Introduction

Vascular ECs organize themselves in a single cellular layer, that is, the endothelium, at the luminal face of all blood vessels. Besides a major function in controlling gas exchanges at the pulmonary level, the endothelium regulates the flow of circulating blood cells and of various bioactive molecules, for example, growth factors, coagulation proteins, lipoproteins and hormones, in particular through the presence of membrane-bound receptors. The endothelium also plays pivotal roles in the regulation of a broad range of physiological processes: it regulates hemostasis through the expression of both anti-thrombotic and prothrombotic factors; in tight cooperation with the underlying extracellular matrix and smooth muscle cells, it controls vascular tone with angiotensin I conversion and bradykinin degradation, and contributes to the metabolism of vasoactive amines; it also participates in inflammatory and immune responses through surface antigens and adhesive molecules and through the synthesis of cytokines^{1, 2, 3}. But ECs, which are in direct contact with plasma and cellular components of blood, are targets of many endogenous molecules and xenobiotics^{4, 5}.

For many years our work focused on the physiology and pathology of ECs, in particular using (from cord of newborns), HUVECs (for human umbilical vein endothelial cells)^{6, 7, 8, 9}. HUVECs represent a model for a large community of researchers all round the world, although properties of HUVECs certainly cannot represent all the metabolic capacities and the responses in physiopathology and toxicity related to the different types of ECs found in an organism. As an example of the interest for the scientific community, more than 100,000 publications have cited ECs in general, and at least 10,000 HUVECs in particular. HUVECs model is useful for any research on general properties of human ECs, but other sources of ECs could be better models for studies on specific pathological areas, for example, atherosclerosis development or metastasis dissemination in particular microvascular areas. Nevertheless, HUVECs are the most simple and available human EC type, accurate for the preparation of large quantities of cells. To increase our knowledge about the protein content and the main biological pathways of this vascular EC model, we undertook the proteomic analysis of HUVECs in primary cultures. Our results are available on the web at http://www.huvec.com and will be soon completed with our recent study employing Fourier transform ion cyclotron resonance mass spectrometry analysis of HUVEC proteins fractionated in two-dimensional (2D) electrophoresis gels^{10, 11, 12}.

We describe in the present paper a protocol for easy isolation and culture of HUVECs to supply every researcher with a method that can be applied in cell biology laboratories with minimum equipment and accessories. This protocol is derived from the original description by Jaffe et al ¹³. The confluent primary cultures of HUVECs can be obtained in 8 days, although you have to wait for an additional 48 h before starting a special treatment, such as the addition of a drug to study a particular effect. As for many other cell types, particularly for adherent cell lines, HUVECs can be used in subcultures, obtained with or without trypsin treatment, for amplification of the number of cells, but most often the primary culture is preferred because of the decrease in expression of many proteins at each passage due to a mechanism of accelerated senescence, with spontaneous apoptosis; for example, both angiotensin I-converting enzyme (ACE) and prostacyclin synthesis decrease as a function of the number of passages^{14, 15, 16, 17}. Moreover, no additional growth factor is required for this protocol, as many growth factors and related products can modify protein synthesis and intracellular trafficking^{18, 19, 20}.

You can characterize HUVECs by using EC markers, such as ACE (peptidyl-dipeptidase EC.3.4.15.1), both in activity and immunofluorescence (CD-143) and Von Willebrand Factor/Factor VIII RA (vWF, immunofluorescence), even if the latter is weakly expressed in these particular ECs. You can also use other EC markers, such as CD-31, CD-34, CD-54 or ICAM-1, CD-62E or E-selectin, CD-106 or VCAM-1, more or less specific for ECs and expressed often in HUVECs after a particular stimulation. The lectin from Ulex europaeus (anti-H blood group specificity) binds to most ECs, normal as well as from neoplastic vessels^{21, 22, 23, 24}. Uptake of labeled acetylated low-density lipoproteins is also a characteristic of EC; but as for CD markers, this uptake is rather applied to flow-cytometry analysis²⁵.

Materials

Reagent

• Umbilical cord
  Caution Appropriate consent must be obtained before processing human tissue.

Equipment

• Hood for cell culture with vertical laminar flow and equipped with UV light for decontamination
• Water-bath with temperature control
• Centrifuge (no temperature control is needed)
• Incubator with both temperature and gas composition controls
• Optical microscope with phase-contrast equipment
• A procedure for material sterilization (Poupinel oven for example)
• Cannulae (i.d. 1 mm 60 mm)
  Critical Cannulae that are not fitted can lead to various problems, see TROUBLESHOOTING; manufacturers specialized in surgical materials can provide appropriate cannulae.

