Clinical Science (1994)
86. 133-139 (Printed in Great Britain)
I33
Secretion of ions and pharmacological responsiveness in
the mouse paw sweat gland
K. SATO, S. CAVALLIN, K. T. SATO and F. SATO
Marshall Dermatology Research Laboratories, Deportment of Dermatology, University of Iowa
College of Medicine. Iowa City, Iowa. U.S.A.
(Received 26 Aprilll2 August 1993; accepted 31 August 1993)
1. Some of the basic functional features of the mouse
paw eccrine sweat gland were delineated to allow
comparison with those of transgenic mice in the
future.
2. The mouse sweat secretory coil responds to methacholine, elaborating a K+-rich ( > 120mmol/l),
Na+-poor ( <70mmol/l) primary fluid as does the
rat paw sweat gland, as previously reported. The
methacholine-induced sweat rate increases with age in
parallel with the growth of the sweat gland over the
first 6 weeks of life.
3. The sweating response to cyclic AMP-elevating
agents, such as isoprenaline or forskolin, is as much
as 40% of the methacholine-induced sweat rate at 1
week of age, but falls to 10% by 6 weeks of age
despite the fact that the agonist-induced tissue
accumulation of cyclic AMP expressed on a per pg
of protein basis triples with age over the same period.
4. A marked K + outflux was also noted in response
to methacholine and a small K + outflux was seen in
response to cyclic AMP-elevating agonists in superfused adult mouse secretory coils in v i m .
5. Since sweat secretion is usually associated with
activation of either K + channels or CI- channels or
both, and since the sweating occurred in response to
cyclic AMPelevating agonists, we speculate that the
cyclic AMP-activated CI- channels (the mouse
version of the cystic fibrosis transmembrane conductance regulator) may also Occur in the mouse
sweat gland, but that the degree of their expression
may be influenced by the age of the mice.
INTRODUCTION
Although the mouse has reigned as the primary
experimental mammal, the importance of the mouse
as a model system has heightened with the advent
of transgenic mouse technology [l]. In order to
evaluate the phenotypic expression of introduced
genes in transgenic mice, knowledge of the normal
function of various tissues and organs in control
animals will be essential. The sweat gland is one of
the target organs in cystic fibrosis. Transgenic mice
for cystic fibrosis which have been generated in
several laboratories have shown some of the symptoms and transport defects typical of cystic fibrosis
C2-41. The study of the sweat gland has been
neglected in these and other transgenic mouse
models simply because the normal function of the
mouse sweat gland has not been well studied. It
would also be of interest to know whether the
mouse sweat gland is similar to the rat sweat gland,
a unique K+-secreting exocrine gland [S]. If the
mouse sweat secretory epithelium is found to be
similar to the rat counterpart, these rodent glands
would serve as a unique epithelial model system for
studying the mechanisms of K secretion. Rodent
sweat glands have also been used as a model system
for studies on innervation, denervation and reinnervation of the glands and on the postnatal development of cholinergic and adrenergic innervation [6,
71. For example, in the rat sweat gland, dynamic
postnatal changes occur in cholinergic and adrenergic fibres of periglandular nerves [6]. Although a
similar study has not been performed in the mouse,
the change in pharmacological responsiveness of the
sweat gland during the postnatal period provides
further insight into the relationship between innervation and pharmacological responsiveness of
the gland. The present study is intended to clarify
some of the basic functional characteristics of the
normal mouse paw sweat gland, namely electrolyte
composition of sweat induced in vioo and in oitro,
pharmacological responsiveness of sweat secretion,
especially that of cyclic AMP-elevating agents, and
the change in cholinergic and adrenergic responsiveness during the postnatal period.
+
METHODS
Collection of sweat induced in vivo
Balb/cAnNHsd mice (Harlan Sprague Dawley,
Inc., Indianapolis, IN, U.S.A.) were used throughout
the study. Three 6-8-week-old male mice were
Key word% cyclic AMP, cptic fibrosis. cystic fibrasis transmembrane conductance regulator. iroprendine. sweu gland.
Abbmviatlonr: CFTR, cptic fibrosis transmembrane conductance regulator; FK. forskolin; IBMX, isobutylmethylxanthine; ISO. isoprendine; KRB. Krebs-Ringer bicarbonate
buffer (for composition, see the text); MCh, methacholine; TH. theophylline.
