- Email: [email protected]
Semiquantitative polymerase chain reaction in RNase-producing tissues: Analysis of the developing pancreas
Semiquantitative Polymerase Chain Reaction in RNase-Producing Tissues: Analysis of the Developing Pancreas By Mark J. Hembree, Krishna Prasadan, Pradip Manna, Barry Preuett, Troy Spilde, Amina Bhatia, Hiroyuki Kobayashi, Brendhan Buckingham, Charles L. Snyder, and George K. Gittes Kansas City, Missouri
Background/Purpose: Many studies in pediatric surgical research use a quantitative analysis of gene expression in microscopic quantities of tissue. The authors describe an analysis of the beta-tubulin mRNA content of the embryonic pancreas, which contains abundant endogenous RNases. A detailed analysis of this RNase-containing system will provide a good template for analysis of other potentially simpler systems. Methods: Embryonic mouse pancreases were harvested at serial gestational ages. DAPI nuclear staining allowed for counting of cells. cDNA was amplified using a fluoresceinated primer and the normalized fluorescence determined. Known numbers of molecules were amplified in parallel as a standard control.
mRNA for beta-tubulin did not increase proportionately. Assuming a yield of 100% at E10.5 when no RNases are present, the yield of expected mRNA was 65.3% at E12.5, 13.8% at E15.5, and 0.9% at E18.5, presumably because of the appearance of RNases.
Conclusions: Several parameters must be considered in performing semiquantitative reverse transcription polymerase chain reaction: (1) the yield of RNA based on the projected amount of mRNA, (2) the number of cells in the tissue, and (3) a known number of template molecules amplified in parallel. J Pediatr Surg 36:1629-1632. Copyright © 2001 by W.B. Saunders Company.
Results: The number of cells increased from 38,000 to 2,700,000 between embryonic day 10.5 (E10.5) and E18.5.
INDEX WORDS: Pancreas, quantitative reverse transcription polymerase chain reaction; ribonuclease, RNase.
P
are given for insulin mRNA over increasing gestational ages with and without accounting for RNase-mediated degradation.
EDIATRIC SURGICAL RESEARCH frequently aims to study processes of differentiation and development during organogenesis. Previously, one of the major hurdles in the study of microscopic developing organs has been the proper analysis of levels of gene expression. With the advent of reverse transcription polymerase chain reaction (RT-PCR) in the late 1980s, there were renewed capabilities for the detection of small quantities of RNA.1 These techniques were a great improvement over previous techniques for quantifying mRNA, including Northern hybridization and RNase protection. Unfortunately, the ability to quantify these levels of mRNA has not progressed very quickly. In particular, the ability to use RT-PCR to quantify mRNA levels in tissues rich in RNases has been a significant problem.2 In this report we present evidence that there can be significant degradation of RNA by RNases in the developing pancreas and, by controlling for the degradation by using a “gold standard” control (ie, counting of the absolute number of cells in a specimen), we can account for this significant degradation and project a more accurate level of mRNA for genes of interest. We report here increasing cell numbers in the developing pancreas over gestational ages, as well as the degree of degradation of mRNA as evidenced by levels of beta-tubulin mRNA. As an example of the significance of this degradation, levels
MATERIALS AND METHODS
Cell Counting of Embryonic Pancreas Embryos were harvested from pregnant CD1 mice corresponding to day 10.5, 12.5, 15.5, and 18.5 of gestation. The embryos were preserved on ice in Dulbecco’s Modified Eagle’s Medium (DMEM). Microdissection was performed on the individual embryos to isolate the pancreas. A 4% paraformaldehyde solution in PBS was prepared and filtered with a 22-m filter. The pancreases were fixed in the paraformaldehyde for 12 to 16 hours at 4°C, transferred to 70% ethanol for 15 minutes at 4°C, and then transferred to 30% sucrose (in PBS) for 4 to 8 hours at 4°C. Specimens were frozen-embedded in OCT and cut into 9-m sections using a Bright cryostat (Huntingdon, Cambs, England). The sections were rehydrated in PBS for 5 minutes and then treated for 10 minutes in 4⬘,6-diamidino-2-phenylindole (DAPI, Sigma, St Louis,
From Children’s Mercy Hospital, Kansas City, MO. Presented at the 34th Annual Meeting of the Pacific Association of Pediatric Surgeons, Kyoto, Japan, April 4-8, 2001. Address reprint requests to George K. Gittes, MD, Associate Professor of Surgery, UMKC, Director, Surgical Research, Children’s Mercy Hospital, Director, Laboratory for Surgical Organogenesis, Children’s Mercy Hospital, 2401 Gillham Rd, HHC 623C, Kansas City, MO 64108. Copyright © 2001 by W.B. Saunders Company 0022-3468/01/3611-0006$35.00/0 doi:10.1053/jpsu.2001.27934
Journal of Pediatric Surgery, Vol 36, No 11 (November), 2001: pp 1629-1632
1629
1630
HEMBREE ET AL
Fig 1. Number of cells per pancreas at serial embryonic ages. Total number of cells per whole pancreas at embryonic ages 10.5, 12.5, 15.5, and 18.5. Cell counting was done using DAPI nuclear staining. There was an exponential increase in the number of cells per pancreas.
