Molec. Biol. Rep. Vol. 5, 3: 145-149, 1979

THE TERMINAL STRUCTURES OF FEATHER KERATIN mRNA

C. Phillip MORRIS & George E. ROGERS

Department of Biochemistry, University of Adelaide, Adelaide, South Australia, 5001

Abstract

Materials and methods

Terminal labeling of embryonic feather keratin mRNA with [3 H] KBH4 followed by digestion with ribonuclease T1 and "1"2,alkaline phosphatase, snake venom phosphodiesterase, and nueleotide pyrophosphatase and analysis of the products by high voltage paper electrophoresis, indicated the presence of the sequence mTG(5')ppp(5')N at the 5'-end of the mRNA. Ribonuclease Tt ~md A digests of the terminally labeled, and also of unlabeled mRNA followed by fractionation on denaturing polyacrylamide gels indicated the presence of polyadenylate tracts ranging in size from 45 to 165 nucleotides at the 3'-end of the mRNA.

Isolation of mRNA

Introduction

Analysis of the 5'-terminus

Embryonic chick feather cells are committed to the production of keratin, a family of fibrous, insoluble intracellular proteins (6, 8). The mRNA for this keratin family has been isolated and partially characterized (11). Toward the further characterization of the mRNA this study investigates the presence of a 5'-terminal cap structure (12, 13), and a 3'terminal polyadenylate tract (3, 12)which are both characteristic of most eukaryotic mRNAs studied so far. The a 2 p labeling of embryonic chick RNA in vivo or in cultured feathers produces RNA of extremely low specific activity (G.A. Partington, unpublished observations), which does not allow analysis of the terminal sequences by the customary procedures used for numerous other eukaryotic RNA molecules (13). It is for this reason that the present paper describes investigations of the terminal structures of keratin mRNA using a post-isolation labeling procedure involving oxidation of the RNA with NalO4 followed by reduction with [3H]KBH4 to label ends containing exposed 2', 3'-hydroxyls.

Digestion of labeled mRNA with 1 unit m1-1 of ribonucleases TI and T2 (Sigma)were performed in 20 mM sodium acetate, 0.2 mM EDTA, pH 5.0 for 2.0 hr at 37~ Digestions with 0.7 mg m1-1 of bacterial alkaline phosphatase (Worthington), with or without 0.2 mg m1-1 of snake venom phosphodiesterase (Sigma), were carried out at 37~ for 2.0 hr in 15 mM magnesium acetate, 15 mM trisHC1, pH 7.9. Digestions with 0.3 units ml -t of nucleotide pyrophosphatase (Sigma), with or without 0.5 mg m1-1 of alkaline phosphatase were performed at 37~ for 150 min in 20 mM tris-acetate, 2 mM magnesium acetate, pH 7.6. High voltage electrophoretic analysis of digests was accomplished using Whatrnan 3 MM paper between water cooled metal plates (Paten Industries, Adelaide) in 2% acetic acid, 1.07% triethylamine, 0.1 mM EDTA, pH 4.1.

12S keratin mRNA was isolated as previously described (10) from the feathers of 14 day old chick embryos by EDTA dissociation of polysomes and the liberation of mRNA by treatment with SDS. [3H] KBH4 labeling of mRNA Between 5 and 10/ag of keratin mRNA were labeled to high specific activity by NalO4 oxidation of the 2',3'-hydroxyls and their reduction with [3 H] KBH4 (15 Ci mmole -1 , Radiochemical Centre, Amersham) as described by Symons (14).

Analysis of the 3'-terminus Poly(A) sequences were released from keratin mRNA (2) by incubation of samples in 1 ml of 0.1 M NaC1, 145

1 mM EDTA, 10 mM tris-HCL pH 7A containing 2/ag of ribonuclease A and 5 units of ribonuclease T1 at 37~ for 30 min. Electrophoresis of poly(A) samples was carried out on cylindrical (8 cm by 0.6 cm) 10% polyacrylamide gels in the presence of 50 mM tris-glycine pH 8.3 and 8 M urea. Electrophoresis was carried out at 2 mA per gel until the bromophenol blue marker reached the bottom of the gel. Molecular weights were calculated from parallel gels containing 5.8S and 5S ribosomal RNAs and tRNA. [a H] poly(A) was detected by slicing the gel into 1 mm slices and measuring the radioactivity eluted from pairs of adjacent slices. Unlabeled RNA was detected by staining with 0.05% toluidine blue in 55 mM sodium acetate, 0.1 mM EDTA, pH 5.5 and the mobility of RNA samples either determined directly or by scanning at 600 nm. All other conditions were as described in the Figures.