C o r v n d c m c : Dr Kenzo k o , Marshall Dermatology, Research Laboratories. Department d Dermatology, University d Iowa College d Medicine, 271 Medial laboratories,
Iowa City, IA 52242-1 181, U.S.A.
K.
134
Sat0
et al.
Fig. I. Illustrative examples of isolated mouse paw sweat glands in different age groups. A, I-week-old (Iwo); 6. 3-week-old
(3wo); C, bweekold (6wo). Abbreviations: S, secretoly coil; D. duct. Note the rapid growth of the gland during the first 6 weeks of
life. Sweat glands from the same animal were apparently uniform in size. The mean body weights were 5.1 g for I-weekold mice,
18.1 g for 3-week-old mice and 3O.Og for b w e e k o l d mice.
anaesthetized with Inactin (Promonta, Hamburg,
Germany, 80 mg/kg body weight) and tracheotomized. 0, (100%) was continuously given at the
orifice of the tracheal catheter. The hind feet were
washed thoroughly with de-ionized water, air-dried
and placed in a parafin oil-filled polyethylene petri
dish with the plantar surface upward. Sweat secretion was stimulated by subcutaneous injection
(10mg/kg body weight) of pilocarpine in the cervical
region [S]. Sweat droplets on the paw tubercles
were counted and collected under oil with an oilfilled glass capillary tube every 10min. The volume
of the collected sweat sample was determined with
an oil-filled constant-bore calibration glass capillary
and
the
sweat
rate
was
expressed
as
nl-' min- ' gland. Thus, at no time was the sweat
sample exposed to air.
Sweat induction in vitro
The method of sweat induction from isolated
sweat glands has been described previously [ 5 , 8, 91.
Briefly, single sweat glands were isolated under a
stereomicroscope from excised paw pads (Fig. 1)
and a short duct segment was removed from the
secretory coil and discarded. The open end of the
secretory coil was held into a siliconized constriction pipette by suction and the pipette was filled
with water-saturated parafin oil. The tissue-glass
junction was sealed with Silgard 184 (Dow Corning,
Midland, MI, U.S.A.). The incubation medium was
Krebs-Ringer bicarbonate buffer (KRB, containing
in mmol/l: NaCI, 145; KCl, 5; NaHCO,, 25;
NaH,PO,,
1.2; MgSO,, 1.2; CaCI,, 1.0; with
5.5 mmol/l glucose) also containing 30% (v/v) fresh
rat serum which had been heat-treated at 56°C for
30 min. This buffer was continuously oxygenated
with a gas mixture of 5% CO, and 95% 0,. Sweat
secretion was induced by the addition of 10-6mol/l
methacholine ( MCh, acetyl-fi-methylcholine; Sigma
Chemical Co., St Louis, MO, U.S.A.) to the
medium. Because of the extreme fragility of the
sweat gland and the small glandular size (and thus
the difficulty of isolating the gland without damage),
only three of more than 15 cannulated glands
showed stable sweat secretion in oitro in response to
! pmol/l MCh. Four other glands responded briefly
but stopped secreting, presumably due to partial
tissue damage during isolation of the glands.
Damage to the gland during sweat induction (e.g.
touching with forceps or turbulent water movement
during medium exchange) also easily resulted in
cessation of sweat secretion. Sweat samples were
collected every 10min with an oil-filled inner
sampling pipette mounted on the pipette holder and
calibrated [8]. After each sweat collection the
sampling pipette was sealed with mineral oil and
stored at 4°C in a hydration chamber to minimize
evaporation of the sweat samples. That the fluid
entering into the pipette does not reflect the leakage
of the incubation medium into the pipette has been
rigorously tested previously [ 5 , 8-10]. Na' and K +
concentrations in the sweat samples were measured
with a picomolar helium glow spectrophotometer
(Aminco, Silver Spring, MD, U.S.A.) as described
previously [ 5 , 8-10].