MD; 1 g/mL in PBS) for nuclear staining. Slides were rinsed in phosphate-buffered saline (PBS) and mounted with Cytoseal 60 (Stephens Scientific, Kalamazoo, MI) mounting media and coverslips. Sections were viewed and imaged with a fluorescence microscope under ultraviolet illumination. Image-Pro Plus (Media Cybernetics, Silver Spring, MD) software was used to count the nuclei in the digital image of each section.
Measuring Beta-Tubulin mRNA Molecules Embryos were harvested as above at days 10.5, 12.5, 15.5, and 18.5 of gestation. The pancreases were placed in a cryogenic vial and then quick-frozen in liquid nitrogen. RNA isolation was performed using the RNeasy Mini Kit (Qiagen, Valencia, CA) immediately following dissection. Reverse transcription was performed on the isolated RNA with Sensiscript Reverse Transcriptase (Qiagen). Quantitative PCR was performed using the Amplifluor Uniprimer (Intergen, Purchase, NY) and primers for beta-tubulin and insulin. Oligonucleotide primers for beta-tubulin were as follows: sense CTGTTCAAGCGCATCTCTGA, antisense CCTCCTCTTCTGCCTCCTCT (202 base pair product). Oligonucleotide primers for insulin were as follows: sense TCTCTACCTGGTGTGTGG, antisense AGTTGCAGTAGTTCTCCAG (211 bp product). Known dilutions (102 to 106) of beta-tubulin PCR product were amplified in parallel to generate a standard curve. PCR parameters were as follows: denaturation at 95°C for 15 seconds, annealing at 55°C for 20 seconds, and elongation at 72°C for 40 seconds, 40 total cycles. PCR product was transferred to a 96-well micro-titer plate. End-point analysis of fluorescent PCR product was performed using CytoFluor II fluorescence plate reader. The fluorescence readings for the known dilutions of beta-tubulin PCR product were used to generate a standard curve. Linear regression analysis of the experimental fluorescence readings was used to determine the initial number of betatubulin mRNA molecules.
