| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INNOVATIVE METHODOLOGY
and MMP-2 expression in capillaries isolated from overloaded skeletal muscles: laser capture microdissection study
School of Kinesiology and Health Science, York University, Toronto, Ontario, Canada
Submitted 23 March 2005 ; accepted in final form 10 May 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
expression was 13.5-fold higher in capillaries subjected to stretch compared with controls (P < 0.05). These findings demonstrate the feasibility of this approach to study angiogenic gene regulation and provide novel evidence of HIF-1 induction in stretched capillary endothelial cells.
endothelium; matrix metalloproteinases; mechanotransduction; transcriptional regulation
Several laboratories including our own have demonstrated that increased muscle activity can trigger angiogenesis mainly as a result of mechanical forces produced during muscle contractions (3, 8). There are few studies that have addressed the relation between mechanical stress and the expression of angiogenesis-related transcription factors and none that have examined changes in the expression of transcription factors within capillaries undergoing angiogenesis in skeletal muscle.
Muscle overload is an established model for stimulating stretch-induced angiogenesis in the absence of an exercise stimulus. The increase in capillary supply in overloaded muscle is associated with upregulation of VEGF protein level both in whole muscle extract and within capillaries as detectable by Western blotting and immunohistochemistry, respectively (15). Transcription of the VEGF gene is driven by hypoxia-inducible factor (HIF)-1 (6, 11). HIF-1 is an 
-heterodimer (18). The intracellular level of the HIF-1 subunit is tightly regulated by oxygen tension, whereas the HIF-1 subunit is expressed constitutively in the nucleus. In oxygenated cells, HIF-1 is subjected to ubiquitination and rapid proteasomal degradation, whereas in hypoxic conditions the proteolytic pathway is inhibited, resulting in accumulation of HIF-1 protein and activation of genes important for adaptation and survival under hypoxia including VEGF (16). However, there is growing evidence that the upregulation of HIF-1 is not limited to hypoxic stress but can be induced by growth factors, cytokines, hormones, and nitric oxide (14). These stimuli often increase levels of HIF-1 gene transcription or mRNA translation, in contrast to the hypoxia-dependent inhibition of HIF-1 protein degradation (14). Enhanced accumulation of HIF-1 protein was observed in cardiac myocytes from nonischemic mechanically stretched myocardium, and this increase in HIF-1 induced an increase in VEGF protein level (9). Endothelial cells also produce VEGF protein in response to mechanical stress (12), but to our knowledge, there are no data on the effect of mechanical stretch on the endothelial cell expression of HIF-1. Therefore, we hypothesized that HIF-1 mRNA would be increased in endothelial cells isolated from overloaded/stretched rat skeletal muscles.
Stretch-induced muscle overload also increases significantly the production and activation of matrix metalloproteinase (MMP)-2, a member of the major protease family responsible for degrading components of basement membrane and interstitial matrix (15). Immunohistochemistry demonstrated MMP-2 protein localization to capillaries. Although we cannot exclude that nonendothelial cells contribute to MMP-2 production in response to this mechanical stimulus, we used analysis of MMP-2 mRNA in LCM samples as a positive control in testing the utilization of this technique to examine mRNA profiles of genes critical to the angiogenesis process in vivo.
