- Email: [email protected]
Optimal method for isolation of human peritoneal mesothelial cells from clinical samples of omentum
TECHNICAL REPORT
Optimal method for isolation of human peritoneal mesothelial cells from clinical samples of omentum Manel Riera School of Clinical Sciences, University of Livelpool, Liverpool L69 3BX Peter McCulloch Nuffield Department of Surgery, John Radcliffe Hospital, Headley Way, Headington, Oxford OX3 9DU Laszlo Pazmany School of Clinical Sciences, University of Liverpool, Liverpool L69 3BX Thomas Jagoe School of Clinical Sciences, University of Liverpool, Liverpool L69 3BX
Key words: mesothelial cells, peritoneum, culture, fibroblasts, contamination Received 19 January 2006, accepted for publication 13 April 2006
Abstract Introduction: Human peritoneal mesothelial cells (HPMC) are a valuable research tool for wlderstanding the molecular biology of several pathologies, in both monolayer and three dimensional models. We compared different methods of HPMC isolation and assessed their outcome as well as fibroblast contamination, a common problem encountered during isolation. Methods: 1-3cm' samples of omentum were collected from 40 consenting patients undergoing elective gastrointestinal surgery. A total of 11 samples were incubated in 0.05% trypsin solution for 20 minutes at 3 lOC (group A) and 29 in 0.25% trypsin (15 samples for 10 minutes (group B) and 14 for 20 minutes (group C)). Following digestion cells were re-suspended and cultured in supplemented Ham's F-12 medium containing 10% foetal calf serum (FCS), penicillinstreptomycin, glutamine, insulin, transferrin and hydrocortisone. Positive outcomes were absence of fibroblast contamination and satisfactory HPMC growth to confluence in a characteristic cobblestone pattern. Cytokeratins 5, 8, 18, Vimentin, Ber-Ep4 and Factor VIII were used to characterise HPMC and fibroblasts by immunohistochemistry. Results: None of the 11 samples in group A yielded HPMC. 14 of 29 samples digested with 0.25% trypsin yielded HPMC: 10 of 14 yielded HPMC in group C versus
© 2006 Tissue Viability Society
22
four of 15 samples in group B (p=0.02). Fibroblast contamination occurred in eight samples in group B versus three in group C. Conclusion: Optimal results are acl1ieved with a 20 minute digestion in 0.25% trypsin. Fibroblast contamination could not be avoided completely. Other factors may minimise fibroblast contamination such as minimal tissue manipulation and early collection during surgery.
Introduction The peritoneum lines the abdominal cavity and viscera and is composed of peritoneal mesothelial cells (HPMC) supported by fibrovascular connective tissue. Diseases of the peritoneum are common and include non-malignant adhesions following abdominal surgery, fibrovascular adhesions after peritoneal dialysis and peritoneal invasion by metastases from tumours of the gastrointestinal tract and the ovary. The early stages of development of peritoneal disease are difficult to detect because of the anatomical localisation of this tissue, and this has hampered studies of pathophYSiology. To address this issue, different tissue culture models of the peritoneum have been developed using isolated human mesothelial cell monolayers alone, or in combination with a variety of connective tissue supports l "'. Though HPMC can be obtained from commercial sources (Coriel Cell Repositories, Camden, New Jersey)', most authors harvest HPMC from omental tissue,,5 or peritoneal effusions6 ,7 obtained from local
VOL 16 NO.4 NOVEMBER 2006
clinical sources. However, satisfactory isolation and growth of HPMC is crucial to the success of any of the tissue culture models used. Unfortunately HPMC isolation is challenging, and often unsuccessful. Commonly encountered problems include poor yield of HPMC from different isolation techniques and contamination of HPMC with fibroblasts that grow much more rapidly and dominate the combined cell preparations, thereby making them unsuitable use in studies aimed at understanding mechanisms of disease in the peritoneum. Isolation of HPMC from omentum usually requires protease digestion to release HPMC from the underlying connective tissue. The use of different concentrations of trypsin for varying lengths of time has been reported. Some studies reported success digesting omentum with trypsin 0.25% for 20 minutes", others used 0.05% trypsin for five minutes9 To our knowledge there are no studies comparing differences in outcome, and there is a pressing need to evaluate the efficacy of the methods currently in use. Thus, we aimed to compare three different methods of HPMC isolation and assessed success of subsequent HPMC growtll in culture and presence of fibroblast contamination.