RESULTS
The number of cells for embryonic day 10.5, 12.5, 15.5, and 18.5 pancreas was 38,000, 110,000, 590,000, and 2,700,000, respectively. These cell counts represent a 2.9-fold increase in cells from E10.5 to E12.5, a
5.4-fold increase in cells from E12.5 to E15.5, and a 4.6-fold increase in cells from E15.5 to E18.5 (Fig 1). Total calculated beta-tubulin mRNA molecules for embryonic day 10.5, 12.5, 15.5, and 18.5 pancreas were 1.9 ⫻ 107, 3.6 ⫻ 107, 4.1 ⫻ 107, and 1.2 ⫻ 107, respectively (Fig 2). Levels of beta-tubulin mRNA were expected to increase proportionately to the increase in cell count because of its function as a structural protein. The E10 specimen was expected to be the least affected by endogenous RNases because pancreatic RNase gene expression does not occur until late in gestation.3 Therefore, the beta-tubulin mRNA molecules for E12.5, E15.5, and E18.5 were normalized against the number of mRNA molecules measured in the E10.5 pancreas. The expected number of mRNA molecules in the absence of RNases was determined by calculating an increase in beta-tubulin mRNA molecules proportional to the increase in the total cell number. The yield of expected beta-tubulin mRNA at increasing gestational ages, assuming a normalized yield of 100% at E10.5, was 65.3% at E12.5, 13.8% at E15.5, and 0.9% at E18.5. The insulin expression at embryonic day 11.5 (E11.5) and E15.5 also was examined by semiquantitative RTPCR. We chose these gestational dates because insulin expression in the embryonic pancreas begins to be upregulated near E11 and shows a secondary burst of expression around day E14 –15. The mRNA expression of insulin increased 157-fold from E11.5 to E15.5. Assuming that the effects of RNases on E11.5 are minimally different from E10.5, then the mRNA yield at E15.5 of 13.8% would apply. Thus, the actual increase in insulin mRNA is over 1,138-fold, when controlling for
Fig 2. Beta-tubulin expression in pancreas at serial embryonic ages. The number of beta-tubulin mRNA molecules for embryonic ages 10.5, 12.5, 15.5, and 18.5 was determined by semiquantitative RT-PCR (F). The expected number of beta-tubulin mRNA molecules (■) was calculated as proportional to the increase in the number of cells and assuming 100% yield at embryonic day 10.5.
QUANTITATIVE PCR IN RNASE-PRODUCING TISSUES
1631
Fig 3. Relative insulin mRNA expression at embryonic ages 11.5 and 15.5. Results from semiquantitative RT-PCR were normalized for E11.5 insulin mRNA molecules equaling 1. Observed results indicated a 157-fold increase in insulin mRNA expression from E11.5 to E15.5. Results adjusted for RNA degradation indicate an 1,138-fold increase in insulin expression from E11.5 to E15.5.
the degradation caused by endogenous RNase activity (Fig 3). DISCUSSION
Our ability to measure gene expression in tissues has increased dramatically over the last 15 years. The advent of Northern blot hybridization and RNase protection assays has allowed us to quantify transcript levels in tissues, but these techniques are not overly sensitive and therefore are not easily applied to studying gene expression in developing embryonic tissues. Dealing with microscopic quantities of tissue is an intrinsic hurdle that must be dealt with in studies of organogenesis pertinent to pediatric surgical research, and, thus, we wished to determine an ideal system for studying gene expression in small amounts of cells. Standard RT-PCR has become routine for studying gene expression in small amounts of tissues, but is notorious for being difficult to quantify. Several techniques for semiquantitative and quantitative PCR have been developed, but these systems tend to be fraught with difficulty.4-8 One of the major difficulties occurs in tissues that express significant levels of RNase
or have significant levels of RNase in them. The pancreas is the most abundant source of RNase in the body and is used as a digestive enzyme secreted by the acinar cells to help digest the RNA within consumed tissues. In addition, pancreatic-type RNases have been shown to be expressed in kidney, liver, spleen, and several other tissues.9 RNases typically are released from intracellular sequestration immediately after tissue harvesting, and, therefore, RNase-rich tissues are particularly difficult to analyze for mRNA expression. Because RNase develops after about day 14 or 15 in the developing mouse pancreas, we wish to see the effect on sq-RT-PCR of the ontogeny of RNase as it relates to sensitivity of the sq-RT-PCR system.3 The results described here show that there is a dramatic fall-off in recovery and sensitivity of the semiquantitative RT-PCR system that correlates well with the expression of RNases. Previous studies have shown that RNase activity peaks around day 18 or 19 of gestation.3 This was the time when we observed a dramatic fall-off in our recovery of RNA. The use of cell counting in this report, although cumbersome, allowed us to determine precisely the true growth of the pancreas and therefore to determine precisely the amount of beta-tubulin mRNA that should have been present. We normalized the quantity of insulin and beta-tubulin mRNA to the whole pancreas instead of the per-cell expression of the 2 genes, because a per-cell comparison would not be accurate because insulin is only expressed in a subset of cells. These results require a basic assumption that there is a 100% yield of RNA at day 10.5 of gestation. Although this assumption intrinsically may be inaccurate, we feel that it is adequate to study the role of RNases, because no pancreatic RNase gene mRNA is detectable at these embryonic ages. The overall results imply that accounting for the presence of RNase is extremely important for proper determination of RNA levels, and that semiquantitative RNA levels should be dependent strictly not only on internal controls such as beta-tubulin, but also predicated on a knowledge of potential degradation by RNases. The results of these studies will help guide us in further studies of embryonic pancreatic gene expression to accurately determine the levels of critical molecules involved in organogenesis patterning.
