Articles in PresS. Am J Physiol Regul Integr Comp Physiol (September 4, 2013). doi:10.1152/ajpregu.00025.2013
1
1
2
Protection against high intravascular pressure in giraffe legs
*
KK Petersen1, *A Hørlyck1, KH Østergaard2,3, J Andresen4, T Broegger4, N Skovgaard2, N
3
Telinius4, I Laher5, MF Bertelsen6, C Grøndahl6, M Smerup7, NH Secher8, E Brøndum4, JM
4
Hasenkam7, T Wang2, U Baandrup3, C Aalkjaer4
5
1) Department of Radiology, Aarhus University Hospital, Skejby, Denmark
6
2) Zoophysiology, Department of Bioscience, Aarhus University, Aarhus, Denmark
7
3) Department of Pathology, Vendsyssel Hospital and Center for Clinical Research, Aalborg University, Denmark
8
4) Department of Biomedicine, Aarhus University, Aarhus, Denmark
9
5) Department of Pharmacology and Therapeutics, University of British Columbia, Vancouver, BC, Canada
10
6) Center for Zoo and Wild Animal Health, Copenhagen Zoo, Frederiksberg, Denmark
11
7) Department of CardioThoracic and Vascular Surgery and Institute of Clinical Medicine, Aarhus University Hospital,
12
Skejby, Denmark
13
8) Department of Anesthesiology, The Copenhagen Muscle Research Center, Rigshopitalet, University of Copenhagen,
14
Denmark
15
*
16
KK Petersen, A Hørlyck, NH Secher, T Wang: Ultrasound and in vivo pressure recordings
17
KH Østergaard, U Baandrup, C Aalkjær: Histology
18
J Andresen, T Broegger, N Skovgaard, N Telinius, I Laher, C Aalkjær: Isolated small arteries tested in
19
vitro
20
MF Bertelsen, C Grøndahl, E Brøndum: Anesthesia, handling of giraffes
21
M Smerup, M Hasenkam: Surgery
22
All authors: Discussion of results, editorial work on the manuscript
Contributed equally
23
24
Running head: Arteries in giraffe legs
25
Corresponding author: Christian Aalkjaer, Department of Biomedicine, Aarhus University, Ole Worms
26
Alle 4, 8000 Aarhus C, Denmark; e-mail: [email protected]
Copyright © 2013 by the American Physiological Society.
2
27
Abstract
28
The high pressure in giraffe leg arteries renders giraffes vulnerable to edema. We investigated in 11
29
giraffes whether large and small arteries in the legs and the tight fascia protect leg capillaries.
30
Ultrasound imaging of foreleg arteries in anesthetized giraffes and ex vivo examination revealed
31
abrupt thickening of the arterial wall and a reduction of its internal diameter just below the elbow. At
32
and distal to this narrowing, the artery constricted spontaneously and in response to norepinephrine
33
and intravascular pressure recordings revealed a dynamic, viscous pressure drop along the artery.
34
Histology of the isolated median artery confirmed dense sympathetic innervation at the narrowing.
35
Structure and contractility of small arteries from muscular beds in the leg and neck were compared.
36
The arteries from the legs demonstrated an increased media thickness to lumen diameter ratio,
37
increased media volume, and increased numbers of smooth muscle cells per segment length and they
38
furthermore contracted more strongly than arteries from the neck (500 ± 49 vs. 318 ± 43 mmHg, n=6
39
legs and neck, respectively). Finally the transient increase in interstitial fluid pressure following
40
injection of saline was 5.5 ± 1.7 times larger (n=8) in the leg than in the neck. We conclude that 1)
41
tissue compliance in the legs is low; 2) large arteries of the legs function as resistance arteries; and 3)
42
structural adaptation of small muscle arteries allow them to develop an extraordinary tension. All
43
three findings can contribute to protection of the capillaries in giraffe legs from a high arterial
44
pressure. (246 words)
45
Keywords: smooth muscle, sympathetic innervation, ultrasound, sphincter, resistance artery
46
structure
3
47
Introduction
48
The giraffe is the mammal with the highest mean blood pressure of approximately 200 mmHg and
49
because of the height of the animal, its lower limbs are at least potentially exposed to a pressure of
50
about 350 mmHg (8; 16). To the extent that this pressure is transmitted to the capillaries, it would
51
likely provoke interstitial edema and yet that does not manifest in the lower leg of the giraffe. Several
52
mechanisms are suggested to prevent edema in the lower limb of the giraffe. The tight fascia in the
53
limb of the giraffe may act as an anti-gravity suit (8; 19) but tissue compliance in the lower limb has
54
not been determined and the interstitial fluid pressure in the lower limbs of a standing giraffe is only
55
44 mmHg (8). Secondly, thickened walls of resistance arteries in giraffe limbs could protect against the
56
pressure load that distal capillaries in the giraffe might be exposed to (7). However, when giraffe
57
limb arteries become smaller than 400 μm, they are reported to be without thickened walls (7) and it
58
appears that no structural differences are present between arteries supplying the skin of the leg and
59
the neck (12). These findings are, however, based on histological sections of arteries not exposed to a
60
well-defined mechanical stress during fixation. It remains therefore not known whether in the giraffe,
61
structure of small arteries is different in the legs and the neck when evaluated under well-defined
62
mechanical conditions. Finally, there is a sphincter-like structure in the isolated anterior tibial artery
63
of the giraffe (9; 16) but whether it provides a relevant dynamic viscous resistance and thus regulates
64
distal vascular pressure in the lower leg is not known.
65
This study aims to provide quantitative assessments of three mechanisms proposed important for
66
prevention of edema in the lower leg of the giraffe. Thus the study evaluated whether: 1) there is a
4
67
sphincter like structure in the median artery of the foreleg, as has been reported for the hindleg
68
(6,13), and whether it affects pressure in the lower leg; 2) whether the structure of small arteries in
69
the leg of giraffes is similar to that of the small arteries of the neck when examined under
70
mechanically well defined conditions; 3) whether tissue compliance is smaller in the lower leg than in
71
the neck of the giraffe.
