Nasal glandular secretory response to cholinergic stimulation in humans and guinea pigs ALEKSANDER AND MICHAEL
Z. GAWIN, BRIAN A. KALINER
E. EMERY,
JAMES
N. BARANIUK,
Laboratory of Clinical Investigation, National Institute of Allergy and Infectious National Institutes of Health, Bethesda, Maryland 20892
Diseases,
GAWIN,ALEKSANDERZ.,BRIAN E. EMERYJAMESNJSARANIUK, AND MICHAEL A. KALINER. Nasal glandular secretory response to cholinergic stimulation in humans and guinea pigs. J. Appl. Physiol. 71(6): 2460-2468,1991.-A guinea pig model of nasalsecretory responseswasdevelopedto assess the contributions of vascular permeability and glandular secretionresponsible for the production of cholinergically stimulated nasalsecretions. The nasalsecretory responsesto provocation with saline, methacholine, and atropine on the ipsilateral (challenged)side and contralateral (reflex) side were analyzed by measurement of total protein (Lowry method), guineapig albumin (enzymelinked immunosorbent assay), lz51-labeledbovine serum albumin after intravenous injection, and alkaline phosphataseenzyme activity in nasal fluid. Alkaline phosphatasewasfound to be localized to submucosalglands by zymography. Topical methacholine challenge increased the secretion of total protein, alkaline phosphataseactivity, and albumin on the ipsilatera1 challenged side, whereasthe percentage of total protein representedby albumin was not increased.This responsewas totally prevented by atropine pretreatment. Serial provocation with methacholine resulted in progressively reduced amounts of both the total protein and alkaline phosphatasein secretions. The observation that repeated challengesproduced progressivelysmaller responseswas alsoexamined employing human nasal provocation. Repeating methacholine (25 mg) challengesfour times at lo-min intervals in six human volunteers revealed that the initial challenge produced the largest responseas reflected in total protein, albumin, lysozyme, lactoferrin, immunoglobulin (Ig) G, IgA, and secretory IgA secretion. When the constituents in secretionswere analyzed in relationship to the total protein, the two vascular proteins, IgG and albumin, demonstrated the greatest decrementswith repeated methacholine challenges.The glandular proteins, lactoferrin, lysozyme, and secretory IgA, either remained constant or increasedin their relative proportion to total protein. Thus, cholinergic stimulation causesglandular secretion from both the guinea pig and human nasal mucosa. Repeated methacholine challengesresultedin progressivelydiminishing macromolecule secretion in guinea pigs and a depletion of plasma proteins in glandular secretionsin humans.
components by use of human models (14, 17, 18). To further amplify our understanding, a guinea pig model of the nasal mucosa secretory responses is also being developed (7; unpublished observations). The model allows topical application of agonists and antagonists and the measurement of secreted total protein (13), albumin, and 1251-labeled bovine serum albumin (BSA) after intravenous injection. The total protein in guinea pig nasal secretions includes both glandular and vascular products. Albumin, which accounts for 50-60% of guinea pig plasma proteins, appears to enter nasal secretions by a combination of vascular permeability, glandular secretion, and epithelial permeability. In contrast to human submucosal glands, a selected subpopulation of guinea pig submucosal gland cells (w 10%) contains immunoreactive albumin, which may be rapidly expelled on stimulation (7). Albumin represents ~5-13% of the total protein in unstimulated human secretions and l-10% of the total protein in unstimulated guinea pig secretion. Plasma albumin (as reflected by movement of 1251-BSA given intravenously) exits vessels and enters both submucosal glandular lumina and the epithelium under baseline conditions. The bulk of the total protein found in guinea pig nasal secretions consists of glandular products. However, glandular products found in human nasal secretion, such as lysozyme, lactoferrin (18), and secretory immunoglobulin (Ig) A (17), are unfortunately not present in guinea pig secretions. In searching for markers of glandular secretion in the guinea pig, alkaline phosphatase activity was localized to guinea pig (but not human) submucosal glands histologically and detected in secretions by calorimetric assay. By measurement of total protein, albumin, 1251-BSA, and alkaline phosphatase activity in nasal secretions, the relative contribution of glandular products and vascular permeability to nasal secretions formed in response to topical cholinergic stimulation could be estimated.
guinea pig nasal mucosa;nasal secretion; methacholine; atropine; guinea pig albumin
METHODS
NASAL MUCOSAL SECRETIONS are the result of several processes that include vascular permeability, glandular permeability, gland and goblet cell exocytosis, and epithelial permeability. Progress is being made in understanding the contribution of vascular, glandular, and epithelial
Male Hartley guinea pigs (Charles River Laboratory, Wilmington, MA) weighing 400-560 g were anesthetized with pentobarbital sodium (50 mg/kg; Somnifer, Richmond Veterinary Supply, Richmond, VA). The femoral vein was dissected and exposed and 1251-BSA, (1 &i/l00 g) was injected intravenously. Both nasal cavities were washed six times with lo-p1 lavages of 0.9% NaCl at room
2460
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
2461 four repeated doses of 25 mg of methacholine. Ten-minute collection periods were interposed between challenges, with a 4-ml saline lavage at the end of the lo-min period. Methacholine was diluted in normal saline and delivered in 0.3-ml volumes. Assay methods were as previously reported for total protein, albumin, total IgA, secretory IgA (l7), IgG (14), lactoferrin, and lysozyme (18). Luvuge fluid assays. Frozen samples were melted, vortexed, and diluted for assays. The total protein in lavage fluid was estimated by the Lowry method (13). Methacholine and atropine do not interfere with this assay. Guinea pig albumin enzyme-linked immunosorbent ussay (ELBA). Lavage fluid guinea pig albumin was measured by competitive ELISA (19). Polypropylene 8 X 12well ELISA plates (Falcon 3913, Bectin Dickinson Labware, Oxnard, CA) were precoated with 100 ~1 of guinea pig albumin (50 pg/ml) in coating buffer (pH 9.3,29.3 ml of 0.1 M Na,CO, in 70.7 ml of 0.1 M NaHCO,) for 2 h at room temperature or 18 h at 4OC. The wells were washed four times with 0.05 M phosphate buffer-0.05% Tween80 (PT, pH 7.4), followed by 250 ~1 of 1% nonimmune goat serum and incubated for 30 min at room temperature. After four washes with PT, 50 ,ul of a standard solution of guinea pig albumin (Cooper Biomedical, Malvern, PA) in PT (l-1,000 pglml) or 50 ~1 of diluted lavage fluid were placed in the wells, followed by 50 ~1 of rabbit antiguinea pig albumin antiserum (Cooper Biochemical) diluted l:l,OOO in PT. The plates were incubated for 90 min at room temperature and then washed four times with PT. Next, 100 ~1 of alkaline phosphatase-conjugated goat anti-rabbit IgG antiserum (Sigma Chemical) diluted l:l,OOO in PT were added, incubated for 90 min at room temperature, and washed with four volumes of PT. The wells were filled with 100 ~1 of phosphatase substrate (Sigma Chemical, 1 mg/ml in 0.1 M coating buffer and 0.02% MgCl,). The reaction was terminated after 30 min by addition of 100 ~1 of 3 N NaOH, and the color intensity was measured on an MR 600 Microplate Reader (Dynatech, Alexandria, VA). Albumin% was calculated as the ratio of guinea pig albumin to total protein in each sample X 100. Standard curves were linear on semilog plots for concentrations between 1 and 1,000 pg/ml. Alkaline phosphutuse. To determine other markers of glandular secretion in the guinea pig, an Apizym assay (Analytab Products, Plainview, NY) was done for guinea pig serum and nasal secretions. Alkaline phosphatase activity was found in nasal secretions. For the assay of alkaline phosphatase activity, guinea pig intestinal alka-
SECRETION
temperature to remove preexisting secretions. Lavage fluid was collected with a Gilson pipette (Rainin Instrument, Emeryville, CA) from the vestibules of each nostril, with care taken to avoid touching the nasal mucosa. Samples were kept on ice during experiments and were stored at -7OOC for later analysis. Each nasal provocation experiment involved unilateral instillation of 10 ~1 of challenge solution. After 10 min, both nostrils were irrigated three times with 10 ~1 of normal saline (pH 6.8) and lavage fluids were collected. The irrigation and collection procedure took ~3 min. Three lavages with 10 ~1 of saline yielded recoveries of 24-30 ~1 (SO-100%) of lavage fluid. In preliminary experiments of repeated saline challenges and lavage, after two lavages (each including 3 applications of 10 ~1 of saline) the protein and albumin levels in the recovered nasal fluid stabilized. In all subsequent experiments, two cycles of saline lavage were carried out and the secretions discarded before with the “baseline” saline challenge. Each set of experiment was done with separate group of six animals. Methacholine dose response. After initial lavages, one nostril (ipsilateral side) was challenged with saline or various doses of methacholine applied sequentially every 10 min (Sigma Chemical, St. Louis, MO). Bilateral lavage was carried out at IO-min intervals. Two sets of experiments were performed, one from 0.1 to 10 pg/lO ~1 and the other from 0.001 to 10 lug/l0 ,ul with a different group of animals. Effect of repeated methacholine challenges. After the initial lavages with saline, one nostril was challenged with saline, followed 10 min later by 1 pg of methacholine. Bilateral lavages were carried out 10 min after each challenge. This procedure was repeated five times. Atropine pretreatment. After the initial lavages, one nostril was challenged with 10 ~1 of saline (ipsilateral), and 10 min later both sides were lavaged. Then, atropine (1 pg/lO ~1) was instilled on the ipsilateral side. After 10 min, both sides were challenged with methacholine (10 pg/lO ~1). Subsequent lavages were done on both sides at 10 and 20 min. Human methucholine nasal challenge. Six adult nonatopic subjects, aged 18-57 yr, who had not had respiratory tract infections in the proceeding 6 wk, were studied after giving informed consent. Repeated methacholine challenges were carried out on both sides of the nose in each subject. Challenge methods were as previously described (14,17,18). After the initial baseline saline wash, each patient was challenged (at lo-min intervals) with
1. Effect of methuch&ne on the cwzcentrution of 1z51-ulbumin, total protein, albumin, in ipsiluterul and contralateral secretiuns
TABLE
‘Y-Albumin,
Baseline Methacholine CM a 1 Peg 10 Pg with
cpm/ml
Total
pg/ml
Albumin, IL
Albumin%
CL
13tQ.9
llt2.3
2,280+238
2,131+255
55t9.5
4221.6
2.4kO.5
1.9*0,3
25+3.0** 21tl.7 20t3.8
7t2.3 12t2.9 17tU.8
5,594+950* 2,373-+488 2,253+681
1,276t225 1,108&167 1,066+236
9ltl3 78tll 62t8.8
25t2.4 30t2.5 29t3.1
1.6t0.5 3.2HI.3 2.7t0.6
1.9tO.5 2.7tU.6 2.7kO.7
CL, contralateral;
CL
pg/ml
IL
Values are means + SE. IL, ipsilateral; previous lavage (unpaired Student’s
IL
Protein,
and ulbumin%
CL
IL
CL
albumin%, percentage of total protein represented as albumin. * P < 0.01, compared
t test).