To determine the individual effect of mechanical stretch on capillary gene expression, we used a previously described animal model in which an ankle flexor muscle was stretched and overloaded because of surgical removal of the synergistic muscle. As the LCM technique has not been implemented for analysis of angiogenesis in skeletal muscle, we adapted and optimized the technique for this particular purpose. We first established the specificity and accuracy of the LCM approach by evaluating MMP-2 mRNA levels in a small number of capillaries harvested from stretched and nonstretched skeletal muscles. We then used this experimental approach to provide novel evidence of capillary-specific HIF-1 expression in skeletal muscle, implicating this transcription factor in mechanical force-induced angiogenesis.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Laser capture microdissection. Cryosections (8 µm thick) were freshly cut, placed on uncoated glass slides, and immediately immersed in cold acetone for 3 min followed by transfer to a –80°C freezer, where they were stored for a maximum of 1 wk. To prevent RNA degradation during histological staining, RNAase inhibitor was added to all sterile PBS solutions (50 U/100 µl; SUPERaseIn, Ambion). Capillaries were prepared for detection by staining (10 min) with isolectin GS-IB4 from Griffonia simplicifolia Alexa Fluor 488 conjugate (1 mg/ml) diluted 1:200 in PBS (Molecular Probes). After a brief wash with PBS, the sections were dehydrated in a graded series of 75, 95, and 2x 100% ethanol for 1 min each, followed by a 5-min immersion in pure xylene and air drying for 20 min (Fig. 1). Slides were pretreated with a PrepStrip (Arcturus Engineering) to remove debris. Capillary laser microdissection was performed with an Arcturus PixCell II system with a power setting of 100 mW, a 7.5-µm spot size, and a pulse duration ranging from 100 to 200 ms (Arcturus Engineering). Capillaries distinguished by size (<10 µm) were collected selectively with CapSure LCM caps (Arcturus Engineering). After the sum of 500 capillaries from each skeletal muscle was captured, the CapSure LCM cap was inserted into a 0.5-ml microcentrifuge tube filled with 50 µl of extraction buffer (PicoPure RNA isolation kit, Arcturus Engineering). The LCM-microcentrifuge tube assembly was inverted to ensure that the membrane of the LCM cap was covered fully with buffer. After incubation at 42°C for 30 min the extraction buffer was collected by a brief spin (800 g for 2 min), the cap was removed, and the microcentrifuge tube was stored at –20°C.
|
Quantitative real-time PCR.
Measurements of cDNA levels were performed by quantitative real-time PCR with an ABI PRISM 7000 Sequence Detection System. cDNA was diluted 1.5-fold with RNAse-free water, and 4 µl were used as a template in each PCR reaction. Primers and TaqMan probe for 18S rRNA (VIC-labeled ready-for-use ribosomal RNA control probe; catalog no. P/N 4308329), FAM-labeled HIF-1 (assay ID Rn00577560), and FAM-labeled MMP-2 were obtained from Applied Biosystems Canada. Primers for HIF-1 spanned exons 4 and 5, generating a 113-bp PCR product. Primers and probe for MMP-2 were designed with Primer Express software. To avoid the possibility of amplifying contaminating DNA, the primer pairs were separated by an intron sequence. Primers for MMP-2 spanned exons 12 and 13, generating a 69-bp PCR product. The sequences of primer pairs and probe for MMP-2 were as follows: MMP-2 probe: 6FAM-caa tgc tga tgg aca gcc ctg ca-MGBNFQ; forward primer: cca tga agc ctt gtt tac ca; reverse primer: ctg gaa gcg gaa cgg aaa. The PCR reaction mixtures (12.5 µl of TaqMan universal PCR Master Mix, 4 µl of cDNA, and an appropriate concentration of primers and probe) were incubated at 50°C for 30 s and 95°C for 10 min and then cycled for 40 times at 95°C for 15 s and 60°C for 60 s. Reactions were performed with appropriate negative control, e.g., template-free control subjects. The parameter threshold cycle (CT) was defined as the fractional cycle number at which the fluorescence generated by cleavage of the probe exceeded a fixed threshold above baseline. 18S rRNA was measured as an endogenous reference to control differences in harvested RNA samples across all experimental groups. To evaluate the amount of target genes, the comparative CT method was utilized as outlined in Applied Biosystems User Bulletin No. 2 (1). First, the mean CT values of triplicate samples from each group were determined and normalized to an endogenous rRNA by calculation of change in (
)CT(sample) according to the equation CT(sample) = average CT(sample) – average CT(samplerRNA). CT(sample) was then related to CT of the experimental control (sham operated) by computing CT, where CT = CT(7days) – CT(sham). The amount of target amplification relative to the experimental control was calculated by the formula 2
.
