recognizes a nuclear antigen associated with cell division in plant cells. To show that ... content of a molecular marker during somatic embryogenesis have been ...
Plant Physiol. (1993) 101: 809-817
lsolation and Characterization of a cDNA Clone for Plant Nuclear Antigen 21 D7 Associated with Cell Division’ Maria W. Smith’, Masaki Ito*, Takaya Yamada, Tatsuo Suzuki, and Atsushi Komamine Biological Institute, Faculty of Science, Tohoku University, Aoba-yama, Sendai 980 (M.W.S., M.I., A.K.); and Kitasato lnstitute Hospital, Shirogane, Minato-ku, Tokyo 108, Japan (T.Y., T.S.)
~
~~~
~
~
~~~~~~
its antibody, Mab 21D7, detected a single, clear band with a molecular mass of 45 kD on blots of SDS-treated carrot proteins (Smith et al., 1985). Analysis of tissue specificity of the 21D7 antigen using Mab 21D7 revealed that only rapidly growing parts of carrot seedlings and dividing cells in carrot suspension cultures contained detectable levels of this protein (Smith et al., 1988); cross-reacting antigens were observed in rapidly proliferating tissues of many other plant species. When plant cells were separated into soluble, nuclear, and membrane fractions, the nuclear fraction was found to be highly enriched in the 21D7 antigen. Indirect immunofluorescence studies confirmed that the antigen was localized in the nucleus, most prominently in the nucleolus (Smith et al., 1988). However, the level of the antigen was too low to purify it by routine methods (Smith et al., 1988). In the present study, the cDNA clone for 21D7 antigen was isolated from a carrot expression library using the Mab 21D7. Here we present the sequence of the carrot cDNA and describe the pattern of expression at the mRNA level in different carrot tissues. To analyze the correlation between cell division and expression of the gene for 21D7 antigen, the corresponding periwinkle (Catharanthus roseus [L.] G Don.) cDNA was isolated from a cDNA library and utilized for northem blot analysis of RNA from cell-suspension cultures grown under various conditions. The results concerning the dependence of expression of the cDNA on the process of cell proliferation are presented. A data base search revealed significant similarity between the deduced amino acid sequence of the carrot 21D7 antigen and the predicted sequence of the mouse tum- transplantation antigen P91A from mutant tumor cells (Lirquin et al., 1989).
A cDNA clone was isolated from a carrot (Daucus carota 1.) cDNA expression library using monoclonal antibody 21D7, which recognizes a nuclear antigen associated with cell division in plant cells. To show that the isolated cDNA encodes the 21D7 antigen, a polyclonal antiserum was raised against a recombinant fusion protein specified by the cDNA. 60th the polyclonal antiserum and the monoclonal antibody 21D7 recognized the same plant protein on immunoblots, in immunoprecipitationexperiments, and in peptide mapping. Analysis of the cDNA revealed that the deduced amino acid sequence has 45% identity to the predicted sequence of the mouse transplantationantigen P91A from mutant tumor cells that is responsible for the immune rejection of the corresponding cell clone in a syngeneic mouse. The expression of the plant cDNA at the mRNA level was highly correlated with cell proliferation. In suspension cultures of Catharanthus roseus (L.) C Don. cells, the highest level of expression was observedduring the midlogarithmic phase of growth. When auxin was added to stimulate cell division of auxin-starved cells arrested in the C, phase, transcription was immediately enhanced, and the level of expression remained high throughout the C1 and S phases and dropped dramatically at the end of DNA replication.
Carrot (Daucus carota L.) cell cultures provide a wellestablished system for the study of morphogenesis in plants using molecular biology methods (Fujimura and Komamine, 1979; Sung et al., 1984). In severa1 studies, changes in the content of a molecular marker during somatic embryogenesis have been correlated with the rate of cell proliferation (Smith et al., 1988; Dudits et al., 1991; Hirt et al., 1991). Smith et al. (1985) used an immunochemical approach to identify antigen markers of low abundance that were preferentially expressed in somatic embryos. A large pane1 of hybridoma clones was generated and screened for production of antibodies that were more reactive with crude homogenates from carrot somatic embryos than with those from carrot cell suspensions. For further work, hybridoma 21D7 was chosen because
MATERIALS A N D METHODS Plant Material
Carrot cell-suspension cultures were derived ,from hypocotyls of domestic carrot (Daucus carota L. cv Kurodagosun). Cells were subcultured at 7-d intervals in a modified Lin and Staba medium (Fujimura and Komamine, 1979), which contained 5 X 10-7 M 2,4-D, and were grown at 27OC on a reciproca1 shaker (90 strokes/min, 5 cm amplitude) in dark-
~~~~~
’
This work was supported by Grants-in-Aid for Scientific Research (02454007) from the Ministry of Education, Science and Culture, Japan, and by a grant from the Ministry of Agriculture, Forestry and Fisheries, Japan, for the Bio-media Program “Research on the mechanism of intemal bio-signals to create new technologies for agriculture, forestry and fisheries.” * Permanent address: Timiriazev Institute of Plant l’hysiology, Russian Academy of Sciences, Botanicheskaya 35, Moscow 127276, Russia. * Corresponding author; fax 81-022-262-5146.
