nociceptive activity on three models of pain in rats. The rat
tail-flick test, incision pain and neuropathic pain were used
to determine the antinociceptive effects of the neomycin
derivatives against acute, post-surgical and neuropathic
pain, respectively. These methods are used widespread and
(
23°C ± 2°C). The coil temperature was then raised by the
passage of electric current through it, and the tail-flick
latency (TFL) was measured. Each trial was terminated after
6
measured at 5-min intervals until a stable baseline was
obtained over three consecutive trials, 10 min after intrathe-
cal injection and then at 10-min intervals for up to 60 min.
s to minimize the probability of skin damage. TFL was
[16]
are essential to progress in pain research. The efficacy of
the derivatives in each test was established. They were in-
Incision pain test
[17]
[3]
jected intrathecally because neomycin and gentamicin
Each animal was anaesthetized with isoflurane (2% for
induction and 0.5% for maintenance) in oxygen flow via a
loose-fitting, cone-shaped mask. A 1-cm longitudinal inci-
sion was made with a surgical blade through the skin and
fascia of the plantar aspect of the right hind paw starting
were more effective in rats by intrathecal than by intra-
cerebroventricular and intraperitoneal routes. The changes
in incision pain induced by tetra-azidoneamine, a neamine
derivative in which the amino groups were replaced by
azido groups (Figure 1), were also evaluated. The effect of
intrathecal administration of calcium chloride against the
antinociceptive activity of neamine and 2-deoxystreptamine
in the incision pain was evaluated as well.
0
.5 cm from the proximal edge of the heel, as described
[19]
elsewhere.
The wound was then closed with two 5-0
nylon sutures, and the animal allowed to recover in the
home cage until the beginning of the experiment.
Materials and Methods
Chronic constriction injury model
Animals
Each animal was anaesthetized as above and subjected to
chronic constriction injury (CCI) of the sciatic nerve as
Male Wistar rats (140–160 g) from the main animal house
of the Ribeirão Preto Campus of the University of São
[20]
proposed elsewhere. A single ligature of 100% cotton
glace thread (Coats Corrente Ltda., São Paulo, Brazil) was
placed around the sciatic nerve proximal to its trifurcation.
The incision was then closed with silk sutures, and the
animal was kept in recovery for 8–10 days.
Paulo were maintained at a controlled temperature
o
(
23 ± 1 C) with free access to food and water. The experi-
ments were approved by the institutional ethics committee
on animal research (No. 116/2012) and followed the guide-
©
2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704
1697

Neomycin derivatives and antinociception
Wiliam A. Prado et al.
Mechanical stimulation
interaction, one-way analysis of variance followed by the
Bonferroni correction was performed for each time point.
The approximate dose producing antinociception in 50%
of the rats in an experimental group (ED50) and 95% confi-
dence limits were calculated as proposed by Litchfield and
Rats with hind paw incision or CCI were placed in a cage
with a nylon mesh bottom, which allowed easy access to the
paw plantar surface. The threshold for mechanical stimula-
tion was measured with an electronic apparatus (IITC Elec-
tronic Equipment, Woodland Hills, CA, USA) consisting of
a hand-held probe unit to which a rigid plastic tip (tip area
[23]
Wilcoxon. The Litchfield–Wilcoxon method was chosen
because it allows the calculation of approximate ED50
[24]
using a small number of animals.
Antinociception
0
.44 mm2) was connected. The plastic tip was applied with
in the tail-flick test was defined whenever TFL > baseline
TFL + (5 × SD). Antinociception in the tests using mechani-
cal stimulation was defined whenever the mechanical
threshold was at least 50% of the baseline. The regression
lines (not shown in the figures) for neamine and
increasing force against the central area of each hind paw of
CCI animals or sites near the heel, 1–2 mm adjacent to the
medial border of the wound in the incised hind paw. The
movement of the probe was stopped when withdrawal of
the stimulated paw occurred. A single trial consisted of
three applications of the tip, once every 5 s, to each hind
paw. The mean of three readings was taken as the threshold
for a particular time. The threshold was determined
immediately before surgery (baseline) and 2 h after hind
paw incision or 8–10 days after CCI. The threshold was
measured again 10 min after intraspinal injection and then
at 10-min intervals for up to 80 min. The mechanical
threshold in non-operated hind paws was measured as well.