REFERENCES 1. Rappolee DA, Mark D, Banda MJ, et al: Wound macrophages express TGF-alpha and other growth factors in vivo: Analysis by mRNA phenotyping. Science 241:708-712, 1988 2. Chirgwin JM, Pryzbyla AE, MacDonald RJ, et al: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:5294-5299, 1979 3. Han JH, Rall L, Rutter WJ: Selective expression of rat pancreatic
genes during embryonic development. Proc Natl Acad Sci U S A 83:110-114, 1986 4. Tyagi S, Kramer FR: Molecular beacons: Probes that fluoresce upon hybridization. Nat Biotechnol 14:303-308, 1996 5. Morrison TB, Weis JJ, Wittwer CT: Quantification of low-copy transcripts by continuous SYBR Green I monitoring during amplification. Biotechniques 24:954-962, 1998 6. Wittwer CT, Herrmann MG, Moss AA, et al: Continuous fluo-
1632
rescence monitoring of rapid cycle DNA amplification. Biotechniques 22:130-138, 1997 7. Gut M, Leutenegger CM, Huder JB, et al: One-tube fluorogenic reverse transcription-polymerase chain reaction for the quantification of feline coronaviruses. J Virol Methods 77:37-46, 1999
HEMBREE ET AL
8. Freeman WM, Walker SJ, Vrana KE: Quantitative RT-PCR: Pitfalls and potential. Biotechniques 26:112-115, 1999 9. Futami J, Tsushima Y, Murato Y, et al: Tissue-specific expression of pancreatic-type RNases and RNase inhibitor in humans. DNA Cell Biol 16:413-419, 1997
Background/Purpose: Many studies in pediatric surgical research use a quantitative analysis of gene expression in microscopic quantities of tissue. The authors describe an analysis of the beta-tubulin mRNA content of the embryonic pancreas, which contains abundant endogenous RNases. A detailed analysis of this RNase-containing system will provide a good template for analysis of other potentially simpler systems. Methods: Embryonic mouse pancreases were harvested at serial gestational ages. DAPI nuclear staining allowed for counting of cells. cDNA was amplified using a fluoresceinated primer and the normalized fluorescence determined. Known numbers of molecules were amplified in parallel as a standard control.
mRNA for beta-tubulin did not increase proportionately. Assuming a yield of 100% at E10.5 when no RNases are present, the yield of expected mRNA was 65.3% at E12.5, 13.8% at E15.5, and 0.9% at E18.5, presumably because of the appearance of RNases.
Conclusions: Several parameters must be considered in performing semiquantitative reverse transcription polymerase chain reaction: (1) the yield of RNA based on the projected amount of mRNA, (2) the number of cells in the tissue, and (3) a known number of template molecules amplified in parallel. J Pediatr Surg 36:1629-1632. Copyright © 2001 by W.B. Saunders Company.
Results: The number of cells increased from 38,000 to 2,700,000 between embryonic day 10.5 (E10.5) and E18.5.
INDEX WORDS: Pancreas, quantitative reverse transcription polymerase chain reaction; ribonuclease, RNase.
P
are given for insulin mRNA over increasing gestational ages with and without accounting for RNase-mediated degradation.