72
Methods
73
Eleven male giraffes belonging to a sub-species (Giraffa camelopardalis giraffa) incompassing at least
74
12.000 animals were studied at an age between 3 and 5 years after being bred in privately owned
75
parks. The lengths of the forelegs (hoof to shoulder joint) and hindlegs (hoof to hip joint) were
76
measured in 10 giraffes. The distance from ground to the sphincter was measured in 4 hindlegs and 6
77
forelegs. Approval for the experiments was obtained from the Animal Ethics Screening Committee at
78
the University of Witwatersrand (Johannesburg) and the Animal Use and Care Committee at
79
University of Pretoria, South Africa. Local animal ethical committee members supervised the
80
experiments and the Gauteng Province of South Africa granted permission for euthanasia.
81
Anesthesia
82
Following overnight deprivation of food and water, giraffes were premedicated with medetomidine
83
(5.5 µg/kg i.m.) and approximately 10 min later anesthesia was induced with etorphine (6.5 µg/kg
84
i.m.) and ketamine (0.65 mg/kg, i.m.). A rope connected to a halter and a pulley allowed control of the
85
head when the giraffes became recumbent after 3-7 min. The giraffes were intubated immediately (ID
86
20 mm) for assisted ventilation with oxygen using a demand valve (Hudson RCI, Pennsylvania).
5
87
Following induction of anesthesia, animals were moved to an adjacent room and placed in right
88
lateral recumbence with the head and neck elevated. Anesthesia was maintained by i.v. α-chloralose
89
(15 mg/ml, KVL-pharmacy, Denmark; 30 mg/kg/h) – a dose that was gradually reduced depending on
90
reflexes and breathing pattern. Anesthesia was monitored using ECG, invasive and non-invasive
91
arterial blood pressure, and recording of the end-tidal CO2 tension (PM9000Vet; E-Vet, Haderslev,
92
Denmark). Additionally, arterial and venous blood gas variables were measured at regular intervals
93
(GEM Premier 3500, Instrumentation Laboratory, Bedford, Massachusetts). Unless otherwise stated
94
the anesthetized giraffes were maintained in the upright position by placing the limbs into straps
95
suspended from the ceiling, leaving the thoracic and abdominal regions free of external pressure as
96
described previously (4).
97
Ultrasound of the foreleg
98
The luminal diameter of the median artery and the thickness of the intima-media complex were
99
determined on both forelegs by ultrasound (GE Healthcare, Vivid I, Israel) using a 12-MHz linear
100
transducer after shaving the medial aspect of the forelegs immediately distal to the elbow by
101
experienced radiologists. Flow velocity was determined by Pulsed Doppler.
102
Foreleg arterial and interstitial fluid pressures
103
Using ultrasound guidance and Seldinger technique, two 20 G catheters were placed in the
104
median artery; one proximal to the sphincter and one about 20 cm proximal to the hoof.
105
Pressures were recroded with Edwards, Baxter disposable pressure transducers (Baxter
106
Healthcare Corporation, Irving, California) and a monitor (Dialogue 2000, IBC-Danica,
6
107
Copenhagen, Denmark), and sampled using a Biopac 100 data acquisition system (Goleta,
108
California). The contribution from gravity to the pressure difference between the two catheters
109
was eliminated. This was done by subtracting the hydrostatic pressure calculated from the
110
distance from the heart to the catheters from the readings.
111
Leg and neck interstitial fluid pressures were measured with an Arrow 14 G catheter (Teflex Inc.,
112
Pennsylvania) that was advanced 15 cm below the fascia or by a 18 G catheter (Periflex, Braun,
113
Melsungen, Germany) that was advanced 8 cm by a Tuohy needle. The Tuohy needle was retracted
114
and the catheters, which have side holes, were left in place for pressure measurements. The
115
transducer reading was set to zero, where after tissue compliance was assessed from the pressure
116
rise following injection of 3 ml saline through the catheter. We verified that the skin surrounding the
117
catheter remained tight when subjected to a pressure of up to 350 mmHg following injection of
118
additional 3 ml of saline.
119
Isolated small arteries
120
Following euthanasia by pentobarbital (50 mg/kg, i.v.), muscle samples from the hind limbs (about 90
121
cm below the heart), the neck (about 30 cm (low neck) and about 130 cm (high neck) above the heart)
122
were obtained together with skin biopsies from the hind limbs (about 140 cm below the heart), and
123
small arteries were dissected under a stereomicroscope. Two arteries (in a few cases only one artery)
124
that were closely size-matched according to lumen diameter were dissected from each muscle biopsy
125
and mounted in a myograph for isometric force recording: arteries were threaded on two wires and
126
the bath temperature was set to 370 C. In one set of experiments the media thickness was measured
7
127
(1) in the arteries from the muscle biopsies. On the basis of the passive length-tension relationship
128
(14), the internal circumference (L100) that the relaxed artery would attain at a transmural pressure of
129
100 mmHg was determined for all arteries. The corresponding diameter (l100) was L100 divided by π.
130
The arteries were set (normalized) to a diameter corresponding to 0.9 x l100. Arteries with measured
131
media thickness were stimulated three times with 10 μM norepinephrine (NE) in K-PSS (NEK; for
132
composition of K-PSS see below) for two minutes at approximately 10 min intervals, after which
133
response curves to increasing NE concentration and acetylcholine (arteries preconstricted to about
134
80% of the maximal response to NE) were obtained. Following this protocol, the arteries were fixed in
135
4% formaldehyde for histology.
136
In another set of experiments two arteries from each site were used to determine the length-tension
137
relationship by stimulating with NEK at different vessel circumferences. The passive (unstimulated)
138
and total tensions were recorded and, by subtraction, active tension was determined at each vessel
139
circumference. After normalization of vessel diameter, the arteries from the skin were stimulated
140
three times with NEK (as in the arteries from the muscles).
141
We dissected median artery segments at the sphincter region from 3 giraffes and they were mounted
142
as ring segments in a myograph for isometric force recording. The diameters were increased to obtain
143
a passive force of about 50-100 mN after which the arteries were stimulated with NEK.