2-deoxystreptamine in each model were compared with
regard to the slope function ratio and corresponding factors
[23]
as suggested elsewhere. The potency ratio of neamine and
-deoxystreptamine in each model was also calculated
2
whenever the regression lines did not deviate significantly
from parallelism.
Preparation of neamine
Neomycin trisulfate (1.0 g, 1.10 mmol) was dissolved in a
saturated solution of methanol (130 ml) and HCl (pH < 1),
and heated at 80°C for 3 h. After cooling to 0°C, the reac-
tion was treated with ethyl ether (40 ml) to precipitate a
white flocculent solid, which was filtered and washed with
ethyl ether. The solid was dried under reduced pressure, and
the resulting crude product was purified as described else-
Intrathecal injection
The injections were performed in rats lightly anaesthetized
as described above.
A
1-inch, 25-G needle was
transcutaneously introduced at the L5–L6 level into the
subarachnoid space as described elsewhere.
received only one intrathecal injection (constant 5-μl
volume) using this method. The animal was submitted to
the algesimetric test 10 min later.
[21]
Each rat
[25]
where. The solid was dissolved in methanol (200 mg/
1
0 ml) followed by titration with ethyl ether until start to
precipitate a white insoluble fraction, which was removed in
glass filter. This procedure was repeated until removal of
impurity. Purity of the solution was monitored by TLC
When two injections per rat were necessary, each rat was
catheterized via a lumbar puncture as described else-
[
22]
where. A 20-gauge Weiss needle was introduced through/
the skin into the L5–L6 intervertebral space. A 12-mm
length of polyethylene tubing (o.d. = 0.4 mm, dead space
(
methanol-ethyl acetate 2 : 1). The solution was evaporated
1
to yield the product in 67%. H NMR (400 MHz, D
1H, q, J 12.8 Hz, H-2’ax); 1.91 (1H, dt, J 4.0; 12.8 Hz,
H-2’eq); 2.66 (1H, ddd, J 4.0; 9.8; 12.8 Hz, H-1′); 2.71–2.84
3H, m, H-2. H-3′, H-6a); 2.98 (1H, dd, J 2.7; 13.6 Hz,
2
O): 1.14
(
=
10 μl) was introduced through the needle to protrude
1
.5 cm into the subarachnoid space in the cranial direction.
(
The needle was removed, and the tubing anchored to the
back skin with a cotton thread suture. A surgical incision
was then performed in the right hind paw as described
above. The experiment was carried out 2 h later. All solu-
tions contained 1% fast green dye to confirm the correct
position of the catheter.
H-6b); 3.09 (1H, t, J 9.8 Hz, H-6′); 3.19–3.28 (2H, m, H-4,
H-4′); 3.44 (1H, t, J 9.8 Hz, H-5′); 3.51 (1H, t, J 9.8 Hz,
H-3); 3.72 (1H, ddd, J 2.7; 7.3; 9.8 Hz, H-5); 5.23 (1H, d, J
3.7 Hz, H-1).
Preparation of 2-deoxystreptamine
A solution of neamine chloride (0.155 g) in aqueous
hydrobromic acid (HBr 48%, 3.2 ml) was heated at 130°C
for 18 h. The solution became dark during this period. The
mixture was then allowed to cool down to room tempera-
ture, evaporated under reduced pressure and repeatedly
treated with water and evaporated three times to remove
acid excess. The residue was dissolved in warm methanol
(3.2 ml), filtered, and treated with activated charcoal for
Statistical analysis
TFL (in seconds) and mechanical threshold (in grams) are
reported as the mean ± SD. Comparisons between control
(
rats given saline intrathecally) and test groups were made
by multivariate analysis of variance with repeated measures.
The factors analysed included treatment, time and treat-
ment × time interaction. In the case of the treatment × time
1
698
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704

Wiliam A. Prado et al.
