Contrasting patterns of c-myc and N-myc expression in proliferating, quiescent, and differentiating cells of the embryonic chicken lens

c-myc and N-myc are nuclear, DNA-binding proteins thought to play a role in cell proliferation and differentiation (reviewed by Alt et al., 1986; Boettinger, 1989). Although their exact functions are not yet known, their ability to bind DNA in a sequence-specific manner (Blackwell et al., 1990) and their structural similarity to known transcription factors (Murre et al., 1989a) suggests that they may be transcription factors. Both proteins contain leucine zipper regions similar to those in the fos- and jun-families of transcription factors (Landschultz et al., 1988), as well as the helix-loop-helix motif found in MyoD, immunoglobulin kappa enhancer binding proteins, and products of the Drosophila genes, achaete scute and daughterless (Murre et al., 1989a). Since these structural motifs have been shown to participate in protein-protein interactions (Murre et al., 1989b; O’Shea et al., 1989), their presence in c-myc and N-myc (as well as other myc family proteins, (Alt et al., 1986)) implies that these proteins have the capacity to form homodimers and heterodimers. Homodimers or higher order oligomers of c-myc have been shown to bind selectively to the sequence CACGTG (Blackwell et al., 1990). In addition, specific proteins have been identified which form heterodimers with both c-myc and N-myc altering the efficiency with which c-myc binds to the CACGTG site (Blackwood and Eisenman, 1991; Prendergast et al., 1991).

Evidence that c-myc and N-myc may be developmentally important regulators of cell proliferation and differentiation has come from studies of their expression in embryonic tissues and cultured cell lines. In vitro studies indicate that c-myc expression is associated with proliferation and is usually down-regulated during differentiation (reviewed by Cole, 1986). Nonetheless, there is no apparent correlation between cell proliferation and c-myc expression in developing embryos (Pfeifer-Ohlsson et al., 1985; Zimmerman et al., 1986; Downs et al., 1989; Schmid et al., 1989), and elevated c-myc expression is seen during differentiation of certain cell types (reviewed by Zelenka, 1990). One possible explanation for the lack of a clear correlation between expression of c-myc and N-myc, on the one hand, and either proliferation or differentiation, on the other hand, may be that c-myc protein has multiple functions stemming from its ability to form a variety of dimeric or oligomeric complexes. Furthermore, since N-myc and c-myc proteins may compete with each other for the same dimerization partners (Blackwood and Eisenman, 1991), the ratio of N-myc to c-myc expression may shift the population of proteins available to dimerize with c-myc. Thus, the relative concentrations of these two proto-oncogene products in developing tissues may be more significant than the absolute concentration of either. The present study investigates the developmental pattern of both c-myc and N-myc expression in the developing chicken lens to determine their relationship to each other during growth and differentiation.

The pattern of growth and differentiation in the lens (reviewed by Piatigorsky, 1981) leads to the spatial separation of quiescent, dividing and differentiating cells. Dividing cells in the epithelial cell layer that covers the anterior surface of the lens continually migrate toward the lens equator, where they differentiate to form lens fiber cells (see Fig. 1). Differentiating cells undergo their last mitosis anterior to the lens equator and begin to elongate to form the “annular pad”. Elongating cells later detach from the lens capsule, which is their basement membrane, and are added to the periphery of the fiber cell mass. The location and number of proliferating and differentiating cells in various regions of the lens has been determined as a function of developmental age (Persons and Modak, 1970). We have used this information to correlate the relative concentrations of c-myc and N-myc mRNA in different regions of the developing lens with cell proliferation and differentiation. While this study has focused on mRNA expression, expression of the corresponding proteins is likely to follow a similar pattern, since both of these mRNAs and proteins have very short half-lives (Hann and Eisenman, 1984; Dani et al., 1984; Slamon et al., 1986; Ramsay et al., 1986; Sawai et al., 1990).

