Cell, Vol.
17.
163-l
73,
May 1979, Copyright 0
1979 by MIT
In Vitro Synthesis of a 5s RNA Precursor Nuclei of Rat Liver and HeLa Cells
Hiroshi Hamada and Masami Muramatsu Department of Biochemistry Cancer Institute Japanese Foundation for Cancer Research Kami-lkebukuro, Toshima-ku Tokyo, Japan Yoshio Urano, Toshio Onishi and Ryo Kominami Department of Biochemistry Tokushima University School of Medicine Tokushima, Japan
Summary Isolated rat liver nuclei were incubated under appropriate conditions in the presence of 0.5 pg/ml LXamanitin and an RNAase inhibitor prepared from cytosol fraction, together with ~Y-~~P-UTP or CU-~‘PCTP and three other nucleoside triphosphates. RNA extracted by an SDS-hot phenol procedure was fractionated with sucrose density gradient centrifugation followed by acrylamide gel electrophoresis. Fingerprint analysis of the in vitro synthesized “5s” RNA, which was slightly larger than mature 5s RNA on gel electrophoresis, showed that it contained all the sequences of mature 5s RNA except for the oligonucleotide at the 3’ end. Instead, it contained two additional spots which were not present in mature 5s RNA. Analysis of the extra spots revealed that they were derived from the 3’ end of the in vitro synthesized “5s” RNA, which was sequenced tentatively as -CUUGAUGCUUUoh (extra sequence underlined). The 5’ end of the product was (p)pGU-. Isolated HeLa cell nuclei synthesized similar sized “5s” RNA under the same conditions. We conclude from these results that in isolated nuclei of these mammalian cells, RNA polymerase Ill starts transcription of 5s RNA gene at the same site as the 5’ end of mature 5s RNA, proceeds toward the 3’ direction and stops at a site probably 8 nucleotides downstream from the 3’ end of mature 5s RNA. Experiments with a short pulse and with various “chases” have demonstrated the presence of a short-lived precursor 5s RNA which is similar in size and sequence to in vitro “5s” RNA, suggesting that 5s RNA is synthesized in vivo as a longer precursor molecule as demonstrated in this in vitro system, and is rapidly processed into mature 5s RNA. Introduction We have studied RNA synthesis in isolated nuclei to elucidate the transcriptional mechanism and its regulation in eucaryotic cells. The synthesis of low molecular weight RNAs, including 5s and pre-4S RNA (precursors of transfer RNA), is of special interest because
by Isolate
RNA polymerase III (or C), which is responsible for their synthesis, can reinitiate transcription extensively in isolated nuclei (McReynolds and Penman, 1974; Udvardy and Seifart, 19761, a situation hardly possible for high molecular RNA species synthesized by RNA polymerase I (or A) and II (or B) (Grummt and Lindigkeit, 1973; Udvardy and Seifart, 1976; Gilboa, Soreq and Ariv, 1977; Onishi, Matsui and Muramatsu, 1977). The high fidelity with which 5s RNA genes were transcribed in isolated nuclei and even with purified chromatin has been demonstrated with various systems utilizing HeLa cells, mouse plasmacytoma 460 cells and Xenopus laevis ovary cells (McReynolds and Penman, 1974; Udvardy and Seifart, 1976; Parker and Roeder, 1977; Sklar and Roeder, 1977; Yamamoto and Seifart, 1977, 1978). In all the reports mentioned above, however, 5s RNA was thought to be synthesized as a molecule identical with or at least indistinguishable from the mature 5s RNA. On the other hand, all other RNA species, including messenger, ribosomal and transfer RNA% are known to be synthesized as larger precursor molecules which are processed during the course of maturation into smaller functional RNA molecules (Perry, 1976). Since previous studies did not seem to be precise enough to identify a precursor 5s RNA molecule with a very small size difference from the mature one, we reexamined the possibility of the presence of a slightly larger precursor to 5s RNA both in vivo and in vitro. In this study, we first compared the size and the primary structure of in vitro synthesized “5s” RNA with those of mature 5s RNA and then tried to demonstrate the presence of a precursor to 5s RNA in vivo. The data clearly showed that in isolated nuclei, 5s RNA was synthesized as a slightly larger molecule than mature 5s RNA, with an extra sequence at the 3’ terminus. Furthermore, in accord with the in vitro