Rapid, fluorometric DNA determination for chick limb ... - Springer Link

In Vitro Cell. Dev. Biol.--Animal 34:158-162, February 1998 © 1998 Society for In Vitro Biology 1071-2690/98 $05.00 + 0.00

RAPID, FLUOROMETRIC DNA DETERMINATION FOR CHICK LIMB-BUD MESENCHYMALCELL MICROCULTURES DOUGLAS F. PAULSEN1, WEI-DANG CHEN, TRESSA SCINEAUX, ANDDOMINIC ADAMS Department of Anatomy, Morehouse School of Medicine, 720 WestviewDrive, S.W., Atlanta, Georgia 30310-1495 (Received 26 February 1997; accepted 24 March 1997)

SUMMARY

Micromass cultures of chick and mouse limb-bud mesenchymal cells are commonly used for in vitro studies of cellular differentiation. Previously, adaptation of these cultures to 96-well plates facilitated analyses of various aspects of cellular behavior and the effects of different media components in these cultures. These adjustments allowed development of a serum-free medium for chick limb-bud mesenchymal ceils and substantially decreased costs associated with media and reagents. Here we report a further development for this model system; a Hoechst 33342-based in situ DNA assay that provides reliable data much more quickly and with considerably less effort than had been feasible in the past. Because it allows quantitation of products of cellular differentiation and DNA in the same cultures, the number of cultures needed to provide the same data is essentially halved and the accuracy of normalized values for quantitative estimates of markers of differentiation is improved. Studies of the effects of retinoic acid on chick limb-bud mesenchymal cells were performed to document the usefulness of this method. Key words: DNA assay; limb-bud: chondrogenesis; cell culture; serum-free medium: retinoic acid. of scraping the cells out of the wells. Complete solubilization of the cultures was not a reliable option because after 4 d in culture, every solubilizer used still left DNA residue behind in the wells. Further, even if solubilization had been possible, obtaining DNA values from companion cultures rather than from the wells from which the other assessments were made seemed a less than optimal approach to normalizing these values. Other investigators have adapted Hoechst- and DAPI-based fluorometric DNA assays to various culture conditions (7,10, 11,19,20,22). As with our previous assay, this new assay involves measurement of fluorescence enhancement due to quantitative interactions between DNA and Hoechst (33258 or 33342). Besides its speed and accuracy (approximately 2 h vs. over 24 h per plate), the method's greatest advantage is that it allows data on both products of differentiation and accumulation of DNA to be measured within the same culture well. This increases the accuracy of the normalized values and decreases the total number of cultures needed to obtain the same information.

INTRODUCTION

The chick limb-bud has proven a useful model for evaluating the role of specific cell behaviors during complex morphogenetic processes. Finch and Zwilling (5) were among the first to grow chick limb-bud mesenchymal cells in culture. They used tissue culture as a way to begin isolating individual cellular activities important in limb-bud morphogenesis. Ahrens et al. (1) advanced the model by developing micromass cultures of these cells, allowing them to undergo cartilage differentiation in vitro without the large cell numbers previously required to achieve the requisite high cell densities (4). Adapting these cultures to 96-well microtiter plates (17) proved to be a valid approach to reducing the amount of expensive media additives and treatments to which the cells were to be exposed in studies of factors affecting their growth and differentiation. DNA assays are often necessary in such studies, either to gauge the effects of various in vitro environments on overall growth, or to normalize data acquired to reflect relative rates of production of differentiation-related antigens and products of cellular synthesis. In earlier studies using the microtiter culture method (14-18), DNA accumulation by chick wing-bud mesenchymal cells exposed to various treatments has been assessed by growing triplicate companion cultures from the same cell suspensions used for other assessments (e.g., chondrogencsis), recovering the tissue from the companion wells using a modified rubber policeman and painstaking washes, followed by sonication of the recovered tissue and fluorometric DNA assays of the sonicates as described by Brunk et al. (3). Initially, the sensitivity of the assay was such that the ability to obtain a separate data point from each culture seemed to offset the tedium