Lenses from 6-, 10-, 14- and 19-day-old embryos of white Leghorn chickens (Truslow Farms, Chestertown, MD) were removed and cleaned of any adhering pigmented iris. The lens capsule/epithelium was separated from the fiber mass by microdissection with sharpened jeweler’s forceps. Each epithelium was further dissected into a central region, as described by Persons and Modak (1970) and an equatorial region, composed of the remainder of the epithelium (Fig. 1A). In certain experiments, the epithelium was dissected into an “inner” region, containing the central region and germinative zone, and an “outer” region, composed of postmitotic annular pad cells (Fig. 1B). Tissues from 12 to 24 embryos were pooled for extraction of RNA as described by Milstone et al. (1976). RNA concentration was determined by absorbance at 260 nm.

A 397 nt region of chicken c-myc mRNA (1018/1414) spanning the exon2/exon3 junction (Watson et al., 1983) and a 195 nt region of chicken N-myc mRNA (1024/1218) (Sawai et al., 1990) were chosen for reverse transcription and amplification by the polymerase chain reaction (PCR) (Mullis and Faloona, 1987; Saiki et al., 1988) (Fig. 2). The RNA secondary structure prediction program, FOLD (Zuker and Stiegler, 1981), was used to ensure that the chosen regions were free of stable secondary structures which might interfere with reverse transcription (Pallansch et al., 1990). Oligonucleotides (20-mers) used for PCR were chosen to allow discrimination between chicken c-myc (Watson et al., 1983) and N-myc (Sawai et al., 1990) mRNAs (Fig. 2). These and other oligonu-cleotides used in this study were synthesized on an Applied Biosystems 380B DNA synthesizer (Foster City, CA).

Reverse transcription was carried out as described in Pallansch et al. (1990) with the following modifications: oligonucleotide concentration was reduced to 1 μM, reaction volume was reduced to 12.5 μl, incubation at 37°C was reduced to 10 minutes, and the samples were heated to 68°C following reverse transcription. Molony murine leukemia virus reverse transcriptase was obtained from Stratagene (LaJolla, CA), and was used at a concentration of 15 units per reaction.

PCR amplification of the cDNA synthesized by reverse transcription of 50 ng of total cytoplasmic RNA was performed as described by Pallansch et al. (1990) in a total reaction volume of 50 μl. Amplification was carried out for 30 cycles in a Perkin Elmer/Cetus thermal cycler programmed for 1.5 minutes at 94°C; 2 minutes at 42°C; 3 minutes at 72°C. The final cycle was followed by a 10 minute extension step at 72°C. DNA internal standards (described below) were included in the PCR reaction at a concentration of 3 fg per reaction (c-myc) and 1 fg per reaction (N-myc). Taq DNA polymerase was obtained from Perkin-Elmer Cetus Corp. (Norwalk, CT). ³²P-labeled PCR products were separated on 8% polyacrylamide/urea gels in parallel with ³²P-labeled molecular weight markers (MspI digest of pBR322 (New England Biolabs, Beverly, MA) or a 123 bp DNA ladder (BRL, Bethesda, MD)). Electrophoresis was carried out at 1750 V for 90 minutes; the gels were dried and