Results and discussion

Nature o f the 5' cap Embryonic chick feather keratin mRNA used in these experiments was isolated from 14 day old embryos and electrophoresed on 4% polyaerylamide gels in 98% formamide as a single discrete peak of Mr 240,000 (7). Labeling of the terminal nucleosides to a high specific activity was achieved by oxidation of the terminal 2',3'-hydroxyls with NalO4 followed by reduction to the dialcohols with high specific activity [3 H] KBH4. For the purposes of interpretation of the data presented in this paper it has been assumed that keratin mRNA has a typical eukaryotic cap and can be written as: 51

poly(A)

3~

mTG(5')ppp(5')NmpNp ......pAp ..... .pA where G is guanosine, A is adenosine and N represents any of the four nucleosides. Terminal labeling of the RNA with [3 H] borohydride will label the 5'-terminal 7-methyl-guanosine and the 3'-terminal adenosine. A mixture of the ribonucleases T1 and T2 will hydrolyse all ribonucleotide 3'-5' phosphodiester linkages, except those containing a 2'-O-methyl group (4) and hence should release radioactive adenosine and mTG(5')ppp(5')NmpNp when the 146

labeled RNA is incubated with such a mixture. Digestion of keratin [a H] mRNA with ribonucleases TI and T2 and analysis of the products by paper electrophoresis at pH 4.1 gave the pattern shown in Figure 1. Three peaks of radioactivity were detected, one of uncharged material which did not move from the origin, one which eoelectrophoresed with the nucleoside trialcohol of adenosine and a broad highly negative peak migrating toward the anode which is assumed to correspond with the postulated mTG(5')ppp(5')NmpNp from the 5'-end of the RNA. The peak of uncharged material at the origin could not be removed from the labeled RNA preparation by phenol extraction or ethanol precipitation and remains unidentified. This material was subsequently found to separate from the peak of RNA when subjected to polyacrylamide gel electrophoresis, running at the top of the gel (data not shown). A similar peak has been reported by Symons (14), although less counts were present in that instance because more extensive procedures were used to isolate the labeled RNA. Greater than 95% of the remaining radioactivity shown in Figure 1 ran with modified adenosine, presumably derived from the poly(A) tract at the 3'end of the mRNA, while the remainder ran with guanosine and uridine. These labeled nucleosides may have arisen from the presence of degraded mRNA as contaminants within the mRNA preparation, from guanosine or uridine present at the end of the poly(A) tract or from mRNA species which had lost the poly(A) tract. These nucleosides may also have ~risen from the labeled 3'-ends of ribosomal RNA breakdown products. The peak of negatively charged radioactivity obtained in Figure 1 was eluted from the paper and a portion incubated with alkaline phosphatase to remove monoesterified phosphates and produce the presumptive mTG(5')ppp(5')NmpN. Analysis of the digestion products by paper electrophoresis at pH 4.1 (Figure 2A) gave a peak running between AMP and GMP while a small peak of the original material can be detected running just faster than UMP due to the partial completion of the reaction with alkaline phosphatase. The observed change in mobility is consistent with the removal of a single phosphate. The 5'-terminal oligonucleotide fragments in Figures 1 and 2A ran as quite broad peaks. This might be due to the presence of several homologous structures of the type m7G(5')ppp(5')NmpNp present in the heterogeneous feather keratin mRNA population (6, 9), as has been described in mouse myeloma cell mRNA (1,5).

i

Up

Gp

Ap

Cp

GIu

A'

C'

m G"

2-0

1"6

~. 1-2 -o

~

0.8

0,4f

~

0 ANODE

10

20

30

STRIP

NUMBER

40

50 CATHODE

Fig. 1. Analysis of the ribonuclease T z and T~ digest of borohydride-labeled keratin mRNA by paper electrophoresis at pH 4.1. Enzymic digestion of the RNA was carried out as described in Materials and Methods. Paper electrophoresis was for 75 min at 45 V em-1 . The electrophoretogram was cut into 1 crn strips and counted in a toluene-based scintillation fluid. The origin is shown by the vertical arrow. The position of markers run at the same time are shown at the top of the Figure. The trialcohols of 7-methylguanosine, adenosine and cytidine are represented by mG', A' and C' respectively.