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
2462
NASAL
SECRETION
200
1 m
El80 %lSO
IPSILATERALSIDE CONTRAIAJERALSIDE
i
I m
I m
IPSILATERALSIDE CONTRALATERALSIDE
'160 tnw e440 sz:l20
$120z loozl T 80
b 100
CONTROL FIG.
E200 kl8O-j
g140
after
*
220
1. Alkaline methacholine
0.1
phosphatase stimulation.
L
1.0 10.0 MEJHACHOLINE(pg)
activity in guinea *p < 0.01, compared
CONTROL 0.001 0.01
pig nasal lavages with control.
2. Alkaline methacholine
phosphatase stimulation.
FIG.
after
0.1 1.0 10.0 MEJHACHOLINE(pg)
activity in guinea *P < 0.02, compared
100.0
pig nasal lavages with control.
line phosphatase (type XXV, Sigma Chemical) was di- the results reflect the means of six subjects per group. Experimental results were compared by Student’s t test. luted in pH 9.3 0.1 M sodium carbonate buffer (coating buffer above) and employed as a standard in a colorimetEach statistical result was also compared by analysis of ric assay (3). Standards or samples (100 ~1) were ali- variance. The results from both the t tests and analysis of quoted into ELISA plates, and 100 ~1 of 10 mM p-nitrovariance were equivalent, and only the t test results are phenyl phosphate (Sigma Chemical) in coating buffer presented. were added. The change in optical density at 410 nm was determined between 0 and 60 min for each well. The RESULTS standard curve was linear from 0.1 to 100 pglml, and Guinea pig nasal lavage fluid was homogeneous and unknown concentrations were interpolated. acellular when collected. The pH was between 6 and 7. 12%BSA content of secretions and serum was assessed After saline, methacholine, and atropine challenges, by gamma spectroscopy (Beckman, Irvine, CA) as de- three lavages with 10 ~1 of saline yielded recoveries of scribed (7; unpublished observations). 24-30 ~1 (85-100%) of lavage fluid. The initial washes removed preexisting secretions, and the total protein levAlkaline phosphatase zymogram. Unchallenged guinea pigs or guinea pigs that had received five administrations els in serial lavages reached a stable level after one or two of 1 pg of methacholine at lo-min intervals were killed by saline washes and remained constant thereafter (7; unpentobarbital overdose. The turbinates were rapidly ex- published observations). Saline challenges and washes cised and immediately placed in 2-methylbutane on dry did not stimulate albumin or protein secretion (7; unpubice for 30 s. Frozen sections, which were prepared by the lished observations). Although these lavages are stanthaw-mount method, were incubated in pH 7.4 50 mM dardized, there are interanimal variations in the baseline tris(hydroxymethyl)aminomethane for 5 min at 25’C conditions after saline lavage. Younger and smaller aniand then incubated with an alkaline phosphatase sub- mals tend to produce lower total protein and albumin strate (Vector Red, Vector Laboratories, Burlingame, concentrations in the baseline state (7; unpublished observations). Concentrations were not corrected or norCA) in pH 8.2 100 mM tris(hydroxymethyl)aminomethane for 40 min at 25*C. Alkaline phosphatase reactions malized for animal age and weight; rather animals of uniwere stopped by washing the slides in water. Deposits of form age and weight were used for all experiments. A insoluble red pigment identified the locations of endogemore difficult variable, however, was the normal nasal nous alkaline phosphatase activity. cycle of vascular congestion, glandular secretion, and naStatistics. Results are shown as mean t SE. Each ex- sal obstruction. Animals could not be synchronized for periment was repeated on six humans or guinea pigs, and their nasal cycles. 2. Effect of methacholine and contralateral secretions
TABLE
on concentration
Total Protein,
pg/ml
IL Baseline Methacholine
of total protein, albumin, Albumin,
CL
2,775+283
3,000+248
0.001pg
2,512+322
0.01 pg a1 ccg 1 Pg 10 LLg
4,350+457* 3,350+666 2,275&415 2,162t-431
2,4753-103 1,900+348
Values
are means
t SE. * P < 0.005,
compared
with
previous
lavage
in ipsilateral
pg/ml
Albumin%
IL
CL
IL
CL
81+11
76,+4
2.8t0.2
2.5kO.l
64t8.4 54-18.6 3Ok3.3 35t5.4 38k6.9
2.6t0.13 3.820.6 3.620.6 4.2t0.55 3.6k0.24
2.6kO.42 2.9kO.34 2v3t0.31 2.5t0.22 2.8t0.41
66Gi.9 162t15.9” 120t25.6 98~~22 80~18
1,3251193 1,212-t51 l,362t-143
and albumin%
(unpaired
Student’s
t test).
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
2463
SECRETION
3. Effect of methacholine repeated five times on the concentration of total protein, albumin, and albumin% in guinea pig nasal secretions
TABLE
Total Protein,
pglml
IL Baseline
Methacholine 1 I% 1 1 1 1
PcL9 ccg P!z clg
Albumin, CL
pg/ml
IL
2,483+177
2,666*189
61t4.8
4,233&688* 2,900+570 2,100+353 2,000+199 2,466+176
2,033+375 1,583+249 1,7OO_t251 1,933+260 1,816+274
155+19.5~ 132t14 88t10.5 7928.3 67t4.3
Albumin% CL
IL
CL
69YZ.6
2.5+0.4
2.WI.29
65t12
3.7t0.29 4.5kO.56 4.2t0.15 3.9kO.06 2.7kO.12
3.2t0.19 3.6t0.43 3.2t0.5 3.0t0.32 2.8t0.14
69216 51t6.7 62t_lO 53t9.6
Values are means f. SE. * P < 0.01; t P < 0.005, compared with previous lavage (unpaired t test).
