| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Microvascular Biology Group, School of Medical Sciences, RMIT University, Bundoora, Victoria 3083, Australia; and 2Department of Medical Physiology, College of Medicine, Texas A&M University, College Station, Texas 77843
Submitted 13 November 2002 ; accepted in final form 23 April 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
myogenic tone; calcium ion; cytoskeleton
Historically, many studies examining pathways activated by contractile agonists have been limited to relatively acute exposures to these agents, whereas studies aimed at delineating events related to compensatory growth responses, such as remodeling in hypertension, have often considered long-term responses to trophic stimuli including catecholamines and angiotensin II (4, 5). Furthermore, growth-related studies have often been performed on cultured cells, which typically exhibit phenotypic changes compared with cells in the in vivo state. It is likely, however, that rather than distinct mechanisms there exists a continuum of responses influenced by both the duration of exposure to the contractile stimulus and possibly the contractile response itself. It has therefore been suggested that the functional properties of smooth muscle are themselves altered by events such as reorganization of the cytoskeleton as the tissue undergoes adaptation to environmental factors (15). As an extension of this, we hypothesized that vascular remodeling may begin at a much earlier time frame and be initially characterized by alterations in cytoskeletal properties and/or the physical relationship between adjacent smooth muscle cells, this latter alteration being considered a property of cell-cell and cell-matrix interactions. Support for the occurrence of such rearrangements is provided by studies of cultured vascular cells exposed to agonist and mechanical stimuli (8, 34).
With respect to the above general hypothesis it would be expected that if arteriolar smooth muscle exhibits time frame-dependent changes in cytoskeletal or cellular arrangement this may be manifested in differences in the rate and extent of relaxation when comparing vascular responses to short- and long-term exposure to a contractile agonist. Therefore, the aim of the present studies was to compare acute (5 min) and prolonged (4 h) exposures to contractile stimuli and to examine the involvement of candidate signaling pathways.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
The studies used 59 male Sprague-Dawley rats weighing between 200 and 350 g. Before experiments, rats were housed in pairs in a dedicated animal facility with a 12:12-h light-dark cycle. During this period rats were allowed free access to a standard rat chow and drinking water. All procedures were approved by the Animal Ethics and Experimentation Committee at RMIT University.
Isolated Arteriole Preparation
Rats were anesthetized with Pentothal Sodium (100 mg/kg ip) after which the right and/or left cremaster muscle was exteriorized, excised from the animal, and placed in a cooled (4°C) chamber containing dissection buffer (in mM: 3 MOPS, 145 NaCl, 5 KCl, 2.5 CaCl2, 1 MgSO4, 1 NaH2PO4, 0.02 EDTA, 2 pyruvate, and 5 glucose with 1% albumin) (12). Segments (1.2–2.2 mm in length) of the main intramuscular arteriole (1A) were dissected from the muscle as previously described (26, 48). Vessel segments were then cannulated with glass micropipettes, secured with 10-0 monofilament suture, and mounted in a custom-designed tissue chamber (volume 5 ml). The vessel preparations were positioned on the stage of an inverted microscope. The cannulated arterioles were continually superfused (0.5–4 ml/min) with a physiological buffer solution containing (in mM) 111 NaCl, 25.7 NaHCO3, 4.9 KCl, 2.5 CaCl2, 1.2 MgSO4, 1.2 KH2PO4, 11.5 glucose, and 10 HEPES. Vessel segments were gradually pressurized to 70 mmHg and warmed to 34°C during a 60-min equilibration period. To be suitable for use in studies vessels had to be free of pressure leaks and develop spontaneous basal tone as evidenced by an active diameter between 40% and 65% of passive diameter at 70 mmHg.
Measurement of vessel diameter (in µm) was performed on-line with an electronic video caliper. Changes in vessel diameter after exposure to a given treatment were typically normalized to passive (maximally dilated with 0 mM Ca2+ buffer containing 2 mM EGTA) diameter at 70 mmHg to account for slight variation in absolute vessel diameters observed between preparations.