MATERIA1,S AND METHOI)S

Basic procedure. Cells from the indicated regions of stage 23-24 chick wing-buds were dissociated and grown for 4 d in defined medium as described previously (14-16,18). The cultures were then incubated for increasing lengths of time with fresh defined medium containing 10 Itg/ml Hoechst 33342 (Polysciences, Warrington, PA). At the end of each time period, the cells were analyzed on a MieroFLUOR96-well plate reader (Dynatech, Chantilly, VA) and the number of fluorescence units recorded. Preparing DNA stocks. Stock preparation began 1 d prior to harvesting microtiter micromass cultures of chick limb-bud mesenchyme that had been prepared as described in previous papers. A l-mg/ml stock solution (DNA Stock A) was prepared fresh from lyophilized chicken blood DNA (Pharmacia, Piscataway, NJ) and water no less than a day and no more than 2 wk prior to

ITo whom correspondence should be addressed. 158

159

DNA ASSAY FOR CHICK LIMB CULTURES TABLE 1

250.

DNA TRIPLICATE STANDARDS IN 96-WELL PLATE T

200.

DNA Col./Rows

p.g/well

Stock B (p.I)

Water (~1)

Medium (lal)

Hoechsl 33342 (p.l)

T / ~ /

~ 150 "E

A - - A C e l I s + Hoechst + Medium

1

O--O

1/A-C 2/A-C 3/A-C 4/A-C 5/A-C 6/A-C 7/A-C 8/A-C 9/A-C

0.0 0.0 4.0 3.0 2.5 2.0 1.5 1.0 0.5

0 0 80 60 50 40 30 20 10

130 100 20 40 50 60 70 80 90

1.70 170 170 170 170 170 170 170 170

0 30 30 30 30 30 30 30 30

the assay. Dissolution overnight is recommended. Hoechst-associated fluorescence is DNA-specific and provides a measure of the number of AdenineThymine (A-T) pairs present in the sample (3,6,12,19,20). Because different species have different A-T:G-C ratios, the DNA used in the standards should derive from the same species as the DNA in the unknowns (in this ease, chicken). Further, A-T pairs can be uncoupled by mechanical shearing associated with vortexing and/or trituration of the DNA to hurry dissolution, or by the progressive degradation that normally accompanies freezing and thawing or simply refrigerating the stocks for more than 2 wk. These procedures reduce Hoechst-associated fluorescence of the DNA standards without markedly affecting absorbance at 260 nm. Thus, they should be avoided to maintain validity of the standards. A 100 pA aliquot of DNA Stock A (approximately 100 ~tg DNA) was transferred to a disposable centrifuge tube containing 1.9 m[ of deionized water to generate a solution of approximately 50 lag/ml DNA. Aliquots of this solution were analyzed for absorbance at 260 nm in a UV spectrophotometer, adjusting the concentration with additional Stock A or water as needed to achieve an absorbance of 1.0 (50 ~tg/mt; DNA Stock B). In this way, the data obtained from successive standard batches could be reliably compared. Washes. The microtiter cultures that had completed their incubation periods were processed for data collection as described below. Sterile 37 ° C Tyrodes Buffered Salt Solution (TBSS) (Sigma, St. Louis, MO) was transferred into a sterile muhiwell pipette solution tray (Nalge, Rochester, NY). Taking care not to let the cultures dry completely, and not to disturb the cell layers, a clean Pasteur pipette was used to aspirate the medium from the culture wells. A multiwell pipetter was then used to deliver 270 I.tl of sterile 37 ° C TBSS into each well containing a culture, taking care not to disturb the cell layer on the well floor. The wells were never left dry for any extended period. This was followed by a second TBSS wash. Preparing Hoechst 33342. A small amount of Hoeehst 33342 (Polyscienees) stock solution (1 mg Hoechst dye per 1 ml deionized water) was made using glove and mask precautions in a darkened room. A second Hoeehst stock was made by diluting the first with TBSS (1:10) to a concentration of 100 ~g/ml and stored in a light-tight tube. Adding Hoechst 33342 and incubating cultures. Once the last wash was aspirated from each culture, 270 p.[ of serum-free medium (14-16,18) was added to each well. In a darkened room, 30 ~1 of the second Hoechst 33342 stock (100 p_l/ml) was added to each culture well (final Hoeehst 33342 concentration = 10 ~tg/ml). The 96-well plate was covered and returned to the tissue culture incubator for the periods indicated. While the unknowns were incubating, DNA standards were often prepared. Preparing DNA standards. For our usual seeding density (2.5 × 10"scells per well), we routinely set up our standards in a clean 96-well plate as shown in Table 1, beginning by adding the appropriate aliquots of DNA Stock B, then water, then the medium, and finally the Hoechst 33342 (100 ttg/ml in medium). The plate containing the standards was then covered and incubated along with the cultures. Fig. 3 A shows that 20-30 min incubation is sufficient to obtain a stable reading from the standards. DNA readings. At the end of the incubation period, the plate was removed from the incubator in a darkened room, and the condensation wiped from the bottom of the plate. The plate was transferred with minimal exposure to light to the MieroFLUOR 96-well-plate fluorometer. The top of the plate was re-