product, short pulse-labeling experiments indicated the presence of a larger 5s RNA precursor in vivo. Results Conditions for in Vitro RNA Synthesis by Isolated Nuclei It has been repeatedly shown that isolated nuclei of various mammalian cells can synihesize all three categories of RNA-that is, heterogeneous nuclear, preribosomal and low molecular weight RNAs-although meaningful initiation occurs only for the last class of RNA under presently available conditions (see Gilboa et al., 1977). To prevent contamination of the degradation products into regions of low molecular weight RNA, we sought conditions in which the degradation of high molecular weight RNA is minimal during incubation. It has also been shown that a low concentration (0.5-l .O pg/ml) of cr-amanitin can efficiently inhibit
Cell 164
RNA polymerase II activity, which is responsible for the synthesis of the first class of RNA, leaving the synthesis of the latter two almost intact for a period of time (Austoker et al., 1974; Udvardy and Seifart, 1975; Sklar and Roeder, 1977). In our in vitro system, 0.5 fig/ml a-amanitin blocked more than half of the incorporation into high molecular weight RNA, indicating almost complete inhibition of RNA polymerase II activity, as mentioned above. At the same time, this reagent suppressed the trailing of the radioactivity of high molecular weight RNA into the low molecular weight region (data not shown). This was at least partially due to the degradation of high molecular weight RNA synthesized by RNA polymerase II. An RNAase inhibitor fraction of rat liver cytoplasm prepared according to Gribnau, Schoenmakers and Bloemendal (1969) was also effective for inhibiting degradation of high molecular weight RNA synthesized under these conditions. An ionic concentration of 0.12 M KCI and an incubation temperature of 25°C (rather than 37’C) were chosen for the same purpose (Onishi and Muramatsu, 1978). Although incorporation into the low molecular weight RNA region increased linearly almost up to 60 min, we analyzed the products at 30 min, since at later times some degradation of high molecular weight RNA was apparent. Detection of the Size Difference between in Vitro “5S” RNA and Mature 5S RNA RNA was extracted from incubated nuclei by a sodium dodecylsulfate (SDS)-hot phenol procedure, as described in Experimental Procedures, and centrifuged on a sucrose gradient. The low molecular weight portion was then further separated on a 13.5 or 15% acrylamide gel under denaturing conditions of 7 M urea. Figure 1 a shows that 32P-labeled in vitro “5s” RNA appears to migrate slightly more slowly than mature 5S RNA on this rather concentrated gel. To prove that the difference was true, a 32P-labeled in vitro product was mixed with a low molecular weight RNA fraction labeled with 3H-uridine for 12 hr in culture and electrophoresed together on the same gel (Figure 1 b). 3H-labeled low molecular weight RNA contained heterogeneous 4S (transfer) RNA and a sharp peak of 5.S RNA. In contrast, 32P-labeled in vitro RNA showed a broad peak of pre-4S RNA (McReynolds and Penman, 1974) which apparently migrated more slowly than mature 45 RNA, and a distinct peak of in vitro “5s” RNA, which migrated “two slices” behind mature 5S RNA under these conditions. Figure lc shows the electrophoretogram in which a product synthesized in vitro without the RNAase inhibitor was migrated. In vitro synthesized “5s” RNA again migrated more slowly than mature 5S RNA, indicating that the presence of the former was not caused by the presence of the RNAase inhibitor. Figure 1 d shows that a similar product was syn-
thesized when c~-~~P-CTP was used instead of (Y-~*PUTP, thus excluding the possibility of terminal addition of U residues to some larger molecules. The size of the in vitro “5S” RNA was determined by co-electrophoresis with 5.8S, 5S and 4S RNAs as internal standards. The linear-logarithmic plots shown in Figure If indicate that this RNA was only about 6 nucleotides longer than mature 5S RNA. The error would not exceed 2 nucleotides, since one slice corresponded to approximately 2 nucleotides and the difference was reproducibly three slices in 15% gels under these conditions. To determine whether the presence of this larger “5s” product is specific for rat liver