Methacholine dose response,The initial dose of methacholine (0.1 pg; Table 1, Fig. 1) induced a significant secretion of total protein, 1251-BSA, and alkaline phosphatase activity on the ipsilateral side. However, progressively higher concentrations (1 and 10 ,ug; Table 1, Fig. 1) induced progressively less protein or alkaline phosphatase secretion, indicating that no dose-dependent relationship was demonstrable. No contralateral effect was produced at any challenge dose. 1251-BSA secretion was increased at the initial dose but not at the higher concentrations of methacholine. Albumin secretion was not significantly affected at any dose, and the albumin% remained constant throughout. When lower initial doses of methacholine were employed (Table 2), significant increases of total protein, albumin, and alkaline phosphatase activity (Fig. 2) in nasal secretions were induced with 0.01 pg of methacholine. Progressively higher concentrations (0.1, 1, and 10 pg) induced less total protein, albumin, and alkaline phosphatase activity in nasal secretions. No contralateral effects were noted, and no change in albumin% occurred. Repeated methacholine challenges (Table 3). Serial application of 1 pg of methacholine led to significant secretion of total protein, albumin, and alkaline phosphatase activity (Fig. 3) after only the initial challenge. In each of four subsequent methacholine provocations, the level of total protein, albumin, and alkaline phosphatase activity
I I
CONTROL
I -
IPSILATERAL SIDE CONTRAL4TERAL SIDE
2
3
METHACHOLINE
4
progressively returned to the baseline. Again, no contralateral effects were noted, and albumin% remained unchanged. Methacholine-atropine interaction (Table 4). The ability of atropine to antagonize the effect of methacholine was studied by instillation of 1 pg of atropine sulfate followed by 10 rug of methacholine and compared with the administration of methacholine alone. Coadministration of atropine ablated the methacholine-induced changes in total protein and albumin. Therefore the effect of methacholine was inhibitable by prior treatment with atropine. Atropine (1 pg) also completely abrogated the alkaline phosphatase secretion (Fig. 4) induced by 1 pg of methacholine. The same inhibitory effect of atropine was seen if 0.1 pg of methacholine was employed as the agonist. Alkaline phosphatase zymogram. Endogenous alkaline phosphatase activity was consistently identified in submucosal glands and ducts (Fig. 5B) as well as in the glycocalyx. Cartilage, perichondrium, and periosteum also possessed endogenous activity. Treatment of guinea pig turbinates with five doses of 1 pg of methacholine resulted in the depletion of alkaline phosphatase activity from the glands and glycocalyx but not from cartilage (Fig. 5C). Repeated methacholine challenge in humans (Figs. 6
and 7). In the human study, the effect of four sequential
BY z \ g w 2 1 g g k 5 3 2 5
b4
FIG. 3. Alkaline phosphatase activity in guinea pig nasal lavages after methacholine stimulation repeated 5 times. Numbers on abscissa indicate order of repeated challenge. *P < 0.01, compared with control.
240 220 200 180 160 140 120 100 80 60 40 20 0
ATROPINE(? ,ug) PRETREATMENT
CONTROL
***
I*0 Pg METHACHOLINE
FIG. 4. Alkaline phosphatase activity after methacholine (1 pg) stimulation after atropine pretreatment vs. methacholine (1 pg) alone. ***P < 0.001, compared with atropine treatment.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
2464
NASAL
TABLE
4. Comparison of the effect of topical atropine plus methacholine vs. methacholine alone Total A+M
Baseline Stimulated
2,625+711 3,583?522
Protein,
SECRETION
pglml
Albumin, M
A+M
2,041?298 4,250+422*
Values are means + SE. A, atropine; M, methacholine.
82219 127+13
pglml
Albumin% M
7729.3 183?23t
A+M
3.1f0.7 3.5+0.9
M
3.7kO.8 4.3kO.8
* P < 0.01; t P < 0.02, compared with baseline lavage (unpaired Student’s t test).
methacholine challenges was examined. The initial methacholine challenge caused the largest increased secretion of total protein (Fig. 6A), albumin (Fig. 6B), lysozyme (Fig. 6D), secretory IgA (Fig. 6F), nonsecretory IgA (Fig. 6G), lactoferrin (Fig. 6C), and IgG (Fig. 6E) compared with subsequent challenges. Repeated challenges with methacholine caused sustained secretion of all the glandular products (lysozyme, lactoferrin, and secretory IgA). By contrast, the secretion of the plasma proteins, albumin and IgG, returned to baseline levels by the fourth challenge. The level of nonsecretory IgA, which is also a plasma protein, remained minimally elevated through the series of challenges. Although expression of the secreted proteins in absolute amounts provided useful data, presentation of the results in relationship to total protein provided additional insights into the various contributions of glandular and vascular proteins (Fig. 7). Repeated methacholine challenges resulted in progressive decrements in albumin% (Fig. 7A) and IgG% (Fig. 70) below baseline values (both achieved significant reductions by the 4th challenge). In contrast, the ratios for glandular products (lactoferrin, lysozyme, and secretory IgA) were not significantly different from baseline after repeated challenges. DISCUSSION
Nasal secretory responses are coordinated by the autonomic nervous system (1,2,6,14,17,18) with the major effector control through cholinergic nerves (4,10,17). To help our understanding of nasal secretory responses, dependable in vivo models that allow quantitative measure-
ment of the secreted substances are very useful. The development of a guinea pig model of nasal secretion may prove helpful in this respect. The guinea pig model allows the collection of secretions from both sides of the nose without irritation. Repeated saline lavages do not affect the total protein and albumin content of secretions, indicating that neither the procedure nor the repeated washing by themselves has an effect on nasal mucosa secretion. Recovery of washing averages 85-100%. Muscarinic receptors are found on submucosal glands, blood vessels, and airway epithelium (9, 11, 17), and methacholine stimulates submucosal gland secretion, ion transport across the epithelial surfaces (9, 17, 18), and arteriolar vasodilation (1,17). Methacholine stimulation of the guinea pig mucosa provoked secretion of protein and albumin, whereas albumin% was not altered. These findings are similar to cholinergic responses in humans (17). Atropine, a potent cholinergic antagonist, abrogated the total secretory response induced by methacholine. Thus the response to topical methacholine in guinea pigs is mediated through muscarinic receptors. Alkaline phosphatase is an enzyme that hydrolyzes organic esters of phosphoric acid and transfers the inorganic phosphate group to other compounds (transphosphorylation) (3, 5, 8, 15, 19, 21). Histochemical studies employing alkaline phosphatase substrate revealed alkaline phosphatase activity in the submucosal glands of the guinea pig nasal mucosa. Immunohistochemical localization of alkaline phosphatase antigenic activity could not be determined because, to our knowledge, no suitable antiserum to guinea pig alkaline phosphatase exists. Alkaline phosphatase activity was recovered in nasal secre-
FIG. 5. A: normal guinea pig nasal turbinate mucosa. Beneath squamous epithelium (e) are many submucosal glands (g), vessels (v), and cartilage (c). Hematoxylin and eosin stain. B: alkaline phosphatase zymogram. Precipitated products from alkaline phosphatase (dark stain) are localized to glands (g), glycocalyx of epithelium (e), and cartilage (c). Vessels (v) did not contain detectable alkaline phosphatase activity. C: turbinates from animals treated with repeated doses of methacholine did not display alkaline phosphatase activity in their epithelium (e) and had reduced activity in glands (g), although cartilage (c) still possessed activity. Bars, 100 pm.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
04
I 1
0
80
70
1
I
2 YcHAupIoEs
I
B
3
4
B
2465
SECRETION
80
1
c
60 50 40 30 20 10 I
0
1
I
2 PCHAUPIGES
I
I
3
4
80
E 1
7oJ
609 50= 4oa 309
80~
F
709 609 5Om
FIG.
tory IgA patients.
6. Total protein, albumin, lactoferrin, lysozyme, immunoglobulin (non s-IgA) in baseline nasal lavages and after 4 sequential *‘P < 0.05; **P < 0.01; ***P < 0.001, compared with baseline.
(Ig) G, secretory IgA (s-IgA), and nonsecrebilateral 25mg methacholine challenges in 6
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
2466
0
207
u
1
2 # CHALlBlGES
3
4
1
2 3 Y CHAUENGES
4
SECRETION
c
20
F 3
FIG. 7. Results of repeated methacholine challenges expressed as albumin%, lactoferrin%, lysozyme%, IgG%, secretory IgA%, and nonsecretory IgA% (means t SE). *.P < 0.05; **Fp < 0.01, compared with baseline.