Neomycin derivatives and antinociception
3
min and filtered again. Finally, the system was filtered and
(a)
*
*
#
*
washed with water. The obtained solution was dried under
reduced pressure, yielding 2-deoxystreptamine (98%). H
NMR (400 MHz, D
*
6
5
4
3
2
1
0
#
1
2
O): 1.66 (1H, q, J 12.3 Hz, H-2ax); 2.31
*
*
(1H, dt, J 4.2; 12.3 Hz, H-2eq); 3.14–3.22 (2H, m, H-1,
H-3); 3.28 (1H, t, J 9.3 Hz, H-5); 3.39 (2H, t, J 9.3 Hz, H-6,
H-4).
Preparation of tetra-azidoneamine
saline
0.4 mg
0.8 mg
A solution of neamine chloride (0.15 g, 0.35 mmol) in
water (6 ml) was treated with sodium carbonate (0.38 mg,
0
.2 mg
2.3 mmol) then copper sulfate (2.8 mg, 0.02 mmol) and
–
20
(b)
0
20
*
40
60
methanol (2.0 ml). Finally, trifluoromethanesulfonyl azide
solution (prepared procedure described below) was added.
Then, methanol was added to make a homogenous solu-
tion, which was stirred at room temperature for 24 h. After
that, solvents were evaporated and residue partitioned in
ethyl acetate and water. To eliminate copper salts, the
organic layer was washed with EDTA solution (10%).
*
6
5
4
3
2
1
0
*
*
*
*
Organic layer was dried under MgSO , filtered and evapo-
rated, yielding neamine tetra-azide in 71%.
4
Preparation of trifluoromethanesulfonyl
saline
.5 mg
1.0 mg
2.0 mg
azide (TfN
A solution of sodium azide (1.20 g, 0.18 mmol) in water
3 ml) was cooled to 0°C, and dichloromethane (5 ml)
was added, followed by dropwise addition of trifluoro-
methanesulfonic anhydride (Tf O) (600 μL) during 5 min.
3
)
0
–
20
0
20
Time (min)
40
60
(
2
Figure 2 The time course of the effects of intrathecal administration
(arrow) of saline (10 μl), neamine or 2-deoxystreptamine on the tail-
flick latency of rats. Points are the mean (± SD) of six rats/curve. Doses
of neamine (a) and 2-deoxystreptamine (b) are given in mg/10 μl.
P < 0.05 vs saline-treated rats (*) or vs all the remaining groups (#).
After 2 h, the organic layer was separated and aqueous layer
extracted with dichloromethane (5 ml × 2). Organic layers
were combined and washed with sodium carbonate solu-
tion (Na
1
2
CO
3
). H NMR (400 MHz, DMSO-d
6
/D O): δ 1.40
2
(1H, q, J 12.3 Hz, H-2’ax); 2.29 (1H, dt, J 4.0; 12.3 Hz,
H-2’eq); 3.21 (1H, dd, J 3.7; 10.6 Hz, H-2); 3.24–3.35 (1H,
m, H-1′, H-3′); 3.45–3.62 (6H, m, H-6a, H-6b, H-4′, H-4,
H-6′, H-5′); 3.79 (1H, dd, J 8.8; 10.6 Hz, H-3); 4.08 (1H,
ddd, J 2.7; 6.3; 8.8 Hz, H-5), 5.66 (1H, d, J 3.7 Hz, H-1).
neamine, but not 2-deoxystreptamine, was also dose-
dependent. No animal showed gross disturbance of motor
coordination, which was judged by visual inspection after
intrathecal administration. The curves in Figure 2a and 2b
were significantly different in terms of treatment F3,20 = 28.0
and 14.4, respectively) and time (F8,160 = 27.8 and 36.5,
respectively) and have a significant treatment × time inter-
action (F = 57.8 and 36.9, respectively) (P < 0.0001 in all
cases).