Reagent Setup

• 1 PBS Dilute 100 ml of 10 PBS, without calcium nor magnesium (Eurobio, cat. no. CS3PB01-04), in 900 ml of distilled water plus 2 g of glucose (Sigma, cat. no. G7520); pass through a syringe equipped with sterile Millex GV 0.22 m filter (Millipore, cat. no. SLGV025BS) fractionate in 50 ml aliquots and freeze at - 20 °C until used.
• Buffer for conservation and transport of umbilical cords Add 1 ml of 1 million unit penicillin G (Peni G, Panpharma) and 1 ml of 1 million unit colistine (Colimycine, Aventis) to 50 ml of 1 PBS; prepare fresh buffer for each cord collection.
• Collagenase solution (0.2% wt/vol) Dissolve 0.2 g of collagenase from Clostridium histolyticum (Roche, cat. no. 103586) in 100 ml of 1 PBS and gently stir for 10 min at laboratory temperature (20 °C) and in the dark (cover with an aluminum foil); then adjust the pH at 7.4 with 1 N NaOH; filter through 0.22 m as before; fractionate in 10 ml aliquots and freeze at - 20 °C until used.
  Critical The use of another collagenase or another proteolytic enzyme may modify the conditions for ECs isolation, such as concentration of the enzyme and time for incubation (Step 10).
• FCS Thaw the FCS (Biosepra, a reserved lot checked for the absence of toxicity toward HUVECs); heat for 30 min at 56 °C (for decomplementation; keep the bottle closed); fractionate in 25 ml aliquots and freeze at - 20 °C until used.
  Critical The lot of the FCS can be critical for ECs growth; thus, keep the same lot for one set of experiments and check each new lot with a viability test.
• Culture medium M199 "stock": dilute 100 ml of 10 M199 with Earles' salts²⁶ (Gibco, cat. no. 21180-021) in 900 ml of distilled water.
• "Complete M199 medium" (with 20% FCS) Make up to 100 ml with M199 "stock" a solution containing 1 ml of 0.2 M glutamine (Sigma, cat. no. P0781), 1.5 ml of 1 M HEPES (Eurobio, cat. no. CSTHEPOO-OP), 1.8 ml of 7% (w/v) NaHCO₃ (Eurobio, cat. no. CSTBICOO-OU), 1 ml of penicillin/streptomycin (10,000 U per 10,000 U; Sigma, cat. no. P0781) and 20 ml of FCS (the defrost aliquot is centrifuged for 10 min at 1,500g at 15 °C before using 20 ml of the supernatant). "Complete M199 medium" can be conserved for 1 week at 4 °C.
• Fibronectin solution Dissolve in 32 ml of distilled water 1 mg of fibronectin from human plasma (Sigma, cat. no. F2006). Keep the solution at 4 °C for 6 months; do not freeze. In place of fibronectin, collagen can also be used (for example from rat tail, type I) or, for certain studies, no coating is preferred; nevertheless, fibronectin obviously gives the best results²⁷.