Abbreviations: IgG, immunoglobulin G; Mab 21D7, the monoclonal antibody produced by the hybridoma 21D7; ORF, openreading frame; TSTA, tumor-specific transplantation antigen; tum-, cell lines that fail to form tumors in syngeneic animals.
809
81O
Smith et al.
ness. For RNA or protein isolation, log phase cells (at the 4th d after subculturing) were harvested. Carrot seeds were germinated in the dark for 10 d and then dissected into six parts: cotyledons, shoot meristem with surrounding tissues, hypocotyl, upper part of the root with growing adventitious roots, lower part of the root, and root tip. Two cell cultures of Catharanthus roseus (L.) G Don. were used. Cultivation of strain B was performed as described by Amino et al. (1983) and by Kodama et al. (1989); initiation and cultivation of the second strain, TN21, is described in detail by Nishida et al. (1992). These strains have been maintained at 27OC in the dark in Murashige-Skoog medium with 3% (w/v) Suc supplemented with 2.2 X 10-6or 4.4 X 10T6M 2,4-D for strain B or strain TNZI, respectively. Cells were subcultured at 7-d intervals. Strain B was utilized for the analysis of expression during growth of asynchronously proliferating cells. For studies on growth limitation by reduced levels of total nitrogen or Suc in the medium, 7-dold cells of strain B were transferred to media containing various concentrations of total nitrogen and SUC. For studies on growth arrest by total deprivation of nitrogen or SUC,2-d-old cells were washed severa1 times with nitrogen- or Suc-free medium and then transferred. to the same medium at the same density as in the original 2-d-old culture. To compensate for the effect of fresh m e l u m , control cells were also washed and transferred into fresh medium (Kodama et al., 1991a). The method for synchronization of strain TN21 cells by starvation and subsequent addition of 2,4-D was performed as described in detail by Nishida et al. (1992).Determinations of cell number and incorporation of [3H]thymidine were performed by the method of Amino et al. (1983). Chemicals
A11 restriction enzymes, DNA-modifying enzymes, and nucleic acids were purchased from commercial sources. Nylon membranes (Biodyne A) were bought from Pall Ultrafine Filtration Corp. (New York, NY). Nitrocellulose membranes (Hybond C), [ ( U - ~ ~ P I ~ (3000 C T P Ci/mmol), and [3H]thymidine (45 Ci/mmol) were obtained from Amersham Japan.
Plant Physiol. Vol. 101, 1993
Nucleotide Sequencing
Nucleotide sequences were determined by the dideoxychain-termination method (Sanger et al., 1977)using doublestranded DNA as a template. Northern Blot Analysis
Total RNA (10-16 pg/lane) was subjected to electrophoresis on 1%(w/v) agarose gels containing 2% (w/v) formaldehyde by a modified procedure of Lechrach et al. (1977). RNA was blotted onto nylon membranes and hybridized to probes labeled using the Multiprime DNA Labeling System (Amersham Japan) (specific activity 3 X 10' to 7 X 10' cpm/ pg). A11 steps were performed according to the instructions of the manufacturer of nylon membranes (Pall Ultrafine Filtration Corp.). lmmunoscreening of Carrot cDNA Expression Library
HPLC-purified Mab 21D7 (a generous gift from J.A. Smith and Z.R. Sung) was utilized for screening the carrot cDNA expression library. The library (kindly provided by M. Matsumoto) was constructed in bacteriophage Xgtll from poly(A)+ RNA isolated from roots of carrot seedlings. Screening was camed out using the pico-Blue Immunoscreening Anti-Mouse Kit (Stratagene) by the protocol of the manufacturer. Production of Polyclonal Antibodies
After analysis of cDNA inserts contained in selected phages, clones of interest were expressed as lysogens in bacterial strain Y1089 following the method of Huynh et al. (1986). Large-scale cultivation of lysogen containing DCl3 cDNA insert was performed according to Carro11 and Laughon (1987). Proteins were extracted with 2X SDS-Laemmlibuffer and subjected to SDS-PAGE (Laemmli, 1970). Recombinant fusion protein was isolated from gel slices by electroelution with subsequent dialysis and concentration, and was utilized for immunization of rabbits, as described by Harlow and Lane (1988). The resulting polyclonal antiserum was named FP13.