g ioo g h~

Hoechst + Medium

z~--Z~Cells + Medium O I O Medium Only

•

50

~0

0

~

o

~==----o~

I

I

I

I

I

I

15

30

45

60

75

90

Incubotion Time (Minutes) FIG. 1. Hoechst 33342-associated fluorescence in situ in 4-d cultures of stage 23-24 chick wing-bud mesenchymal cells and controls. Note the dramatic fluorescence enhancement in the wells containing both the cells and the Hoechst dye as compared to either separately• Values are mean -+ standard error.

moved and the bottom positioned in the tray. Reading the plate takes less than 1 min. Two readings were routinely taken for each plate to assure that the readings were stable. Once the DNA readings were taken, light exposure was no longer an issue and the cultures were fixed and stained with Alcian blue 8GX (Sigma, St. Louis, MO) to determine the amount of chondrogenesis (2,8,9,14-18). Alternatively, at this point they could have been subjected to immunohistochemistry to assess the level of expression of developmentally regulated antigens (17). RESULTS AND DISCUSSION

Dye uptake and range of values. Initial studies assisted in determining the kinetics of dye uptake by the cultured cells. Fig. 1 shows that fluorescence levels in 4-d cultures are notably higher in cultures containing cells and Hoechst together than with any of the components individually. It further indicates that dye uptake/fluorescence levels off by about 45 rain at about 200 fluorescence units, at this cell density. It is conceivable that addition of a mild detergent to the Hoechst could enhance membrane permeability and speed equilibration. However, if studies to localize or quantitate specific intracellular or extracellular substances are to accompany the DNA determinations, higher cell-membrane permeability, and additional washes or incubations prior to fixation, could result in redistribution or loss of the substance in question, including the DNA itself. The next step was to estimate the range of fluorescence units that would be encountered in a typical assay. Since using this culture protocol the ceils are seeded at or near confluency, and since in previous work DNA content for cultures grown in defined medium typically approximately doubled, it was estimated that a range of cell numbers from zero to four times the seeding density should encompass most values likely to be encountered. Since 2.5 X l 0 s cells are typically seeded into each well (14--16,18) of a Coming 96 X 1/2well plate, a series of densities of dissociated chick wing-bud mesenchymal cells was prepared by serial dilution and then by checking and adjusting the densities with the aid of a hemacytometer. This allowed analysis of whether cell density dramatically affected the incubation period needed fbr dye uptake. Further, readings were taken for mesenchymal cells isolated from the proximal and distal halves of stage 2 3 - 2 4 wing buds. Because ceils in the distal half of

160

PAULSEN ET AL.

(xl~ cdl,#,~l

t' A o~.O

o ~

A 2000-

Z

e--lO.O 1500i

1000,

•

O

~

/*A~

i

-

A

i

~

O

~

O

~

A ~ A

-,

/~3 DN/VWell - - 10.0

O

_

• --

5.0

/--

, 0

15

30

45

30

'~"

, Lo,o

75

gO

105

0.

120

I 5

0

Incubation Time (Minutes) CelIII/~V_ell

(xl~

i

1"/;

; .It,

e--lO.O

~

--

30

45

30

75

I 25

I 30

I 35

40

90

B

j=.

x_ 0.0 15

I 20

1000-

/ - 1.0

0 0

I 15

1300.

-,,--s.o

/',I".

,r;" " " ' T . "

2000.