In experiments requiring measurement of changes in intracellular Ca2+, vessels were incubated (60 min, room temperature) with 2 µM fura 2-AM (Molecular Probes, Eugene, OR) in buffer containing 0.5% DMSO and 0.01% Pluronic. The abluminal surface of the vessel segment was exposed to the fura 2-AM solution to restrict dye loading to the vascular smooth muscle layer (26). The dye loading procedure was followed by a 30-min washout period with the physiological salt solution. Fura 2-loaded vessels were exposed to epiillumination (75-W xenon source) with light of alternating excitation wavelengths (340 and 380 nm) by using a motor-controlled filter wheel (20 Hz). Fluorescence emission was transferred from the microscope to a photomultiplier tube (Hamamatsu, Bridgewater, NJ) coupled to a signal processor (Texas A&M University, College Station, TX). Simultaneous transillumination with wavelengths >610 nm provided a nonfluorescent image that enabled measurement of internal arteriolar diameter while fluorescence data were collected. This procedure did not interfere with measurements of Ca2+-related fluorescence. Fluorescence emission intensities were expressed as the 340 nm-to-380 nm ratio to allow quantitative estimates of changes in arteriolar wall intracellular Ca2+. Details of these procedures have been presented in previous publications (17, 26, 48).
Experimental Protocols
Acute and prolonged exposure to norepinephrine. After baseline
diameter was established, arterioles were exposed to norepinephrine (NE, 5
µM) until a maximal contraction was achieved (
5 min). The superfusate
was then rapidly changed, and vessel diameter and time were recorded as the NE
was washed out. After return to basal diameter, or after 15-min superfusion
with fresh physiological salt solution (chosen as a predetermined surrogate
end point), vessels were subsequently exposed to adenosine (3 x
10–4 M) and then 0 mM Ca2+
buffer containing 2 mM EGTA to obtain a measurement of maximal passive
diameter. Vessels were then superfused with
Ca2+-containing buffer for 20 min to reestablish basal
conditions. The exposure to NE was then repeated with the exception that the
contraction was maintained for 4 h before washout as described above. Dilator
responses to adenosine and 0 mM Ca2+-2 mM EGTA buffer
were similarly repeated.
Time control studies were conducted to determine whether vessels responded similarly to acute NE exposures performed 4 h apart. A further set of control experiments were conducted to determine whether the initial acute exposure to NE exerted a preconditioning effect. In these experiments, relaxation responses to adenosine (10 µM) were compared before exposure to NE (that is, in the presence of spontaneous myogenic tone alone at 70 mmHg intraluminal pressure) and to a similar adenosine exposure after 4 h of NE (5 µM) exposure. NE was washed out, as for the above experiments, before addition of adenosine.
In a separate set of vessels (n = 7) loaded with fura 2 (2 µM), the above protocol was repeated to examine Ca2+ responses during the acute and prolonged NE exposures. Experiments were performed in both the presence and absence of fura 2 loading to determine any possible buffering effects due to dye loading.
Short- and long-term exposure to KCl. To examine the effect of contraction per se, the studies described in Acute and prolonged exposure to norepinephrine were repeated in a separate set of arterioles (n = 5), substituting KCl (75 mM) as the contractile agent.
Effect of microtubule depolymerization on arteriolar responses to acute and prolonged exposure to NE. Previous studies in our laboratories (37) showed modulation of arteriolar contractile responses to NE by the polymerization state of the microtubular network. Therefore, to examine the possible involvement of the microtubular network in the differences in contractile responses to acute and prolonged NE exposure, experiments were performed in the absence and presence of the depolymerizing agent demecolcine (10 µM). The response of arterioles to acute NE (5 µM) exposure was studied in the absence of the agent, after which vessels were exposed to demecolcine for 60 min and the acute NE response was repeated. The prolonged response (4 h) to NE was then determined in the continued presence of demecolcine. Conditions for these experiments were established in previous studies (37, 38).