Results
The antinociceptive effects of intrathecal
injection of neamine and
2
-deoxystreptamine on the tail-flick test
The time course of the changes induced in the TFL by
intrathecal neamine (0.2–0.8 mg) and 2-deoxystreptamine
The estimated ED50 of neamine and 2-deoxystreptamine
in the test were 1.35 mg (confidence limits: 0.61 and 2.95)
and 1.05 (confidence limits: 0.68 and 1.60), respectively.
The regression lines obtained for neamine and 2-
deoxystreptamine did not deviate significantly from paral-
lelism (slope ratio = 1.74 < slope ratio factor = 4.0), nor did
they differ significantly regarding the calculated potency in
this test (potency ratio = 1.28; confidence limits: 0.52 and
3.13).
(
0.5–2.0 mg) is shown in Figure 2a and 2b, respectively. The
effects of the smaller doses of each derivative did not differ
from the effect of saline (10 μl). However, higher doses pro-
duced a significant increase in the TFL compared with the
effect of saline. The peak effect was reached 30–40 min after
administration of neamine and 20 min after administration
of 2-deoxystreptamine. The duration of the effect of
©
2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704
1699

Neomycin derivatives and antinociception
Wiliam A. Prado et al.
The antinociceptive effects of intrathecal
injection of neamine and
saline. The duration of the effects of neamine and
2-deoxystreptamine was also dose-dependent. No animal
showed gross disturbance of motor coordination, which
was judged by visual inspection after intrathecal adminis-
tration. The curves in Figure 3a and 3b were significantly
different in terms of treatment (F3,20 = 48.1 and 231.6,
respectively) and time (F9,180 = 475,2 and 1508.0, respec-
tively) and have a significant treatment × time interaction
2
-deoxystreptamine on incision allodynia
None of the rats had spontaneous pain 2 h after surgery.
However, all rats had a significant reduction of the
mechanical threshold of the incised hind paw measured 2 h
after surgery. The time course of the changes induced by
intrathecal neamine (0.1–0.8 mg) and 2-deoxystreptamine
(F
27,180 = 44.18 and 152.2, respectively) (P < 0.0001 in all
cases).
The estimated ED50 of neamine and 2-deoxystreptamine
were 0.20 mg (confidence limits: 0.14 and 0.276) and
.78 mg (confidence limits: 0.776 and 0.783), respectively.
The regression lines obtained for neamine and
-deoxystreptamine did not deviate significantly from par-
(
0.5–2.0 mg) in the mechanical threshold is shown in
Figure 3a and 3b, respectively. The effects of the smaller
doses of neamine did not differ from the effect of saline
(
10 μl). However, the higher doses of neamine and all doses
0
of 2-deoxystreptamine produced a significant increase of
the mechanical threshold compared with the effect of
2
allelism (slope ratio = 1.38 < slope ratio factor = 3.48).
Neamine was 3.9 (confidence limits: 1.39 and 11.48) times
more potent than 2-deoxystreptamine. No change in the
mechanical threshold of the non-incised hind paw of
saline-, neamine- or 2-deoxystreptamine-treated rats was
detected throughout the period of experimentation (data
not shown).
(
a)
5
4
3
2
1
0
0
0
0
0
0
#
*
#
*
#
*
#
*
#
*
#
*
#
*
#
*
The antinociceptive effects of intrathecal
injection of neamine and
2-deoxystreptamine on neuropathic pain
saline
0.2 mg
0.8 mg
None of the rats exhibited autotomy or any motor abnor-
mality after surgery. All animals had a significant reduc-
tion of the mechanical threshold measured 8–10 days after
surgery in the operated hind paw compared with preop-
erative thresholds. None of the rats had spontaneous pain
during this period. The time course of the changes
induced in the mechanical threshold of operated rats by
intrathecal neamine (0.2–0.8 mg) and 2-deoxystreptamine
(0.5–2.0 mg) is shown in Figure 4a and 4b, respectively. All
doses of neamine and 2-deoxystreptamine produced a sig-
nificant and dose-dependent increase of the mechanical
threshold compared with the effect of saline (10 μl).