144
The responses were expressed as a) active wall tension, which is the force measured divided by 2 x
145
segment length, b) active media stress, the active wall tension divided by the media thickness, and c)
8
146
as equivalent active pressure, the active wall tension divided by the arterial radius as a measure of the
147
transmural pressure the artery would be able to constrict against.
148
The physiological salt solution (PSS) used for the in vitro experiments contained (in mM): NaCl 119,
149
KCl 4.7, KH2PO4 1.18, MgSO4 1.17, NaHCO3 25, CaCl2 1.6, EDTA 0.026, and glucose 5.5. The solution
150
was gassed with 5% CO2 in air and adjusted to a pH of 7.4. K-PSS was PSS in which NaCl was
151
substituted with KCl on an equimolar basis.
152
Histology
153
Arteries (7 from the median artery, 10 from tibial artery) from 9 giraffes were obtained. The sphincter
154
area was located and 8 transverse sections of the arteries were cut from 5 cm proximal to 5 cm distal
155
to this area, with approximately 1 cm between each block. The blocks were fixed in 4% buffered
156
formaldehyde, embedded in paraffin and two 3 µm sections were cut from each block. For
157
immunohistochemical detection of nerves, one section was stained with S100 (1:2000 polyclonal Z
158
0311, protein kinase K S3020, DAKO, autostainer LAB VISION) and one with anti-tyrosine-hydroxylase
159
(TH) (1:2500, Abcam, ab112). The sections were examined using an Olympus MVX10 macroscope
160
equipped with a camera connected to a PC. For the number and area of nerves in the arteries, the 2D
161
nucleator software from newCAST (Visiopharm, Copenhagen, Denmark) was used.
162
The small muscular arteries from the leg and upper part of the neck were embedded in plastic and
163
two longitudinal sections produced. The stereological technique “disector” (13) provided an unbiased
164
estimate of smooth muscle cell (SMC) volume, number per segment length, length, cross sectional
9
165
area and cell layers. The estimates of the variables listed are based on the assumption that each SMC
166
contains only one nucleus.
167
Statistics
168
When several arteries from the same giraffe biopsy were investigated, the mean value was taken to
169
represent data from that giraffe. The small arteries from the muscle biopsies (leg, proximal neck,
170
distal neck) were compared with ANOVA. When the ANOVA was significant a test for a linear trend
171
between the values at the three sites was performed (GraphPad Prism 5). The effect of NE on
172
median/tibial artery structure and pressure differences between the catheters as well as the indices
173
of innervation were tested with a paired t-test. For tissue pressures the values at the lowest and
174
highest positions were compared with a paired t-test. Data are presented as mean±SE with n
175
indicating the number of giraffes and p-value <0.05 was considered statistically significant.
176
Results
177
Ultrasound and pressure recordings of the median artery
178
The length of giraffe forelegs (hoof to shoulder joint) was 206±19 cm and of the hindlegs (hoof to hip
179
joint) 191±14 cm, n=10. The median artery (the extension of brachial artery) was visualized by
180
ultrasonography (Fig. 1). The lumen diameter was reduced and a thick intima-media complex became
181
apparent about 5-10 cm below the elbow (Figs. 1 and 2). This structural change took place over less
182
than 1 cm and was maintained in the distal part of the artery. Figure 1C shows the isolated median
10
183
artery where the structural change can be seen. The mean distance from ground to sphincter was
184
134±8 cm (n=6, forelegs) and 123±13 cm (n=4, hindlegs)
185
With ultrasound we observed what appeared to be a fully dilated artery (Fig. 1A), but on one occasion
186
the artery was constricted (Fig. 1B) and the media-intima complex thickness increased indicating a
187
spontaneous constriction. Pressure recordings revealed a pressure difference between the distal and
188
proximal catheters of about 10 mmHg when the artery was dilated (Fig. 3). Sometimes a
189
spontaneous, transient (lasting 5-15 min) increase in the pressure difference was seen (Fig. 3A). Such
190
a transient was seen in the artery which was constricted based on the ultrasound recording (Fig. 3A)
191
supporting the conclusion that the artery can constrict spontaneously. During an orthostatic challenge
192
no consistent changes in the pressure difference or artery diameter were apparent.
193
Intra-arterial injection of 1 μg NE proximal to the elbow had no effect on the sphincter but in 2 out of
194
7 cases 10 μg NE caused a constriction. Intra-arterial injection of 100 μg NE, however, caused
195
constriction in all cases (Figs. 2 and 4). Constriction was on occasions restricted to the sphincter area
196
(Fig. 2B), while in most cases the area distal to the sphincter also constricted (Fig. 2C and
197
Supplementary video 1). Within 10 sec after injection of NE, the mean flow velocity in the artery
198
decreased to almost zero (Supplementary video 1) and then slowly returned to re-establish pre-
199
injection values over 10-15 min as the artery constriction waned. On the other hand, there was no
200
constriction in response to administration of NE proximal to the sphincter, no constriction in the
201
contralateral foreleg and no effect on the blood pressure indicating no systemic effect.
11
202
Intra-arterial NE injection was associated with an approximately 2 min decrease in the pressure
203
difference between the two sites of recording in 6 of 7 giraffes (Fig. 3), followed by an increase to an
204
average of about 30 mmHg (Figs. 3 and 4). The increase of the pressure difference developed slowly
205
(over 5-10 min) and then slowly waned but closely followed the reduction of the lumen diameter and
206
thickening of the media-intimal complex (Fig. 3B).
207
Histology of median/tibial arteries
208
Staining of the arteries revealed TH positive (Figs. 5A and C) and S100 positive cells at the sphincter,
209
with very little staining proximal to the sphincter (Fig. 5B). The number of TH positive structures in the
210
most proximal slice, i.e. proximal to the sphincter, was 0.8 ± 0.2, while in a slice at the sphincter it was
211
10.2 ± 1.4 (Fig. 6A). The area of TH positive structures at the proximal slice was 2.0x10-3 ± 0.3x10-3
212
mm2, while at the sphincter it was 32x10-3 ± 7x10-3 mm2 (Fig. 6B).