Purification of RNA and Construction of the cDNA Library
Total RNA was purified by the method of Schmidt et al. (1981). Poly(A)+ RNA was prepared from total RNA by chromatography on oligo(dT)-cellulose(Pharmacia). A Xgtll cDNA expression library was constructed from poly(A)+RNA isolated from carrot cell-suspension cultures in the logarithmic phase of growth. Double-stranded cDNA was prepared as described by Watson and Jackson (1985). The cDNA was ligated to the vector and packaged in vitro using the cDNA Cloning System Xgtll (Amersham). Another cDNA library was used that was prepared in previous work by Kodama et al. (1991a) from periwinkle cells in the S phase of the cell cycle. Both libraries were screened with DNA probes as described by Kodama et al. (1991b). Selected cDNAs were subcloned into the unique EcoRI site of plasmid vector pBluescript KS M13' (Stratagene) (Sambrook et al., 1989).
Western Blots
Total plant proteins were extracted with 2~ SDS-Laemmli buffer, electrophoresed using 9% (w/v) separation gels, and transferred to nitrocellulose--membranesby the method of Towbin et al. (1979). When the Mab 21D7 (10-5 dilution of the ascitic fluid) was to be used as a primary reagent to probe blots, goat anti-mouse IgG coupled to horseradish peroxidase (Bio-Rad) was applied as a secondary antibody, and a11 procedures were performed according to Smith et al. (1988). If the polyclonal antiserum FP13 (10-4 dilution of the rabbit serum) was utilized as a primary reagent, the secondary reagent was protein A conjugated with horseradish peroxidase (Bio-Rad). In the latter case, Tween 20 in both washing and blocking buffers was replaced with O. 1% (w/v) SDS and 0.05% (w/v) Triton X100. The development of immunoblots
A cDNA for Plant Cell Division-Associated Antigen 21D7 was performed with an ECL Western Blot Detection System (Amersham Japan). Immunoprecipitation
Plant proteins were extracted from 0.5 g (fresh weight) of carrot suspension cultures at the logarithmic phase of growth with 2 mL of RIPA buffer (Harlow and Lane, 1988). Aliquots (50 /iL each) of these extracts were immunoprecipitated, in parallel, with 2, 5, or 10 ^L of polyclonal antiserum FP13, or preimmune antiserum as a control, under the conditions recommended by Harlow and Lane (1988). Immune complexes were collected on protein A Sepharose CL-4B beads (Pharmacia). Both supernatants after removal of protein A beads and immunoprecipitates were subjected to SDS-PAGE. Following transfer of proteins to nitrocellulose, the blots were probed with Mab 21D7. Peptide Mapping One-dimensional peptide mapping by limited proteolysis with Staphylococcus aureus V8 protease (ICN Immunobiologicals) in SDS-polyacrylamide slab gels was carried out by the method of Cleveland et al. (1977) as modified by Luna et al. (1979). After SDS-PAGE of carrot protein extracts performed as above, gel slices corresponding to the position around the 43-kD molecular mass marker (Bio-Rad) were excised and equilibrated with buffer A: 125 ITIM Tris-HCl (pH 6.8), 0.1% (w/v) SDS, 1 mM EDTA. The gel pieces were then inserted at the bottom of the sample wells atop an SDSpolyacrylamide gel, composed of a long 5% (w/v) stacking gel (5.5 cm in length) and a 12% (w/v) separation gel (7 cm). All gel solutions contained 1 mM EDTA. Gel pieces in each well were overlaid with 10 ^L of buffer A containing 20% (v/v) glycerol. Finally, 10 nL of buffer A with 10% (v/v) glycerol and a given amount (20, 40, or 100 ng) of V8 protease were applied as a layer over the buffer in each slot. SDSPAGE was run at 15 mA (constant current) until the dye (bromphenol blue) entered the separation gel, and then the current was increased to 25 mA. After electrophoresis and blotting, digested polypeptides were probed either with Mab 21D7 or with polyclonal antiserum FP13 as described above.