The effects of different doses of neamine and 2-
deoxystreptamine remained significantly above those of
the control throughout the period of observation. No
animal showed gross disturbance of motor coordination,
which was judged by visual inspection after intrathecal
administration. The curves in Figure 4a and 4b were sig-
nificantly different in terms of treatment (F3,20 = 552.4 and
0
.1 mg
–
122 –2
0
20
40
60
80
#
(
b)
5
4
3
2
1
0
0
0
0
0
0
#
#
#
#
*
*
#
*
#
*
*
*
*
*
*
*
*
*
*
*
*
*
saline
.5 mg
1.0 mg
2.0 mg
0
–
122 –2
0
20
40
60
80
Time (min)
1
358.0, respectively) and time (F9,180 = 1263.0 and 2877.0,
Figure 3 The time course of the effects of intrathecal administration
arrow) of saline (10 μl), neamine or 2-deoxystreptamine on the
respectively) and have a significant treatment × time inter-
action (F27,180 = 116.9 and 545.8, respectively) (P < 0.0001
in all cases).
The estimated ED50 of neamine and 2-deoxystreptamine
were 0.28 mg (confidence limits: 0.12 and 0.63) and
0.79 mg (confidence limits: 0.46 and 1.34), respectively.
(
mechanical threshold (in grams) of rats before and 2 h after a surgical
incision of a hind paw. Points are the mean (± SD) of six rats/curve.
Doses of neamine (a) and 2-deoxystreptamine (b) are given in
mg/10 μl. P < 0 .05 vs saline-treated rats (*) or vs all the remaining
groups (#).
1
700
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704

Wiliam A. Prado et al.
Neomycin derivatives and antinociception
(
a)
50
40
5
% DMSO (10 µl)
4
3
2
1
0
0
0
0
0
tetra-azidoneamine (3 mg/10 µl)
#
*
#
*
#
*
#
*
#
*
#
*
#
*
#
*
3
0
0
*
*
*
*
*
*
*
*
2
10
0
saline
.2 mg
0.4 mg
.8 mg
0
0
–122 –2
0
10 20 30 40 50 60 70 80
Time (min)
a
b
0
20
40
60
80
#
(
b)
50
40
30
20
10
0
Figure 5 The time course of the effects of intrathecal administration
(arrow) of DMSO (10 μl), or tetra-azidoneamine (3 mg/10 μl) on the
mechanical threshold (in grams) of rats before and 2 h after a surgi-
cal incision of a hind paw. Points are the mean (± SD) of six rats/
curve.
#
#
#
#
#
*
#
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
surgical incision of a hind paw. Neither this neamine
derivative nor the vehicle produced a significant change in
the post-incision mechanical allodynia (Figure 5).
saline
.5 mg
1.0 mg
2.0 mg
0
Antagonism of the effects of intrathecal
injection of neamine and
-deoxystreptamine on incision allodynia by
a
b
0
20
40
60
80
Time (min)
2
intrathecal calcium chloride
Figure 4 The time course of the effects of intrathecal administration
arrow) of saline (10 μl), neamine or 2-deoxystreptamine on the
mechanical threshold (in grams) of rats before (a) and 8–10 days after
b) the chronic ligature of the sciatic nerve. Points are the mean (± SD)
of six rats/curve. Doses of neamine (a) and 2-deoxystreptamine (b) are
given in mg/10 μl. P < 0.05 vs saline-treated rats (*) or vs all the
remaining groups (#).
(
All rats exhibited a significant reduction of the mechanical
threshold of the incised hind paw measured 2 h after
surgery. The intrathecal injection of saline did not produce
significant changes in the incision allodynia. Neamine
(0.8 μg, Figure 6a) and 2-deoxystreptamine (2.0 μg,
Figure 6b) produced a significant increase in the mechanical
threshold compared with the effect of saline. A further
intrathecal injection of calcium chloride (4 and 8 μg)
(
The regression lines obtained for neamine and 2-
deoxystreptamine did not deviate significantly from paral-
lelism (slope ratio = 1.28 < slope ratio factor = 4.0).