213
In the S100 stained arteries the number of nerves proximal to the sphincter was 0.9 ± 0.2 while at the
214
sphincter it was 17.2 ± 3.2 (Fig. 6C). The corresponding nerve areas were 2x10-3 ± 0.4x10-3 mm2 and
215
40x10-3 ± 8x10-3 mm2 (Fig. 6D). All indices of innervation were significantly greater at the sphincter
216
compared to proximal to the sphincter.
217
Interstitial fluid pressure in the foreleg
218
The interstitial fluid pressure in the hindleg (20-40 cm above the hoof) in four giraffes suspended in an
219
upright position was 47±10 mmHg. In eight additional giraffes the interstitial fluid pressure was
220
measured (Fig. 7A) at 4 positions along the foreleg (13-15, 30-40, 60-80 and 120-150 cm from the
12
221
hoof) and in two positions in the neck (approximately 200 and 350 cm from the hoof) with giraffes in
222
right recumbent position on an oblique table with the head placed on a 70 cm high support, such that
223
the head was about 150 cm above the hoofs. The interstitial fluid pressure varied from 20-25 mmHg at
224
the two lowest positions and was close to zero at the other positions (Fig. 7B). Thus, the pressures at
225
the lowest and highest positions were significantly different. At the same sites injection of 3 ml saline
226
caused pressure to increase for 5-20 sec. After the transient increase the pressure returned to a value
227
close to the preinjection value, probably as a consequence to saline spreading into the tissue, which
228
supports that interstitial fluid pressure was recorded by our approach. The peak pressure increase
229
was larger at the lowest position compared to the highest position (Fig. 7C).
230
Function of isolated arteries
231
Arteries with a normalized internal diameter of about 200 μm were dissected from muscle biopsies.
232
Arteries were easy to distinguish from the accompanying veins in the biopsies from all three sites.
233
The small arteries responded to NEK with rapid force development reaching a steady state force
234
within 10-15 sec (Fig. 8A). In contrast, the median artery constricted much slower and the time from
235
application of NEK to reaching steady state force varied from 2-3 to 10 min or more (Fig. 8B).
236
The length tension curves for small muscular arteries are shown in figures 8C and D. These curves
237
demonstrate that with the circumference set to 0.9xL100, all arteries developed near maximal active
238
tension.
239
The morphological characteristics of the small arteries from the muscles are shown in figure 9.
240
Although the arteries dissected had the same normalized diameter (Fig. 9A), the media thickness
13
241
increased inversely with the distance from the ground. The media was about twice as thick (and with
242
a significantly increased media area; not shown) in arteries from the leg compared to those from near
243
the head (Fig. 9B), resulting in a significantly increased media thickness to lumen diameter ratio in the
244
arteries near the ground (Fig. 9C).
245
Morphometric analysis indicated that the large media area in the arteries from the legs resulted from
246
an increased number of smooth muscle cells per segment length. Smooth muscle cells were organized
247
in an increased number of layers, with no increase in size (length and cross sectional area) of each cell
248
(Table 1).
249
The high active tension developed in response to NEK in arteries from the lower leg was related to
250
increased media thickness as there was no increase in media stress development to NEK in these
251
arteries compared with the arteries from the neck (Fig. 10A). In contrast, the equivalent active
252
pressure was larger in arteries closer to the ground, indicating that the arteries from the limbs could
253
constrict against transmural pressures of more than 500 mmHg and significantly more than the
254
arteries from the upper part of the giraffe (transmural pressures about 300 mmHg) (Fig. 10B).
255
All arteries from the leg muscles responded to NE with an EC50 between 1 and 5 μM. In contrast, only
256
50% of the arteries from the neck developed tension to cumulative NE concentrations. For those
257
arteries that developed tension in response to administration of NE, the EC50 also varied between 1
258
and 5 μM. All the arteries relaxed to increasing concentrations of acetylcholine (most of the arteries
259
relaxed fully) with EC50 between 30 to 100 nM.
14
260
Skin arteries with internal diameters larger than 500 μm ran parallel to and just outside the dense
261
connective tissue of the skin. Branches of these arteries running obliquely into the skin were
262
dissected and mounted in myographs. The arteries had a mean diameter of 363 ± 43 μm (n=8) and
263
produced a maximal tension of 4.70 ± 0.83 N/m, which is equivalent to an effective pressure of 190 ±
264
28 mmHg. It was not possible to visualize the media in the myographs as was the case for arteries
265
from the muscular bed.
266
Discussion
267
Of all mammals, giraffes have the highest blood pressure, which combined with their extreme height,
268
challenges the cardiovascular systems and especially so in their lower extremities. In anaesthetized
269
giraffes, similar to the giraffes used in the present study, we have reported a mean blood pressure of
270
193 mmHg in the upright position (2) and a mean arterial pressure of up to 350 mmHg in the lower
271
limbs (13), which are values similar to values reported in awake giraffes (5). Such a pressure might
272
lead to edema, but healthy giraffes do not experience edema in their lower limbs.
273
A large viscous resistance in the arteries would reduce the capillary pressure and the likelihood of
274
edema development in the lower legs of giraffes. A sphincter-like structure in the (anterior) tibial
275
artery, immediately beneath the knee, has been reported in isolated arteries (9; 16). Whether a
276
similar sphincter like structure also exists in the forelegs and whether it provides a significant viscous
277
resistance was not known previous to this study. In addition, a substantial precapillary viscous
278
resistance could be developed by a large wall thickness to lumen diameter ratio in small limb arteries.
279
A large wall thickness to lumen diameter ratio is established in limb small arteries of the giraffe, but
15
280
apparently not in arteries with diameters less than 400 μm and not in those at the level of the ankle
281
(7). These measurements were obtained without definition of the mechanical strain the arteries were
282
exposed to during measurements though. Here we determined wall thickness of arteries <400 μm
283
outer diameter in giraffe legs exposed to a well-defined strain. We tested the hypothesis that the
284
structural and/or functional characteristics of small arteries in the legs were different to the
285
corresponding variables in small arteries in the neck and could contribute to protection of leg
286
capillaries against a high transmural pressure. Low tissue compliance in the lower leg can also protect
287
against edema formation. But this variable has not previously been assessed in giraffe legs. We
288
therefore tested the hypothesis that tissue compliance is lower in legs than in the neck of the giraffe.