Oligonucleotides containing the reverse complement of sequences from c-myc (1242/1261) and N-myc (1102/1121) mRNAs were end-labeled with [γ-³²P]ATP (sp. act. >7000 Ci/mmol, ICN, Irvine, CA) using T4 polynucleotide kinase (Pharmacia, Piscataway, NJ) as described by Sambrook et al. (1989). Approximately 5 μg of unlabeled c-myc and N-myc PCR products were electrophoresed on a 2% agarose gel at 50 V for 2-3 hours and subsequently transferred to GeneScreen Plus nylon membranes (Dupont, Boston, MA) by capillary blotting. Hybridization with ³²P-labeled oligonucleotides was performed according to standard protocols (Sambrook et al., 1989) at 57°C (c-myc) or 59°C (N-myc). After appropriate washes, the membranes were autoradiographed on Kodak XAR 5 film overnight at −80°C. autoradiographed at −80°C on Kodak XAR 5 film. After one to three days exposure, the film was developed and the bands were quantitated by scanning densitometry. Experiments were performed a minimum of five times for each of the three regions of the lens at each developmental age, using at least four separate RNA preparations. Controls lacking reverse transcriptase were included in each experiment to ensure that products were derived from RNA. DNA internal standards were synthesized from the c-myc and N-myc PCR products, using the method of sequence overlap extension (Ho et al., 1989) to delete the regions corresponding to nucleotides 1198 to 1269 of c-myc mRNA, and 1083 to 1157 of N-myc mRNA. Thus, the internal standards could be amplified with the same oligonucleotides used to generate the original PCR products, but yielded shorter products, 324 nt for c-myc, 120 nt for N-myc. The concentration of internal standard DNA needed in the PCR reaction was determined by titrating a given amount of RNA (50 ng) with increasing amounts of internal standard DNA (0.01-10 fg). The ratio of PCR products derived from reverse transcribed mRNA and the internal standard was used to compare the relative levels of mRNA from different areas of the lens at different developmental ages in each experiment (Gilliland et al., 1990). Because of possible differences in reverse transcriptase activity, the results of different experiments were compared by normalizing the results of each experiment. Thus, concentrations of mRNA are expressed in arbitrary units.

Pulse-labeling of embryonic lenses with [³H]thymidine was performed as described by Nath et al. (1987). Epithelia were removed and the epithelial cells lysed with 1% sodium dodecyl sulfate in 0.01 M Tris, pH 7.4. [³H]thymidine-labeled DNA was purified by extraction with CHCl₃:phenol:isoamyl alcohol 50:50:1 and ethanol precipitation (Sambrook et al., 1989). Duplicate aliquots of DNA were removed for determination of tritium incorporation by scintillation counting, and triplicate determinations of DNA content were made on the remainder, using the ELISA assay of Richardson and McAvoy (1989).

Lenses to be used for in situ hybridization were oriented in OCT compound (Lab Tek) and quick frozen in a dry ice/hexane bath. Sections (8 μm) were taken from the center of the lens, parallel to the lens axis (anterior to posterior pole) on a Reichert-Jung 2800E Frigocut microtome, collected onto silane-treated slides and rapidly refrozen on a block of dry ice. Sections were fixed for 5 minutes at 4°C in 4% glutaraldehyde (Polysciences, Inc., Warrington, PA) and 20% ethylene glycol (Tousimis, Inc., Rockville, MD) in 0.1 M sodium phosphate buffer at pH 7.4 and rinsed in 2× standard saline citrate solution (SSC, 0.16 M NaCl, 0.016 M sodium citrate) as previously described (Thomas et al. 1990). Pretreatment, hybridization and post-treatment were carried out as described in the Oncor in situ procedure (Oncor, Gaithersbutg, MD). Briefly, fixed tissues were pretreated with acetic anhydride, then glycinated, before dehydration and overnight hybridization at 52°C in a humidified chamber. ³⁵S-labeled sense and antisense riboprobes were used at a concentration of 5×10² cts/minute/μl in 0.33 M NaCl, 50% formamide hybridization solution (Oncor, Gaithersburg, MD). Following hybridization, tissues were washed for 10 minutes in 50% formamide/2× SSC at 52°C, rinsed in 2× SSC, digested with RNAase, and again washed for 10 minutes in 50% formamide/2× SSC and rinsed in 2× SSC. Tissue sections were dehydrated in a graded alcohol series and air dried. Slides were dipped in a 1:1.5 dilution of emulsion (Ilford G-5) and exposed for 2–5 days at room temperature. Emulsion was developed in Dektol developer (Kodak, Rochester, NY), fixed and counterstained in hematoxylin and eosin. Coverslips were mounted with Permount. Micrographs were taken with Kodak Tmax film on an Olympus BH2 microscope fitted for both bright-field and dark-field optics.