Mouse sweat gland
DvN,
+
To moisture analyser
sed
Hind paw
Fig. 2. Evaporation chamber for measuring the sweat rate from paw
pads. Dry N, was continuously introduced into the chamber at 50ml/min
and the moisture in the outflow N, was continuously monitored by a Meeco
moisture analyser. The inner diameter of the chamber measured Smm and
the chamber was completely sealed with silastic glue (Elasticon: Kerr
Manufacturing Co.. Romulus. Mi, U.S.A.). When 0.0Sml of MCh was
injected intradermally into the centre of the paw, sweating was noted only
on the tubercles, but not on the fingertips as visualized by the method of
Hayashi [El. Nevertheless, the fingertips were placed outside the chamber,
or when not possible, totally embedded in the silastic glue. Thus the
registered moisture reflects only sweating from the glands in the tubercles.
Effect of age on cholinergic and hdrenergic sweat
secretion
Although the sweat volume after cholinergic
stimulation in adult mice was sufficient to allow for
collection of sweat droplets from the skin surface
under mineral oil, this was not possible for younger
mice and during 8-adrenergic stimulation because of
the much lower sweat rates. Thus a microevaporation chamber was constructed, as shown in
Fig. 2, to continuously monitor the sweat rate.
Briefly, dry N, gas was introduced into the chamber
(5 mm internal diameter, about 8 mm long) at 50 ml/
min and the moisture in the outflow N, gas was
continuously monitored with an electrolytic water
vapour analyser (Meeco, Warrington, PA, U.S.A.),
essentially as described by Van Gasselt and Vierhout [111 with modifications [123.
Superfusion of isolated secretory coils
Since the isoprenaline (IS0)-induced sweat rate
was extremely small, we further tested the specificity
of the small I S 0 responsiveness using another sensitive assay system, i.e. K + outflux from isolated adult
mouse sweat glands using the superfusion technique
reported previously [131. The superfusion chamber,
which was constructed from a 0.8mm outer diameter glass capillary on a microforge, was about
2 mm long and had a 0.4 mm internal diameter with
a constriction at its orifice (to achieve a tight seal
with a superfusion tubing) and a flared distal portion (to accommodate ion-sensitive electrodes). The
entire unit was secured to the floor of the incubation chamber with epoxy glue. About 50 isolated
secretory coils placed inside the chamber were
secured to the thorns created on a polyethylene rod
(about 0.15 mm outer diameter). Polystyrene perfusion pipettes containing different solutions were
I35
sealed tightly to the orifice of the superfusion
chamber one at a time (also see Fig. 7a). The sweat
glands were stimulated with the different solutions
by quickly changing the perfusion pipette to the
appropriate one. Perfusion speed was shown to be
constant (usually 0.5 or 1 pl/min) and identical in all
the perfusion channels. The superfusion chamber
was submerged in the bath, kept at 37"C, and the
bath was continuously refreshed by infusing fresh
Ringer's solution (KRB without added serum). The
K +-sensitive electrode was constructed by fusing a
small porous granule of alumina (Sigma Chemical
Co.), about 50pm in diameter, to the tip of the
electrode using a microforge [8]. After siliconization, the electrode tip was filled to a point 500pm
from the tip with a K+-sensitive resin (WPI, New
Haven CT, U.S.A.). The electrode was then filled
with l00mmol/l KCI and connected to a Keithley
amplifier (model 602 or 604) via a small Calomel
electrode.
Determination of tissue cyclic AMP level
Since the relative ISO-induced sweat rate (ISOinduced sweat rate/MCh-induced sweat rate, which
increased with the age and thus the growth of the
mice) decreased with age during the first 6 weeks of
life, we examined whether ISO-stimulated tissue
cyclic AMP levels (expressed on a per pg of protein
basis) also decreased with age. The method of cyclic
AMP determination was the same as that described
previously [14, 151. Briefly, 10-20 secretory coils
(depending on the size of the glands) were incubated
in each tube in KRB at 37°C for 5min with or
without drugs. The sweat glands were then picked
up as a mass with a pair of forceps, snap-frozen in
liquid N, and transferred to 50p1 of 0.1 mol/l HCl.
The glands were boiled for 5min to extract cyclic
AMP, freeze-dried to remove HCI and reconstituted
for the determination of cyclic AMP using a radioimmunoassay kit (New England Nuclear, Boston,
MA, U.S.A.). The adequacy of the method has been
documented repeatedly [14, 151.
RESULTS
Like the rat paw sweat gland [5], the mouse
gland secretes sweat with high K + and low N a +
concentrations on a persistent basis in response to
systemic administration of pilocarpine in uivo (Fig.