289
Large arteries in giraffe legs function as resistance arteries
290
By ultrasonography we visualized thickening of the arterial wall and a distinct intima-media complex
291
with a reduction in the lumen diameter over a short distance below the elbow. Thus we provide in
292
vivo evidence for a structural adaptation in the vasculature of forelegs akin to that observed in
293
isolated arteries from the hind legs of giraffes (9; 16) and the finding was confirmed in isolated
294
arteries. The ultrasound evaluation also revealed spontaneous contraction of the artery at the site
295
where the arterial lumen became small, and we demonstrated a dose-dependent constriction to NE
296
that often extended to the distal part of the median artery. Importantly this structural feature
297
contributed viscous resistance to blood flow since varying pressure differences between the proximal
298
and distal sites of the median artery developed concomitant with a reduction in flow when the artery
299
constricted either spontaneously or in response to NE.
16
300
Without activation of the sphincter and the thick median artery distal to the sphincter, the viscous
301
pressure drop between the two catheter sites was about 10 mmHg, which is insufficient to account
302
for the pressure increase from the proximal to the distal site (the contribution from gravity between
303
the two sites is about 90 mmHg). Even after maximal activation with NE, the viscous pressure drop
304
was at most 70 mm Hg (and on average 25 mmHg), implying that although the structural modification
305
of the large leg arteries could reduce the high pressure in the legs, it is insufficient alone to prevent
306
the high arterial pressure in the legs, consistent with previous pressure recordings (8; 16).
307
The initial decrease of the pressure difference between the proximal and distal site after injection of
308
NE was associated with an almost complete cessation of blood flow, suggesting that the resistance
309
arteries distal to the distal catheter constricted more rapidly than the median artery, and so indicating
310
an extraordinarily powerful regulatory capacity of the leg arteries. This suggestion derived from
311
measurements showing that the distal arteries constricted faster than the median arteries, an
312
observation that was supported by the much faster force development of small muscular arteries
313
compared to the median artery observed in vitro. However, we did not visualize the dynamics of a
314
spontaneous constriction in vivo to see whether this was equally slow in the median arteries. Arteries
315
from giraffes have a rich sympathetic innervation (10; 15) as confirmed here by staining for the
316
sympathetic nerve marker tyrosine hydroxylase in the adventitia and the staining was most
317
pronounced near the sphincter, suggesting that the sphincter is richly innervated by sympathetic
318
nerve fibers. In support, staining of Schwann cells with S100, followed a similar pattern and confirms
319
previous observations on the hind leg of giraffes (9; 16). Our findings thus suggest that giraffes can
17
320
increase viscous resistance in the main arteries of the legs by sympathetic activation and hence
321
reduce capillary pressure in the lower legs.
322
Structural adaptation of resistance arteries in the leg has functional consequences
323
Our functional data on small muscular arteries in the legs of the giraffes provide further evidence for
324
their role in protection against interstitial edema in the lower leg. We compared the structure and
325
function of small arteries from muscular tissue at three levels above the hoof of the giraffe. Media
326
thickness correlated with the distance of the artery from the ground as did the ratio of media
327
thickness to lumen diameter, lending additional support to the hypothesis that vascular structure
328
varies with transmural pressure. The increased media thickness (and media volume per segment
329
length) was due to an increased cell number organized in a greater number of layers in the arteries
330
from the lower leg of the giraffe, while cell length and cross sectional area were similar at all 3
331
positions from which the arteries were obtained. This hyperplastic response in arteries exposed to a
332
high transmural pressure is different from that seen in human essential hypertension, for which an
333
eutrophic increase in media thickness to lumen diameter ratio is associated with unchanged smooth
334
muscle numbers and size (11). A similar wall-to-lumen ratio of skin arteries from the leg and neck of
335
giraffes was reported by Mitchell and Skinner (12), suggesting that small arteries of the skin do not
336
adapt structurally when exposed to the high pressures in the legs. In agreement with this notion is our
337
observation that small arteries of the leg skin had low contractility compared to those from the
338
skeletal muscles of the leg. The skin arteries may be protected from the high blood pressure in the
339
lower leg of giraffes by the stiff collagen content of the skin.
18
340
Arteries from giraffes studied in vitro constricted to NE and relaxed to acetylcholine, demonstrating
341
that these regulatory functions are maintained in small arteries from giraffes. We observed that only
342
some arteries from the neck responded to NE (although they all constricted to K-PSS), suggesting that
343
sympathetic innervation is more pronounced in the legs of giraffes than in the neck. The ability of the
344
arteries to constrict against a transmural pressure was related to their distance from the ground, i.e.
345
arteries from the legs were able to constrict against a higher transmural pressure than those from the
346
neck. This increased contractility of the arteries from the legs was not caused by stronger constriction
347
per smooth muscle cross-section per se, since there was no change in media stress and distance from
348
the ground and is explained by the altered vascular morphology. Furthermore, length-tension curves
349
confirmed that arteries from the 3 sites were investigated under conditions where maximal active
350
force production was obtained. We therefore conclude that small arteries with diameter < 400 μm
351
are structurally changed and that this has functional consequences which are likely to contribute to
352
the high precapillary resistance, which explains a cardiac output that is not larger than that of animals
353
of similar weight and a high blood pressure (5). We also suggest that these structural characteristics
354
make it possible for NE to almost completely stop blood flow in the legs as reported here.