EDIATRIC SURGICAL RESEARCH frequently aims to study processes of differentiation and development during organogenesis. Previously, one of the major hurdles in the study of microscopic developing organs has been the proper analysis of levels of gene expression. With the advent of reverse transcription polymerase chain reaction (RT-PCR) in the late 1980s, there were renewed capabilities for the detection of small quantities of RNA.1 These techniques were a great improvement over previous techniques for quantifying mRNA, including Northern hybridization and RNase protection. Unfortunately, the ability to quantify these levels of mRNA has not progressed very quickly. In particular, the ability to use RT-PCR to quantify mRNA levels in tissues rich in RNases has been a significant problem.2 In this report we present evidence that there can be significant degradation of RNA by RNases in the developing pancreas and, by controlling for the degradation by using a “gold standard” control (ie, counting of the absolute number of cells in a specimen), we can account for this significant degradation and project a more accurate level of mRNA for genes of interest. We report here increasing cell numbers in the developing pancreas over gestational ages, as well as the degree of degradation of mRNA as evidenced by levels of beta-tubulin mRNA. As an example of the significance of this degradation, levels
MATERIALS AND METHODS
Cell Counting of Embryonic Pancreas Embryos were harvested from pregnant CD1 mice corresponding to day 10.5, 12.5, 15.5, and 18.5 of gestation. The embryos were preserved on ice in Dulbecco’s Modified Eagle’s Medium (DMEM). Microdissection was performed on the individual embryos to isolate the pancreas. A 4% paraformaldehyde solution in PBS was prepared and filtered with a 22-m filter. The pancreases were fixed in the paraformaldehyde for 12 to 16 hours at 4°C, transferred to 70% ethanol for 15 minutes at 4°C, and then transferred to 30% sucrose (in PBS) for 4 to 8 hours at 4°C. Specimens were frozen-embedded in OCT and cut into 9-m sections using a Bright cryostat (Huntingdon, Cambs, England). The sections were rehydrated in PBS for 5 minutes and then treated for 10 minutes in 4⬘,6-diamidino-2-phenylindole (DAPI, Sigma, St Louis,
From Children’s Mercy Hospital, Kansas City, MO. Presented at the 34th Annual Meeting of the Pacific Association of Pediatric Surgeons, Kyoto, Japan, April 4-8, 2001. Address reprint requests to George K. Gittes, MD, Associate Professor of Surgery, UMKC, Director, Surgical Research, Children’s Mercy Hospital, Director, Laboratory for Surgical Organogenesis, Children’s Mercy Hospital, 2401 Gillham Rd, HHC 623C, Kansas City, MO 64108. Copyright © 2001 by W.B. Saunders Company 0022-3468/01/3611-0006$35.00/0 doi:10.1053/jpsu.2001.27934
Journal of Pediatric Surgery, Vol 36, No 11 (November), 2001: pp 1629-1632
1629
1630
HEMBREE ET AL
Fig 1. Number of cells per pancreas at serial embryonic ages. Total number of cells per whole pancreas at embryonic ages 10.5, 12.5, 15.5, and 18.5. Cell counting was done using DAPI nuclear staining. There was an exponential increase in the number of cells per pancreas.
MD; 1 g/mL in PBS) for nuclear staining. Slides were rinsed in phosphate-buffered saline (PBS) and mounted with Cytoseal 60 (Stephens Scientific, Kalamazoo, MI) mounting media and coverslips. Sections were viewed and imaged with a fluorescence microscope under ultraviolet illumination. Image-Pro Plus (Media Cybernetics, Silver Spring, MD) software was used to count the nuclei in the digital image of each section.