,t,

:t°°l;" ,,A ....... ,, .__A--

I 10

Incubotion Time (Minutee)

IB ,rio

1.0

105

o o.o

120

I 5°0

Incubation "nine CMlnutu)

! lO.O

DN/~/Well

300

300

250-

_-- 4OO

~200300 c8 150200

o

100-

iz

50

O:

0.0

: 2.5

I

5.0

I

7.5

I

10.0

lOO o

-1.o

0:0

I

I

I

I

1.0

2.0

3.0

4.0

5.0

/~ DNA/Well FIG. 2. Hoechst 33342 dye uptake and fluorescence over time as a function of cell number. A, Ceils obtained from the distal halves of stage 23-24 chick wing buds. B, Cells obtained from the proximal halves of stage 23-24 chick wing buds. C, Regression plot of fluorescence values as a function of cell number at the 45-rain time point for the proximal cuhures. Values are mean + standard error.

the wing bud are generally understood to divide more rapidly than those in the proximal half (13,21), we examined whether the source of the ceils had any influence on dye uptake. Fluorescence readings similar to those described above were then taken at different cell densities. Fig. 2 A and B show that again equilibrium was reached by about 45 rain of incubation even at the highest cell density, and regardless of the limb region of origin. Further, there was a clear indication that fluorescence was proportionate to the number of cells

FIG. 3. Hoechst 33342 dye uptake and fluorescence over time as a function of DNA concentration. A, Chicken blood DNA used as standard, after initial flash fluorescence stabilizes. B, Regression plot of fluorescence values as a function of DNA concentration over a broad range (0-15 rtg/well). C, Regression plot of fluorescence values as a function of DNA concentration over a narrow range covering the values expected in these cultures. Values are mean + standard error.

present. This was borne out in the regression analysis plotted in Fig. 2 C. Because the cell seeding at a density of 2.5 × 105 per well is at or near confluence for these wells, any increase in cell numbers over the culture period is likely to result in some cells overlying others. The reasonable concern regarding the assay's reliability in cases where cell numbers exceed those found in a monolayer is in part addressed by the fact that even at cell numbers fourfold as high

161

DNA ASSAY FOR CHICK LIMB CULTURES

TABLE 2

250-

TOTAL DNA PER WELL IN UNTREATED 4-D CULTURES OF CHICK WING-BUD MESENCHYME

20O

A

Stage23-24 Wing-BudRegion Distal Proximal

Average~g DNA/well-+ SD (n= 6) PreviousMethod CurrentMethod 0.311 + 0.098 0.259 + 0.097

8c

0.482 + 0.049 0.572 + 0.083

as those needed to reach confluency, the regression shows that the fluorescence intensity per cell remains linear (Fig. 2 C). Standardization and confirmation. To improve the usefulness of the assay and to help standardize the values obtained from one experiment to the next, known amounts of purified DNA were delivered to defined medium in identical 96-well plates, after which the same final concentration of Hoechst 33342 (10 ~tg/ml) was added to each well as was added to those containing the cultured cells. Fig. 3 A shows that after temporary flash of fluorescence that was most pronounced at the higher DNA concentrations at 5 min of incubation, the fluorescence levels stabilized by about 20 min. Because, in the standards, the Hoechst need not penetrate cell membranes to reach the DNA, it is not surprising that stable fluorescence levels were reached more quickly in these wells. The regression plotted in Fig. 3 B shows that fluorescence is proportionate to the amount of DNA present over a wide range of DNA values. To further enhance the usefulness of the method, a standard curve (Fig. 3 C) was generated that included more standard DNA concentrations straddling the fluorescence values expected based on the cultures examined in Fig. 2 and on our experience with DNA values obtained from the Brunk et al. (3) assay for similar cultures (14-18). This shows that fluorescence is proportionate to DNA content even over a relatively narrow range of DNA concentrations. It is this series of standards that we now typically use along with each assay. Table 2 compares the absolute DNA amounts per culture obtained using the method described in this article with those obtained using our earlier method. While the values overlap allowing both to be correct, the new method gives higher values with smaller standard errors. Thus, by this measure as well, the new method is preferable to our previous approach. Finally, to confirm the utility of this assay in obtaining quantitative estimates of the effects of particular treatments on developmentally related behaviors of chick wing-bud mesenchymal cells, we used this model to examine the effects of the potent Vitamin A analog and teratogen, retinoic acid (RA), on growth and cartilage differentiation in microtiter cultures of chick wing-bud mesenchymal cells. Fig. 4 A shows that the two RA concentrations used both enhanced growth of cells from the distal halves of stage 23-24 chick wing buds, while they inhibited growth in cuhures of cells from the proximal halves from these same limbs. Fig. 4 B shows the results of colorimetric assays of chondrogenesis in these cultures (17) based on the extraction of Alcian-blue-stained cartilage matrix. Here, companion cultures from the same cell suspensions were assayed for chondrogenesis with or without prior growth analysis using the Hoechst 33342-based assay. Clearly, this figure shows that there is no difference in the values obtained for chondrogenesis in the paired cultures despite the differences in the treatments or the regions from which

S

!

