Figure 1.
Gene expression of complement components in HUVECs.
HUVECs were maintained for 24 hours in serum-free medium prior to RNA extraction. Total RNA was isolated, reverse transcribed, and the cDNA was analyzed by real-time PCR using TaqMan probes with GAPDH as the reference gene. The graph shows fold differences (log scale) of complement component expression in unstimulated HUVECs relative to VWF expression (marked by asterisk). The line at 1 is the boundary between increased and decreased expression. Data shown are means plus SD, N = 4. Values for ADAMTS13 (A13) are shown for comparison.
Figure 2.
Specificity of antibodies to human complement components.
(A) Denatured, non-reduced samples were separated by 4–15% SDS-PAGE and transferred blots were detected with polyclonal goat antibodies to single human complement components. Each blot contains lanes with: 50 ng of a purified complement protein (Std); normal serum (NS) containing 50 ng of the specific complement component; and an equal volume of specific complement component-depleted serum (Dep). Arrows show relative molecular mass of each protein migrating in SDS and MW indicates molecular weight markers in kDa. (B) FD and FP were analyzed by Western blots as described in (A) except: In the FD blot the Std lane contains 159 ng FD and NS lane contains 4 ng FD (FD serum conc. 1–2 ng/µl); and in the FP blot the NS lane contains 10 ng FP (FP serum conc. 4–6 ng/µl). The Coomassie stained gels show the high levels of protein (∼100 µg/lane) in the serum samples that were applied to the gels.
Figure 3.
AP-specific components FB, FD and FP attach to ULVWF strings secreted by, and anchored to, stimulated HUVECs.
HUVECs were stimulated with 100 µM histamine and stained with rabbit anti-VWF plus secondary fluorescent anti-rabbit IgG-488 (green). Cells were then p-formaldehyde-fixed and stained with goat IgG antibody to human FB (A), antibody to human FD (B), and antibody to human FP (C) plus secondary fluorescent anti-goat IgG-594 (red). The HUVEC nuclei were labeled with DAPI (blue). In (A) are: (1) ULVWF (488-nm, green); (2) FB (594-nm, red); and (3) ULVWF and FB combined image; (4) Graph of fluorescent intensities (y-axis) measured from identical locations in ULVWF string images (488-nm, green) and in complement component proteins images (594-nm, red) are plotted against the ULVWF string length (in microns, x-axis). The black line indicates the background intensities measured in the 594-nm images. (5) ULVWF intensities were measured along lines of ULVWF strings detected at 488-nm (shown by dotted line); (6) FB intensities were measured in 594-nm images along lines at identical locations (shown by lower dotted line) as determined in (5). Background intensities were also measured in 594-nm (red) images at parallel locations (shown by upper dotted line) away from the area of interest. Similar types of images are shown using antibody to human FD in (B) panels 1–6 and antibody to human FP in (C) panels 1–6 to identify the complement component attached to the HUVEC-secreted/anchored ULVWF strings. In (C) panel 6, only the locations of the background intensities are identified by the dotted line. The white arrows in (3) indicate FB (A), FD (B) and FP (C) attachment to the strings. Images were selected from 5 (FB and FD) and 4 (FP) independent experiments.
Figure 4.
AP-specific negative regulatory components FH and FI attach to ULVWF strings secreted by, and anchored to, stimulated HUVECs.
HUVECs were stimulated and stained as in the legend for Fig. 3, except that antibody to human FH was used in (A) and antibody to human FI in (B) to identify complement component attachment to the ULVWF strings. In (A) panel 6, only the locations of the background intensities are identified by the dotted line. In (B) panel 6, the upper dotted line shows the location of the FI intensity measurements and the lower dotted line identifies the locations of background intensity measurements. The white arrows in (3) indicate FH (A) and FI (B) attachment to the strings. Images were selected from 12 (FH) and 4 (FI) independent experiments.
Figure 5.
Complement components C3 and C5 attach to ULVWF strings secreted by, and anchored to, stimulated HUVECs.
HUVECs were stimulated and stained as in the legend for Fig. 3, except that antibody to human C3 was used in (A) and antibody to human C5 in (B) to identify complement component attachment to the ULVWF strings. In (A) panel 6, the upper dotted line shows the location of the C3 intensity measurements and the lower dotted line identifies the locations of background intensity measurements. In (B) panel 6, only the locations of the background intensities are identified by the dotted line. The white arrows in (B) panel 3 indicate C5 attachment to the strings. Images were selected from 6 (C3) and 5 (C5) independent experiments.