Neamine was 2.82 (confidence limits: 1.07 and 7.44) times
more potent than 2-deoxystreptamine. No change in the
mechanical threshold of the non-operated hind paw of
saline-, neamine- or 2-deoxystreptamine-treated rats was
detected throughout the period of experimentation (data
not shown).
3
9 min after the derivatives produced a transitory and dose-
dependent reduction of the antiallodynic effect of neamine
and 2-deoxystreptamine. Calcium chloride (8 μg) injected
39 min after saline did not produce effects different from
baseline values. The curves in Figure 6a and 6b were signifi-
cantly different in terms of treatment (F2,15 = 702.8 and
9
respectively) and have a significant treatment × time inter-
action (F14,105 = 114.8 and 495.3, respectively) (P < 0.0001 in
all cases).
10.0, respectively) and time (F7,105 = 458.0 and 1755.0,
The effects of intrathecal injection of
tetra-azidoneamine on the incision
allodynia
Discussion
The presented results demonstrate that intraspinal admin-
istration of the neomycin derivatives neamine and
2-deoxystreptamine in rats has an antinociceptive effect in
Tetra-azidoneamine (3 mg/10 μl) or 5% DMSO (10 μl) was
injected intrathecally in groups of six rats each 2 h after
©
2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704
1701

Neomycin derivatives and antinociception
Wiliam A. Prado et al
(
a)
due to an overall antinociceptive effect produced by the
derivatives.
Aminoglycoside antibiotics are antinociceptive following
intracerebroventricular administration in mice (20–80 μg;
5
4
3
2
1
0
0
0
0
0
0
*
*
*
*
*
*
4
0–160 μg and 80–320 μg, for neomycin, gentamicin and
[2]
kanamycin, respectively) and rat (ED50 = 50.12 μg, for
gentamicin) tail-flick test. Neomycin is antinociceptive in
[3]
the rat tail-flick (ED50 = 9.22 μg; confidence limits 6.98 and
saline
calcium 8 µg
calcium 4 µg
calcium 8 µg
1
2.17) and incision pain tests (ED50 = 83.17 μg; confidence
neamine 0.8 µg
neamine 0.8 µg
[17]
limits 51.6 and 133.9). Neomycin applied in situ is also
antinociceptive against formalin- or carrageenan-induced
inflammatory pain rats, but much higher doses of the
1
2
[26]
antibiotic were necessary (34 and 58 mg, respectively).
–
122 –2
0
10 20 30 40 50 60 70 80
The intraspinal administration of morphine was also
(
b)
50
40
30
20
10
0
antinociceptive in the rat tail-flick test (ED50 = 0.29 μg; con-
[22]
fidence limits 0.19 and/0.47). Thus, the neomycin deriva-
tives seem to be weaker than neomycin and morphine
regarding antinociception. Neamine was more potent than
*
*
*
*
*
2-deoxystreptamine in the incision and neuropathic pain
*
models, but these derivatives had similar potency in the tail-
flick test.
*
Tetra-azidoneamine, a neamine derivative in which free
amino groups are replaced with azido groups, did not
change the mechanical allodynia displayed by rats 2 h after
the surgical incision of a hind paw. We conclude that the
presence of free amino groups in the structure of neomycin
and its derivatives is fundamental for its antiallodynic
effect. However, the aminoglycoside unit seems to be
very important for the antinociceptive potency of these
compounds.
saline
calcium 8 µg
calcium 4 µg
calcium 8 µg
2
2
-deoxy 2.0 µg
-deoxy 2.0 µg
1
2
–
122 –2
0
10 20 30 40 50 60 70 80
Time (min)
Figure 6 Reduction of the antiallodynic effect of neamine (a) and
-deoxystreptamine (b) by intrathecal administration of calcium chlo-
2
We have also shown that the reduction of incisional
allodynia by intrathecal neamine and 2-deoxystreptamine
was antagonized by further intrathecal administration of
calcium chloride in a transitory and dose-dependent
manner. Intrathecal gentamicin-induced antinociception in
the rat tail-flick test is also dose-dependently antagonized
ride in rats. Thresholds (in grams) to mechanical stimulation were
measured before and 2 h after a surgical incision of a hind paw.