Methods All HPMC were isolated from 1-3cm3 samples of omentum collected from consenting patients undergoing elective gastrointestinal surgery (local ethical committee approved). Omental samples were inlmediately placed in fresh Ham's Fl2 Medium (Sigma-Aldrich, Dorset, UK) in theatre, then taken to the laboratory and washed in phosphate buffered saline (pH7.4) (Sigma-Aldrich, Dorset, UK). Every effort was made to standardise and minimise manipulation of omental samples. The washed omental samples were incubated at 3]0 in a hUmidified atmosphere of 95% air and 5% CO 2 and treated in one of the following three ways: method A (n=ll) involved incubation in 0.05% trypsin (Gibco, Invitrogen Ltd Paisley, UK) solution for 20 minutes; method B (n=15) involved incubation in 0.25% trypsin for 10 minutes; and method C (n=14) involved incubation in 0.25% trypsin for 20 minutes. Thereafter all samples were centrifuged in 50ml flasks (Greiner Bio-One Ltd, Stonehouse, UK) at 500g for five minutes. The resulting cell pellet was then re-suspended in Ham's F-12 medium supplemented with 10% foetal calf serum, penicillin streptomycin (5~g/ml), glutamine (Gibco, Invitrogen Ltd Paisley, UK), insulin (O.lIU/ml) (Novo Nordisk, Bagsv:erd, Denmark), transferrin (5~g/ml) and hydrocortisone (O.4~g/ml) (Solucortef®, Pfizer, New York, US). Finally the isolated cells were grown to confluence in 0.2% gelatin coated 25cm 2 tissue culture flasks (Falcon, Beckton Dickinson, Oxford, UK). Tissue culture and cell characterisation were assessed by light microscopy and immunohistochemistry with the follOWing antibodies: Cytokeratins 5, 8, 18, virnentin, factor VIII and Ber EP4 (DakoCytomation, Ely, UK). Satisfactory
VOL 16 NO.4 NOVEMBER 2006
FIGURE 1 LIGHT MICROSCOPE PICTURE (40X MAGNIFICATION) OF HPMC
This shows the typical cobblestone pattern of HPMC grown to confluence (10 days in culture)
growth to confluence in a characteristic cobblestone pattern and absence of fibroblast contamination (Figure I, Figure 2) were the expected outcome.
Results HPMC stained positive with cytokeratins and virnentin and negative with Factor VIII and Ber EN, while fibroblasts were positive with virnentin only. No HPMC were recovered using method A (0.05%
FIGURE 2 LIGHT MICROSCOPE PICTURE (lOX MAGNIFICATION) DEMONSTRATING FIBROBLAST CONTAMINATION
'.
•
This shows the fibroblast contamination (2) of HPMC (1) after isolation and culture (nine days in CUlture)
23
TECHNICAL REPORT
TABLE 1 GROWlrH OF MESOTHELIAL CELLS RECOVERED FROM HUMAN OMENTAL BIOPSIES USING THREE DIFFERENT DIGESTION METHODS
Number Satisfactory growth Absent growth Fibroblast contamination
Method A
Method B
Method C
11
15
14
o
4 (26%)
10 (71%)
11 (100%)
3 (21%)
4 (%)
o
8 (53%)
3 (21%)
trypsin for 20 minutes), whereas 71% of samples using method C (0.25% trypsin for 20 minutes), and 26% of samples using method B (0.25% trypsin for 10 minutes) yielded HPMC Furthermore the proportion of samples yielding HPMC was significantly higher for method C compared to method B (C versus B, Fisher's exact test p=0.02) (Table 1). The presence of fibroblast contamination was also more common for method B (53%) than method C (21 %) although this difference was not statistically significant (Fisher's exact test p=0.13). Therefore, we have shown that there are considerable differences in HPMC yield using different published methods.