811
Carrot DC13 cONA Does Encode for 21D7 Antigen
Immunoblotting demonstrated that recombinant fusion proteins, specified by all isolated phage clones with different cDNA inserts, were recognized by Mab 21D7 (data not shown). Therefore, it was necessary to show that the DC 13 cDNA indeed encoded for the 21D7 antigen. For this purpose, the fusion protein was purified from the culture of a recombinant lysogen expressing the DC 13 cDNA insert, and a new polyclonal antiserum (FP13) against this protein was generated. To compare plant antigens recognized by two different kinds of antibodies, Mab 21D7 and polyclonal FP13, proteins were extracted from carrot-suspension cultures at the logarithmic phase of growth, separated by SDS-PAGE, blotted, and probed with either Mab 21D7 or polyclonal antiserum FP13 (Fig. 1). Two different antibodies reacted with bands of identical molecular mass of about 45 kD. Preimmune antiserum did not recognize any carrot proteins when used to probe blots (data not shown). To confirm that the 45-kD band on different immunoblots represented the same antigen, carrot extracts were precipitated with polyclonal antiserum FP13, and resultant immune complexes were collected using protein A beads. Immunoprecipitates were electrophoresed, transferred to nitrocellulose, and probed with the Mab 21D7, followed by incubation with peroxidase-labeled anti-mouse IgG (Fig. 2). Three bands were detected on blots, one of which had a molecular mass of 45 kD, the same as 21D7 antigen in control entire extracts. Thus far, the identity of the lower two bands is unclear.
Mab21D7
FP13
kD
—— 97.4
66.2
RESULTS Isolation of the Carrot cDNA Clone with Mab 21D7
Mab 21D7 was used to screen the Xgtll cDNA expression library prepared from poly(A)+ RNA extracted from roots of carrot seedlings. Seven positive clones were isolated from a total of 3 x 105 recombinant phages. It was shown by crosshybridization experiments that four clones contained an identical cDNA of about 0.7 kb, which was named DC 13 (DC from Daucus carota). Other inserts had no similarity either with DC 13 cDNA or with each other (data not shown). Sequencing of terminal regions of selected cDNA inserts revealed that only the DC 13 cDNA was integrated in the vector in frame with the ORF of the lacZ gene.
.42.7
-31.0 Figure 1. Blot analysis of carrot proteins with different antibodies. Mab 21D7 was used at a 10~5 dilution of the ascitic fluid; for detection, a 10~" dilution of goat anti-mouse IgC coupled with horseradish peroxidase was applied. Polyclonal antiserum FP13 were used at a 1CT4 dilution of the whole rabbit serum, with subsequent detection by a 1CT4 dilution of protein A conjugated with peroxidase. Positions of the molecular mass standards expressed in kD are shown at the right side.
Smith et al.
812
PI
antiserum FP13. Both kinds of antibodies recognized the same polypeptides on the blots, and two very similar peptide maps were obtained (Fig. 3). The upper band appeared to be nondigested protein, whereas the lower bands represented partially digested polypeptides, and their intensity increased with the amount of protease applied.
FP13
4
Plant Physiol. Vol. 101, 1993
5
97.4 —
Sequence Analysis of Carrot cDNA for 21D7 Antigen
IgG
Figure 2. Immunoprecipitation of carrot proteins. Lane 1, The entire protein extract; lane 2, proteins precipitated with protein A Sepharose beads only; lane 3, precipitate obtained with 5 ^L of preimmune antiserum; lanes 4 and 5, the precipitates obtained with 5 and 2 ^L of polyclonal antiserum FP13, respectively. In lanes 3, 4, and 5, immune complexes were collected on protein A beads. Immunoprecipitates were resolved by SDS-PACE, blotted, probed with Mab 21D7 (10~5 dilution of ascitic fluid), and incubated with a 10~4 dilution of anti-mouse IgG conjugated with peroxidase. Molecular mass markers are indicated on the left side; the dark arrowhead shows the position of the 45-kD band. The upper band in immunoprecipitates is the heavy chain of rabbit IgC from the antiserum used for precipitation.