355
Low tissue compliance in the leg
356
In addition to providing a means of limiting edema formation due to high arterial pressure, it is
357
important to note that blood vessels in giraffe legs also possess a thick basement membrane (21),
358
appear impermeable to protein (8), and are enclosed within relatively thin (compared to the trunk)
359
but inflexible skin (12) due to a high collagen density (18). These features are likely to all contribute to
19
360
preventing edema formation in the lower legs (8). However, the functional consequence of an
361
inflexible skin, i.e. subcutaneous compliance has not been assessed previously. The antigravity suit-
362
like function of the tight skin in the leg (8) is strongly supported by the finding that injection of a small
363
volume of saline into the interstitial space caused larger pressure increases in the leg than in the neck
364
of giraffes. The low compliance that we demonstrate is likely to contribute to the reported large
365
fluctuations in interstitial fluid pressure in the leg of walking giraffes (8).
366
Limitations
367
The evaluation of the sphincter function and its response to NE were made in anesthetized giraffes
368
and it is important to confirm the pressure recordings in the legs of free-ranging animals.
369
It may be controversial to measure interstitial pressure by a perforated catheter in order to assess
370
vascular transmural pressure and tissue compliance. Guyton et al. (6) suggested that interstitial
371
pressure represents the interstitial fluid pressure and the pressure exerted by solid tissue elements
372
that together constitute the total tissue pressure exerting the outside pressure on the vessel walls.
373
Later, Aukland and Reed (2), however, concluded that a distinction between the interstitial fluid
374
pressure and the pressure in solid tissue elements is irrelevant since solid tissue pressure is likely to
375
be zero. The interstitial fluid pressure of importance is the pressure measured with “fluid
376
equilibration techniques”, i.e. in a saline-filled column brought in contact with the interstitium that
377
represents the filling pressure of the lymph vessels. In their giraffe study, Hargens et al. (8) used wick
378
catheters to measure interstitial fluid pressure. With this technique, the wick secures contact with
379
the interstitium, but this group also demonstrated that a solid state microtransducer inserted into the
20
380
tissue (17) provides, over a range of pressures similar results as techniques involving wicks (20). Thus
381
a wick is not needed to increase the contact area with interstitial fluid. Leakage in the recording
382
system might have resulted in loss of continuity in the saline-filled column between the interstitium
383
and the transducer, but for all experiments it was secured that there was no leakage. Taken together,
384
these considerations support that the catheters used in this study recorded the interstitial fluid
385
pressure.
386
We report tissue compliance as ΔV/ΔP, where ΔV was the injected volume of isotonic saline (3 ml)
387
and ΔP is the resulting immediate increase in interstitial fluid pressure. Since the local interstitial fluid
388
volume was not determined, ΔV/ΔP is not the true compliance of the interstitium (for discussion see
389
(20) since ΔV/ΔP is influenced by the ability of the tissue to distribute volume vs. structures
390
restricting tissue expansion including the tight skin ensheathing the limb of the giraffe as, e.g. in the
391
rat tail (3). True tissue compliance could be provide by inflating a balloon within the interstitium but
392
ΔV/ΔP provides at least some information on the counter-pressure exerted by the interstitium to
393
volume expansion at different levels of the giraffe and thereby an indication for how edema is
394
prevented in their lower leg.
395
It should be further stressed that our measurements are made in anaesthetized giraffes and although
396
the interstitial fluid pressure we measure is similar to that measured in awake giraffes (8) it would be
397
important to obtain compliance measurements also in awake giraffes.
398
399
21
400
Conclusions
401
Vascular adaptations in the giraffe limbs appear to protect distal capillaries against high hydrostatic
402
pressures. The large leg arteries of the giraffe have a pronounced and richly, sympathetically
403
innervated narrowing that can regulate viscous resistance to the extent that it eliminates 20%, or
404
more of the pressure facing the distal arteries. In addition, isolated small resistance arteries
405
demonstrate structural adaptations that enable them to contract against transmural pressures
406
reaching 500 mmHg, or more than reported for any isolated resistance artery in the animal kingdom.
407
The steady state interstitial fluid pressure e of 40-50 mmHg in combination with low tissue compliance
408
in the legs confers vascular protection. However, the role of these adaptations for hemodynamic
409
control in free-ranging giraffes has yet to be examined.
410
Acknowledgement: We would like to thank all members of the DaGir expedition and Dr Charles van
411
Niekerk and Wildlife Assignment International, South Africa for assistance with numerous details in
412
relation to handling of giraffes, Helge Wiig, Department of Biomedicine, University of Bergen, Bergen,
413
Norway for valuable discussions on interstitial pressure and Jane Rønn, Aarhus University, Denmark
414
for assistance with stereology on small arteries.
415
Grants: The project was supported by grants from Aarhus University, Aarhus University Hospital
416
(Skejby), Danish Heart Foundation, Lundbeck Foundation, Aase og Ejner Danielsen Foundation,
417
Carlsberg Foundation
418
Disclosures: None
22
419
420
421
Reference List
422
423
1. Aalkjaer C, Heagerty AM, Petersen KK, Swales JD and Mulvany MJ. Evidence for increased media
424
thickness, increased neuronal amine uptake, and depressed excitation--contraction coupling in isolated
425
resistance vessels from essential hypertensives. Circ Res 61: 181-186, 1987.
426
427
428
429
2. Aukland K and Reed RK. Interstitial-lymphatic mechanisms in the control of extracellular fluid volume.
Physiol Rev 73: 1-78, 1993.
3. Aukland K and Wiig H. Hemodynamics and interstitial fluid pressure in the rat tail. Am J Physiol 247: H80H87, 1984.
430
4. Brondum ET, Hasenkam JM, Secher NH, Bertelsen MF, Grondahl C, Petersen KK, Buhl R, Aalkjaer C,
431
Baandrup U, Nygaard H, Smerup M, Stegmann F, Sloth E, Oestergaard KH, Nissen P, Runge M,
432
Pitsillides K and Wang T. Jugular venous pooling during lowering of the head affects blood pressure of
433
the anesthetised giraffe. Am J Physiol Regul Integr Comp Physiol 297: R1058-R1065, 2009.
434
435
436
5. Goetz RH, Warren JV, Gauer OH, Patterson JL, Jr., Doyle JT, Keen EN and McGregor M. Circulation of the
giraffe. Circulation 8: 1049-1058, 1960.