Measuring Beta-Tubulin mRNA Molecules Embryos were harvested as above at days 10.5, 12.5, 15.5, and 18.5 of gestation. The pancreases were placed in a cryogenic vial and then quick-frozen in liquid nitrogen. RNA isolation was performed using the RNeasy Mini Kit (Qiagen, Valencia, CA) immediately following dissection. Reverse transcription was performed on the isolated RNA with Sensiscript Reverse Transcriptase (Qiagen). Quantitative PCR was performed using the Amplifluor Uniprimer (Intergen, Purchase, NY) and primers for beta-tubulin and insulin. Oligonucleotide primers for beta-tubulin were as follows: sense CTGTTCAAGCGCATCTCTGA, antisense CCTCCTCTTCTGCCTCCTCT (202 base pair product). Oligonucleotide primers for insulin were as follows: sense TCTCTACCTGGTGTGTGG, antisense AGTTGCAGTAGTTCTCCAG (211 bp product). Known dilutions (102 to 106) of beta-tubulin PCR product were amplified in parallel to generate a standard curve. PCR parameters were as follows: denaturation at 95°C for 15 seconds, annealing at 55°C for 20 seconds, and elongation at 72°C for 40 seconds, 40 total cycles. PCR product was transferred to a 96-well micro-titer plate. End-point analysis of fluorescent PCR product was performed using CytoFluor II fluorescence plate reader. The fluorescence readings for the known dilutions of beta-tubulin PCR product were used to generate a standard curve. Linear regression analysis of the experimental fluorescence readings was used to determine the initial number of betatubulin mRNA molecules.
RESULTS
The number of cells for embryonic day 10.5, 12.5, 15.5, and 18.5 pancreas was 38,000, 110,000, 590,000, and 2,700,000, respectively. These cell counts represent a 2.9-fold increase in cells from E10.5 to E12.5, a
5.4-fold increase in cells from E12.5 to E15.5, and a 4.6-fold increase in cells from E15.5 to E18.5 (Fig 1). Total calculated beta-tubulin mRNA molecules for embryonic day 10.5, 12.5, 15.5, and 18.5 pancreas were 1.9 ⫻ 107, 3.6 ⫻ 107, 4.1 ⫻ 107, and 1.2 ⫻ 107, respectively (Fig 2). Levels of beta-tubulin mRNA were expected to increase proportionately to the increase in cell count because of its function as a structural protein. The E10 specimen was expected to be the least affected by endogenous RNases because pancreatic RNase gene expression does not occur until late in gestation.3 Therefore, the beta-tubulin mRNA molecules for E12.5, E15.5, and E18.5 were normalized against the number of mRNA molecules measured in the E10.5 pancreas. The expected number of mRNA molecules in the absence of RNases was determined by calculating an increase in beta-tubulin mRNA molecules proportional to the increase in the total cell number. The yield of expected beta-tubulin mRNA at increasing gestational ages, assuming a normalized yield of 100% at E10.5, was 65.3% at E12.5, 13.8% at E15.5, and 0.9% at E18.5. The insulin expression at embryonic day 11.5 (E11.5) and E15.5 also was examined by semiquantitative RTPCR. We chose these gestational dates because insulin expression in the embryonic pancreas begins to be upregulated near E11 and shows a secondary burst of expression around day E14 –15. The mRNA expression of insulin increased 157-fold from E11.5 to E15.5. Assuming that the effects of RNases on E11.5 are minimally different from E10.5, then the mRNA yield at E15.5 of 13.8% would apply. Thus, the actual increase in insulin mRNA is over 1,138-fold, when controlling for
Fig 2. Beta-tubulin expression in pancreas at serial embryonic ages. The number of beta-tubulin mRNA molecules for embryonic ages 10.5, 12.5, 15.5, and 18.5 was determined by semiquantitative RT-PCR (F). The expected number of beta-tubulin mRNA molecules (■) was calculated as proportional to the increase in the number of cells and assuming 100% yield at embryonic day 10.5.
QUANTITATIVE PCR IN RNASE-PRODUCING TISSUES
1631
Fig 3. Relative insulin mRNA expression at embryonic ages 11.5 and 15.5. Results from semiquantitative RT-PCR were normalized for E11.5 insulin mRNA molecules equaling 1. Observed results indicated a 157-fold increase in insulin mRNA expression from E11.5 to E15.5. Results adjusted for RNA degradation indicate an 1,138-fold increase in insulin expression from E11.5 to E15.5.