For preparation of RNA riboprobes, EcoR1 and BamH1 restriction sites were incorporated into the upstream and downstream PCR oligonucleotides, respectively, and the resulting PCR products were directionally cloned into pBlueScript(KS+) (Stratagene, LaJolla, CA) following EcoR1 and BamH1 digestion. The identity of the cloned products was confirmed by sequencing using the dideoxy method of chain termination (Sambrook et al., 1989). ³⁵S-labeled sense and antisense riboprobes were synthesized from the T3 and T7 promoters, respectively, using ³⁵S-UTP (New England Nuclear, sp.act. >1000 Ci/mmole). Riboprobes were purified over a Nick column (Pharmacia) before dilution in Oncor RNA hybridization buffer.

Since c-myc and N-myc mRNAs share regions of sequence homology (Fig. 2), it was necessary to confirm the specificity of the PCR reactions used to measure these proto-oncogene mRNAs in the developing lens. Reverse transcription of embryonic chicken lens RNA followed by PCR using oligonucleotide primers specific for chicken c-myc sequences in the presence of the c-myc internal standard generated products of the expected sizes (397 bp for c-myc; 324 bp for the c-myc internal standard) (Fig. 3A). Similarly, oligonucleotide primers specific for chicken N-myc sequences generated the expected 195 bp N-myc product and a 120 bp N-myc internal standard product (Fig. 3B). Sequencing of the subcloned c-myc and N-myc products confirmed that each was derived from the expected region of the corresponding mRNA. Oligonucleotides complementary to sequences within the amplified regions of c-myc and N-myc mRNA hybridized specifically to their respective PCR products and showed no evidence of minor products derived from the other mRNA (Fig. 3C). Thus, we conclude that the c-myc PCR product is derived exclusively from c-myc mRNA, the N-myc product exclusively from N-myc mRNA.

A representative autoradiogram of the PCR products generated using RNA from the central lens epithelium from chicken embryos at 6, 10, 14 and 19 days of development is shown in Fig. 4A. The compiled data from several such experiments are presented as a bar graph in Fig. 4B, along with the results obtained for the equatorial lens epithelium and the lens fibers. Superimposed on the bar graphs are line graphs indicating the labeling index in these regions of the lens as a function of developmental age (data taken from Persons and Modak (1970)). The level of c-myc mRNA in the central region of the lens epithelium was proportional to the percentage of labeled nuclei in this region throughout development (Fig. 4B). In the equatorial region, however, the levels of c-myc mRNA expression were relatively constant throughout development despite a decrease in the percentage of proliferating cells (Fig. 4B). This suggested that c-myc might continue to be expressed in cells that are in the early stages of cytodifferentiation. Significant levels of c-myc mRNA were also found in the differentiated lens fiber cells (Fig. 4B).

The pattern of expression of N-myc mRNA in the developing lens was distinctly different from that of c-myc. At each age examined N-myc mRNA was predominantly associated with the equatorial region of the lens (Fig. 5A,B). The levels of N-myc mRNA showed no correlation with the changes in cell proliferation that accompany development in either the central or equatorial epithelium. As with c-myc, N-myc expression seemed to continue in the differentiated fiber cells.

To determine the extent to which c-myc and N-myc mRNAs were expressed in the differentiating cells of the annular pad at the lens equator, lenses from 14-day-old embryos were dissected as shown in Fig. 1B. This method of dissection separates almost all the proliferating cells of the lens epithelium into a single fraction, which we have called the “inner” epithelium. The remainder of the epithelium represents the annular pad, an almost pure population of cells in the early stages of cytodifferentiation. To test whether the dissection was successful, epithelia were labeled with [³H]thymidine prior to dissection, and the percentage of labeled DNA recovered in each tissue fraction was determined (Fig. 6). Approximately 93% of the labeled DNA was associated with the “inner” epithelium (average of 5 determinations). Levels of c-myc and N-myc mRNA were then measured in similarly dissected tissues. 50% of the total c-myc mRNA of the epithelium and over 80% of the N-myc mRNA were found in the annular pad, indicating that both proto-oncogenes are actively expressed in cells that are in the early stages of cytodifferentiation (Fig. 6).