Procedure

1. Points from here (point 1) up to and including point 2 are related to The day before isolation of HUVECsCoat 35-mm-diameter Petri dishes with one drop of fibronectin solution per plate. Allow seven dishes per umbilical cord. Dry the dishes under the hood, with lid open.
2. Take a sterile container, with 50 ml of buffer for conservation and transport, to the maternity hospital. One container is needed for each umbilical cord of 10–30 cm; the container must be kept at 4 °C until cord collection and again immediately afterwards. Provide enough containers, if necessary, for preparing several umbilical cords, and processing for culture within 6 h.
3. Points from here (point 3) up to and including point 16 are related to The day of HUVECs isolation and cultureFill up another container with 500 ml of sterile 0.15 M NaCl ("physiologic serum" or 0.09% saline) and maintain at 37 °C in a water-bath.
4. Under the hood, prepare (for one cord) two syringes of 50 ml, one syringe of 30 ml and one syringe of 10 ml, one surgical clamping clip, two cannulae, string, sterile compresses, sterile scalpels and sterile gloves.
5. Spread onto the working area, under the hood, for the subsequent cord manipulation one aluminum foil, one paper sheet and the sterile paper envelope of gloves (strictly in this order).
6. Check whether cords are suitable for use and then treat each cord one after another as follows.
   Critical step Hematic or damaged cords should be discarded and also those from infected parturient women (such as HBV+, HIV+, and so on)²⁸.
7. Tidily cut both ends of the cord with a scalpel.
8. Introduce a cannula at each extremity of the vein and tightly maintain them with string.
   Critical step Umbilical vein (the widest vessel) must not be confounded with the two arteries²⁹ and both cannulae must be carefully fitted.Troubleshooting
9. Wash the umbilical vein with 1 PBS using the 50 ml syringe until the effluent buffer is transparent or slightly pink.
   Critical step All red blood cells must be removed.
10. Inject the 0.2% collagenase solution at one extremity of the vein using the 30 ml syringe; when it leaks out of the other extremity, tightly clamp it with the surgical clamp; maintain the syringe and protect both cord extremities with a clean aluminum foil; incubate the cord in the water-bath for 10 min.
11. Afterwards, under the hood, onto the sterile envelope of gloves, gently squeeze the cord to facilitate cell detachment.
12. Fill up a sterile 50 ml tube with 10 ml of "complete M199 medium"; collect the cells in this tube by washing the vein with 40 ml of 1 PBS.
    Critical step Do not lose cell suspension during this manipulation.
13. Centrifuge the closed tube at 750g for 10 min without temperature control.
14. Carefully discard the supernatant and suspend the pellet of cells in 14 ml of "complete M199 medium"; dissociate the cells by gentle aspiration and repulsing (three times) with the 30 ml syringe equipped with a needle (i.d. 0.7 mm 30 mm).
15. Count the cells from an EC aliquot under an optical microscope with a Malassez's cell; adjust the volume for seeding each Petri dish with 40,000 cells. Typically, when one 20–30 cm cord can be used entirely, you can isolate enough ECs for seeding seven dishes with 14 ml, thus allowing 2 ml per dish. For some studies needing several cords, pool the cells and mix before seeding; but, in some cases, pooling is rejected because HUVECs express HLA-DR antigens that may cause problems of immunocompatibility³⁰. You can use smaller dishes (24- or 96-well plates), or larger dishes or Falcon flasks (Becton-Dickinson); however, the 35 mm Petri dishes are particularly adapted to fibronectin coating with a cell scraper, examination by both optical and fluorescent microscopy and lysis of the cells with a cell scraper or a lysis solution.
16. Incubate ECs cultures at 37 °C in a 95% air/5% CO₂ atmosphere saturated with H₂O.
17. Points from here (point 17) up to and including point 18 are related to The days after isolation of HUVECsThe following day (after less than 24 h), remove non-adherent cells by changing the culture medium. Replace with 1.5 ml culture medium instead of 2 ml. The result must be examined under phase-contrast microscopy; if some remaining blood clots contaminate the ECs, change the culture medium again. Check for contamination. At this stage, you can also count the adherent EC islets or determine a yield of adhesion.Troubleshooting
18. Change culture medium every 2 days (replacing with 1.5 ml medium each time). Typically, cell confluency is achieved in 6–8 days, although it is dependent on the number of cells seeded, the yield of adhesion and the growth capabilities of these peculiar cells, in particular when mixtures of several umbilical cords are used. When confluent ECs show contact inhibition, the EC cultures will present a "cobblestone appearance" (tightly packed polygonal ECs) in phase-contrast microscopy (Fig. 1). Often you have to wait 2 days more before treating the cells, because real confluency (with no mitosis) is achieved later than can be seen via microscope. Typically, in a 35 mm Petri dish, the constituted endothelium (organotypical culture) contains about 1 million cells.

    Figure 1: HUVECs at confluency observed under phase-contrast microscopy.

    

    Full size image (72 KB)
    