6. Guyton AC, Granger HJ and Taylor AE. Interstitial fluid pressure. Physiol Rev 51: 527-563, 1971.
23
437
438
439
440
441
442
443
444
445
446
447
448
7. Hargens AR, Gershuni DH, Danzig LA, Millard RW and Petterson K. Tissue adaptations to gravitational
stress: Newborn versus adult giraffes. The Physiologist 31, Suppl. 1: S-110-S-113, 1988.
8. Hargens AR, Millard RW, Pettersson K and Johansen K. Gravitational haemodynamics and oedema
prevention in the giraffe. Nature 329: 59-60, 1987.
9. Kimani JK, Mbuva RN and Kinyamu RM. Sympathetic innervation of the hindlimb arterial system in the
giraffe (Giraffa camelopardalis). Anat Rec 229: 103-108, 1991.
10. Kimani JK and Opole IO. The structural organization and adrenergic innervation of the carotid arterial
system of the giraffe (Giraffa camelopardalis). Anat Rec 230: 369-377, 1991.
11. Korsgaard N, Aalkjaer C, Heagerty AM, Izzard AS and Mulvany MJ. Histology of subcutaneous small
arteries from patients with essential hypertension. Hypertension 22: 523-526, 1993.
12. Mitchell G and Skinner JD. Giraffe thermoregulation: a review. Transactions of the Royal Society of South
Africa 59: 109-118, 2004.
449
13. Mulvany MJ, Baandrup U and Gundersen HJ. Evidence for hyperplasia in mesenteric resistance vessels
450
of spontaneously hypertensive rats using a three-dimensional disector. Circ Res 57: 794-800, 1985.
451
14. Mulvany MJ and Halpern W. Contractile properties of small arterial resistance vessels in spontaneously
452
hypertensive and normotensive rats. Circ Res 41: 19-26, 1977.
24
453
454
15. Nilsson O, Booj S, Dahlstrom A, Hargens AR, Millard RW and Pettersson KS. Sympathetic innervation of
the cardiovascular system in the giraffe. Blood Vessels 25: 299-307, 1988.
455
16. Ostergaard KH, Bertelsen MF, Brondum ET, Aalkjaer C, Hasenkam JM, Smerup M, Wang T, Nyengaard
456
JR and Baandrup U. Pressure profile and morphology of the arteries along the giraffe limb. J Comp
457
Physiol B 181: 691-698, 2011.
458
459
460
461
17. Ozerdem U. Measuring interstitial fluid pressure with fiberoptic pressure transducers. Microvasc Res 77:
226-229, 2009.
18. Sathar F, Ludo BN and Manger PR. Variations in the thickness and composition of the skin of the giraffe.
Anat Rec (Hoboken ) 293: 1615-1627, 2010.
462
19. Warren JV. The physiology of the giraffe. Sci Am 231: 96-100, 105, 1974.
463
20. Wiig H and Swartz MA. Interstitial fluid and lymph formation and transport: physiological regulation and
464
465
466
467
468
469
roles in inflammation and cancer. Physiol Rev 92: 1005-1060, 2012.
21. Williamson JR, Vogler NJ and Kilo C. Regional variations in the width of the basement membrane of
muscle capillaries in man and giraffe. Am J Pathol 63: 359-370, 1971.
25
470
471
Figure captions
472
Fig. 1 A and B. Ultrasonography imaging of the sphincter region of the median artery of the giraffe.
473
(A) no apparent contraction (B) spontaneous contraction that is most pronounced distal to the
474
sphincter region. The composite images (indicated by broken lines in A and B) are composed of scans
475
obtained at adjacent positions along the leg with the ultrasound transducer (which has a limited field
476
of view) (C) the isolated median artery.
477
478
Fig. 2. Ultrasonography imaging of the sphincter region of the median artery of the giraffe. (A)
479
control; (B) and (C) after injection of 100 µg norepinephrine.
480
481
Fig. 3. (A and B) Hydrostatic pressure immediately proximal to the sphincter region (top, proximal) of
482
the median artery of a giraffe and at a location about 20 cm above the hoof (middle, distal). The
483
contribution to hydrostatic pressure from the effect of gravity from the heart to the catheter tip was
484
subtracted from the recorded pressures, so that differences between the recordings from the two
485
catheters represent a viscous pressure drop between the catheters. The lower curve (pressure
486
difference) shows the distal minus the proximal pressures (a measure of the viscous pressure drop
487
between the two catheters). (A) The values indicate the internal diameter of the median artery
488
measured with ultrasonography distal to the sphincter region at the two time points indicated by the
26
489
arrows. (B) Arrows indicate intra-arterial injections of norepinephrine just proximal to the sphincter
490
region. (C) Records (typical of 7 experiments) of the distal minus the proximal pressure of the median
491
artery (pressures recorded as described in (3B)), and proximal median artery lumen diameter and
492
intima-media complex thickness. At time zero, 100 µg norepinephrine was injected intra-arterially.
493
Fig. 4. Diameter (A) and intima-media complex thickness (B) of the sphincter region of the median
494
artery of the giraffe before, 5-10 min after (at maximal contraction) and 20-30 min after (fully relaxed)
495
intra-arterial injection of 100 µg of norepinephrine just proximal to the sphincter. (C) shows the
496
pressure approximately 20 cm above the hoof minus the pressure immediately proximal to the
497
sphincter region. The pressure difference between the two points due to gravity was subtracted to
498
obtain a value that was solely dependent on viscous resistance between the two points. The p values
499
indicate the level of statistical significant differences between the points in the presence of
500
norepinephrine and the mean of the values before and after norepinephrine.
501
Fig. 5. Tyrosine hydroxylase staining of the median artery at the sphincter location (A and C) and 5 cm
502
proximal to the sphincter location (B) of the giraffe foreleg. The innervation was almost absent at
503
proximal sites. The darkly stained nerve structures are highlighted in (C).