tions and probably represents glandular secretion. This conclusion is based on the following lines of evidence: 1) It is unlikely that serum alkaline phosphatase is transudated from plasma, inasmuch as methacholine does not cause vascular permeability because the albumin% did not change (Tables 1 and 2). Moreover, the levels of alkaline phosphatase in serum are only 400-550 pg/ml,
much the same levels present in nasal secretions. In contrast, albumin is found at 25 mg/ml in serum and only 250 pg/ml in secretions. Thus the ratio of plasma albumin to albumin in secretions indicates that serum proteins are diluted -IOO-fold in the lavage protocol. In addition, alkaline phosphatase activity is found diffusely in submucosal gland cells and cuboidal cells of the ducts. In
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
guinea pigs, albumin is found in -10% of submucosal gland cells, whereas alkaline phosphatase activity was widely distributed in gland acini. In contrast, in human submucosal gland histochemistry, plasma proteins are never seen within gland cells. Instead plasma proteins are distributed throughout the interstitium and occasionally in gland lumina. 2) The dose-response effects of methacholine on the secretion of alkaline phosphatase activity have a pattern similar to the secretion of total protein. 3) Atropine completely abrogates the secretion of total protein and alkaline phosphatase activity. 4) Methacholine challenge appeared to deplete the glands of alkaline phosphatase activity. These data indicate that alkaline phosphatase is secreted as a glandular product from guinea pig nasal submucosal glands and suggest that it can be employed to monitor glandular secretion. In addition, these data suggest that glandular secretion in response to methacholine rapidly depletes the available alkaline phosphatase, resulting in rapidly decreasing quantities of this glandular product. The pattern of the initial methacholine challenge causing the most dramatic secretory response was seen whether total protein, albumin, or alkaline phosphatase was monitored. The dose-response experiments suggest that the guinea pig rapidly depletes its glands of protein, albumin, and alkaline phosphatase on initial stimulation, as reflected in an outpouring of secretions. This suggestion was confirmed by the loss of alkaline phosphatase activity in guinea pig nasal glands on histochemical analysis after methacholine stimulation. Perhaps the guinea pig’s nasal mucosa is designed for a rapid brief secretory response, which cannot be sustained. As an obligate nasal breather, it might be difficult to breathe if excessive secretion continued. Stimulation of cholinergic receptors in the human nasal mucosa has also been shown to be an important mechanism for producing glandular secretions (17). Earlier human studies have demonstrated that lysozyme, lactoferrin, and s-IgA present in nasal secretions come from glandular sources (17, 18). Both IgG and albumin have been found in periglandular interstitial spaces and in the lumen of submucosal glands (14). It is believed that methacholine-induced increases in IgG and albumin in secretions most likely represent transport of these proteins with glandular secretions. Methacholine has been demonstrated to selectively increase glandular secretion without affecting vascular permeability (17), but the resultant secretions always contain increased amounts of plasma proteins as well. Glandular secretions are relatively enriched in lysozyme, lactoferrin, and secretory IgA, whereas secretions resulting from increased vascular permeability are relatively enriched in albumin and IgG. To determine whether the pattern of repeated cholinergic stimulation inducing depletion of secreted proteins seen in the guinea pig had relevance to human secretory patterns, six volunteers were stimulated with repeated methacholine challenges. Much as seen in the guinea pig, the initial methacholine challenge in humans caused the greatest secretory response, but serial challenges continued to stimulate secretion of total protein, albumin, IgG, lysozyme, lactoferrin, and IgA. It was noted, however,
SECRETION
2467
that the secretion of the vascular proteins (albumin, IgG) returned to baseline after four challenges, while the glandular proteins (lysozyme, lactoferrin, and secretory IgA) remained increased. Nonsecretory IgA remained at a modestly increased level throughout. Taken as a percentage of total protein, the serial methacholine challenges resulted in a significant reduction for the two plasma proteins (albumin and IgG), suggesting that the glands became relatively depleted of the plasma proteins with repeated stimulation. These data indicate that a nasal secretory response can be studied in guinea pigs and that the response to cholinergic stimulation is primarily glandular. It was found that glandular secretions include the enzyme alkaline phosphatase. In contrast to the data reported here, other work with this model indicates that histamine causes a plasma protein-rich secretion (unpublished observations). The depletion of glandular responses seen in the guinea pig after repeated methacholine challenges was not seen in humans, although plasma proteins in human nasal secretions were progressively reduced. Thus one must conclude that the guinea pig provides a useful and easy model for the study of nasal physiology but that the results must be interpreted carefully and that extrapolations to human responses made cautiously. This study was supported in part by grants from the Merck Foundation and Merck Pharmaceuticals (A. 2. Gawin) and by fellowships from Proctor and Gamble, Inc, (B. E. Emery and J. N, Baraniuk). Address for reprint requests: A. 2. Gawin, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, Bldg. 10, Room 11-C-209, National Institutes of Health, Bethesda, MD 20892. Received 8 May 1990; accepted in final form 7 August 1991. REFERENCES A. The effect of parasympathetic nerve stimulation on the microcirculation and secretion in the nasal mucosa of the cat. Acta Otohyngol. 78: 98-105, 1974. BARNES, P. Neural control of human airways in health and disease. Am. Rev. Respir. Dis, 134: 1289-1314, 1986. BESSEY, 0. A., 0. H, LOWRY, AND M. J, BROCK. A method for the rapid determination of alkaline phosphatase with five cubic millimeters of serum. J. Biol. Chem. 164: 321-329, 1946. BORUM, P. Nasal methacholine challenge: a test for measurement of nasal reactivity. J. Allergy Clin. Immunol. 63: 253-257, 1979. BUURNE, G. H. Alkaline phosphatase in taste buds and nasal mucosa. Nature Lsnd. 161: 445-446,1948. DRUCE, H. M., R. H. WRIGHT, D. KOSSOF, AND M. KALINER. Cholinergic nasal hyperactivity in atopic subjects. J. Allergy Clin. hamunol. 76: 445-452,1985. GAWIN, A. Z., J. N. BARANIUK, AND M. A. KALINER. Guinea pig nasal mucosa secretory responses (Abstract). J. Allergy Clin. Immunoz. 83: 304, 1989. IIIZUMI, Y., K. HIRANO, M. SUGIURA, S. 11~0, H. SUZUKI, AND T. ODA. Evidence for the physiological function of alkaline phosphatase. Chem. Pharm. Butl31: 772-775,1983. JACKSON, R. T. Evidence for presynaptic parasympathetic receptors on nasal blood vessels. Ann. Otol. 91: 216-219, 1982. KLAASEN, A.B.,W. KNIJEPERS, I.F. RODRIGEZDEMIRANDA,AND A. J. BELD. The influence of neuropharmaca on the nasal glands. Rhinology 23: 191-194, 1985. KLAASEN, A. B., W. KNIJPERS, I. F. RODRIGEZ DE MIRANDA, AND J. BELD. Muscarinic receptors in rat nasal glands. Cell BioL. ht. Rep. 9: 521-521,1985. KONNO, A., AND K. TAGAWA. The mechanisms involved in onset of
1. ANGGARD,
2. 3. 4. 5. 6. 7. 8. 9. 10.
11.
12.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
2468
NASAL
SECRETION
allergic manifestation
in nasal allergy. Eur. J. Respir. Dis. 64, Suppt. 128: 155~166,1983. 13. LOWRY, 0. H., N. J. RUSENBROUGH, A. L. FARR, AND R. 9, RANDALL. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 165-1’75,195l. 14. MEREDITH, S. ID., G. II. RAPHAEL, J. N, BARANIUK, S. M. BANKS, AND M. A. KALINER. The pathophysiology of rhinitis. III. The control of IgG secretion. J. Allergy Clin. lmmunol. 84: 920-930, 1989. 15. MIKI, A., Y. TANAKA, S. OGATA, AND Y. IKEHARA. Selective prepa-
18. RAPHAEL, G. D., E. V. JENEY, J. N. BARANIUK, I. KIM, S. D. MEREDITH, AND M. A. KALINER. Pathophysiology of rhinitis. Lactoferrin and lysozyme in nasal secretions. J. Clin. Inuest. 84: 1528-1533, 1989. 19. ROSENBAUM, D., S. R. HIMMELHOCH, E. A. PATERSON, W. H. EVANS, AND M. G. MAGE. Purification of alkaline phosphatase from guinea pig bone marrow. Arch. Biochem. Biophys. 141: 303312, 1070. 20. SHASBY, M. D., AND R. L. ROBERTS. Transendothelial transfer of
ration and characterization of membrane soluble forms of alkaline phosphatase from rat tissue. A comparison with the serum enzyme.
macromolecules in vitro. Federation Proc. 46: 2506-2510, 1987. 21. SHELHAMER, J. M., 2. MAROM, AND M. KALINER. Immunologic and neuropharmacologic stimulation of mucous glycoprotein release from human airways in vitro. J. Clin. Invest. 66: 1400-1408, 1980. 22. SJOGREN, S., B. EK-RYLANDER, AND L. HAMMARSON. Characterization of alkaline phosphatase buccal mucosa in young rats. Acta
Eur. J. Biochem. 160: 41-48, 1986. 16. NAKAMURA, R., Z. OGITA, AND T. MATSUMARA.
Isozymes of alkaline phosphatase in gingiva and alveolar bone of the guinea pig.
Arch. Oral Bio!. 11: 131-133, 1966. 17. RAPHAEL, G. II., H. M. DRUCE, R.
H. WRIGHT, AND M. KALINER. Pathophysiology of rhinitis. I. Assessment of the sources of protein in induced nasal secretions. Am. Rev. Respir. Dis. 138: 413-420, 1988.
Odontol. 23. WANG,
Stand.
46: 75-82,
1988.