In addition to assessing HPMC yield our data suggest a trend towards increased fibroblast contamination in method B versus method C The differences observed did not reach statistical significance and need further corroboration. However, these results do reinforce the result that the method C for HPMC isolation (0.25% trypsin for 20 minutes) is superior.
Acknowledgements This work was funded by the Aintree Research on Cancer charity. We would like to thank Mr M Aguirreburualde of the Walton Centre for Neurology and Neurosurgery in Liverpool for his supervision and analysis of the immunohistochemical experiments..
Discussion Our results suggest that digestion with 0.25% trypsin for 20 minutes is the best method among those tested. The presence of fibroblast contamination in both method Band C illustrates that this is a significant problem that must be borne in mind when performing studies with isolated HPMC The fibroblast contamination observed might indicate the peritoneal mesothelial cell monolayer present in the omentum was disrupted, increasing the chances that subsequent trypsin digestion and centrifugation released fibroblasts in addition to HPMC In our study all omental samples were immediately washed, kept moist, and manipulated as little as possible in an effort to minimise potential trauma to the mesothelial cell monolayer after removal from the peritoneal cavity. However, it is possible that trauma to the peritoneal mesothelial cell monolayer occurred during surgery, and indeed this type of damage is thought to explain the development of peritoneal adhesions after abdominal surgery. We did not collect data on the timing of collection of omental samples in relation to the abdominal surgery being performed, but it might be appropriate to ensure that omental samples for HPMC isolation should be collected at the beginning of the surgical procedure, to reduce the chances of peritoneal trauma due to the surgical manipulation.
© 2006 Tissue Viability Society
24
References 1. Sawada M, Shii J, Akedo H, Tanizawa O. An experimental model for ovarian tumor invasion of cultured mesothelial cell monolayer. Laborat01Y Investigation 1994; 70(3): 333-8. 2. Jayne DG, Perry SL, Morrison E, Farmery SM, Guillou Pl. Activated mesothelial cells produce heparin-binding growth factors: implications for tumour metastases. British Joumal of Cancer 2000; 82(6): 1233-8. 3. Casey RC, Koch KA, Oegema TR Jr., Skubitz KM, Pambuccian SE, et al. Establishment of an in vitro assay to measure the invasion of ovarian carcinoma cells through mesothelial cell monolayers. Clinical and Experimental Metastasis 2003; 20(4): 343-56. 4. Stylianou E, Jenner LA, Davies M, Coles GA, Williams JD. Isolation, culture and characterization of human peritoneal mesothelial cells. Kidney Intemational1990; 37(6): 1563-70. 5. Zhang XY, Pettengell R, Nasiri N, Kalia V, Dalgleish AG, Barton DP. Characteristics and growth patterns of human peritoneal mesothelial cells: comparison between advanced epithelial ovarian cancer and non-ovarian cancer sources. Joumal of the Society for Gynecologic Inuestigation 1999; 6(6): 333-40. 6. Rieppi M, Vergani V, Gatto C, Zanetta G, Allavena P, Taraboletti G, et al. Mesothelial cells induce the motility of human ovarian carcinoma cells. Intemational Joumal of Cancer 1999; 80(2): 303-7. 7. Sakata K, Shigemasa K, Uebaba Y, Nagai N, Ohama K. Expression of matrix metalloproteinases-2 and -9 by cells isolated from the peritoneal fluid of women with ovarian carcinoma. Acta Cytologica 2002; 46(4): 697-703. 8. Mizutani K, Kofuji K, Shirouzu K. The significance of MMP1 and MMP-2 in peritoneal disseminated metastasis of gastric cancer. SurgeJy Today 2000; 30(7): 614-21. 9. Bittinger F, Brochhausen C, Skarke C, Kohler H, Kirkpatrick Cl. Reconstruction of peritoneal-like structure in threedimensional collagen gel matrix culture. Experimental Cell Research 1997; 236(1): ] 55-60.