nuclei or a general phenomenon found in other mammalian cells, isolated HeLa cell nuclei were incubated in the same manner. The results shown in Figure 1 e indicate that a similar in vitro “5s” RNA was synthesized in HeLa cell nuclei under these conditions, and suggest that the synthesis of a larger “5s” RNA in vitro is a more or less general phenomenon in mammalian (and possibly in other eucaryotic) cells. Structural Analysis of in Vitro “5s” RNA To clarify the structural relationship between the in vitro “5s” RNA and mature 5S RNA, we compared the primary structure of these two RNAs with the fingerprint analysis of RNAase T, oligonucleotides. Since the whole nucleotide sequence of mature 5S RNA is known to be the same for various mammalian cells (Forget and Weissman, 1969; Labrie and Sanger, 1969; Williamson and Brownlee, 1969; Takai, Hashimoto and Muramatsu, 19751, and that of the rat is the same as the mouse (H. Hamada, unpublished observations), the extra sequences, if any, that are present in in vitro “5s” RNA should be detected, identified and located on the mature 55 RNA sequence by appropriate analysis (Takai et al., 1975). Figure 2 shows the RNAase T, fingerprints of the in vitro synthesized “5s” RNA in the presence of either cw-3ZP-UTP (Figures 2a and 2c) or a-32P-CTP (Figures 2b and 2d). The molar yield of 3’P-labeled nucleotide and the nearest neighbors of the U residue were then determined for each spot (Table 1). When the product was labeled with a-3ZP-UTP, all the spots that should have appeared with this labeled precursor were detected at nearly theoretical molar yields, while any spots not to be labeled with it were not found on the fingerprint. The 5’ terminal nucleotides of the in vitro “5s” RNA were ppGp and pGp, which were found only when a-32P-UTP was used as the labeled precursor (Figures 2a and 2c; Table 1). This indicated that the transcription in this system started from pppGpUp---, which was the same sequence as the 5’ terminus of mature 5S RNA. The absence of any extra sequence which could be placed at the 5’ end of mature 5S RNA supports the idea that the starting point was exactly the 5’ end of mature 5S
5.S RNA Precursor 165
Synthesis
in Isolated
Nuclei
>uce
Figure Weight
number
1. Electrophoresis of in Vitro Synthesized Low Molecular RNA on 13.5% or 15% Acrylamide-7 M Urea Gels
Nuclei were incubated in vitro with either a-32P-UTP or a-32P-CTP for 30 min as described in Experimental Procedures. Phenol-extracted RNA was purified with sucrose density gradient centrifugation followed by Sephadex-G50 gel filtration, and electrophoresed together with the low molecular weight RNA labeled for 12 hr in viva with 3H-uridine as described in Experimental Procedures. After autoradiography. gels were sliced at 1 mm width, digested with Soluene and counted for radioactivity with separate windows for 32P and 3H. (a-l) Autoradiograph of the low molecular weight RNA prepared from MHI 34/C-c cells labeled in viva for 12 hr with 32P-orthophosphate. (a-2) Autoradiograph of the low molecular weight RNA synthesized in vitro with ~z-~~P-UTP. (b) Electrophoretogram of a-2. (c) Low molecular weight RNA synthesized in vitro with a-32P-UTP in the absence of RNAase inhibitor. (d) Low molecular weight RNA synthesized in vitro with e-32P-CTP (RNA in a-2, b, c, d was synthesized with isolated rat liver nuclei). (e) Low molecular weight RNA synthesized with isolated HeLa cell nuclei in the presence of u-32P-CTP. Acrylamide concentration of the gel was 13.5% in (a. b and d) and 15% in (c and e). Electrophoresis was carried out at 100 V for 48 hr in (b and d) and for 60 hr in (c and e). (M) 32P; (0- - -0) 3H. (f) Linear-logarithmic plots of electrophoretic mobilities and nucleotide lengths. The electrophoretic mobilities of 5.85. in vitro “5S”, mature 5.S and a group of tRNA were plotted from the electrophoretogram of (c). 5.8s having 158 nucleotides was on slice 71 (outside of c). The slowest peak of tRNA complex contained mostly tRNALe” and tRNASe’, whose nucleotide lengths were 85 (tRNA*). The in vitro “55” RNA lies on about 126 nucleotides.
170X0-
5.0 s
1souo130-
in vitro”5S”
i
110-
g 1 Y =
loo-
55
\
120-
goti?NA* \.\ 80-
707 41
m
00
90 Mobility
100
110
in mm
120
1Xl
Cell 166
C
a
P g@ 5g"
636
-30-A@& a7 , \ 8 1. _)
17.