3). K + concentration was constant at around 150160mmol/l independent of the sweat rate over the
sweat collection period of 30min. In contrast, N a +
concentration was about 60-79 mmol/l. As shown in
Fig. 4, the primary sweat collected from the secretory portion in oitro also showed a high K + concentration (at around 120mmol/l) and a low N a +
concentration (at 50-70 mmol/l) over the sweat collection period of 40min in the presence of MCh.
Four other glands that briefly responded to MCh
contained 50-78 mmol/l K + and 80-1 10mmol/l
K. Sat0
I36
q , ,
2 0.0
.
10
0
,
.
,
,
30
20
Time (min)
et
al.
,
40
Fig. 3. K + ( 0 )and N + (0)
concentrations in mouse paw sweat in
rim (sweat collected f r o m the skin surface). (a) and sweat rate (6).
At time zero, pilmarpine was injected in the cervical area and sweat was
collected after IOmin. The average number of droplets was 17 on each
tubercle. The sweat rate was expressed on a per gland basis. Since the
sweat droplets were collected in pardin oil, they were not in contact with
the skin and therefore not contaminated by epidermal ingredients. Values
are means fSEM (n = 3 mice studied).
4min
Fig. 5. Illustrative traces for sweating in rho in mponr t o lntradermal local injection (into the centre of the paw) of M C h o r I S 0
in different age groups as determined w i t h a M m o moisture
analyser. (a) I-weekold (5-7dayold mice included in this group) mouse;
(b) 2-weekold ( I 2 4 4 d a y o l d mice) mouse; (c) &weekold or older mouse
(note that mice reach adult size at 6 4 weeks old according to the data
provided by Harlan Spngue Dawley. Inc.; most of the mice used in this y e
group were from 6 to 10 weeks d ye). At the small arrows, IOpnol/l
MCh was injected intradermally. At the large empty arrows, 0.OSml of a
mixture containing 0.5 mmol/l ISO. IOmmol/l TH (5 mmol/l aminophylline)
and IOpmol/l atropine was injected (muming that the final periglmduhr
drug concentration would be diluted about l M o l d by tissue water and
capillary blood flow). Injection of Ringer's solution alone failed to register a
detectable increase in moisture level. Propranolol (I mmol/l) added t o the
IS0 mixture completely blocked ISO-induced sweat secretion (resulu not
shown).
O J
z
J
,
,
,
,
10
20
30
40
0.0
Time (min)
Fig.4. K + ( 0 )and Na+ (0)
concentrations in primary fluid
coll&ed from the isolated secretory p o r t i o n o f the mouse paw
sweat gland in vitm ( a ) and sweat rate (b). Values are means fSEM
(n = 3 mice studied).
N a + (these glands were not included in the tabulation because they most likely sustained partial
glandular damage during dissection). Thus the data
are consistent with the notion that mouse and rat
sweat glands represent exocrine epithelia capable of
secreting a predominantly K +-rich primary fluid.
We have not addressed whether the short duct
segment (see Fig. 1) accounts for the additional
30mmol/l K + secretion (i.e. the difference in the K +
concentration between the primary fluid collected in
uitro and that of the final skin surface sweat). As in
the rat sweat gland [ S ] , none of the mouse sweat
glands responded to I S 0 (50pmol/l) in oitro ( n = 3 ,
results not shown).
Fig. 5 shows illustrative traces of sweat secretion
in different age groups of mice in response to
cholinergic and P-adrenergic stimulation in oioo. The
MCh-induced sweat rate in the 1-week-old mouse
was rather modest. However, MCh-induced sweating significantly increased in the 2-week-old mouse
(Fig. 5b and Fig. 6). By 6 weeks of age, the maximal
MCh-induced sweat rate had increased about 5-fold
(Figs. 5 and 6). Note that the glandular size also
increased many fold (see Fig. 1) during the same
period. Body weight also increased linearly until
adult size was reached at 6 7 weeks of age (which is
about 4-5 times that of the I-week-old mouse
according to the data provided by Harlan Sprague
Dawley, Inc.). Thus the observed increase in the
MCh-induced sweat rate with age reflects the en-
Mouse sweat Eland
I37
7
-
4
0
I
2
3
4
5
6
7
I
+
Y
W
MCh 3 x IO-'mol/l
Fig. 7. K + outflux from superfused mouse sweat secretory coils in
vitro. (a) Schematic illustration of a superfusion chamber. Abbreviations:
PP, perfusion pipette made of polypropylene tubc; R, rod with thorns; SC.