VOL 16 NO.4 NOVEMBER 2006
Optimal method for isolation of human peritoneal mesothelial cells from clinical samples of omentum Manel Riera School of Clinical Sciences, University of Livelpool, Liverpool L69 3BX Peter McCulloch Nuffield Department of Surgery, John Radcliffe Hospital, Headley Way, Headington, Oxford OX3 9DU Laszlo Pazmany School of Clinical Sciences, University of Liverpool, Liverpool L69 3BX Thomas Jagoe School of Clinical Sciences, University of Liverpool, Liverpool L69 3BX
Key words: mesothelial cells, peritoneum, culture, fibroblasts, contamination Received 19 January 2006, accepted for publication 13 April 2006
Abstract Introduction: Human peritoneal mesothelial cells (HPMC) are a valuable research tool for wlderstanding the molecular biology of several pathologies, in both monolayer and three dimensional models. We compared different methods of HPMC isolation and assessed their outcome as well as fibroblast contamination, a common problem encountered during isolation. Methods: 1-3cm' samples of omentum were collected from 40 consenting patients undergoing elective gastrointestinal surgery. A total of 11 samples were incubated in 0.05% trypsin solution for 20 minutes at 3 lOC (group A) and 29 in 0.25% trypsin (15 samples for 10 minutes (group B) and 14 for 20 minutes (group C)). Following digestion cells were re-suspended and cultured in supplemented Ham's F-12 medium containing 10% foetal calf serum (FCS), penicillinstreptomycin, glutamine, insulin, transferrin and hydrocortisone. Positive outcomes were absence of fibroblast contamination and satisfactory HPMC growth to confluence in a characteristic cobblestone pattern. Cytokeratins 5, 8, 18, Vimentin, Ber-Ep4 and Factor VIII were used to characterise HPMC and fibroblasts by immunohistochemistry. Results: None of the 11 samples in group A yielded HPMC. 14 of 29 samples digested with 0.25% trypsin yielded HPMC: 10 of 14 yielded HPMC in group C versus
© 2006 Tissue Viability Society
22
four of 15 samples in group B (p=0.02). Fibroblast contamination occurred in eight samples in group B versus three in group C. Conclusion: Optimal results are acl1ieved with a 20 minute digestion in 0.25% trypsin. Fibroblast contamination could not be avoided completely. Other factors may minimise fibroblast contamination such as minimal tissue manipulation and early collection during surgery.
Introduction The peritoneum lines the abdominal cavity and viscera and is composed of peritoneal mesothelial cells (HPMC) supported by fibrovascular connective tissue. Diseases of the peritoneum are common and include non-malignant adhesions following abdominal surgery, fibrovascular adhesions after peritoneal dialysis and peritoneal invasion by metastases from tumours of the gastrointestinal tract and the ovary. The early stages of development of peritoneal disease are difficult to detect because of the anatomical localisation of this tissue, and this has hampered studies of pathophYSiology. To address this issue, different tissue culture models of the peritoneum have been developed using isolated human mesothelial cell monolayers alone, or in combination with a variety of connective tissue supports l "'. Though HPMC can be obtained from commercial sources (Coriel Cell Repositories, Camden, New Jersey)', most authors harvest HPMC from omental tissue,,5 or peritoneal effusions6 ,7 obtained from local
VOL 16 NO.4 NOVEMBER 2006
clinical sources. However, satisfactory isolation and growth of HPMC is crucial to the success of any of the tissue culture models used. Unfortunately HPMC isolation is challenging, and often unsuccessful. Commonly encountered problems include poor yield of HPMC from different isolation techniques and contamination of HPMC with fibroblasts that grow much more rapidly and dominate the combined cell preparations, thereby making them unsuitable use in studies aimed at understanding mechanisms of disease in the peritoneum. Isolation of HPMC from omentum usually requires protease digestion to release HPMC from the underlying connective tissue. The use of different concentrations of trypsin for varying lengths of time has been reported. Some studies reported success digesting omentum with trypsin 0.25% for 20 minutes", others used 0.05% trypsin for five minutes9 To our knowledge there are no studies comparing differences in outcome, and there is a pressing need to evaluate the efficacy of the methods currently in use. Thus, we aimed to compare three different methods of HPMC isolation and assessed success of subsequent HPMC growtll in culture and presence of fibroblast contamination.