163-l
73,
May 1979, Copyright 0
1979 by MIT
In Vitro Synthesis of a 5s RNA Precursor Nuclei of Rat Liver and HeLa Cells
Hiroshi Hamada and Masami Muramatsu Department of Biochemistry Cancer Institute Japanese Foundation for Cancer Research Kami-lkebukuro, Toshima-ku Tokyo, Japan Yoshio Urano, Toshio Onishi and Ryo Kominami Department of Biochemistry Tokushima University School of Medicine Tokushima, Japan
Summary Isolated rat liver nuclei were incubated under appropriate conditions in the presence of 0.5 pg/ml LXamanitin and an RNAase inhibitor prepared from cytosol fraction, together with ~Y-~~P-UTP or CU-~‘PCTP and three other nucleoside triphosphates. RNA extracted by an SDS-hot phenol procedure was fractionated with sucrose density gradient centrifugation followed by acrylamide gel electrophoresis. Fingerprint analysis of the in vitro synthesized “5s” RNA, which was slightly larger than mature 5s RNA on gel electrophoresis, showed that it contained all the sequences of mature 5s RNA except for the oligonucleotide at the 3’ end. Instead, it contained two additional spots which were not present in mature 5s RNA. Analysis of the extra spots revealed that they were derived from the 3’ end of the in vitro synthesized “5s” RNA, which was sequenced tentatively as -CUUGAUGCUUUoh (extra sequence underlined). The 5’ end of the product was (p)pGU-. Isolated HeLa cell nuclei synthesized similar sized “5s” RNA under the same conditions. We conclude from these results that in isolated nuclei of these mammalian cells, RNA polymerase Ill starts transcription of 5s RNA gene at the same site as the 5’ end of mature 5s RNA, proceeds toward the 3’ direction and stops at a site probably 8 nucleotides downstream from the 3’ end of mature 5s RNA. Experiments with a short pulse and with various “chases” have demonstrated the presence of a short-lived precursor 5s RNA which is similar in size and sequence to in vitro “5s” RNA, suggesting that 5s RNA is synthesized in vivo as a longer precursor molecule as demonstrated in this in vitro system, and is rapidly processed into mature 5s RNA. Introduction We have studied RNA synthesis in isolated nuclei to elucidate the transcriptional mechanism and its regulation in eucaryotic cells. The synthesis of low molecular weight RNAs, including 5s and pre-4S RNA (precursors of transfer RNA), is of special interest because
by Isolate
RNA polymerase III (or C), which is responsible for their synthesis, can reinitiate transcription extensively in isolated nuclei (McReynolds and Penman, 1974; Udvardy and Seifart, 19761, a situation hardly possible for high molecular RNA species synthesized by RNA polymerase I (or A) and II (or B) (Grummt and Lindigkeit, 1973; Udvardy and Seifart, 1976; Gilboa, Soreq and Ariv, 1977; Onishi, Matsui and Muramatsu, 1977). The high fidelity with which 5s RNA genes were transcribed in isolated nuclei and even with purified chromatin has been demonstrated with various systems utilizing HeLa cells, mouse plasmacytoma 460 cells and Xenopus laevis ovary cells (McReynolds and Penman, 1974; Udvardy and Seifart, 1976; Parker and Roeder, 1977; Sklar and Roeder, 1977; Yamamoto and Seifart, 1977, 1978). In all the reports mentioned above, however, 5s RNA was thought to be synthesized as a molecule identical with or at least indistinguishable from the mature 5s RNA. On the other hand, all other RNA species, including messenger, ribosomal and transfer RNA% are known to be synthesized as larger precursor molecules which are processed during the course of maturation into smaller functional RNA molecules (Perry, 1976). Since previous studies did not seem to be precise enough to identify a precursor 5s RNA molecule with a very small size difference from the mature one, we reexamined the possibility of the presence of a slightly larger precursor to 5s RNA both in vivo and in vitro. In this study, we first compared the size and the primary structure of in vitro synthesized “5s” RNA with those of mature 5s RNA and then tried to demonstrate the presence of a precursor to 5s RNA in vivo. The data clearly showed that in isolated nuclei, 5s RNA was synthesized as a slightly larger molecule than mature 5s RNA, with an extra sequence at the 3’ terminus. Furthermore, in accord with the in vitro