100

.rr

50

o

m

DISTN..

PROXIMAL HALF"

DISTN. HAlF

PROXIMAl. HALF"

0.'40"

[3 0.;30-

r-'qW'rchout Hoech~ 33342 P".'.'.uWlthHoechst 33342

i

i== 0.20 0.10U

I*R

0.00

5O

0

0

5

5O

[Retinolc Acid] (ng,/ml) 2.50 "C"

---8

OISTN..HALF

PROXIMALHALF"

1.~

0.50 0.0{

0

5

5O

0

5

5O

[Retdnolc Acld] (ng,/ml)

FIG. 4. Applicability of in situ fluorometric DNA determinations to microcultures of stage 23-24 chick wing-bud mesenchymal cells grown for 4 d in the presence or absence of retinoic acid (RA). A, Hoechst 33342-associated fluorescence in treated and untreated cultures from different wing-bud regions and at different RA concentrations. RA stimulates DNA accumulation in cultures of the distal cells, but appears to have a dose-related inhibitory effect on DNA accumulation in cultures of the proximal cells. B, Quantitative estimates of chondrogenesis in treated and untreated companion cultures. Comparison of values obtained for chondrogenesis in cultures with or without previous in situ fluorometric DNA determinations. Although there are dramatic differences in chondrogenesis depending on the source of the mesenchymal cells and the treatments they received, there are no significant differences between values obtained with or without the DNA determinations. C, Data expressed as the relative amount of cartilage matrix produced (chondrogenesis) per cell under the different experimental conditions. Because both measurements can be made in situ in the same culture, quantitative estimates of cartilage differentiation can be more reliably normalized to the cell numbers present than when values from companion cultures are artificially combined. Values are mean + standard error.

162

PAULSEN ET AL.

the cells were obtained. Since both the DNA values and chondrogenesis values could be obtained from the same culture without changing the chondrogenesis values, it is now possible to obtain and graph values that reflect the effects of these treatments on the degree of phenotypic expression (chondrogenesis) per cell in these cultures (Fig. 4 C). These findings confirm our previous findings of regiondependent and concentration-dependent differences in the effects of RA on the growth of these cells (14,18). Thus, we expect this assay to become exceptionally useful as we and others explore the effects of a variety of growth factors, growth inhibitors, hormones, and nutrients on developmentally related behaviors by these cells. This method for quantitating total DNA in microcuhures has proven rapid and reliable and facilitates normalization of various measures of cellular differentiation to the n u m b e r of cells present. While this method was developed for use with primary cultures of chick limb-bud mesenchymal cells, it can be relatively easily adapted to any cell type and is likely to even be easier to use in studies where measurements are made on treated and untreated cell lines. Studies are currently underway to adapt whole mount in situ hybridization methods to these microcultures to allow sensitive quantitative estimates to be made of the relative level of expression of developmentally regulated genes per cell in various in vitro conditions. ACKNOWLEDGMENTS The authors thank Dr. Karla Daniels for initially directing us toward related methods in the literature. The studies described were supported by NIH Minority Biomedical Research Support Grant GM08248 and by NASA grant NAG 9-644. Tressa Scineaux is a medical student at Louisiana State University and was a Minority Access to Research Careers scholar at Spelman College while contributing to this study. Some of the work was done with equipment in core research facilities supported by NIH Research Centers in Minority Institutions Grant RR03034. REFERENCES 1. Ahrens, P. B.; Solursh, M.; Reiter, R. S. Stage-related capacity for limb chondrogenesis in cell culture. Dev. Biol. 60:69-82; 1977. 2. Biddulph, D. M.; Sawyer, L. M.; Dozier, M. M. Chondrogenesis in chick limb mesenchyme in vitro derived from distal limb bud tips: changes in cyclic AMP and in prostaglandin responsiveness. J. Cell. Physiol. 136:81-87; 1988. 3. Brunk, C. E; Jones, K. C.; James, T. W. Assay for nanogram quantities of DNA in cellular homogenates. Anal. Biochem. 92:497-500; 1979. 4. Daniels, K.; Reiter, R.; Solursh, M. Micromass cultures of limb and other mesenchyme. In: Bronner-Fraser, M., ed. Methods in avian embryology. New York: Academic Press; 1996:237-247.