Methods All HPMC were isolated from 1-3cm3 samples of omentum collected from consenting patients undergoing elective gastrointestinal surgery (local ethical committee approved). Omental samples were inlmediately placed in fresh Ham's Fl2 Medium (Sigma-Aldrich, Dorset, UK) in theatre, then taken to the laboratory and washed in phosphate buffered saline (pH7.4) (Sigma-Aldrich, Dorset, UK). Every effort was made to standardise and minimise manipulation of omental samples. The washed omental samples were incubated at 3]0 in a hUmidified atmosphere of 95% air and 5% CO 2 and treated in one of the following three ways: method A (n=ll) involved incubation in 0.05% trypsin (Gibco, Invitrogen Ltd Paisley, UK) solution for 20 minutes; method B (n=15) involved incubation in 0.25% trypsin for 10 minutes; and method C (n=14) involved incubation in 0.25% trypsin for 20 minutes. Thereafter all samples were centrifuged in 50ml flasks (Greiner Bio-One Ltd, Stonehouse, UK) at 500g for five minutes. The resulting cell pellet was then re-suspended in Ham's F-12 medium supplemented with 10% foetal calf serum, penicillin streptomycin (5~g/ml), glutamine (Gibco, Invitrogen Ltd Paisley, UK), insulin (O.lIU/ml) (Novo Nordisk, Bagsv:erd, Denmark), transferrin (5~g/ml) and hydrocortisone (O.4~g/ml) (Solucortef®, Pfizer, New York, US). Finally the isolated cells were grown to confluence in 0.2% gelatin coated 25cm 2 tissue culture flasks (Falcon, Beckton Dickinson, Oxford, UK). Tissue culture and cell characterisation were assessed by light microscopy and immunohistochemistry with the follOWing antibodies: Cytokeratins 5, 8, 18, virnentin, factor VIII and Ber EP4 (DakoCytomation, Ely, UK). Satisfactory
VOL 16 NO.4 NOVEMBER 2006
FIGURE 1 LIGHT MICROSCOPE PICTURE (40X MAGNIFICATION) OF HPMC
This shows the typical cobblestone pattern of HPMC grown to confluence (10 days in culture)
growth to confluence in a characteristic cobblestone pattern and absence of fibroblast contamination (Figure I, Figure 2) were the expected outcome.
Results HPMC stained positive with cytokeratins and virnentin and negative with Factor VIII and Ber EN, while fibroblasts were positive with virnentin only. No HPMC were recovered using method A (0.05%
FIGURE 2 LIGHT MICROSCOPE PICTURE (lOX MAGNIFICATION) DEMONSTRATING FIBROBLAST CONTAMINATION
'.
•
This shows the fibroblast contamination (2) of HPMC (1) after isolation and culture (nine days in CUlture)
23
TECHNICAL REPORT
TABLE 1 GROWlrH OF MESOTHELIAL CELLS RECOVERED FROM HUMAN OMENTAL BIOPSIES USING THREE DIFFERENT DIGESTION METHODS
Number Satisfactory growth Absent growth Fibroblast contamination
Method A
Method B
Method C
11
15
14
o
4 (26%)
10 (71%)
11 (100%)
3 (21%)
4 (%)
o
8 (53%)
3 (21%)
trypsin for 20 minutes), whereas 71% of samples using method C (0.25% trypsin for 20 minutes), and 26% of samples using method B (0.25% trypsin for 10 minutes) yielded HPMC Furthermore the proportion of samples yielding HPMC was significantly higher for method C compared to method B (C versus B, Fisher's exact test p=0.02) (Table 1). The presence of fibroblast contamination was also more common for method B (53%) than method C (21 %) although this difference was not statistically significant (Fisher's exact test p=0.13). Therefore, we have shown that there are considerable differences in HPMC yield using different published methods.