To determine whether c-myc and N-myc mRNAs are expressed in the same population of annular pad cells, in situ hybridization was performed using ³⁵S-labeled riboprobes derived from the cloned PCR products (Fig. 7A-F). Hybridization of N-myc mRNA to the the antisense N-myc riboprobe was observed in cells throughout the annular pad and in the perinuclear region of the fiber cells (Fig. 7C,E). Hybridization to the sense riboprobe in control sections was not above background (not shown). Hybridization of c-myc mRNA to the c-myc antisense riboprobe was also seen throughout the annular pad region, although the intensity of the signal was consistently lower than for N-myc using comparably labeled probes with identical exposure times (Fig. 7D,F). Again, control sections hybridized to the c-myc sense riboprobe showed no hybridization above background (not shown). In the fiber cells, hybridization to the c-myc antisense riboprobe was not above background, although c-myc mRNA was detected in these cells by the more sensitive PCR assay. These results indicate that both c-myc and N-myc mRNAs are expressed in the same population of annular pad cells of the 14-day embryonic chick lens, in a region that contains only postmitotic cells (Persons and Modak, 1970).

The developing lens provides an opportunity to observe a single population of proliferating epithelial cells as it undergoes two types of cell cycle arrest: one leading to quiescence in the central epithelium, and one leading to terminal differentiation at the lens equator. The present results indicate that c-myc mRNA expression is distinctly different in these two instances. In the central region of the lens epithelium, the level of c-myc mRNA was proportional to the percentage of proliferating cells throughout development. This implies that expression of c-myc mRNA is associated predominantly with proliferating lens epithelial cells and ceases or is greatly diminished as these cells enter quiescence. The decreased expression of c-myc mRNA is a specific effect, since expression of c-fos and c-jun mRNAs increases in the central lens epithelium as cells become quiescent (Rinaudo and Zelenka, 1992). The finding that c-myc mRNA decreases as lens cells cease dividing in vivo is consistent with studies showing that c-myc mRNA is absent or at very low levels in quiescent mouse lens epithelial cells in culture (Rao and Church, 1989), as in many other quiescent cell types in vitro (reviewed by Cole, 1986). In contrast, we observed no correlation between c-myc expression and cell proliferation in the equatorial epithelium of the developing lens, where the non-proliferating cell population consists of differentiating, rather than quiescent, cells. This suggested that c-myc mRNA might be expressed in postmitotic lens epithelial cells in the early stages of differentiation in vivo, as was previously found in vitro (Nath et al., 1987). This was confirmed by direct PCR analysis of c-myc mRNA in the isolated annular pad cells and by in situ hybridization. Continued expression of c-myc mRNA during differentiation has also been observed in a limited number of other cell types, including PC12 cells (Greenberg et al., 1985), chronic lymphocytic leukemia cells (Larsson et al., 1987), and postmitotic cerebellar neurons (Ruppert et al., 1986).

The present data further indicate that expression of N-myc mRNA is strongly correlated with the initial stages of lens fiber cell formation. The highest concentrations of N-myc mRNA were found in the annular pad cells at all developmental stages, although lower levels of N-myc mRNA may also be present in the proliferating or quiescent cells of the central epithelium (this study; Sawai et al., 1990; Hirning et al., 1991). This pattern of expression implies that N-myc mRNA levels increase as proliferating lens epithelial cells withdraw from the cell cycle to differentiate. Thus, although c-myc expression continues during the early stages of lens fiber cell differentiation, its expression relative to N-myc decreases, because of the increased N-myc expression. This shift in the relative levels of c-myc and N-myc expression may alter the array of proteins available to complex with c-myc, and contribute to uncoupling the usual association between c-myc expression and cell proliferation.

We thank Dr G. Hyatt for preparing the c-myc internal standard and Drs B. Brewitt, J. Piatiagorsky, P. Russell and G. Wistow for critically reading the manuscript.