H. W., AND R. T. JACKSON. Do cholinergic neurons directly innervate nasal blood vessels? Rhinology 26: 139-141, 1988.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
Z. GAWIN, BRIAN A. KALINER
E. EMERY,
JAMES
N. BARANIUK,
Laboratory of Clinical Investigation, National Institute of Allergy and Infectious National Institutes of Health, Bethesda, Maryland 20892
Diseases,
GAWIN,ALEKSANDERZ.,BRIAN E. EMERYJAMESNJSARANIUK, AND MICHAEL A. KALINER. Nasal glandular secretory response to cholinergic stimulation in humans and guinea pigs. J. Appl. Physiol. 71(6): 2460-2468,1991.-A guinea pig model of nasalsecretory responseswasdevelopedto assess the contributions of vascular permeability and glandular secretionresponsible for the production of cholinergically stimulated nasalsecretions. The nasalsecretory responsesto provocation with saline, methacholine, and atropine on the ipsilateral (challenged)side and contralateral (reflex) side were analyzed by measurement of total protein (Lowry method), guineapig albumin (enzymelinked immunosorbent assay), lz51-labeledbovine serum albumin after intravenous injection, and alkaline phosphataseenzyme activity in nasal fluid. Alkaline phosphatasewasfound to be localized to submucosalglands by zymography. Topical methacholine challenge increased the secretion of total protein, alkaline phosphataseactivity, and albumin on the ipsilatera1 challenged side, whereasthe percentage of total protein representedby albumin was not increased.This responsewas totally prevented by atropine pretreatment. Serial provocation with methacholine resulted in progressively reduced amounts of both the total protein and alkaline phosphatasein secretions. The observation that repeated challengesproduced progressivelysmaller responseswas alsoexamined employing human nasal provocation. Repeating methacholine (25 mg) challengesfour times at lo-min intervals in six human volunteers revealed that the initial challenge produced the largest responseas reflected in total protein, albumin, lysozyme, lactoferrin, immunoglobulin (Ig) G, IgA, and secretory IgA secretion. When the constituents in secretionswere analyzed in relationship to the total protein, the two vascular proteins, IgG and albumin, demonstrated the greatest decrementswith repeated methacholine challenges.The glandular proteins, lactoferrin, lysozyme, and secretory IgA, either remained constant or increasedin their relative proportion to total protein. Thus, cholinergic stimulation causesglandular secretion from both the guinea pig and human nasal mucosa. Repeated methacholine challengesresultedin progressivelydiminishing macromolecule secretion in guinea pigs and a depletion of plasma proteins in glandular secretionsin humans.
components by use of human models (14, 17, 18). To further amplify our understanding, a guinea pig model of the nasal mucosa secretory responses is also being developed (7; unpublished observations). The model allows topical application of agonists and antagonists and the measurement of secreted total protein (13), albumin, and 1251-labeled bovine serum albumin (BSA) after intravenous injection. The total protein in guinea pig nasal secretions includes both glandular and vascular products. Albumin, which accounts for 50-60% of guinea pig plasma proteins, appears to enter nasal secretions by a combination of vascular permeability, glandular secretion, and epithelial permeability. In contrast to human submucosal glands, a selected subpopulation of guinea pig submucosal gland cells (w 10%) contains immunoreactive albumin, which may be rapidly expelled on stimulation (7). Albumin represents ~5-13% of the total protein in unstimulated human secretions and l-10% of the total protein in unstimulated guinea pig secretion. Plasma albumin (as reflected by movement of 1251-BSA given intravenously) exits vessels and enters both submucosal glandular lumina and the epithelium under baseline conditions. The bulk of the total protein found in guinea pig nasal secretions consists of glandular products. However, glandular products found in human nasal secretion, such as lysozyme, lactoferrin (18), and secretory immunoglobulin (Ig) A (17), are unfortunately not present in guinea pig secretions. In searching for markers of glandular secretion in the guinea pig, alkaline phosphatase activity was localized to guinea pig (but not human) submucosal glands histologically and detected in secretions by calorimetric assay. By measurement of total protein, albumin, 1251-BSA, and alkaline phosphatase activity in nasal secretions, the relative contribution of glandular products and vascular permeability to nasal secretions formed in response to topical cholinergic stimulation could be estimated.
guinea pig nasal mucosa;nasal secretion; methacholine; atropine; guinea pig albumin
METHODS
NASAL MUCOSAL SECRETIONS are the result of several processes that include vascular permeability, glandular permeability, gland and goblet cell exocytosis, and epithelial permeability. Progress is being made in understanding the contribution of vascular, glandular, and epithelial
Male Hartley guinea pigs (Charles River Laboratory, Wilmington, MA) weighing 400-560 g were anesthetized with pentobarbital sodium (50 mg/kg; Somnifer, Richmond Veterinary Supply, Richmond, VA). The femoral vein was dissected and exposed and 1251-BSA, (1 &i/l00 g) was injected intravenously. Both nasal cavities were washed six times with lo-p1 lavages of 0.9% NaCl at room
2460
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
2461 four repeated doses of 25 mg of methacholine. Ten-minute collection periods were interposed between challenges, with a 4-ml saline lavage at the end of the lo-min period. Methacholine was diluted in normal saline and delivered in 0.3-ml volumes. Assay methods were as previously reported for total protein, albumin, total IgA, secretory IgA (l7), IgG (14), lactoferrin, and lysozyme (18). Luvuge fluid assays. Frozen samples were melted, vortexed, and diluted for assays. The total protein in lavage fluid was estimated by the Lowry method (13). Methacholine and atropine do not interfere with this assay. Guinea pig albumin enzyme-linked immunosorbent ussay (ELBA). Lavage fluid guinea pig albumin was measured by competitive ELISA (19). Polypropylene 8 X 12well ELISA plates (Falcon 3913, Bectin Dickinson Labware, Oxnard, CA) were precoated with 100 ~1 of guinea pig albumin (50 pg/ml) in coating buffer (pH 9.3,29.3 ml of 0.1 M Na,CO, in 70.7 ml of 0.1 M NaHCO,) for 2 h at room temperature or 18 h at 4OC. The wells were washed four times with 0.05 M phosphate buffer-0.05% Tween80 (PT, pH 7.4), followed by 250 ~1 of 1% nonimmune goat serum and incubated for 30 min at room temperature. After four washes with PT, 50 ,ul of a standard solution of guinea pig albumin (Cooper Biomedical, Malvern, PA) in PT (l-1,000 pglml) or 50 ~1 of diluted lavage fluid were placed in the wells, followed by 50 ~1 of rabbit antiguinea pig albumin antiserum (Cooper Biochemical) diluted l:l,OOO in PT. The plates were incubated for 90 min at room temperature and then washed four times with PT. Next, 100 ~1 of alkaline phosphatase-conjugated goat anti-rabbit IgG antiserum (Sigma Chemical) diluted l:l,OOO in PT were added, incubated for 90 min at room temperature, and washed with four volumes of PT. The wells were filled with 100 ~1 of phosphatase substrate (Sigma Chemical, 1 mg/ml in 0.1 M coating buffer and 0.02% MgCl,). The reaction was terminated after 30 min by addition of 100 ~1 of 3 N NaOH, and the color intensity was measured on an MR 600 Microplate Reader (Dynatech, Alexandria, VA). Albumin% was calculated as the ratio of guinea pig albumin to total protein in each sample X 100. Standard curves were linear on semilog plots for concentrations between 1 and 1,000 pg/ml. Alkaline phosphutuse. To determine other markers of glandular secretion in the guinea pig, an Apizym assay (Analytab Products, Plainview, NY) was done for guinea pig serum and nasal secretions. Alkaline phosphatase activity was found in nasal secretions. For the assay of alkaline phosphatase activity, guinea pig intestinal alka-
SECRETION
temperature to remove preexisting secretions. Lavage fluid was collected with a Gilson pipette (Rainin Instrument, Emeryville, CA) from the vestibules of each nostril, with care taken to avoid touching the nasal mucosa. Samples were kept on ice during experiments and were stored at -7OOC for later analysis. Each nasal provocation experiment involved unilateral instillation of 10 ~1 of challenge solution. After 10 min, both nostrils were irrigated three times with 10 ~1 of normal saline (pH 6.8) and lavage fluids were collected. The irrigation and collection procedure took ~3 min. Three lavages with 10 ~1 of saline yielded recoveries of 24-30 ~1 (SO-100%) of lavage fluid. In preliminary experiments of repeated saline challenges and lavage, after two lavages (each including 3 applications of 10 ~1 of saline) the protein and albumin levels in the recovered nasal fluid stabilized. In all subsequent experiments, two cycles of saline lavage were carried out and the secretions discarded before with the “baseline” saline challenge. Each set of experiment was done with separate group of six animals. Methacholine dose response. After initial lavages, one nostril (ipsilateral side) was challenged with saline or various doses of methacholine applied sequentially every 10 min (Sigma Chemical, St. Louis, MO). Bilateral lavage was carried out at IO-min intervals. Two sets of experiments were performed, one from 0.1 to 10 pg/lO ~1 and the other from 0.001 to 10 lug/l0 ,ul with a different group of animals. Effect of repeated methacholine challenges. After the initial lavages with saline, one nostril was challenged with saline, followed 10 min later by 1 pg of methacholine. Bilateral lavages were carried out 10 min after each challenge. This procedure was repeated five times. Atropine pretreatment. After the initial lavages, one nostril was challenged with 10 ~1 of saline (ipsilateral), and 10 min later both sides were lavaged. Then, atropine (1 pg/lO ~1) was instilled on the ipsilateral side. After 10 min, both sides were challenged with methacholine (10 pg/lO ~1). Subsequent lavages were done on both sides at 10 and 20 min. Human methucholine nasal challenge. Six adult nonatopic subjects, aged 18-57 yr, who had not had respiratory tract infections in the proceeding 6 wk, were studied after giving informed consent. Repeated methacholine challenges were carried out on both sides of the nose in each subject. Challenge methods were as previously described (14,17,18). After the initial baseline saline wash, each patient was challenged (at lo-min intervals) with
1. Effect of methuch&ne on the cwzcentrution of 1z51-ulbumin, total protein, albumin, in ipsiluterul and contralateral secretiuns
TABLE
‘Y-Albumin,
Baseline Methacholine CM a 1 Peg 10 Pg with
cpm/ml
Total
pg/ml
Albumin, IL
Albumin%
CL
13tQ.9
llt2.3
2,280+238
2,131+255
55t9.5
4221.6
2.4kO.5
1.9*0,3
25+3.0** 21tl.7 20t3.8
7t2.3 12t2.9 17tU.8
5,594+950* 2,373-+488 2,253+681
1,276t225 1,108&167 1,066+236
9ltl3 78tll 62t8.8
25t2.4 30t2.5 29t3.1
1.6t0.5 3.2HI.3 2.7t0.6
1.9tO.5 2.7tU.6 2.7kO.7
CL, contralateral;
CL
pg/ml
IL
Values are means + SE. IL, ipsilateral; previous lavage (unpaired Student’s
IL
Protein,
and ulbumin%
CL
IL
CL
albumin%, percentage of total protein represented as albumin. * P < 0.01, compared
t test).