the degradation caused by endogenous RNase activity (Fig 3). DISCUSSION
Our ability to measure gene expression in tissues has increased dramatically over the last 15 years. The advent of Northern blot hybridization and RNase protection assays has allowed us to quantify transcript levels in tissues, but these techniques are not overly sensitive and therefore are not easily applied to studying gene expression in developing embryonic tissues. Dealing with microscopic quantities of tissue is an intrinsic hurdle that must be dealt with in studies of organogenesis pertinent to pediatric surgical research, and, thus, we wished to determine an ideal system for studying gene expression in small amounts of cells. Standard RT-PCR has become routine for studying gene expression in small amounts of tissues, but is notorious for being difficult to quantify. Several techniques for semiquantitative and quantitative PCR have been developed, but these systems tend to be fraught with difficulty.4-8 One of the major difficulties occurs in tissues that express significant levels of RNase
or have significant levels of RNase in them. The pancreas is the most abundant source of RNase in the body and is used as a digestive enzyme secreted by the acinar cells to help digest the RNA within consumed tissues. In addition, pancreatic-type RNases have been shown to be expressed in kidney, liver, spleen, and several other tissues.9 RNases typically are released from intracellular sequestration immediately after tissue harvesting, and, therefore, RNase-rich tissues are particularly difficult to analyze for mRNA expression. Because RNase develops after about day 14 or 15 in the developing mouse pancreas, we wish to see the effect on sq-RT-PCR of the ontogeny of RNase as it relates to sensitivity of the sq-RT-PCR system.3 The results described here show that there is a dramatic fall-off in recovery and sensitivity of the semiquantitative RT-PCR system that correlates well with the expression of RNases. Previous studies have shown that RNase activity peaks around day 18 or 19 of gestation.3 This was the time when we observed a dramatic fall-off in our recovery of RNA. The use of cell counting in this report, although cumbersome, allowed us to determine precisely the true growth of the pancreas and therefore to determine precisely the amount of beta-tubulin mRNA that should have been present. We normalized the quantity of insulin and beta-tubulin mRNA to the whole pancreas instead of the per-cell expression of the 2 genes, because a per-cell comparison would not be accurate because insulin is only expressed in a subset of cells. These results require a basic assumption that there is a 100% yield of RNA at day 10.5 of gestation. Although this assumption intrinsically may be inaccurate, we feel that it is adequate to study the role of RNases, because no pancreatic RNase gene mRNA is detectable at these embryonic ages. The overall results imply that accounting for the presence of RNase is extremely important for proper determination of RNA levels, and that semiquantitative RNA levels should be dependent strictly not only on internal controls such as beta-tubulin, but also predicated on a knowledge of potential degradation by RNases. The results of these studies will help guide us in further studies of embryonic pancreatic gene expression to accurately determine the levels of critical molecules involved in organogenesis patterning.
REFERENCES 1. Rappolee DA, Mark D, Banda MJ, et al: Wound macrophages express TGF-alpha and other growth factors in vivo: Analysis by mRNA phenotyping. Science 241:708-712, 1988 2. Chirgwin JM, Pryzbyla AE, MacDonald RJ, et al: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:5294-5299, 1979 3. Han JH, Rall L, Rutter WJ: Selective expression of rat pancreatic
genes during embryonic development. Proc Natl Acad Sci U S A 83:110-114, 1986 4. Tyagi S, Kramer FR: Molecular beacons: Probes that fluoresce upon hybridization. Nat Biotechnol 14:303-308, 1996 5. Morrison TB, Weis JJ, Wittwer CT: Quantification of low-copy transcripts by continuous SYBR Green I monitoring during amplification. Biotechniques 24:954-962, 1998 6. Wittwer CT, Herrmann MG, Moss AA, et al: Continuous fluo-
1632
rescence monitoring of rapid cycle DNA amplification. Biotechniques 22:130-138, 1997 7. Gut M, Leutenegger CM, Huder JB, et al: One-tube fluorogenic reverse transcription-polymerase chain reaction for the quantification of feline coronaviruses. J Virol Methods 77:37-46, 1999
HEMBREE ET AL
8. Freeman WM, Walker SJ, Vrana KE: Quantitative RT-PCR: Pitfalls and potential. Biotechniques 26:112-115, 1999 9. Futami J, Tsushima Y, Murato Y, et al: Tissue-specific expression of pancreatic-type RNases and RNase inhibitor in humans. DNA Cell Biol 16:413-419, 1997