Statistical analysis. Treatment effects were determined by factorial ANOVA, with intergroup comparisons assessed by post hoc tests (Fisher’s protected least significant difference), and P < 0.05 was considered to be significant. Data are presented as means ± SE transcript amounts relative to control values.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
CT) can replace the standard curve method as long as a validation experiment is performed to ensure that amplification efficiencies of the target gene and internal reference are approximately equal, i.e., CT [CT(target) – CT(internal reference)] is invariable over a range of serial template dilutions. Therefore, the efficiencies of the target genes and the reference (control rRNA) amplifications were carried out at five different dilutions of total cDNA. A plot of log input amount vs. CT was then prepared. If the efficiencies of the two amplicons are almost equal, the plot of log input amount vs. CT has a slope of approximately zero. The validation experiments confirmed that the CT between MMP-2 and rRNA (Fig. 3A) and HIF-1 and rRNA (Fig. 3B) remained constant independent of cDNA concentration, because in both cases the absolute value of the slope of log input amount vs. CT was <0.1 (0.004 and 0.0019, respectively). These slopes are consistent with equal efficiencies of amplification regardless of template concentration, thus demonstrating that the CT calculation method was valid for use in our system.
|
expression and capillary supply.
After 7 days of stretch induced by overloading the EDL muscle, the MMP-2 mRNA level within capillaries was elevated almost 2.5-fold (P < 0.05 vs. controls; Fig. 4A). At the same time point the HIF-1 mRNA level was 14-fold increased compared with controls (P < 0.05; Fig. 4B). Stretch-induced overload of the EDL muscles for 7 days did not cause a significant change in capillary-to-fiber ratios (1.49 ± 0.06 vs. 1.41 ± 0.02 in controls; not significant).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
mRNA level can be increased in nonhypoxic endothelial cells by mechanical stretch. 3) Mechanical stretch of muscle results in MMP-2 gene activation in capillary endothelium. Laser capture microdissection. Development of the LCM technology allows for a detailed analysis of the molecular mechanisms behind neovascularization in small and valuable samples. Furthermore, a distinct population of cells can be separated and analyzed with high specificity that is unachievable through other currently available techniques. In the present adaptation of the LCM technique for the study of angiogenesis within skeletal muscles we showed the feasibility of the isolation of a pure population of capillaries from muscle tissue. To eliminate the risk of accumulation of other cells, a very small (7.5 µm) spot size of the laser beam was used and only blood vessels smaller than 10 µm, i.e., capillaries, were captured. We cannot exclude the possibility that periendothelial cells such as pericytes and fibroblasts that remain in close proximity to the microvascular endothelial cells could be accumulated unintentionally. Nevertheless, endothelial cells were dominant in the collected cell population, because in a cross section of skeletal muscle only 10% of capillaries are accompanied by adjacent pericyte or fibroblast nuclei (3).
There are a number of important considerations with regard to successful laser capture and subsequent gene analysis. For the successful performance of LCM, tissue sections must be mounted on plain, untreated glass slides because slide pretreatment increases the strength of adhesion and thus interferes with cell retrieval. Furthermore, the sections must be dehydrated thoroughly because any moisture in the tissue can prevent cell capture. Visualization of the capillary network was achieved by brief staining with isolectin GS-IB4 conjugated to Alexa Fluor 488 that very nicely and clearly depicted capillaries within muscle tissue. Initially, we used enzymatic staining for endogenous capillary alkaline phosphatase. However, this resulted in a poor efficiency of capillary capture, likely due to the insoluble blue precipitating product.