Effect of tyrosine phosphorylation inhibition on arteriolar responses to acute and prolonged exposure to NE. To examine whether tyrosine phosphorylation-dependent events contribute to the differences in contractile responses to acute and prolonged NE exposure, experiments were performed in the absence and presence of genistein (3, 10, and 30 µM). The response of arterioles to acute NE (5 µM) exposure was studied in the absence of the inhibitor, after which vessels were exposed to genistein for 20 min and the acute NE response was repeated. The prolonged response (4 h) to NE was then determined in the continued presence of the tyrosine kinase inhibitor. Only one concentration of genistein was examined in any given arteriolar segment.
In addition to the studies of the nonspecific tyrosine kinase inhibitor genistein, studies were conducted in the presence of the Src inhibitor PP1 (10 µM) and the MEK inhibitor PD-98059 (50 µM) to prevent activation of p42/44 MAP kinase. Studies were performed according to the protocol used for genistein. The rationale for examining the effects of these inhibitors was based on studies implicating the involvement of Src and MAP kinase activation in both agonist- and pressure-induced smooth muscle responses (24, 25, 46).
Drugs and Chemicals
Buffer salts and chemicals were obtained from Sigma unless otherwise stated. NE (Arterenol) was dissolved in 1 mM ascorbic acid as a stock solution at 10–2 M and diluted in Krebs solution to 5 µM in 100-ml quantities as required throughout the protocol. Working solutions of NE were protected from light, and all solutions were kept below room temperature (either on ice or on a "cold" block). A stock solution of adenosine was prepared in Krebs solution at a concentration of 10–2 M and added directly to the arteriole bath (5 ml) to achieve a concentration of 3 x 10–4 M. PD-98059 (Biomol Research Laboratories, Plymouth Meeting, PA) dissolved in DMSO at a concentration of 10–1 M was stored at –20°C in 50-µl aliquots and diluted to 50 µM in Krebs solution containing DMSO (600 µl/100 ml) for each experiment. Frozen 100-µl aliquots of PP1 (Biomol Research Laboratories) in DMSO at a concentration of 10–2 M were diluted in Krebs solution to 10 µM for each experiment. Demecolcine dissolved in DMSO (1 mM) was further diluted in Krebs solution to 10 µM.
The concentrations of NE, adenosine, genistein, PD-98059, PP1, and demecolcine used were based on previous studies (26, 30, 32). In addition, PD-98059 was shown to inhibit phosphorylation of p42/44 MAP kinase in response to epidermal growth factor or increases in intraluminal pressure as determined by gel electrophoresis/Western transfer (data not shown).
Statistics and Data Handling
To account for differences in starting vessel diameters, data were normalized according to the following equation: diameter (% max) = (diameter under a given experimental condition/diameter in 0 mM Ca2+ buffer containing 2 mM EGTA) x 100. All measurements were taken at an intraluminal pressure of 70 mmHg, and maximal diameters were measured before prolonged NE exposure.
Differences in group data for the effect of a given treatment were determined with ANOVA. When ANOVA indicated an overall significant change, differences between individual groups were determined with the t-test. Results are presented as means ± SE, and differences were considered significant when P < 0.05 (n represents both the number of arteriole segments and the number of rats studied).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
Acute and Prolonged Exposure to NE
Acute exposure to NE (5 µM) resulted in a rapid contraction from 46.9 ± 3.9% to 23.9 ± 1.7% of maximal diameter, which was subsequently maintained during the prolonged exposure [26.9 ± 1.7% of maximal diameter at 4 h; not significant (NS) compared with acute exposure] (Figs. 1 and 2). Despite reaching a similar level of vasoconstriction, subsequent washout of the mechanical response with fresh buffer was significantly (P < 0.05) delayed in arterioles subjected to the prolonged NE exposure. In 10 arterioles acutely exposed to NE, baseline diameter was reached in 2 min, 137 ± 45 s after washout was initiated with fresh buffer. In contrast, 7 of the 10 vessels remained constricted for periods of >15 min when washout followed the prolonged (4 h) NE exposure. After 4-h exposure to NE arterioles also exhibited an impaired relaxation response to topically applied adenosine (3 x 10–4 M), although full relaxation was attained on removal of extracellular Ca2+ (Figs. 1 and 3).