secretory coils; Su. superfusion chamber; T, epoxy table to secure Su to the
floor; K. potassiumsensitive electrode; V, amplifier. The bathing medium
outside the superfusion chamber, to which the superfurate from all the
i
2
3
4
Age (weeks)
5
6
1
Fig. 6. Summary of agerelated changes in the MCh- and ISO-induced sweating response. Maximal (peak) levels of sweating responses are
shown in (0). (b) shows the relative (to MCh) ISO- and ISO- plus FKinduced
sweat rates. The number at each symbol is the number of mice studied. FK
was added to IS0 to maximally enhance tissue levels of cyclic AMP (see also
Fig. 8). 0 ,MCh; 0,IS0 plus TH; A,IS0 plus FK plus TH.
largement of the sweat gland over the 6 weeks after
birth. In contrast, agents that increase the tissue
cyclic AMP levels, such as I S 0 plus theophylline
(TH) or I S 0 plus forskolin (FK) plus TH, stimulated sweat secretion in 1-week-old mice to an
extent that was about 40% that of the MCh-induced
sweat rate. However, the ISO-induced sweat rate
failed to increase along with the increase in the
glandular size over the next 5 weeks (Fig. 6 4 . Thus
the relative ISO- or I S 0 plus FK-induced sweat
rate [I21 declined as the mouse grew to adulthood
(Fig. 6b). In order to examine the specificity of the
observed small, although reproducible, sweat secretory responses after stimulation with cyclic AMPelevating agents such as I S 0 or I S 0 plus FK, we
then studied whether these agents also stimulated
K + outflux in a superfusion system in uitro used
earlier for the Rhesus monkey sweat gland [13]. As
shown in Fig. 7, K + outflux was indeed stimulated
by cyclic AMP-elevating agents, although it was to
a much smaller extent than during MCh stimulation. Thus, although the mechanism of K + outflux
in the mouse sweat gland is unknown (for the
mechanisms of K + outflux in Rhesus monkey sweat
glands, see [13, 16-18], the data serve to confirm
the presence of cyclic AMP responsiveness in the
mouse sweat gland. Lastly, we addressed whether
the decrease in the relative 1SO-stimulated sweat
rate from 1 week to 6 weeks of age was due to
pipettes was allowed to drain, was continuously refreshed by adding fresh
KRB. The superfusion rate in different pipettes was identical and perfusates
could be switched in a b u t a second. Calibration of the K+-electrode was
performed by perfusing U)mmol/l K + in Ringer's solution using the same
perfusion chamber, but empty. See [I31 for further details of the method*
logy. (b) Illustrative trace showing K + oufflux from superfused mouse
secretory coils as determined with a K'-sensitive electrode. Both cyclic
AMP-elevating drug mixtures contained 0.1 pmol/l atropine. Note that
during MCh stimulation K + outflux was followed by K + uptake. After
cessation of MCh stimulation. a small rebound K + outflux (arrow) is seen.
Four other traces had similar results (not shown).
decreased tissue cyclic AMP accumulation during
stimulation. Contrary to our expectation, tissue
levels of cyclic AMP (expressed on a per pg of
protein basis) increased with age after stimulation
by all of the cyclic AMP-elevating agents, namely
isobutylmethylxanthine (IBMX) alone, IBMX plus
ISO, IBMX plus FK and IBMX plus I S 0 plus FK
(Fig. 8). In particular, cyclic AMP accumulation in
response to IBMX plus I S 0 plus F K was highest in
the mice 6 weeks of age of older, yet the sweating
response to TH plus I S 0 plus FK was the lowest,
indicating that the tissue cyclic AMP level is not the
mechanism for the decrease in cyclic AMP-induced
sweating during the growth of the animal.