Guanosine and utidine are represented by G and U while nucleoside monophosphates are represented by Ap, Cp, Gp and U). A Up

Gp

AO

Cp

A'

C'

A further portion o f the oligonucleotide peak in Figure 1 was digested with both alkaline phosphatase and snake venom phosphodiesterase in order to hydrolyse the oligonucleotide to nucleosides. Analysis by paper electrophoresis at pH 4.1 (Figure 2B) revealed the presence of one major peak of radioactivity which co-migrated with the trialcohol o f 7-methylguanosine. Two minor peaks can be detected which correspond with the monophosphate and diphosphate of 7-methylguanosine, which arose due to the incomplete removal of phosphates from the products of the snake venom phosphodiesterase cleavage b y alkaline phosphatase. These data indicate that 7-methylguanosine must be at the 5'-end o f the feather keratin m R N A with its 2',3'-hydroxyls exposed, and that it is connected through a 5'phosphate to the rest of the molecule. The presence of the triphosphate bridge linking 7-methylguanosine to the RNA was confirmed by digestion of the terminal oligonucleotide fragment (Figure 1) with nucleotide pyrophosphatase, which hydrolyses nucleotide pyrophosphate linkages, and a mixture o f nucleotide pyrophosphatase and alkaline phosphatase. The products were analysed by paper electrophoresis at pH 4.1 (Figure 3A and B). NucleoA Up

Gp

Ap

Cp

A'

pinG

C'

rnG'

pprnG

mG'

I

?

2 :E L r

1 I

x

I

i

i

B

B

:S a. (J

3: 3

4 2:

2

3

1

~

1

'" 20

ll0 ANODE STRIP 10

20

30

40

50

CATHODE

NUMBER

40

ANODE

CATHODE STRIP

A 30

NUMBER

Fig. 2. Electrophoretic analysis at pH 4.1 digests of the

Fig. 3. Electrophoretic analysis at pH 4.1 of digests of the terminal oligonucleotide fragment (see Figure 1) produced by the action of nucleotide pyrophosphatase (A)

terminal oligonueleotide fragment (see Figure 1) produced by the action of alkaline phosphatase (A) and alkaline phosphatase plus snake venom phosphodiesterase (B). The oligonucleotide was isolated from the electrophoretogram in Figure 1 by elution with 0.1 mM EDTA after extensive washing with 1:1 ethanol[ether to remove any scintillant. Digestion conditions were as described in Materials and

The oligonucleotide was isolated as in Figure 2 except that water was used for elution instead of 0.1 mM EDTA. Digestion conditions were as described in Materials and Methods. Conditfons for electrophoresis, counting and running of markers were as described in Figure 1 except that 7-methyl-

Methods.

guanosine-diphosphate (ppmG) and monophosphate (pmG)

Conditions for

eleetrophoresis,

counting and

running of markers were as described in Figure 1.

and nucleotide pyrophosphatase plus alkaline phosphatase (B).

were also included as markers.

147

tide pyrophosphatase cleavage produced both 7methylguanosine monophsophate and diphosphate as predicted. The further action of nucleotide pyrophosphatase on the diphosphate yielding the monophosphate resulted in the appearance of the majority of the counts in the monophosphate peak. The inclusion of alkaline phosphatase resulted in the removal of any phosphates to produce a single peak co-migratin~ with 7-methylguanosine. These results lead to the conclusion that embryonic feather keratin mRNA contains a cap structure at its 5'-terminus. However, the data do'not indicate the number of nucleosides present in the terminal oligonucleotide fragment, although there must be at least two, i.e., m?G(5')ppp(5')Np. Studies on the charge of the terminal nucleotide fragment would be needed to clarify this point, but from evidence of other higher eukaryotes (13) it is most likely that it contains at least one 2'-O-methylated nucleoside.

The length of the 3'-polyadenylitte sequence Determination by [aH]KBI-Lt labeling. Keratin mRNA labeled using [aH]KBH4 as previously described was digested with ribonucleases A and T1 under conditions which left the poly(A) tract intact but reduced the remainder of the mRNA to nucleo-

tides and small oligonucleotides (2). The length of poly(A) sequences so produced was determined by electrophoresis on a denaturing polyacrylamide gel. The gel system used was very similar to a system which has been shown to give correct estimates of poly(A) length when rRNA and tRNA used as markers (15). Since the [3H]KBH4 labeling procedure labeled the 3'-terminal adenosine the [3H]poly(A) produced by digestion of labeled mRNA could be detected on the gels by slicing and counting. Figure 4A shows the length distribution of poly(A) sequences so obtained. There is a range of length from 45 to 165 nucleotides. Since each poly(A) sequence is given equal weighting regardless of its molecular weight, by this labeling procedure, the profile obtained lends itself to the calculation of a number-average length for the poly(A) of 65 nucleotides. Determination by gel staining. The length of poly(A) produced by ribonuclease T1 and A digestion of sufficient keratin mRNA to be detected on a gel by staining, was determined by electrophoresis on an identical gel system to that ,already described. Figure 4B shows the distribution of poly(A) lengths so obtained. This method reveals exactly the same range of poly(A) lengths as the labeling procedure, but since each molecule is weighted by its molecular

B

S i

5.8S

SS

~

4

X

~ 3 r 2 Z

\

/

/

j./

? r c~

MJ

/\

O :p < m tv

\

O (/) m