5. Finch, R. A.; Zwilling, E. Culture stability of morphogenetic properties of chick limb-bud mesoderm. J. Exp. Zool. 176:397408; 1971. 6. Kapuscinski, J.; Skoczylas, B. Simple and rapid fluorimetric method for DNA microassay. Anal. Biochem. 83:252-257; 1977. 7. Labarca, C.; Paigen, K. A simple, rapid, and sensitive DNA assay. Anal. Biochem. 102:344-352; 1980. 8. Leonard, C. M.; Bergman, M.; Frenz, D. A., et al. Abnormal ambient glucose levels inhibit prnteoglycan core protein gene expression and reduce proteoglycan accumulation during chondrogenesis: possible mechanism for teratogeuic effects of maternal diabetes. Proc. Natl. Acad. Sci. USA 86:10113-10117; 1989. 9. Lev, R.; Spicer, S. S. Specific staining of sulfate groups with alcian blue at low pH. J. Histochem. Cytochem. 12:309; 1964. 10. Meyer, J. Ch.; Grundmann, H. Fluorometric determination of DNA in epidermis and cultured fibroblasts using 4'-6-diamidino-2-phenylindole (DAPI). Arch. Dermatol. Res. 276:52-56; 1984. l 1. Nagata, Y.; Yokota, H.; Kosuda, O., et al. Quantification of picogram levels of specific DNA immobilized in microtiter wells. FEBS Lett. 183:379382; 1985. 12. Otto, F.; %ou, K. C. A comparative study of DAPI, DIPI and Hoechst 33258 and 33342 as chromosomal DNA stains. Stain Technol. 60:711; 1985. 13. Paulsen, D. E Retinoic acid in limb-bud outgrowth: review and hypothesis. Anat. Embryol. 190:399~!~15; 1994. 14. Paulsen, D. E; Chen, W.-D.; Okello, D., et al. Stage- and region-dependent responses of chick wing-bud mesenchymal cells to retinoic acid in serum-free microcultures. Dev. Dyn. 201:310-323; 1994. 15. Paulsen, D. E; Chen, W.-D.; Pang, L., et al. Stage- and region-dependent chondrogenesis and growth of chick wing-bud mesenchyme in serumcontaining and defined tissue-culture media. Dev. Dyn. 200:39-52; 1994. 16. Paulsen, D. E; Langille, R. M.; Dress, V., et aI. Selective stimulation of in vitro limb-bud chondrogenesis by retinoic acid. Differentiation 39:123-130; 1988. 17. Paulsen, D. E; Solursh, M. Microtiter micromass cultures of limb-bud mesenchymal cells. In Vitro Cell. Dev. Biol. 24:138-147; 1988. 18. Paulsen, D. F.; Solursh, M.; Langille. R. M., et al. Stable position-related responses to retinoie acid by chick limb-bud mesenchymal cells in serum-free cultures. In Vitro Cell. Dev. Biol. 30A:181-186; 1994. 19. Plummer, H. K.; Dillard, C. Y.; Murray, S. A. Hoechst 33342 for microfluorometric measurement of adrenocortical tumor cell proliferation. In Vitro Cell. Dev. Biol. 28A:4-6; 1992. 20. Richards, W. L.; Song, M.-K.; Krutsch, H., et al. Measurement of cell proliferation in microculture using Hoechst 33342 for the rapid semiautomated microfluorometric determination of chromatin DNA. Exp. Cell Res. 159:235-246; 1985. 21. Summerbell, D.; Wolpert, L. Cell density and cell division in the early morphogenesis of the chick wing. Nat. New Biol. 239:24-26; 1972. 22. West, D. C.; Sattar, A.; Kumar, S. A simplified in situ solubilization procedure for the determination of DNA and cell number in tissue cultured mammalian cells. Anal. Biochem. 147:289-295, 1985.