Figure 6.
Complement components C3 and C5 attach to the same positions along HUVEC secreted/anchored ULVWF strings.
HUVECs were stimulated and stained as in the legend for Fig. 3, except that the cells were simultaneously stained for C3 and C5 (in addition to VWF and DAPI). Individual fluorescent channels detected: (A) rabbit anti-VWF plus anti-rabbit IgG-488 (turquoise); (B) a combination of two mouse monoclonal antibodies to human C3 (clone 755 against C3b and clone 10A1 against C3) plus anti-mouse IgG-647 (red); and (C) goat anti-human C5 plus anti-goat IgG-594 (green). (D) Simultaneous detection of C3 (red) and C5 (green) is colored yellow in the combined image from 647- and 594-nm channels. White arrows indicate points along the ULVWF strings where high intensity levels of C3 and C5 were detected. (E) Graph of fluorescent intensities (y-axis) along the ULVWF string (488-nm, turquoise), C3 (647-nm, red) and C5 (594-nm, green) are plotted against the ULVWF string length (in microns, x-axis). The black numbered arrows correspond to the white numbered arrows in (D) and point to the C3 and C5 peak intensity locations. Images were selected from 9 experiments with simultaneous VWF, C3 and C5 staining.
Figure 7.
C4, a component of the classical and lectin pathways, does not attach to ULVWF strings secreted by, and anchored to, stimulated HUVECs.
HUVECs were stimulated and stained as in the legend for Fig. 3, except the antibody to human C4 was used to identify complement component attachment to the ULVWF strings. In panel 6, the upper dotted line shows the location of the C4 intensity measurements and the lower dotted line identifies the locations of background intensity measurements. Images were selected from 4 experiments.
Figure 8.
Quantification of HUVEC-released complement components attached to HUVEC-secreted/anchored ULVWF strings.
Intensities of each HUVEC-released complement proteins were measured along histamine-stimulated HUVEC-secreted/anchored ULVWF strings, as described in the legend for Fig. 3. Shown are the complement component fluorescent intensities per micron of ULVWF string length after background subtraction. Values are means plus SD; N = 7–12 strings for each complement component from 4 to 12 experiments and were compiled from 130 fluorescent images. Some data were collected from images within the same experiment at a different location on the coverslip.
Figure 9.
Complement components C3 and C5 do not attach to unstimulated HUVECs.
HUVECs were washed once with PBS, incubated in 25% heated serum/PBS for 5 min, and washed 4X with PBS before staining with rabbit anti-VWF plus anti-rabbit IgG-488 (green), goat antibodies either to (A) C3 or (B) C5 and anti-goat IgG-594 (red). Combined images with DAPI-stained nuclei are shown at 600X and are representative of 3–4 experiments.
Figure 10.
Attachment to HUVEC-secreted/anchored ULVWF strings of C3, FB and C5 released from HUVECs +/− added in heated normal serum.
Intensities of C3, FB and C5 were measured along histamine-stimulated HUVEC-secreted/anchored ULVWF strings as described in the legend for Fig. 3. Light gray bars represent the binding of exclusively HUVEC-released C3, FB and C5 to ULVWF strings (shown for comparison from Fig. 8), and dark gray bars show the binding of the same components per micron of ULVWF string length in the presence of normal heated serum. Values are means plus SD; N = 8–11 strings for each complement component from 5 to 7 experiments for each of C3, FB and C5 and data were compiled from 46 fluorescent images.
Figure 11.
Fluorescent emission “bleed-through” controls: Weibel-Palade bodies (WPBs) contain a high concentration of VWF but are devoid of complement components.
Unstimulated HUVECs were fixed with p-formaldehyde and treated with Triton-X to allow intracellular fluorescent staining. VWF in WPBs was detected with rabbit anti-VWF plus anti-rabbit IgG-488; and FB was detected with goat anti-FB plus anti-goat IgG-594. (A) The merged image was combined from 488-nm (green) and 594-nm (red) channels at 600X magnification. Single channel emissions of the circled area are shown in the inset images: (a) 488-nm and (b) 594-nm. (B) Graph of fluorescent intensities measured at points along the white lines in inset images (a) and (b) shows that extremely low intensities at 594-nm were measured at the same locations as high intensities were measured in the 488-nm channel, i.e., there was little or no green-to-red “bleed-through”.