Saline, neamine or 2-deoxystreptamine (2-deoxy) was injected (arrow
1
) 39 min before calcium chloride (arrow 2). Points are the mean
(
± SD) of six rats/curve. *P < 0.05 vs threshold measured at time
t = 30 min.
[3]
by intrathecal administration of calcium chloride. We
conclude that the effect of neamine and 2-deoxystreptamine
may be due to a calcium antagonist property, a mechanism
known to alter the effects of aminoglycoside antibiotics on
the tail-flick test and models of incisional and neuropathic
pain. The effect was not accompanied by changes of motor
coordination in all cases. The approximate ED50 (confi-
dence limits in parenthesis) were 1.35 mg (0.61 and 2.95),
[6]
various peripheral tissues.
The earlier demonstration that intrathecal aminoglyco-
[3]
side antibiotics produce analgesia in rodents is an interest-
ing finding, suggesting the eventual use of these drugs
in clinical pain management. Neamine and 2-
deoxystreptamine seem to be less potent than neomycin
and morphine. However, the present data showing that neo-
mycin derivatives that do (neamine) or do not
(2-deoxystreptamine) exhibit ototoxicity retain the
antinociceptive property of the entire molecule may
promote these derivatives as alternative targets for further
studies on the management of clinical post-surgical and
neuropathic pain.
0.20 mg (0.14 and 0.27) and 0.28 mg (0.12 and 0.63),
respectively, and for 2-deoxystreptamine they were 1.05 mg
(
(
0.68 and 1.60), 0.78 mg (0.776 and 0.783) and 0.79 mg
0.46 and 1.34), respectively. The effects of both derivatives
lasted less than 60 min in the tail-flick test but reduced
incisional and neuropathic pain for at least 80 min. The
mechanical threshold measured in the non-incised and
non-operated hind paw after intrathecal injection did not
change throughout the period of observation. Therefore,
the reduction of incisional or neuropathic pain is not
1
702
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704

Wiliam A. Prado et al.
Neomycin derivatives and antinociception
Conclusions
Declarations
Our results demonstrate that neamine and 2-
deoxystreptamine, but not tetra-azidoneamine, have an
antinociceptive property in different models of pain
in rats. Thus, the presence of free amino groups
in the structure of neomycin derivatives is fundamental to
their antiallodynic effect. The reduction of incisional
mechanical allodynia by intrathecal neamine and
Conflict of interest
The Author(s) declare(s) that they have no conflicts of
interest to disclose.
Funding
This study was supported by FAPESP. ACR was the recipient
of an FAPESP fellowship.
2
-deoxystreptamine was antagonized by intrathecal admin-
istration of calcium chloride in a transitory and dose-
dependent manner. Therefore, the antiallodynic effect of
the derivatives may be due to their action as calcium
antagonists.
Acknowledgement
The authors greatly appreciate the technical assistance of
M.A. Carvalho, P.R. Castania and I.A. Daniel.
and control. Antimicrob Agents
Chemother 2003; 47: 1010–1016.
9. Corrado AP et al. Antagonismo
competitivo
animal models of pain in drug
discovery. Neurosci Lett 2013; 557:
65–72.
References
1
2
3
. Prado WA et al. Antinociception
induced by intraperitoneal injection
of gentamicin in rats and mice. Pain
y
non competitivo 16. Barrett JE. The pain of pain: chal-
calcio-antibioticos. In: Mejia RH,
lenges of animal behavior models. Eur
J Pharmacol 2015; 753: 183–190.
1990; 41: 365–371.
Moguilevsky JA, eds. Recientes
. Ocaña M, Baeyens JM. Analgesic
effects of centrally administered
Adelantos en Biologia. Buenos Aires: 17. Prado WA, Machado Filho EB.
Bona SCA, 1971: 385–391.
Antinociceptive potency of amino-
glycoside antibiotics and magnesium
aminoglycoside antibiotics in mice. 10. Pittinger CB, Adamson R. Antibiotic
Neurosci Lett 1991; 126: 61–70.
blockade of neuromuscular function.
