Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 589 (Pt 19), 4681-96

Temporal Changes in Microvessel Leakiness During Wound Healing Discriminated by in Vivo Fluorescence Recovery After Photobleaching


Temporal Changes in Microvessel Leakiness During Wound Healing Discriminated by in Vivo Fluorescence Recovery After Photobleaching

Maria J C Machado et al. J Physiol.


Regeneration of injured tissue is a dynamic process, critically dependent on the formation of new blood vessels and restructuring of the nascent plexus. Endothelial barrier function, a functional correlate of vascular restructuring and maturation, was quantified via intravital microscopic analysis of 150 kDa FITC-dextran-perfused blood vessels within discrete wounds created in the panniculus carnosus (PC) muscle of dorsal skinfold chamber (DSC) preparations in mice. Time to recovery of half-peak fluorescence intensity (t(1/2)) within individual vessel segments in three functional regions of the wound (pre-existing vessels, angiogenic plexus and blind-ended vessels (BEVs)) was quantified using in vivo fluorescence recovery after photobleaching (FRAP) and linear regression analysis of recovery profiles. Plasma flux across the walls of new vessel segments, particularly BEVs, was greater than that of pre-existing vessels at days 5-7 after injury (P < 0.05). TNP-470 reduced the permeability of BEVs at the leading edge of the advancing vascular plexus as measured by the decrease in luminal t(1/2) (P < 0.05), confirming the utility of FRAP as a quantitative measure of endothelial barrier function. Furthermore, these data are suggestive of a role for TNP-470 in selection for less leaky vascular segments within healing wounds. Increased FITC-dextran leakage was observed from pre-existing vessels after treatment with TNP-470 (P < 0.05), consistent with induction of transient vascular damage, although the significance of this finding is unclear. Using in vivo FRAP this study demonstrates the relationship between temporal changes in microvascular macromolecular flux and the morphology of maturing vascular segments. This combination of techniques may be useful to assess the therapeutic potential of angiogenic agents in restoring pre-injury levels of endothelial barrier function, following the establishment of a functional vascular plexus such as in models of wounding or tumour development.