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
2462
NASAL
SECRETION
200
1 m
El80 %lSO
IPSILATERALSIDE CONTRAIAJERALSIDE
i
I m
I m
IPSILATERALSIDE CONTRALATERALSIDE
'160 tnw e440 sz:l20
$120z loozl T 80
b 100
CONTROL FIG.
E200 kl8O-j
g140
after
*
220
1. Alkaline methacholine
0.1
phosphatase stimulation.
L
1.0 10.0 MEJHACHOLINE(pg)
activity in guinea *p < 0.01, compared
CONTROL 0.001 0.01
pig nasal lavages with control.
2. Alkaline methacholine
phosphatase stimulation.
FIG.
after
0.1 1.0 10.0 MEJHACHOLINE(pg)
activity in guinea *P < 0.02, compared
100.0
pig nasal lavages with control.
line phosphatase (type XXV, Sigma Chemical) was di- the results reflect the means of six subjects per group. Experimental results were compared by Student’s t test. luted in pH 9.3 0.1 M sodium carbonate buffer (coating buffer above) and employed as a standard in a colorimetEach statistical result was also compared by analysis of ric assay (3). Standards or samples (100 ~1) were ali- variance. The results from both the t tests and analysis of quoted into ELISA plates, and 100 ~1 of 10 mM p-nitrovariance were equivalent, and only the t test results are phenyl phosphate (Sigma Chemical) in coating buffer presented. were added. The change in optical density at 410 nm was determined between 0 and 60 min for each well. The RESULTS standard curve was linear from 0.1 to 100 pglml, and Guinea pig nasal lavage fluid was homogeneous and unknown concentrations were interpolated. acellular when collected. The pH was between 6 and 7. 12%BSA content of secretions and serum was assessed After saline, methacholine, and atropine challenges, by gamma spectroscopy (Beckman, Irvine, CA) as de- three lavages with 10 ~1 of saline yielded recoveries of scribed (7; unpublished observations). 24-30 ~1 (85-100%) of lavage fluid. The initial washes removed preexisting secretions, and the total protein levAlkaline phosphatase zymogram. Unchallenged guinea pigs or guinea pigs that had received five administrations els in serial lavages reached a stable level after one or two of 1 pg of methacholine at lo-min intervals were killed by saline washes and remained constant thereafter (7; unpentobarbital overdose. The turbinates were rapidly ex- published observations). Saline challenges and washes cised and immediately placed in 2-methylbutane on dry did not stimulate albumin or protein secretion (7; unpubice for 30 s. Frozen sections, which were prepared by the lished observations). Although these lavages are stanthaw-mount method, were incubated in pH 7.4 50 mM dardized, there are interanimal variations in the baseline tris(hydroxymethyl)aminomethane for 5 min at 25’C conditions after saline lavage. Younger and smaller aniand then incubated with an alkaline phosphatase sub- mals tend to produce lower total protein and albumin strate (Vector Red, Vector Laboratories, Burlingame, concentrations in the baseline state (7; unpublished observations). Concentrations were not corrected or norCA) in pH 8.2 100 mM tris(hydroxymethyl)aminomethane for 40 min at 25*C. Alkaline phosphatase reactions malized for animal age and weight; rather animals of uniwere stopped by washing the slides in water. Deposits of form age and weight were used for all experiments. A insoluble red pigment identified the locations of endogemore difficult variable, however, was the normal nasal nous alkaline phosphatase activity. cycle of vascular congestion, glandular secretion, and naStatistics. Results are shown as mean t SE. Each ex- sal obstruction. Animals could not be synchronized for periment was repeated on six humans or guinea pigs, and their nasal cycles. 2. Effect of methacholine and contralateral secretions
TABLE
on concentration
Total Protein,
pg/ml
IL Baseline Methacholine
of total protein, albumin, Albumin,
CL
2,775+283
3,000+248
0.001pg
2,512+322
0.01 pg a1 ccg 1 Pg 10 LLg
4,350+457* 3,350+666 2,275&415 2,162t-431
2,4753-103 1,900+348
Values
are means
t SE. * P < 0.005,
compared
with
previous
lavage
in ipsilateral
pg/ml
Albumin%
IL
CL
IL
CL
81+11
76,+4
2.8t0.2
2.5kO.l
64t8.4 54-18.6 3Ok3.3 35t5.4 38k6.9
2.6t0.13 3.820.6 3.620.6 4.2t0.55 3.6k0.24
2.6kO.42 2.9kO.34 2v3t0.31 2.5t0.22 2.8t0.41
66Gi.9 162t15.9” 120t25.6 98~~22 80~18
1,3251193 1,212-t51 l,362t-143
and albumin%
(unpaired
Student’s
t test).
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
2463
SECRETION
3. Effect of methacholine repeated five times on the concentration of total protein, albumin, and albumin% in guinea pig nasal secretions
TABLE
Total Protein,
pglml
IL Baseline
Methacholine 1 I% 1 1 1 1
PcL9 ccg P!z clg
Albumin, CL
pg/ml
IL
2,483+177
2,666*189
61t4.8
4,233&688* 2,900+570 2,100+353 2,000+199 2,466+176
2,033+375 1,583+249 1,7OO_t251 1,933+260 1,816+274
155+19.5~ 132t14 88t10.5 7928.3 67t4.3
Albumin% CL
IL
CL
69YZ.6
2.5+0.4
2.WI.29
65t12
3.7t0.29 4.5kO.56 4.2t0.15 3.9kO.06 2.7kO.12
3.2t0.19 3.6t0.43 3.2t0.5 3.0t0.32 2.8t0.14
69216 51t6.7 62t_lO 53t9.6
Values are means f. SE. * P < 0.01; t P < 0.005, compared with previous lavage (unpaired t test).