DISCUSSION
The advent of transgenic mouse technology has
prompted us to study some of the normal characteristics of the mouse sweat gland. We have observed that the mouse sweat secretory coil elaborates a K +-rich (concentration > 120 mmol/l) and
Na+-poor (concentration < 70 mmol/l) primary
fluid, as does the rat paw sweat gland [S]. The
ultrastructural similarity of the mouse sweat gland
to that of the rat sweat gland is already known
[191. Interestingly, the secretory coils of these
K. Sat0 et al.
I38
mmol/l Na', see Fig. 4), the K + and CI channels
may be located mainly in the luminal membrane
(note that in primate and possibly human secretory
cells, the K channels are present in the basolateral
membrane and CI- channels in the luminal membrane which is consistent with the Na+-K+-2CIco-transport model) [1 8, 211. The modest sweating
response in uiuo and the small K + outflux from the
superfused mouse secretory coils in uitro also suggest that K + channels, CI- channels or both are
regulated by cyclic AMP in the mouse sweat gland.
In light of the recent cloning of the cystic fibrosis
gene [cystic fibrosis transmembrane conductance
regulator (CFTR)] and the subsequent demonstration that CFTR is the same as cyclic AMPregulated C 1 ~channels [22, 231, our observation
supports the notion that the mouse sweat secretory
cell is also equipped with (the mouse version of)
CFTR [24]. The present study thus provides a basis
for future molecular biological and electrophysiological studies on the regulation of CFTR in normal
as well as transgenic mice.
+
C
lBMX
lBMX +Is0
IBMX 3- FK IBMX
+ IS0 + FK
Fig. 8. Tissue levels of cyclic A M P after stimulation w i t h different
combinations of cyclic AMP-elevating agents. Atropine (I p o l / l ) was
added to all the tubes. Sweat glands were incubated at 37 C for Smin
Note that the cyclic AMP level is expressed on a per p g of protein basis (as
determined by the method of Lowry et al. [26]) and the increase in the
glandular size with age does not reflect an increase in tissue cyclic AMP
levels. Mouse ages:
2 weeks old; 0.
3 weeks old: g.2 6 weeks old.
Drug concentrations. IBMX. 0.5mmol/l: ISO. SO~tmol/l: FK. IOpnol/l.
Values are means iSEM (n = 6).
m.
rodent sweat glands consist of only the clear and
myoepithelial cells [19] and lack the dark-cell type.
It remains unknown whether the lack of dark cells
is related to the unique K+-secreting characteristic
of the rodent sweat gland. The present study also
disclosed that the mouse sweat gland is predominantly cholinergic and that its responsiveness
to cyclic AMP-elevating agents, such as I S 0 or FK,
is modest and declines with age. For example. cyclic
AMP-induced sweating is as much as 407; of the
MCh-induced sweat rate in 1-week-old mice, but
falls to less than 10% by 6 weeks of age. This is in
contrast to the cholinergic sweat rate, which increases in parallel with the glandular enlargement
during the growth of the animal. Nevertheless, the
decline of responsiveness to cyclic AMP-elevating
agents with age is not due to decreased tissue
accumulation of cyclic AMP. Rather, the stimulated
tissue level of cyclic AMP expressed on a per pg of
protein basis almost tripled from 1 week to > 6
weeks of age.
The mechanism of K ' secretion in the rodent
sweat gland is not well understood. In the rat sweat
secretory coil, a marked depolarization of the membrane potential was noted in response to MCh
stimulation in a dose-dependent manner [20]. A
marked K' outflow was also noted in response to
MCh in the mouse secretory coil (Fig. 7). We
therefore speculate that cholinergic stimulation in
mouse sweat glands is associated with a marked
increase in the K + and CI- channels as in Rhesus
monkey secretory coils [ 131. Since the primary fluid
(i.e. the secreted product) contains a cytoplasm-like
ionic composition (i.e. 120mmol/l K + and 60
ACK N0W LEDGMENTS
This study was supported in part by NIH grants
DK 27857 and AR 25339 and a Cystic Fibrosis
Foundation grant G124. Neal Kane assisted in the
preparation of the manuscript.
REFERENCES
I Hogan B. Costantini F. Lacy E. Manipulating the mouse embryo: a laboratory
manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1986.
1 SnOUWaeK JN. Brigman KK, Latour AM, et al. An animal model for cystic
fibrosis made by gene targeting. Science (Washington DC) 1992 257: 1083-8.