product, short pulse-labeling experiments indicated the presence of a larger 5s RNA precursor in vivo. Results Conditions for in Vitro RNA Synthesis by Isolated Nuclei It has been repeatedly shown that isolated nuclei of various mammalian cells can synihesize all three categories of RNA-that is, heterogeneous nuclear, preribosomal and low molecular weight RNAs-although meaningful initiation occurs only for the last class of RNA under presently available conditions (see Gilboa et al., 1977). To prevent contamination of the degradation products into regions of low molecular weight RNA, we sought conditions in which the degradation of high molecular weight RNA is minimal during incubation. It has also been shown that a low concentration (0.5-l .O pg/ml) of cr-amanitin can efficiently inhibit
Cell 164
RNA polymerase II activity, which is responsible for the synthesis of the first class of RNA, leaving the synthesis of the latter two almost intact for a period of time (Austoker et al., 1974; Udvardy and Seifart, 1975; Sklar and Roeder, 1977). In our in vitro system, 0.5 fig/ml a-amanitin blocked more than half of the incorporation into high molecular weight RNA, indicating almost complete inhibition of RNA polymerase II activity, as mentioned above. At the same time, this reagent suppressed the trailing of the radioactivity of high molecular weight RNA into the low molecular weight region (data not shown). This was at least partially due to the degradation of high molecular weight RNA synthesized by RNA polymerase II. An RNAase inhibitor fraction of rat liver cytoplasm prepared according to Gribnau, Schoenmakers and Bloemendal (1969) was also effective for inhibiting degradation of high molecular weight RNA synthesized under these conditions. An ionic concentration of 0.12 M KCI and an incubation temperature of 25°C (rather than 37’C) were chosen for the same purpose (Onishi and Muramatsu, 1978). Although incorporation into the low molecular weight RNA region increased linearly almost up to 60 min, we analyzed the products at 30 min, since at later times some degradation of high molecular weight RNA was apparent. Detection of the Size Difference between in Vitro “5S” RNA and Mature 5S RNA RNA was extracted from incubated nuclei by a sodium dodecylsulfate (SDS)-hot phenol procedure, as described in Experimental Procedures, and centrifuged on a sucrose gradient. The low molecular weight portion was then further separated on a 13.5 or 15% acrylamide gel under denaturing conditions of 7 M urea. Figure 1 a shows that 32P-labeled in vitro “5s” RNA appears to migrate slightly more slowly than mature 5S RNA on this rather concentrated gel. To prove that the difference was true, a 32P-labeled in vitro product was mixed with a low molecular weight RNA fraction labeled with 3H-uridine for 12 hr in culture and electrophoresed together on the same gel (Figure 1 b). 3H-labeled low molecular weight RNA contained heterogeneous 4S (transfer) RNA and a sharp peak of 5.S RNA. In contrast, 32P-labeled in vitro RNA showed a broad peak of pre-4S RNA (McReynolds and Penman, 1974) which apparently migrated more slowly than mature 45 RNA, and a distinct peak of in vitro “5s” RNA, which migrated “two slices” behind mature 5S RNA under these conditions. Figure lc shows the electrophoretogram in which a product synthesized in vitro without the RNAase inhibitor was migrated. In vitro synthesized “5s” RNA again migrated more slowly than mature 5S RNA, indicating that the presence of the former was not caused by the presence of the RNAase inhibitor. Figure 1 d shows that a similar product was syn-
thesized when c~-~~P-CTP was used instead of (Y-~*PUTP, thus excluding the possibility of terminal addition of U residues to some larger molecules. The size of the in vitro “5S” RNA was determined by co-electrophoresis with 5.8S, 5S and 4S RNAs as internal standards. The linear-logarithmic plots shown in Figure If indicate that this RNA was only about 6 nucleotides longer than mature 5S RNA. The error would not exceed 2 nucleotides, since one slice corresponded to approximately 2 nucleotides and the difference was reproducibly three slices in 15% gels under these conditions. To determine whether the presence of this larger “5s” product is specific for rat liver