|
|
|
Control experiments comparing acute responses to NE at time 0 and
4 h later indicated that the delayed relaxation following the prolonged NE
exposure was not due to time-dependent changes in the preparation independent
of the action of the 
-agonist. Similarly, an additional series of
vessels (n = 6) were studied to examine the effect of 4-h NE exposure
in the absence of an acute preexposure to the agonist to exclude any effect
resulting from preconditioning. Consistent with impaired relaxation after
prolonged exposure to NE, after 4-h superfusion with 5 µM NE vessels
dilated to 78.3 ± 3.7% of maximal diameter in response to adenosine
(10–4 M) compared with 92.2 ± 0.9% before
NE exposure (P < 0.05). As above, arterioles exposed to NE for 4 h
dilated to maximum when superfused with 0 mM Ca2+-2 mM
EGTA buffer.
In contrast to NE, exposure to KCl (75 mM) did not impair subsequent relaxation responses (Fig. 2). Furthermore, unlike the NE-induced responses, KCl contractions tended to wane over the 4-h time period (Fig. 2).
Vessels loaded with the Ca2+-sensitive dye fura 2
showed a similar impairment in relaxation after prolonged NE exposure. NE (5
µM) caused a transient increase in Ca2+ that declined
toward baseline over the period of prolonged agonist exposure despite the
maintained vasoconstriction. Comparison with time controls not exposed to NE
demonstrated that after 30- to 60-min exposure to the -agonist
intracellular Ca2+
levels were not
statistically different from baseline (Fig.
4).
|
The Ca2+ response did not appear to be a function of preexposure to NE or time, as peak changes were 178.9 ± 7.3% (relative to values in 0 Ca2+ buffer) during the first NE treatment compared with 183.0 ± 10.9% in response to the second NE exposure (NS). Similarly, after 5-min exposure to NE Ca2+ had plateaued to 139.4 ± 4.2% and 134.2 ± 6.3% (NS), respectively.
Effect of Microtubular Depolymerization
One-hour demecolcine (10 µM) treatment caused a decrease in baseline diameter and potentiated NE-induced constriction as seen in our previous studies (Refs. 36, 37; Table 1). Despite the presence of the microtubule depolymerizing agent, arterioles maintained the prolonged NE constrictor response and continued to show the impaired relaxation on washout of the agonist (Fig. 5; Table 1).
|
|
Effect of Tyrosine Kinase Inhibitors
The initial series of studies examined the effect of the general tyrosine kinase inhibitor genistein at 3, 10, and 30 µM. Genistein caused a concentration-dependent loss of basal tone but did not significantly inhibit the acute contractile response to NE (5 µM) (Table 1). In contrast, genistein treatment resulted in a concentration-dependent loss of the NE-induced constriction over time despite the continued presence of the agonist (Fig. 6; Table 1).
|
The Src inhibitor PP1 (10 µM; n = 4) did not affect either the
basal level of myogenic tone or the acute contraction to NE
(Fig. 6B;
Table 1). Similarly to
genistein, however, vessels treated with PP1 could not maintain the
constrictor response to NE over the 4-h time period
(Fig. 6B;
Table 1). At 1 h into the
4-h time course arterioles had returned to control diameters. The MEK
inhibitor PD-98059 (50 µM; n = 6) similarly showed little effect
on the spontaneous level of myogenic tone or the acute contractile response to
NE (Fig. 6B;
Table 1). However, in the
presence of PD-98059 contraction to NE waned over the 4-h exposure period
(Fig. 6B;
Table 1).