504
Fig. 6. Number (A and C) and area (B and D) of nerves before (slice 1-4), at (slice 5-6), and after (slice
505
7-8) the sphincter location of the giraffe foreleg. Slices were cut about 1 cm apart. The number of
506
nerves with tyrosine hydroxylase (TH) (A) and S100 staining (C) had a similar pattern with an abrupt
507
increase within the sphincter region. The same pattern was also seen when the area of nerves was
508
measured on TH (B) and S100 stained tissues (D). Horizontal lines are SEM.
27
509
Fig. 7. (A) Trace showing measurement of interstitial fluid pressure and compliance under the skin in
510
the lower leg of a giraffe. At the arrows 3 ml saline was injected. (B) Interstitial fluid pressure in 8
511
giraffes. The pressure was measured at 6 positions as defined in the text (1 being closest to the
512
ground). (C) Peak increases in tissue pressure to injection of 3 ml saline at the different positions in 8
513
giraffes. Vertical lines indicate SEM.
514
Fig. 8. Recordings of force development to 10 µM norepinephrine in a high K+ containing solution
515
(NEK) of (A) leg small muscular artery of the giraffe and (B) median artery at the sphincter region of
516
the foreleg. Note the different ordinate scaling of the two traces. Passive (C) and active (D) length-
517
tension relationships of small arteries from the leg muscles (circles, punctuate line; n=4), lower neck
518
(diamonds, hatched line; n=3) and upper neck (squares, full line; n=3). Tension is presented relative to
519
the passive and active (C and D, respectively) tension at 0.9xL100 (L100 is the circumference the relaxed
520
artery would attain at a transmural pressure of 100 mmHg). Bars indicate SEM.
521
Fig. 9. (A) Diameter (ANOVA: non-significant for position), (B) media thickness (ANOVA: p<0.01 for
522
position), and (C) media thickness to lumen diameter ratio (ANOVA: p<0.01 for position) of arteries
523
from muscular beds in the leg, proximal neck, and upper neck as indicated. Each point represents the
524
mean of two arteries from one giraffe and the horizontal lines indicate mean and SEM.
525
Fig. 10. (A) Media stress produced by arteries from the muscles of the giraffe leg (ANOVA: non-
526
significant). (B) The transmural pressure against which arteries from muscular beds in the leg,
527
proximal neck, and upper neck would be able to constrict against (ANOVA: p<0.05 for position). Each
28
528
point represents the mean of two arteries from one giraffe and the horizontal lines indicate mean and
529
SEM.
530
Supplementary video 1
531
29
532
Table 1. Morphological characteristics of smooth muscle cells from resistance arteries in the lower
533
leg and upper part of the neck of giraffes.
534
Cell cross
Cells per unit
Leg arteries
sectional
Cell volume
length
Cell length
area
μm3
μm-1
μm
μm2
2462±178
5.88±0.55
95.7±13.4
29.2±4.0
4.19±0.75
2425±353
2.98±1.03
85.5±6.9
27.6±2.3
2.24±0.12
ns
P<0.05
ns
ns
P<0.05
Cell layers
Upper part of
neck
p-value
535
536
Vaules are mean±SEM
A
B
C
1 cm
1 cm
Fig. 1
A
B
C
1 cm
Fig. 2
Proximal
pressure
mmHg
275
225
175
125
Distal
pressure
mmHg
275
225
175
125
B
Proximal
pressure
mmHg
5 min
Distal
pressure
mmHg
4.8 mm
Pressure
difference
mmHg
1.7 mm
20
-20
-60
-100
Pressure
difference
mmHg
6
C
Δpressure ((mmHg)
0
diameter
4
-20
Δpressure
2
-40
Intima-media thickness
-60
0
-5
Fig. 3
0
5
10
time (minutes)
15
20
Lumen diametter/intima-media
thickne
ess (mm)
A
C
A
-10
4
P<0.0001
2
0
B
P=0.0165
Δ pressure (mmHg)
Diameter (mm
m)
6
Before
norepinephrine
Max. effect of
norepinephrine
After
norepinephrine
-30
-50
-70
Before
norepinephrine
Intima-media tthickness (mm)
2
1
P<0.0001
0
Before
norepinephrine
Max.
Max effect of
norepinephrine
After
norepinephrine
Fig. 4
Max. effect of
norepinephrine
After
norepinephrine
C
A
Fig. 5
10 —m
100 μm
B
100 μm
Fig. 6
A
B
C
D
B
Interstitial fluid
d pressure (mm
mHg)
Interstitial fluid
d pressure (mm
mHg)
A
200
150
100
50
0
48 2 min 50
46
52
54
40
30
20
10
0
Near ground
Upper neck
-10
1
2
3
4
5
Position along the giraffe
Δinterstital fluid pressure (mmH
Δ
Hg)
C
175
150
125
100
75
50
25
Near ground
Upper neck
0
1
2
3
4
5
Position along the giraffe
6
Fig. 7
6
A
B
NEK
20 mN
50 mN
5 minutes
Fig. 8
NEK
5 minutes
C
D
60
40
Relative tensio
R
on
Relative tensio
R
on
1.2
1.2
20
0.8
0.8
0.4
0.8
0.0
0.0
0
0.7
0.9
1.1
1.3
R l ti circumference
Relative
i
f
((xL
L100)
Fig. 8
0.7
0.9
1.1
1.3
R l ti circumference
Relative
i
f
((xL
L100)
Fig. 9
pp
er
N
ec
k
ec
k
0
U
10
al
N
20
Le
g
30
Media/lumen
40
xi
m
N
ec
k
ec
k
B
Pr
o
pp
er
50
al
N
100
Le
g
150
Mediia thickness (μm))
200
U
ec
k
g
rN
ec
k
al
N
pp
e
m
Le
250
xi
m
U
ox
i
300
Pr
o
Pr
Diameter (μm)
D
A
C
0.3
0.2
0.1
0.0
Fig. 10
U
pp
er
n
ec
k
ec
k
200
Lo
w
er
n
400
Equivale
ent pressure (mmHg)
600
Le
g
Le
Pr
g
ox
im
al
N
ec
k
U
pp
er
ne
ck
Media
a stress (m
mN/mm 2)
A
B
800
600
400
200
0