Analysis of mRNA levels in LCM-captured cells is hindered by difficulties associated with isolating RNA from very small numbers of cells. Initially, we tried to use a mixture of guanidine thiocyanate and phenol for RNA isolation (TRI Reagent, Sigma), but because of an inefficient isolation of RNA the genes represented by a minute number of copies were undetectable. The Pico Pure RNA isolation kit dramatically improved the amount and quality of total RNA that was ready to be used for reverse transcription reaction with Sensiscript reverse transcriptase. Both random decamers and oligo(dT) primers were used in reverse transcription reactions to represent more accurately all sequences of the RNA population and to prevent skewing the cDNA yield in favor of 3' ends. Combining these two procedures provided an efficient and sensitive method for cDNA synthesis of small amounts of RNA. It was not feasible to estimate the cDNA concentration in these small samples; therefore, to standardize the amount of sample cDNA added to each PCR reaction the amplification of an endogenous control (18S rRNA) was performed simultaneously. The target and endogenous controls were amplified in separate tubes in triplicate to further eliminate error associated with unequal loading. The efficiencies of the target and endogenous control amplifications were nearly the same regardless of cDNA concentration, thus validating the use of the CT method for relative quantification of gene expression in LCM-captured cells. The amount of time required to carry out microdissection, particularly if a large number of tissue samples need to be analyzed, is an inconvenience of using the LCM technique. However, this is balanced by the capacity of the LCM technique to retrieve a specific population of endothelial cells from in vivo experimental models for molecular analyses.
MMP-2 expression in stretched capillaries. To prove that this technique can distinguish differential gene expression during the process of angiogenesis, we first compared MMP-2 expression in stretched and control nonstretched muscle tissue. The capillary network is physically tethered to the myocytes by extracellular matrix, and thus it experiences tensile forces and longitudinal stretch during repetitive shortening and relaxation of the muscle fibers (4). Although it has not been demonstrated that stretch may directly stimulate capillary endothelium to produce MMP-2, it is known that MMP-2 mRNA and protein levels are increased in 7-day stretched EDL, and immunostaining localized it to capillaries (15). Furthermore, cultured skeletal muscle endothelial cells increased production of MMP-2 mRNA in response to stretch (19). Thus it appeared likely that we would see an elevated expression of MMP-2 in capillaries of the stretched muscles. In this study, because of the use of LCM we were able to show that capillaries isolated from skeletal muscle produced MMP-2. Moreover, it was observed that mechanical stretch led to a significant increase in MMP-2 expression in capillaries.
HIF-1 gene activation by mechanical stretch.
Using the LCM technique, we found that HIF-1 mRNA level drastically increased in nonhypoxic endothelial cells subjected to enhanced mechanical stretch. Our data are in agreement with previous reports on mechanical stress-mediated induction of HIF-1 in nonendothelial cells, such as cardiac myocytes, in the nonischemic myocardium (9) and mechanically overloaded bovine cartilage disks (13). To our knowledge, this phenomenon has not been observed in endothelial cells. Owing to the LCM technique we were able to address a previously unanswerable question, i.e., whether endothelial cells forming capillaries within muscle tissue can produce HIF-1 in response to stretch. Future work will determine the signaling pathways responsible for increasing HIF-1 mRNA levels and assess whether HIF-1 mRNA translation also is affected by stretch, as was observed in myocardial cells (9).
The induction of VEGF in stretched myocardium is mediated by stretch-activated ion channel/phosphatidylinositol 3-kinase/Akt/FK506-binding protein-rapamycin-associated protein/HIF-1 pathways (9). Stretch-induced VEGF production was demonstrated in myocardial, mesangial, and smooth muscle cells (7, 10, 17). Likewise, the overload of the EDL muscle for 7 days led to increased VEGF protein production, as detected by both Western blotting and immunostaining, concurrently with enhanced endothelial cell proliferation (15). On the basis of our earlier results we hypothesized that the observed peak in VEGF protein production coincided with activation of the HIF-1 gene; therefore, we chose the 7-day time point to analyze stretch-induced HIF-1 expression. In the current study we showed that capillaries produced HIF-1 in response to mechanical stretch, and the timing of this increase was consistent with the previously documented increase in VEGF production (15). Seven days of stretch did not increase the number of new capillaries based on capillary-to-muscle fiber ratios. However, this is consistent with our previous report showing a slow onset of capillary growth in overloaded muscle that was apparent after 14 days and was preceded by increased endothelial cell proliferation and VEGF expression (2, 15). Overall, our results, together with already published reports, imply that HIF-1 plays important roles in the adaptation not only to ischemia but also to mechanical stress and that mechanically stretched endothelial cells may be the source of this transcription factor.