Methacholine dose response,The initial dose of methacholine (0.1 pg; Table 1, Fig. 1) induced a significant secretion of total protein, 1251-BSA, and alkaline phosphatase activity on the ipsilateral side. However, progressively higher concentrations (1 and 10 ,ug; Table 1, Fig. 1) induced progressively less protein or alkaline phosphatase secretion, indicating that no dose-dependent relationship was demonstrable. No contralateral effect was produced at any challenge dose. 1251-BSA secretion was increased at the initial dose but not at the higher concentrations of methacholine. Albumin secretion was not significantly affected at any dose, and the albumin% remained constant throughout. When lower initial doses of methacholine were employed (Table 2), significant increases of total protein, albumin, and alkaline phosphatase activity (Fig. 2) in nasal secretions were induced with 0.01 pg of methacholine. Progressively higher concentrations (0.1, 1, and 10 pg) induced less total protein, albumin, and alkaline phosphatase activity in nasal secretions. No contralateral effects were noted, and no change in albumin% occurred. Repeated methacholine challenges (Table 3). Serial application of 1 pg of methacholine led to significant secretion of total protein, albumin, and alkaline phosphatase activity (Fig. 3) after only the initial challenge. In each of four subsequent methacholine provocations, the level of total protein, albumin, and alkaline phosphatase activity
I I
CONTROL
I -
IPSILATERAL SIDE CONTRAL4TERAL SIDE
2
3
METHACHOLINE
4
progressively returned to the baseline. Again, no contralateral effects were noted, and albumin% remained unchanged. Methacholine-atropine interaction (Table 4). The ability of atropine to antagonize the effect of methacholine was studied by instillation of 1 pg of atropine sulfate followed by 10 rug of methacholine and compared with the administration of methacholine alone. Coadministration of atropine ablated the methacholine-induced changes in total protein and albumin. Therefore the effect of methacholine was inhibitable by prior treatment with atropine. Atropine (1 pg) also completely abrogated the alkaline phosphatase secretion (Fig. 4) induced by 1 pg of methacholine. The same inhibitory effect of atropine was seen if 0.1 pg of methacholine was employed as the agonist. Alkaline phosphatase zymogram. Endogenous alkaline phosphatase activity was consistently identified in submucosal glands and ducts (Fig. 5B) as well as in the glycocalyx. Cartilage, perichondrium, and periosteum also possessed endogenous activity. Treatment of guinea pig turbinates with five doses of 1 pg of methacholine resulted in the depletion of alkaline phosphatase activity from the glands and glycocalyx but not from cartilage (Fig. 5C). Repeated methacholine challenge in humans (Figs. 6
and 7). In the human study, the effect of four sequential
BY z \ g w 2 1 g g k 5 3 2 5
b4
FIG. 3. Alkaline phosphatase activity in guinea pig nasal lavages after methacholine stimulation repeated 5 times. Numbers on abscissa indicate order of repeated challenge. *P < 0.01, compared with control.
240 220 200 180 160 140 120 100 80 60 40 20 0
ATROPINE(? ,ug) PRETREATMENT
CONTROL
***
I*0 Pg METHACHOLINE
FIG. 4. Alkaline phosphatase activity after methacholine (1 pg) stimulation after atropine pretreatment vs. methacholine (1 pg) alone. ***P < 0.001, compared with atropine treatment.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
2464
NASAL
TABLE
4. Comparison of the effect of topical atropine plus methacholine vs. methacholine alone Total A+M
Baseline Stimulated
2,625+711 3,583?522
Protein,
SECRETION
pglml
Albumin, M
A+M
2,041?298 4,250+422*
Values are means + SE. A, atropine; M, methacholine.
82219 127+13
pglml
Albumin% M
7729.3 183?23t
A+M
3.1f0.7 3.5+0.9
M
3.7kO.8 4.3kO.8
* P < 0.01; t P < 0.02, compared with baseline lavage (unpaired Student’s t test).
methacholine challenges was examined. The initial methacholine challenge caused the largest increased secretion of total protein (Fig. 6A), albumin (Fig. 6B), lysozyme (Fig. 6D), secretory IgA (Fig. 6F), nonsecretory IgA (Fig. 6G), lactoferrin (Fig. 6C), and IgG (Fig. 6E) compared with subsequent challenges. Repeated challenges with methacholine caused sustained secretion of all the glandular products (lysozyme, lactoferrin, and secretory IgA). By contrast, the secretion of the plasma proteins, albumin and IgG, returned to baseline levels by the fourth challenge. The level of nonsecretory IgA, which is also a plasma protein, remained minimally elevated through the series of challenges. Although expression of the secreted proteins in absolute amounts provided useful data, presentation of the results in relationship to total protein provided additional insights into the various contributions of glandular and vascular proteins (Fig. 7). Repeated methacholine challenges resulted in progressive decrements in albumin% (Fig. 7A) and IgG% (Fig. 70) below baseline values (both achieved significant reductions by the 4th challenge). In contrast, the ratios for glandular products (lactoferrin, lysozyme, and secretory IgA) were not significantly different from baseline after repeated challenges. DISCUSSION
Nasal secretory responses are coordinated by the autonomic nervous system (1,2,6,14,17,18) with the major effector control through cholinergic nerves (4,10,17). To help our understanding of nasal secretory responses, dependable in vivo models that allow quantitative measure-
ment of the secreted substances are very useful. The development of a guinea pig model of nasal secretion may prove helpful in this respect. The guinea pig model allows the collection of secretions from both sides of the nose without irritation. Repeated saline lavages do not affect the total protein and albumin content of secretions, indicating that neither the procedure nor the repeated washing by themselves has an effect on nasal mucosa secretion. Recovery of washing averages 85-100%. Muscarinic receptors are found on submucosal glands, blood vessels, and airway epithelium (9, 11, 17), and methacholine stimulates submucosal gland secretion, ion transport across the epithelial surfaces (9, 17, 18), and arteriolar vasodilation (1,17). Methacholine stimulation of the guinea pig mucosa provoked secretion of protein and albumin, whereas albumin% was not altered. These findings are similar to cholinergic responses in humans (17). Atropine, a potent cholinergic antagonist, abrogated the total secretory response induced by methacholine. Thus the response to topical methacholine in guinea pigs is mediated through muscarinic receptors. Alkaline phosphatase is an enzyme that hydrolyzes organic esters of phosphoric acid and transfers the inorganic phosphate group to other compounds (transphosphorylation) (3, 5, 8, 15, 19, 21). Histochemical studies employing alkaline phosphatase substrate revealed alkaline phosphatase activity in the submucosal glands of the guinea pig nasal mucosa. Immunohistochemical localization of alkaline phosphatase antigenic activity could not be determined because, to our knowledge, no suitable antiserum to guinea pig alkaline phosphatase exists. Alkaline phosphatase activity was recovered in nasal secre-
FIG. 5. A: normal guinea pig nasal turbinate mucosa. Beneath squamous epithelium (e) are many submucosal glands (g), vessels (v), and cartilage (c). Hematoxylin and eosin stain. B: alkaline phosphatase zymogram. Precipitated products from alkaline phosphatase (dark stain) are localized to glands (g), glycocalyx of epithelium (e), and cartilage (c). Vessels (v) did not contain detectable alkaline phosphatase activity. C: turbinates from animals treated with repeated doses of methacholine did not display alkaline phosphatase activity in their epithelium (e) and had reduced activity in glands (g), although cartilage (c) still possessed activity. Bars, 100 pm.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
04
I 1
0
80
70
1
I
2 YcHAupIoEs
I
B
3
4
B
2465
SECRETION
80
1
c
60 50 40 30 20 10 I
0
1
I
2 PCHAUPIGES
I
I
3
4
80
E 1
7oJ
609 50= 4oa 309
80~
F
709 609 5Om
FIG.
tory IgA patients.
6. Total protein, albumin, lactoferrin, lysozyme, immunoglobulin (non s-IgA) in baseline nasal lavages and after 4 sequential *‘P < 0.05; **P < 0.01; ***P < 0.001, compared with baseline.
(Ig) G, secretory IgA (s-IgA), and nonsecrebilateral 25mg methacholine challenges in 6
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
2466
0
207
u
1
2 # CHALlBlGES
3
4
1
2 3 Y CHAUENGES
4
SECRETION
c
20
F 3
FIG. 7. Results of repeated methacholine challenges expressed as albumin%, lactoferrin%, lysozyme%, IgG%, secretory IgA%, and nonsecretory IgA% (means t SE). *.P < 0.05; **Fp < 0.01, compared with baseline.