Figure 1
Figure 1. Examples of vessel segments chosen for FRAP imaging
At 5 days after thermal injury and following intravenous injection of 150 kDa FITC-dextran, the PC muscle wound site and surrounding vasculature in a CD1 mouse was imaged by low magnification epifluorescence microscopy (2× objective: A) and individual vessel segments identified for bleaching via confocal microscopy (10× objective: BD). In the micrographs, the wound boundary is demarcated by a dotted line (in A, B and C). With the use of confocal microscopy, blind-ended vessels (BEVs) were identified immediately around the edge of the wound and chosen for bleaching (rectangular dashed ROI in A and B). Vessel segments from the angiogenic plexus were identified immediately behind the leading edge of the regenerating vascular plexus, at a distance of 500–750 μm from the wound centre (rectangular dashed ROI in A and C). At a distance of 1000–1250 μm from the wound centre, pre-existing vessel segments (rectangular dashed ROI in A and D) were also bleached. Scale bar, 1000 μm (A), 100 μm (B) and 250 μm (C and D).
Figure 2
Figure 2. Representative bleach series in a BEV at 6 days after injury
A confocal image series of a growing vascular plexus within a DSC at day 6 after injury in a male CD1 mouse. A single representative vessel segment within a ROI (dashed box) was chosen for bleaching. The pre-bleaching series (A) comprises 10 frames acquired at 1 s intervals (from t = 0 s to t = 9 s) and sets the baseline fluorescence intensity. The dashed box indicates the individual BEV used for FRAP analysis. During the bleaching sequence (B), the image was zoomed in on the chosen region of interest and laser intensity augmented as indicated in the Methods section. The bleach was performed from t = 10 to 16 s and 10 frames captured during the sequence with a minimized time interval (0.393 s) between successive frame captures. C, a series of images was acquired every second post-bleaching (up to 120 s) by focusing out on the identical image area as the pre-bleach sequence. Scale bar in A and C, 100 μm; B, 25 μm. Offline analysis of the BEV segment allowed two ROI to be traced (D): one limiting the border of the blood vessel (black arrow indicates the line demarcating the BEV inlet) and one 10 pixels distal from vessel lumen (white line) which delimited a strip of the interstitium used to calculate fluorescence recovery outside the vessel. Average pixel intensity (0–255 arbitrary units), taken as a determinant of brightness of fluorescence, was measured within the two ROIs over the frame sequence (E) of the BEV in D. The intensity of fluorescence in the vessel lumen is represented by the filled circles and in the adjacent interstitium by the open circles. F, normalized fluorescence intensity (Fn) of each frame of the post-bleach sequence in the vascular lumen region of the same BEV was plotted as a function of τ (logarithm of base 10 of time in seconds). Linear regression of the points in this plot (dashed-dot grey line) was used to estimate the value of the x-intercept when Fn = 0.5 (or τ1/2) with a 95% confidence interval.
Figure 3
Figure 3. Analysis of microvascular morphological parameters in a wound at 6 days post-injury
The geometric centre of the wound (0) was determined and counts performed at 250 μm intervals, as previously described (Machado et al. 2011). White rectangles indicate the distance markers (0, 250 and 500 μm in this micrograph) in images selected for analysis. Vessel width was quantified by measuring the diameter of the closest five perfused vessels to the marker (asterisks) at each sampling distance. Scale bar, 250 μm.
Figure 4
Figure 4. A temporal series of confocal images from a single animal showing vessel growth into the centre of a wounded PC muscle
Microvessels run parallel to myofibres in uninjured panniculus carnosus (PC; A), where 150 kDa FITC-dextran labels flowing vessels. At 15 min following thermal injury (B), prominent FITC-dextran leakage is observed (star). Blind-ended vessels (BEVs, arrows) can be observed by day 5 (C), in addition to dilated large-calibre vessels and numerous irregular vessel segments close to the edge of the wound. By day 6 post-injury (D), BEVs are closer to the wound centre and new anastomoses are formed (arrowhead), whilst by day 7 (E), FITC leakage has diminished. Scale bars, 250 μm.
Figure 5
Figure 5. Fluorescence intensity profiles of individual vessel segments in a single healing wound
Pixel intensity values were plotted for different vessel segments (insets) imaged on days 5–7 in the same animal, within three different functional areas of the vascular plexus (pre-existing vessels, angiogenic plexus and blind-ended vessels) as described in the Methods. The discrete vessel segments were labeled with a ROI on days 5–7, bleached and quantified to give rise to the fluorescence recovery profile plots. Baseline (pre-bleach) Fi is much higher in BEVs than surrounding flowing vessels. The difference between pre-bleach Fl and post-bleach Fi is also reduced in angiogenic vessel segments and BEVs compared to pre-existing vessels. Fl recovers to pre-bleach levels instantaneously in pre-existing vessels, where peaks of around 16 to 22 s periodicity are observed. In the angiogenic plexus, recovery profiles are highly variable but, by day 7, tend to resemble those of pre-existing vessels. Between days 5 and 6, BEVs never recover to pre-bleach Fl and on day 7 recovery to pre-bleach levels takes about 115 s.
Figure 6
Figure 6. Distinct fluorescence intensity profiles and t1/2 values discriminate specific vessel segment types
A, in each ROI (lumen or interstitium) from the vessel segment chosen, the average Fn (±SEM) was calculated pre- and post-bleaching (n = 5). Following bleaching (dark grey band) there was a significant reduction in the luminal Fn in both BEVs and angiogenic vessel segments. Pre-existing vessel Fn recovered to pre-bleach intensity much faster than the angiogenic plexus. In the interstitium on days 5 or 6 post-injury, Fn did not reach pre-bleach levels during the recording period (∼120 s). By day 7, Fn in both pre-existing plexus and BEVs recovered to pre-bleach levels by ∼85 s post-bleaching but in angiogenic vessel segments return to pre-bleach Fn did not occur during the recording period. B, in the lumen of BEVs between days 5 to 7 post-wounding t1/2 was highest of the 3 vessel segment types and significantly slower than in pre-existing vessels (P < 0.05; Dunń;s multiple comparison test). In pre-existing vessels, t1/2 on day 7 was essentially instantaneous (7.7 × 10−5 s). By day 6, there was a wide variability in t1/2 in angiogenic segments. In the interstitium adjacent to the vessel segments, t1/2 did not vary significantly over time or distance from the wound (P = 0.3611; Friedman test). The t1/2 value for each individual vessel segment was calculated within a 95% confidence interval, as described in Methods. Values plotted are mean ± SEM (n = 5).
Figure 7
Figure 7. TNP-470 differentially affects t1/2 in pre-existing vessels, angiogenic plexus and BEVs
In vehicle-injected animals, t1/2 was essentially instantaneous in pre-existing vessels, although it took longer than 20 s to recover to t1/2 in BEVs. Within the vascular lumen (A), TNP-470 injection induced a significantly faster recovery time in BEVs, whilst significantly delaying the recovery time for pre-existing vessels. In the interstitial space (B), there were no significant differences until day 7, when the t1/2 value adjacent to flowing vessels (pre-existing or angiogenic) was significantly delayed) when compared to vehicle-injected animals. Values plotted are mean ± SEM (n = 4 for each group), *P < 0.05; **P < 0.01.
Figure 8
Figure 8. TNP-470 injection decreases plasma leakage from BEVs and increases the diameter and leakage of plasma from pre-existing vasculature
A, in a vehicle-only injected mouse sprouting began at 500 μm (depicted by the asterisk inside rectangle) from the wound centre on day 3, when vessels were maximally dilated. The largest numbers of sprouts were observed on day 6 and the newly formed plexus began to orientate in parallel with the regenerating myofibres by day 9. No significant differences were observed in the area distal from the wound centre (1000 μm, depicted by the asterisk inside rectangle). B, in an animal injected with 30 mg kg−1 TNP-470 every 2 days, sprouting at 500 μm from the wound centre was only observed by day 6, when it was markedly reduced (compared with the vehicle control in A) and failed to remodel by day 9. Robust vasodilatation and FITC-dextran extravasation was observed in areas distal to the wound centre (1000 μm, depicted by the asterisk inside rectangle) by days 3 and 6. By day 9, the perfusion in the tissue distal to the wound had been significantly reduced in the animal treated with TNP-470. C, the average internal microvessel diameter in both groups of animals varied according to the distance from the wound centre (P < 0.0001) but was not affected by time elapsed since injury. On day 6, newly formed vessel segments were significantly smaller in diameter in the group treated with TNP-470 (P < 0.01); however, pre-existing vessels of the surviving rim by day 9 (P < 0.05) were wider than those observed in the vehicle-only control group. The variation in average vessel internal diameter by day 9 was due to distance from the wound centre (36.12% of total variation, P = 0.0008) and also due to TNP-470 treatment (5.58% of total variation, P = 0.0478). *P < 0.05; **P < 0.01.

Similar articles

See all similar articles

Cited by 6 PubMed Central articles

See all "Cited by" articles

MeSH terms