Many questions remain to be addressed concerning gene activation in the context of angiogenesis induced by mechanical stimuli. The roles of other angiogenic transcription factors need to be investigated. This approach, when coupled with pharmacological or gene-based modulations in signaling pathways, will provide a sensitive measurement of the roles of individual molecules in the coordinated upregulation of angiogenic genes. In particular, this technique provides the potential to compare capillary-specific patterns of gene regulation in response to mechanical stimuli that are known to trigger diverse mechanisms of the capillary growth, thus providing clarification of the molecular regulation of capillary network patterning (3).
In conclusion, we have demonstrated the feasibility of isolating a valuable population of endothelial cells from heterogeneous tissue, allowing further cell-specific analysis of target mRNA. LCM-assisted analysis of transcriptional changes was shown to be a powerful tool for measurement of gene regulation in a discrete endothelial cell population involved in angiogenesis. Use of the LCM technique followed by real-time quantitative PCR analysis will permit a more accurate exploration of capillary-specific gene regulation in tumor environments or in ischemic tissue to better understand pathological activation or disruption of the angiogenic process.
|
ACKNOWLEDGMENTS |
|---|
|
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 |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
in response to mechanical stresses: regulation by stretch-activated channels and the phosphatidylinositol 3-kinase signaling pathway. Circ Res 90: E25–E33, 2002. in vascular smooth muscle cells. J Biol Chem 275: 26765–26771, 2000. (HIF-1) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J Biol Chem 272: 22642–22647, 1997.This article has been cited by other articles:
|
|
 |
|
  E. Rullman, J. Norrbom, A. Stromberg, D. Wagsater, H. Rundqvist, T. Haas, and T. Gustafsson Endurance exercise activates matrix metalloproteinases in human skeletal muscle J Appl Physiol, March 1, 2009; 106(3): 804 - 812. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J D Raffetto and R A Khalil Mechanisms of varicose vein formation: valve dysfunction and wall dilation Phlebology, April 1, 2008; 23(2): 85 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Kivela, H. Kyrolainen, H. Selanne, P. V. Komi, H. Kainulainen, and V. Vihko A single bout of exercise with high mechanical loading induces the expression of Cyr61/CCN1 and CTGF/CCN2 in human skeletal muscle J Appl Physiol, October 1, 2007; 103(4): 1395 - 1401. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Milkiewicz, J. L. Doyle, T. Fudalewski, E. Ispanovic, M. Aghasi, and T. L. Haas HIF-1{alpha} and HIF-2{alpha} play a central role in stretch-induced but not shear-stress-induced angiogenesis in rat skeletal muscle J. Physiol., September 1, 2007; 583(2): 753 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Zhang, D. J. Bellotto, P. Ravikumar, O. W. Moe, R. T. Hogg, D. C. Hogg, A. S. Estrera, R. L. Johnson Jr., and C. C. W. Hsia Postpneumonectomy lung expansion elicits hypoxia-inducible factor-1{alpha} signaling Am J Physiol Lung Cell Mol Physiol, August 1, 2007; 293(2): L497 - L504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Ispanovic and T. L. Haas JNK and PI3K differentially regulate MMP-2 and MT1-MMP mRNA and protein in response to actin cytoskeleton reorganization in endothelial cells Am J Physiol Cell Physiol, October 1, 2006; 291(4): C579 - C588. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