In addition to assessing HPMC yield our data suggest a trend towards increased fibroblast contamination in method B versus method C The differences observed did not reach statistical significance and need further corroboration. However, these results do reinforce the result that the method C for HPMC isolation (0.25% trypsin for 20 minutes) is superior.
Acknowledgements This work was funded by the Aintree Research on Cancer charity. We would like to thank Mr M Aguirreburualde of the Walton Centre for Neurology and Neurosurgery in Liverpool for his supervision and analysis of the immunohistochemical experiments..
Discussion Our results suggest that digestion with 0.25% trypsin for 20 minutes is the best method among those tested. The presence of fibroblast contamination in both method Band C illustrates that this is a significant problem that must be borne in mind when performing studies with isolated HPMC The fibroblast contamination observed might indicate the peritoneal mesothelial cell monolayer present in the omentum was disrupted, increasing the chances that subsequent trypsin digestion and centrifugation released fibroblasts in addition to HPMC In our study all omental samples were immediately washed, kept moist, and manipulated as little as possible in an effort to minimise potential trauma to the mesothelial cell monolayer after removal from the peritoneal cavity. However, it is possible that trauma to the peritoneal mesothelial cell monolayer occurred during surgery, and indeed this type of damage is thought to explain the development of peritoneal adhesions after abdominal surgery. We did not collect data on the timing of collection of omental samples in relation to the abdominal surgery being performed, but it might be appropriate to ensure that omental samples for HPMC isolation should be collected at the beginning of the surgical procedure, to reduce the chances of peritoneal trauma due to the surgical manipulation.
© 2006 Tissue Viability Society
24
References 1. Sawada M, Shii J, Akedo H, Tanizawa O. An experimental model for ovarian tumor invasion of cultured mesothelial cell monolayer. Laborat01Y Investigation 1994; 70(3): 333-8. 2. Jayne DG, Perry SL, Morrison E, Farmery SM, Guillou Pl. Activated mesothelial cells produce heparin-binding growth factors: implications for tumour metastases. British Joumal of Cancer 2000; 82(6): 1233-8. 3. Casey RC, Koch KA, Oegema TR Jr., Skubitz KM, Pambuccian SE, et al. Establishment of an in vitro assay to measure the invasion of ovarian carcinoma cells through mesothelial cell monolayers. Clinical and Experimental Metastasis 2003; 20(4): 343-56. 4. Stylianou E, Jenner LA, Davies M, Coles GA, Williams JD. Isolation, culture and characterization of human peritoneal mesothelial cells. Kidney Intemational1990; 37(6): 1563-70. 5. Zhang XY, Pettengell R, Nasiri N, Kalia V, Dalgleish AG, Barton DP. Characteristics and growth patterns of human peritoneal mesothelial cells: comparison between advanced epithelial ovarian cancer and non-ovarian cancer sources. Joumal of the Society for Gynecologic Inuestigation 1999; 6(6): 333-40. 6. Rieppi M, Vergani V, Gatto C, Zanetta G, Allavena P, Taraboletti G, et al. Mesothelial cells induce the motility of human ovarian carcinoma cells. Intemational Joumal of Cancer 1999; 80(2): 303-7. 7. Sakata K, Shigemasa K, Uebaba Y, Nagai N, Ohama K. Expression of matrix metalloproteinases-2 and -9 by cells isolated from the peritoneal fluid of women with ovarian carcinoma. Acta Cytologica 2002; 46(4): 697-703. 8. Mizutani K, Kofuji K, Shirouzu K. The significance of MMP1 and MMP-2 in peritoneal disseminated metastasis of gastric cancer. SurgeJy Today 2000; 30(7): 614-21. 9. Bittinger F, Brochhausen C, Skarke C, Kohler H, Kirkpatrick Cl. Reconstruction of peritoneal-like structure in threedimensional collagen gel matrix culture. Experimental Cell Research 1997; 236(1): ] 55-60.
VOL 16 NO.4 NOVEMBER 2006