Ann Rev Pharmacol 1972; 12: 169–
184.
chloride: a comparative study on
. Rego EM et al. Antinociception
induced by intracerebroventricular or
models of phasic and incisional pain
in rats. Braz J Med Biol Res 2002; 35:
395–403.
intrathecal administration of gen- 11. Vital Brazil O et al. Neuromuscular
tamicin in rats. Gen Pharmacol 1992;
3: 481–485.
blockade produced by streptomycin 18. Zimmermann M. Ethical guidelines
2
and some of its degradation products.
for investigations of experimental pain
in conscious animals. Pain 1983; 16:
109–110.
4
. Wagner JA et al. Calcium antagonist
receptors. Clinical and physiological
relevance. Ann NY Acad Sci 1988; 522:
In: Bovet D et al., eds. Curare and
Curare-Like
Agents.
Amsterdam:
Elsevier, 1957: 415–421.
19. Brennan TJ et al. Characterization of a
rat model of incisional pain. Pain
1996; 64: 493–501.
116–133.
12. Prado WA et al. Competitive antago-
5
. Yamada
K
et al.
Neurophar-
of
nism between calcium and antibiotics
macological
characterization
at the neuromuscular junction. Arch 20. Bennett GJ, Xie YK. A peripheral
voltage-sensitive calcium channels:
possible existence of neomycin-
Int Pharmacodyn Ther 1978; 231: 297–
mononeuropathy in rat that produces
disorders of pain sensation like those
seen in man. Pain 1988; 33: 87–107.
307.
sensitive, w-conotoxin GVIA- and 13. Silva JG et al. Aminoglycoside antibi-
dihydropyridine-resistant calcium
channels in the rat brain. J Pharmacol
993; 63: 423–432.
otics derivatives: preparation and 21. Mestre C et al. A method to perform
evaluation of toxicity on cochlea and
vestibular tissues and antimicrobial
activity. Bioorg Medic Chem 2007; 15:
3624–3634.
direct transcutaneous intrathecal
injection in rats. J Pharmacol Toxicol
Methods 1994; 32: 197–200.
1
6
. Prado WA. Involvement of calcium in
pain and antinociception. Braz J Med
Biol Res 2001; 34: 449–461.
22. Prado WA. Antinociceptive potency of
intrathecal morphine in the rat tail
flick test: a comparative study using
acute lumbar catheter in rats with or
without a chronic atlanto-occipital
catheter. J Neurosci Methods 2003; 129:
33–39.
14. Rauck RL et al. Comparison of the
efficacy of epidural morphine given by
intermittent injection or continuous
infusion for the management of post-
operative pain. Reg Anesth 1994; 19:
316–324.
7
. Dulon D et al. Comparative uptake of
gentamicin, netilmicin, and amikacin
in the guinea pig cochlea and vesti-
bule. Antimicrob Agents Chemother
1986; 30: 96–100.
8. Rougier
F
et al. Aminoglycoside 15. Whiteside GT et al. An industry per- 23. Litchfield JT Jr, Wilcoxon F. A simpli-
nephrotoxicity: modeling, simulation,
spective on the role and utility of
fied method of evaluating dose-effect
©
2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704
1703

Neomycin derivatives and antinociception
Wiliam A. Prado et al.
experiments. J Pharmacol Exp Ther 25. Botton RE, Coxon B. Nitrogen-15
produces analgesia in inflammatory
pain models in the rat. Inflammation
2015; 38: 649–657.
1949; 96: 99–113.
nuclear magnetic resonance spectros-
copy of neomycin B and related
aminoglycosides. J Am Chem Soc 1983;
105: 1021–1028.
2
4. Paumgartten FJR et al. Comparison of
five methods for the determination of
lethal dose in acute toxicity studies.
Brazilian J Med Biol Res 1989; 22: 987– 26. Mercado F et al. Inhibition of periph-
91. eral nociceptors by aminoglycosides
9
1
704
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1696–1704