3 Dorin JR, Dickinson P. Alton EWFW. et al. Cystic fibrosis in the mouse by
targeted insertional mutagenesis. Nature (London) 1992 359: 21 1-15,
4 Clark LL. Grubb BR. Gabriel SE, Smithies 0, Koller BH,Boucher RC.
Defective epithelial chloride transport in a genetargeted mouse model of
cystic fibrosis. Science (Washinton DC) 1992; 7.57: 1125-8.
5 Sato F. Sato K. Secretion of K-rich. low-Na primary sweat by the secretory
coil of the rat paw eccrine sweat gland. J Physiol (London) 1978; 274: 37-50,
6 Stevens LM. Landis SC. Development interactions between sweat glands and
the sympathetic neurons which innervate them: effects of delayed innervation
on neurotransmitter plasticity and gland maturation. Dev Biol 1988; 130:
70S20.
7 Kennedy WR, Sakuta M. Collateral reinnervation of sweat glands. Ann Neurol
1984: IS: 73-0.
8 Sato K. Sweat induction from isolated eccrine sweat gland. Am J Physiol 1973;
W: 1147-52.
109 Sato K, Sato F. Pharmacologic responsiveness of isolated eccrine sweat glands.
Am J Physiol 1981; 2& R44-51.
Sat0 K. Sato F. Nonirotonicity of simian eccrine primary sweat induced in
vitro. Am J Phpiol 1987; UZ:R1099-IOS.
I I Van Glcselt HRM, Vierhout RR. Registration of the insensible perspiration of
small quantities of sweat. Dermatologica 1%); 127: 255-9.
I 2 Sat0 K. Sat0 F. Variable reduction in Padrenergic sweat secretion in cystic
fibrosis in heterozygotes. J Lab Clin Invest 1988; Ill: 511-18.
13 Sato KT. Saga K. Ohtsuyama M, et al. Movement of cellular ions during
stimulation with isoproterenol in simian eccrine clear cells. Am K Dhysiol
1991; 261: R87-93.
14 Sato K. Sato F. Cyclic AMP accumulation in the bettadrenergic mechanism of
eccrine sweat secretion. M i i g e n Arch 1981; 390: 49-53.
15 Sato K. Sato F. Effect of VIP on sweat secretion and CAMP accumulation in
simian eccrine glands. A, J Physiol 1987: 35): R 9 3 H l
M o w sweat gland
16. Sato K, Kang HW, Saga K, Sat0 KT. Biology of the eccrine sweat gland. I:
Mechanism of sweat secretion. J Am Acad Dermatol 1989; ID: 53765.
17. Suruki Y. Ohuuyama M. Samman C. Sato F, Sat0 K. Ionic basis d
methacholine-induced shrinkage of dissociated eccrine clear cells. J Membr Biol
1991; 123: 33-41,
18. Sat0 K. Ohuuyama M, Sat0 F. Whole cell K and CI currents in dissociated
eccrine secretory coil cells during stimulation. J Membr Biol 1993; IM
93-106.
19. Quick DC. Kennedy WR, Yoon K. Ultrastructure of the secretory epithelium.
nerve fibers, and capillaries in the mouse sweat gland. Anat Rec 1984;
491-9.
20. Sat0 K. Electrochemical driving forces for K secretion by the rat paw eccrine
sweat gland. Am J Physiol 1980; k CW7.
I39
21. OCrady SM. Palfrey HC. Field M. Characteristics and functions of Na-K-Cl
COtranSpoK in epithelial tissues. Am J Physiol 1987; 253: Cl77-92.
22. Collins FS. Cystic fibrosis: molecular and therapeutic implications. Science
(Washington DC) 1992; 2% 774-9.
23. Higgins CF. Cystic fibrmis transmembrane conductance regulator. Br Med Bull
1992; Q 754-65.
24. Tau F. Stanier P. Wicking C. et d. Cloning the mouse homolog of the human
cystic fibrosis transmembrane conductance regulator gene. Genomics 1991; 10:
301-7.
25. Hayashi H. Functional activity of the sweat glands of the mouse. Tohoku J Exp
Med 1968; PI: 289-95.
26. Lowry OH, Rmebrough NJ.Farr AL, Randall RJ. Protein measurement with
the Folibphenol reagent. J Biol Chem 1951; I93 265-77.