tions and probably represents glandular secretion. This conclusion is based on the following lines of evidence: 1) It is unlikely that serum alkaline phosphatase is transudated from plasma, inasmuch as methacholine does not cause vascular permeability because the albumin% did not change (Tables 1 and 2). Moreover, the levels of alkaline phosphatase in serum are only 400-550 pg/ml,
much the same levels present in nasal secretions. In contrast, albumin is found at 25 mg/ml in serum and only 250 pg/ml in secretions. Thus the ratio of plasma albumin to albumin in secretions indicates that serum proteins are diluted -IOO-fold in the lavage protocol. In addition, alkaline phosphatase activity is found diffusely in submucosal gland cells and cuboidal cells of the ducts. In
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
NASAL
guinea pigs, albumin is found in -10% of submucosal gland cells, whereas alkaline phosphatase activity was widely distributed in gland acini. In contrast, in human submucosal gland histochemistry, plasma proteins are never seen within gland cells. Instead plasma proteins are distributed throughout the interstitium and occasionally in gland lumina. 2) The dose-response effects of methacholine on the secretion of alkaline phosphatase activity have a pattern similar to the secretion of total protein. 3) Atropine completely abrogates the secretion of total protein and alkaline phosphatase activity. 4) Methacholine challenge appeared to deplete the glands of alkaline phosphatase activity. These data indicate that alkaline phosphatase is secreted as a glandular product from guinea pig nasal submucosal glands and suggest that it can be employed to monitor glandular secretion. In addition, these data suggest that glandular secretion in response to methacholine rapidly depletes the available alkaline phosphatase, resulting in rapidly decreasing quantities of this glandular product. The pattern of the initial methacholine challenge causing the most dramatic secretory response was seen whether total protein, albumin, or alkaline phosphatase was monitored. The dose-response experiments suggest that the guinea pig rapidly depletes its glands of protein, albumin, and alkaline phosphatase on initial stimulation, as reflected in an outpouring of secretions. This suggestion was confirmed by the loss of alkaline phosphatase activity in guinea pig nasal glands on histochemical analysis after methacholine stimulation. Perhaps the guinea pig’s nasal mucosa is designed for a rapid brief secretory response, which cannot be sustained. As an obligate nasal breather, it might be difficult to breathe if excessive secretion continued. Stimulation of cholinergic receptors in the human nasal mucosa has also been shown to be an important mechanism for producing glandular secretions (17). Earlier human studies have demonstrated that lysozyme, lactoferrin, and s-IgA present in nasal secretions come from glandular sources (17, 18). Both IgG and albumin have been found in periglandular interstitial spaces and in the lumen of submucosal glands (14). It is believed that methacholine-induced increases in IgG and albumin in secretions most likely represent transport of these proteins with glandular secretions. Methacholine has been demonstrated to selectively increase glandular secretion without affecting vascular permeability (17), but the resultant secretions always contain increased amounts of plasma proteins as well. Glandular secretions are relatively enriched in lysozyme, lactoferrin, and secretory IgA, whereas secretions resulting from increased vascular permeability are relatively enriched in albumin and IgG. To determine whether the pattern of repeated cholinergic stimulation inducing depletion of secreted proteins seen in the guinea pig had relevance to human secretory patterns, six volunteers were stimulated with repeated methacholine challenges. Much as seen in the guinea pig, the initial methacholine challenge in humans caused the greatest secretory response, but serial challenges continued to stimulate secretion of total protein, albumin, IgG, lysozyme, lactoferrin, and IgA. It was noted, however,
SECRETION
2467
that the secretion of the vascular proteins (albumin, IgG) returned to baseline after four challenges, while the glandular proteins (lysozyme, lactoferrin, and secretory IgA) remained increased. Nonsecretory IgA remained at a modestly increased level throughout. Taken as a percentage of total protein, the serial methacholine challenges resulted in a significant reduction for the two plasma proteins (albumin and IgG), suggesting that the glands became relatively depleted of the plasma proteins with repeated stimulation. These data indicate that a nasal secretory response can be studied in guinea pigs and that the response to cholinergic stimulation is primarily glandular. It was found that glandular secretions include the enzyme alkaline phosphatase. In contrast to the data reported here, other work with this model indicates that histamine causes a plasma protein-rich secretion (unpublished observations). The depletion of glandular responses seen in the guinea pig after repeated methacholine challenges was not seen in humans, although plasma proteins in human nasal secretions were progressively reduced. Thus one must conclude that the guinea pig provides a useful and easy model for the study of nasal physiology but that the results must be interpreted carefully and that extrapolations to human responses made cautiously. This study was supported in part by grants from the Merck Foundation and Merck Pharmaceuticals (A. 2. Gawin) and by fellowships from Proctor and Gamble, Inc, (B. E. Emery and J. N, Baraniuk). Address for reprint requests: A. 2. Gawin, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, Bldg. 10, Room 11-C-209, National Institutes of Health, Bethesda, MD 20892. Received 8 May 1990; accepted in final form 7 August 1991. REFERENCES A. The effect of parasympathetic nerve stimulation on the microcirculation and secretion in the nasal mucosa of the cat. Acta Otohyngol. 78: 98-105, 1974. BARNES, P. Neural control of human airways in health and disease. Am. Rev. Respir. Dis, 134: 1289-1314, 1986. BESSEY, 0. A., 0. H, LOWRY, AND M. J, BROCK. A method for the rapid determination of alkaline phosphatase with five cubic millimeters of serum. J. Biol. Chem. 164: 321-329, 1946. BORUM, P. Nasal methacholine challenge: a test for measurement of nasal reactivity. J. Allergy Clin. Immunol. 63: 253-257, 1979. BUURNE, G. H. Alkaline phosphatase in taste buds and nasal mucosa. Nature Lsnd. 161: 445-446,1948. DRUCE, H. M., R. H. WRIGHT, D. KOSSOF, AND M. KALINER. Cholinergic nasal hyperactivity in atopic subjects. J. Allergy Clin. hamunol. 76: 445-452,1985. GAWIN, A. Z., J. N. BARANIUK, AND M. A. KALINER. Guinea pig nasal mucosa secretory responses (Abstract). J. Allergy Clin. Immunoz. 83: 304, 1989. IIIZUMI, Y., K. HIRANO, M. SUGIURA, S. 11~0, H. SUZUKI, AND T. ODA. Evidence for the physiological function of alkaline phosphatase. Chem. Pharm. Butl31: 772-775,1983. JACKSON, R. T. Evidence for presynaptic parasympathetic receptors on nasal blood vessels. Ann. Otol. 91: 216-219, 1982. KLAASEN, A.B.,W. KNIJEPERS, I.F. RODRIGEZDEMIRANDA,AND A. J. BELD. The influence of neuropharmaca on the nasal glands. Rhinology 23: 191-194, 1985. KLAASEN, A. B., W. KNIJPERS, I. F. RODRIGEZ DE MIRANDA, AND J. BELD. Muscarinic receptors in rat nasal glands. Cell BioL. ht. Rep. 9: 521-521,1985. KONNO, A., AND K. TAGAWA. The mechanisms involved in onset of
1. ANGGARD,
2. 3. 4. 5. 6. 7. 8. 9. 10.
11.
12.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
2468
NASAL
SECRETION
allergic manifestation
in nasal allergy. Eur. J. Respir. Dis. 64, Suppt. 128: 155~166,1983. 13. LOWRY, 0. H., N. J. RUSENBROUGH, A. L. FARR, AND R. 9, RANDALL. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 165-1’75,195l. 14. MEREDITH, S. ID., G. II. RAPHAEL, J. N, BARANIUK, S. M. BANKS, AND M. A. KALINER. The pathophysiology of rhinitis. III. The control of IgG secretion. J. Allergy Clin. lmmunol. 84: 920-930, 1989. 15. MIKI, A., Y. TANAKA, S. OGATA, AND Y. IKEHARA. Selective prepa-
18. RAPHAEL, G. D., E. V. JENEY, J. N. BARANIUK, I. KIM, S. D. MEREDITH, AND M. A. KALINER. Pathophysiology of rhinitis. Lactoferrin and lysozyme in nasal secretions. J. Clin. Inuest. 84: 1528-1533, 1989. 19. ROSENBAUM, D., S. R. HIMMELHOCH, E. A. PATERSON, W. H. EVANS, AND M. G. MAGE. Purification of alkaline phosphatase from guinea pig bone marrow. Arch. Biochem. Biophys. 141: 303312, 1070. 20. SHASBY, M. D., AND R. L. ROBERTS. Transendothelial transfer of
ration and characterization of membrane soluble forms of alkaline phosphatase from rat tissue. A comparison with the serum enzyme.
macromolecules in vitro. Federation Proc. 46: 2506-2510, 1987. 21. SHELHAMER, J. M., 2. MAROM, AND M. KALINER. Immunologic and neuropharmacologic stimulation of mucous glycoprotein release from human airways in vitro. J. Clin. Invest. 66: 1400-1408, 1980. 22. SJOGREN, S., B. EK-RYLANDER, AND L. HAMMARSON. Characterization of alkaline phosphatase buccal mucosa in young rats. Acta
Eur. J. Biochem. 160: 41-48, 1986. 16. NAKAMURA, R., Z. OGITA, AND T. MATSUMARA.
Isozymes of alkaline phosphatase in gingiva and alveolar bone of the guinea pig.
Arch. Oral Bio!. 11: 131-133, 1966. 17. RAPHAEL, G. II., H. M. DRUCE, R.
H. WRIGHT, AND M. KALINER. Pathophysiology of rhinitis. I. Assessment of the sources of protein in induced nasal secretions. Am. Rev. Respir. Dis. 138: 413-420, 1988.
Odontol. 23. WANG,
Stand.
46: 75-82,
1988.
H. W., AND R. T. JACKSON. Do cholinergic neurons directly innervate nasal blood vessels? Rhinology 26: 139-141, 1988.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