    
19. Points from here (point 19) up to and including point 20 are related to HUVECs treatmentIf you wish to harvest the cells, wash three times with PBS. Then scrape the cells into PBS, water or lysis buffer. ECs adhere onto fibronectin and secrete at their basal membrane molecules of the extracellular matrix, that is, collagens and glycosaminoglycans; thus, this is also harvested with the cells³¹. Use of a lysis solution is recommended, in particular for the solubilization of membranes. An example of a suitable lysis solution for the recovers of proteins and enzymatic activities (cell proteins and enzymes) is 1% Triton X-100 solution in PBS containing protease inhibitors, 2 mM phenymethyl-sulfonylfluoride to inhibit serine-proteases and 1 mM N-ethylmaleinimide to inhibit cysteine-proteases^{6, 8, 23}. For our proteomic analysis of HUVECs^{10, 11} we washed three times with PBS and then harvested directly in 2D electrophoresis buffer (6 M urea, 2 M thiourea, 2% w/v CHAPS, 0.3% w/v dithiothreitol, 0.5% immobilized pH gradient buffer, 2 mM phenymethyl-sulfonylfluoride and 5 mM N-ethylmaleinimide).
20. If you wish to subculture the cells, use a solution with 0.25% trypsin and 0.02% EDTA. At the end of the treatment inhibit trypsin activity by washing the harvested cells in complete medium (calcium and proteins in FCS inhibit trypsin and trypsin-like activities).

Timing

Day 1, Steps 1 and 2, preparation of plates: 15 min for seven dishes
Day 2, Steps 3–15, initial processing of the cord: 2 h for one umbilical cord, add 30 min for each additional cord
Days 3–8, Steps 17 and 18,replacing medium and monitoring cell growth: 10 min for seven dishes without counting or particular washing

Troubleshooting

Contamination

Bacteria and fungi can be identified by regular optical microscopy; however, it might be necessary to detect low-level infection by incubation of cell cultures in microbiological broth (fluid thioglycolate medium and tryptone soya broth). This test procedure should be carried out in a microbiology laboratory away from the cell culture laboratory.

Mycoplasma, which is not usually detected by light microscopy because it is intracellular, can be detected either by culture on specific medium (also in microbiology laboratory, giving a result in 3–4 weeks) or by indirect DNA stain (Hoechst 33258) and fluorescence examination (in the cell culture laboratory, giving a result in 2–3 days). Using Hoechst is quicker, but less sensitive.

Fibroblasts and vascular smooth muscle cells can contaminate EC cultures. They have characteristic features that enable them to be identified on microscopic examination. The use of a panel of markers (such as vimentin, -smooth muscle actin, desmin and calponin) is difficult as HUVECs also express these proteins^{10, 11}; only fibroblast specific protein-1, the matrix metalloproteinases-2 and -3, and CD-90 could be more specific for fibroblastic cells. Moreover, HUVECs retain the potential to differentiate into smooth muscle-like cells by deprivation of growth factors³².

Culture of HUVECs without serum

HUVECs can survive in serum-free medium (M199 without FCS) for up to 12 h without losing their phenotypic features. However, for longer experiments, in particular those lasting 24 h, the absence of growth factors triggers apoptosis, an irreversible mechanism. This process can be circumvented by the use of an enriched medium but with a low concentration of proteins such as the commercial medium Ultroser G (Biosepra, cat. no. 259515) (see refs. 6,8,23) at 1% in M199 without FCS; in this medium, HUVECs can survive for 48 h. Some other low-serum media that support HUVECs viability can also be used.

Troubleshooting advice can be found in Table 1.

Table 1: Troubleshooting table.

Full table

Anticipated results

In a series of experiments (n=4), 2 days after confluency, HUVECs contained 1.56 0.49 U mg^{- 1} ACE activity, as determined on synthetic substrate^{23, 33} and expressed as a function of protein content, whereas they secreted in culture medium (as determined on supernatant) 0.643 0.054 U per 24 h per 10⁶ cells. These results correlate with immunofluorescence examination showing fine fluorescent granulates on cell surface using a personal anti-ACE antibody and after paraformaldehyde fixation. In these conditions but after methanol cell permeation, HUVECs were poorly stained with anti-vWF antibody. Vascular smooth muscle cells, as cultured from porcine aortic or pulmonary artery explants, did not contain or secrete ACE activity and were not stained by either anti-ACE or anti-vWF antibodies.

Typically, after isolation of HUVECs from three or four umbilical cords, we could seed twenty-four 35-mm Petri dishes allowing experiments in quadruplates, that is, six conditions such as one control and five concentrations of a drug or a particular molecule, or three controls versus three times of exposition with this particular molecule, that we have made for studying the toxicity and/or apoptosis induced by anticancer drugs^{7, 8, 9}. For our proteomic study using 2D electrophoresis and mass spectrometry, we plated HUVECs on the same dishes but we pooled cell lysates in specific solution, as stated above, from eight dishes (control). The other 16 dishes were used for the examination of apoptosis induced by etoposide, an anticancer drug, at two concentrations, with eight dishes for each concentration, thereafter pooled such as for controls¹¹.