nuclei or a general phenomenon found in other mammalian cells, isolated HeLa cell nuclei were incubated in the same manner. The results shown in Figure 1 e indicate that a similar in vitro “5s” RNA was synthesized in HeLa cell nuclei under these conditions, and suggest that the synthesis of a larger “5s” RNA in vitro is a more or less general phenomenon in mammalian (and possibly in other eucaryotic) cells. Structural Analysis of in Vitro “5s” RNA To clarify the structural relationship between the in vitro “5s” RNA and mature 5S RNA, we compared the primary structure of these two RNAs with the fingerprint analysis of RNAase T, oligonucleotides. Since the whole nucleotide sequence of mature 5S RNA is known to be the same for various mammalian cells (Forget and Weissman, 1969; Labrie and Sanger, 1969; Williamson and Brownlee, 1969; Takai, Hashimoto and Muramatsu, 19751, and that of the rat is the same as the mouse (H. Hamada, unpublished observations), the extra sequences, if any, that are present in in vitro “5s” RNA should be detected, identified and located on the mature 55 RNA sequence by appropriate analysis (Takai et al., 1975). Figure 2 shows the RNAase T, fingerprints of the in vitro synthesized “5s” RNA in the presence of either cw-3ZP-UTP (Figures 2a and 2c) or a-32P-CTP (Figures 2b and 2d). The molar yield of 3’P-labeled nucleotide and the nearest neighbors of the U residue were then determined for each spot (Table 1). When the product was labeled with a-3ZP-UTP, all the spots that should have appeared with this labeled precursor were detected at nearly theoretical molar yields, while any spots not to be labeled with it were not found on the fingerprint. The 5’ terminal nucleotides of the in vitro “5s” RNA were ppGp and pGp, which were found only when a-32P-UTP was used as the labeled precursor (Figures 2a and 2c; Table 1). This indicated that the transcription in this system started from pppGpUp---, which was the same sequence as the 5’ terminus of mature 5S RNA. The absence of any extra sequence which could be placed at the 5’ end of mature 5S RNA supports the idea that the starting point was exactly the 5’ end of mature 5S
5.S RNA Precursor 165
Synthesis
in Isolated
Nuclei
>uce
Figure Weight
number
1. Electrophoresis of in Vitro Synthesized Low Molecular RNA on 13.5% or 15% Acrylamide-7 M Urea Gels
Nuclei were incubated in vitro with either a-32P-UTP or a-32P-CTP for 30 min as described in Experimental Procedures. Phenol-extracted RNA was purified with sucrose density gradient centrifugation followed by Sephadex-G50 gel filtration, and electrophoresed together with the low molecular weight RNA labeled for 12 hr in viva with 3H-uridine as described in Experimental Procedures. After autoradiography. gels were sliced at 1 mm width, digested with Soluene and counted for radioactivity with separate windows for 32P and 3H. (a-l) Autoradiograph of the low molecular weight RNA prepared from MHI 34/C-c cells labeled in viva for 12 hr with 32P-orthophosphate. (a-2) Autoradiograph of the low molecular weight RNA synthesized in vitro with ~z-~~P-UTP. (b) Electrophoretogram of a-2. (c) Low molecular weight RNA synthesized in vitro with a-32P-UTP in the absence of RNAase inhibitor. (d) Low molecular weight RNA synthesized in vitro with e-32P-CTP (RNA in a-2, b, c, d was synthesized with isolated rat liver nuclei). (e) Low molecular weight RNA synthesized with isolated HeLa cell nuclei in the presence of u-32P-CTP. Acrylamide concentration of the gel was 13.5% in (a. b and d) and 15% in (c and e). Electrophoresis was carried out at 100 V for 48 hr in (b and d) and for 60 hr in (c and e). (M) 32P; (0- - -0) 3H. (f) Linear-logarithmic plots of electrophoretic mobilities and nucleotide lengths. The electrophoretic mobilities of 5.85. in vitro “5S”, mature 5.S and a group of tRNA were plotted from the electrophoretogram of (c). 5.8s having 158 nucleotides was on slice 71 (outside of c). The slowest peak of tRNA complex contained mostly tRNALe” and tRNASe’, whose nucleotide lengths were 85 (tRNA*). The in vitro “55” RNA lies on about 126 nucleotides.
170X0-
5.0 s
1souo130-
in vitro”5S”
i
110-
g 1 Y =
loo-
55
\
120-
goti?NA* \.\ 80-
707 41
m
00
90 Mobility
100
110
in mm
120
1Xl
Cell 166
C
a
P g@ 5g"
636
-30-A@& a7 , \ 8 1. _)