Time-dependent inhibition of the 4-h NE constriction by the tyrosine phosphorylation inhibitors prevented the possibility of determining the time to washout. Despite this, sufficient tone remained to assess adenosine (3 x 10–4 M)-induced dilation (Fig. 7). Control vessels exposed to NE for 4 h showed an impaired dilator response to adenosine compared with the acute NE exposure, whereas vessels treated with the inhibitors did not show an attenuated adenosine response (Fig. 7).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
Control experiments were performed to demonstrate that changes in the ability to relax after 4-h NE treatment were not a function of either time or a preconditioning phenomenon resulting from acute (5 min) exposure to the agonist. Thus vessels showed reproducible contractile responses to acute NE exposure applied at 4-h intervals, and, furthermore, washout/relaxation was unaffected by this time period. The lack of a preconditioning effect was shown by impaired relaxation to adenosine after 4-h continuous exposure to NE in a group of vessels not subjected to any preexposure to the agonist. It was therefore concluded that the impaired relaxation response was indeed a function of the prolonged agonist exposure.
Remodeling has been used to describe, in particular, structural vascular adaptations whereby alterations in the wall-to-lumen ratio are evident. Such remodeling can be hypotrophic, eutrophic, or hypertrophic depending on the growth response of the medial layer (27–29). These classifications have resulted from the results of many earlier studies, especially those related to changes in resistance vessel structure occurring in hypertension (1, 3, 14) and states of altered blood flow (6, 23). In the case of eutrophic remodeling, it is believed that an initial period of active vasoconstriction is followed by rearrangement of normal smooth muscle cells to maintain a reduced diameter. Although not directly shown, it is conceivable that the results of the present study are consistent with an early stage in such a process. Interestingly, similar signaling pathways (i.e., tyrosine phosphorylation events involving cSrc and MAP) have been implicated in both structural remodeling and the functional remodeling demonstrated in the present studies (for example, see Refs. 20, 24, 25, 45). Similarly, support for early agonist-induced remodeling events in tracheal smooth muscle involving tyrosine phosphorylation has been provided by Gunst and colleagues (15, 42).
A possible explanation for the delayed relaxation response to the
longer-term exposure to NE is that agonist binding initiates a pathway(s), in
addition to Ca2+-calmodulin-myosin light chain
kinase-mediated activation of myosin, that operates in a time-dependent manner
to influence the contraction/relaxation process. Such pathways may require
either a longer onset period or time for washout, resulting in an altered time
course compared with those simply stimulated by a rise in
(as might be expected for
KCl-induced activation). An obvious class of candidate mechanisms involves
processes that alter the standard relation between Ca2+
and contraction, such as Ca2+ sensitization
(40,
41).
Ca2+ sensitization has been demonstrated to occur in
arteriolar smooth muscle and may involve tyrosine phosphorylation, because
tyrosine phosphatase inhibition by pervanadate causes constriction without a
measurable change in
(31).
In addition to a Ca2+ sensitization process, a number
of vasoactive agonists induce cytoskeletal rearrangement in various smooth
muscle preparations (19,
20,
44,
45) and modulate cell
migration and attachment (20).
Of possible relevance to our study, many of these events have been shown to be
dependent on tyrosine phosphorylation-dependent pathways. Of specific
relevance to the present studies the 1-adrenoceptor agonists
NE and phenylephrine have been shown to activate tyrosine phosphorylation
events including those involving focal adhesion kinase, cSrc, and p42 MAP
kinase (2,
21,
47). Such studies have
implicated these signaling events in the adrenergic stimulation of
non-selective cation currents and contraction and mitogenic responses.
An interesting observation in the present study was that arteriolar
constriction to KCl (75 mM) was not maintained throughout the 4-h observation
period. Despite the continued presence of KCl, arterioles had returned to
control diameters by 120 min. This may indicate that maintained long-term
constriction requires the involvement of signaling mechanisms not activated by
simple membrane depolarization. Similarly in the presence of the nonselective
tyrosine kinase inhibitor genistein (10 and 30 µM), the NE constriction
tended to wane after 60–90 min. This occurred although genistein
did not significantly affect the acute contraction to NE (see
Table 1). Thus it is tempting
to speculate that tyrosine phosphorylation-mediated events are required for
maintenance of prolonged arteriolar constriction. Further studies are required
to determine whether these events reflect the action of additional parallel
contractile pathways (for example, through thin filament regulation,
Ca2+ sensitization, or cytoskeletal rearrangement) or
perform a support role in maintaining the contractile pathways.
An obvious question relates to the physiological significance, or advantage, of an apparent time-dependent switch in the mechanisms underlying a maintained arteriolar contraction. It is conceivable that these additional pathways provide a mechanism by which there is maintained tension development at comparatively low energy cost. Such an argument was previously forwarded for the relevance of Ca2+ sensitization (40). Regardless of the underlying mechanism, this may confer an energetic advantage to smooth muscles such as are evident in the arteriolar wall, which exhibit a high level of maintained tone.
An additional consideration is that the functional and growth responses to
vasoconstrictors are often modulated by the generation of paracrine factors
that exert either a positive or a negative effect on the vasomotor event. Of
relevance to the results of the present studies, DeFily et al.
(11) suggested that a
component of the prolonged in vivo 1-adrenergic-mediated
constriction of canine coronary arterioles results from the release of
endothelin. Endothelin has been reported to modulate
Ca2+ sensitivity via tyrosine phosphorylation-dependent
(35,
39) and -independent
(13) mechanisms in arterial
smooth muscle. Furthermore, prolonged in vivo administration of NE induces
remodeling of small arteries via a process dependent on endothelin
(9). Consistent with these
observations, involvement of endothelin in the studies of DeFily et al. was
confirmed by experiments demonstrating that either an ETA-receptor
antagonist or an inhibitor of conversion of prohormone to active endothelin
prevented the sustained constriction to phenylephrine. Interestingly, these
investigators had previously shown
(22) that under in vitro
conditions canine coronary arterioles did not respond to
1-agonists, suggesting that in vivo the source of required
endothelin was not restricted to the wall of the arteriole under study. It
appears likely that in the in vivo situation 1-agonist
stimulation of cardiac myocyte endothelin production contributed
(42). With respect to the
present data it was felt unlikely that in the absence of significant
parenchymal tissue, and in the presence of a superfusion system, sufficient
quantities of a vasoconstrictor would accumulate to explain the prolonged
vasoconstrictor response.
In summary, the results of the present study demonstrate that prolonged exposure of arterioles to the contractile agonist NE leads to the stimulation of tyrosine phosphorylation-dependent mechanisms that impair relaxation on washout of the agonist. Furthermore, the prolonged (over a 4-h period) NE-induced contraction was totally prevented by inhibitors of cSrc, MEK, and general tyrosine phosphorylation, suggesting that such events are critical to force maintenance in arterioles. It is suggested that these observations are consistent with an early phase of the remodeling process and, in particular, represent a functional adaptation whereby force is maintained by mechanisms supporting, or acting in parallel to, Ca2+-mediated contraction. Future experiments are required to determine whether these results are applicable to prolonged myogenic constriction resulting from a sustained increase in intraluminal pressure. In addition, to make these observations relevant to common clinical situations, such as hypertension, interactions that occur between mechanical and neurohumoral stimuli should be considered.
|
DISCLOSURES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
1-adrenergic constriction of coronary arterioles.
Am J Physiol Heart Circ Physiol
276: H1028–H1034,
1999.-Adrenergic responses of isolated
coronary microvessels. Basic Res Cardiol
90: 61–69,
1995.[ISI][Medline]
1-adrenergic constriction.
Circulation 98:
9–12, 1998.[Medline]
1A-adrenergic and growth factor receptors in transfected
PC12 cells. Biochem J 344:
889–894, 1999.[ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  B. E. Spurrell, T. V. Murphy, and M. A. Hill Intraluminal pressure stimulates MAPK phosphorylation in arterioles: temporal dissociation from myogenic contractile response Am J Physiol Heart Circ Physiol, October 1, 2003; 285(4): H1764 - H1773. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, Time
controls (i.e., not exposed to NE; n = 4). See