These proteins were not observed in precipitates obtained with preimmune antiserum or protein A beads alone (Fig. 2), indicating that they could be breakdown products or coprecipitating proteins. If incubation with Mab 21D7 was omitted, and the blot was treated with anti-mouse IgG, it reacted only with the heavy chain of rabbit IgG from antiserum FP13 or preimmune antiserum (data not shown). Taken together, these results demonstrated that the 45-kD antigen precipitated by polyclonal antiserum FP13 was recognized by Mab 21D7. The proteins remaining in the supernatants after completion of the immunoprecipitarion were also subjected to SDSPAGE, blotted, and probed with Mab 21D7. If excess amounts of polyclonal antiserum FP13 were applied for precipitation, the 45-kD antigen recognizable by Mab 21D7 was almost completely depleted from the extracts. However, when the same amounts of preimmune antiserum or protein A beads only were used for precipitation, the 45-kD antigen remained in the extracts and the band could be observed on blots (data not shown). Another approach was also applied to demonstrate that both kinds of antibodies recognize the same carrot protein. Gel slices were excised from the region of SDS-polyacrylamide gels corresponding to the position of the 45-kD molecular marker. The slices were exposed to limited quantities of staphylococcal V8 protease during the second SDS-PAGE, and the digested polypeptides, resolved by electrophoresis, were blotted and probed with either Mab 21D7 or polyclonal
The size of the DC 13 cDNA, which was shown to encode for the 21D7 antigen of carrot, was about 0.7 kb. RNA blot analysis revealed the presence of the mRNA corresponding to DC 13 cDNA in cells of carrot suspension cultures. However, comparison with RNA standards gave a size of about 2.2 kb for the transcript (data not shown). To isolate longer cDNA clones, we constructed a new carrot cDNA library in Xgtll using poly(A)+ RNA extracted from carrot cell-suspension cultures in the logarithmic phase of growth. Screening of 4.7 X 10s recombinant phages from the new library with the probe of DC 13 cDNA resulted in isolation of eight positive clones. The longest cDNA insert, 1.9 kb, was subcloned and sequenced (Fig. 4). It is likely that the isolated and characterized long cDNA insert represents almost the complete coding region of the gene. The clone contains a long ORF from an ATG codon located at position 30 until a stop codon at position 1496, followed by a 357-bp 3' untranslated region with a poly(A) tail (Fig. 4). The long ORF codes for a protein of 488 amino acids with the deduced molecular mass of 55 kD, which is 10 kD larger than the apparent molecular mass of the 21D7 antigen. This protein contains 7 Cys residues, and it is rich in Leu (59 residues, 12% of all amino acids). protease 20
40
1
2
Mab21D7
100
40
3
4
FP13
Figure 3. Peptide maps obtained by limited proteolysis in SDSpolyacrylamide gels. Gel slices containing carrot proteins with the molecular mass around 45 kD were excised from the first gel, placed in the sample wells of the second gel, and exposed to a given amount of staphylococcal V8 protease during subsequent electrophoresis: lane 1, 20 ng; lanes 2 and 4, 40 ng; lane 3, 100 ng of enzyme. Digested polypeptides were blotted and probed with Mab 21D7 (lanes 1, 2, and 3) or polyclonal antiserum FP13 (lane 4) (both maps were done from a single first gel).
A cDNA for Plant Cell Division-Associated Antigen 21 D7 ATTTMCCGGAGCTCCGATCTACACMCAATGACTCMGACGTTGAMTGMA~GGTTC U T- O D V B . - U . . K .. E - V
60
CGGCGCCGGCGCCCTCTMTTCCGTCACGGCCGCTASTCCCTCCACTCTCXAGCATTTGA 120 P A P A P S N S V T A A T P S T L Q H L AGGAGATTGCTTCGTTGATTGAGTCTGGTGCTTATGCTCGCGMGTTCGTCGGATTCTCC 180 K E I A S L I E S G A Y A R E V R R I L GAGCAGTTCGGTTGACGATTGCGTTGAGAMGMGCTTMTGCGTCTGTTGTCMTGCGT 240 R A V R L T I A L R I K L I V A S V V R A TTCTCMTTTCTCTCTGGTTCCTGGATCTGAGGTTCATGCTCGTTTGGCTTCATATCTTC 300 P L R P S L V P G S E V H A R L A S Y L &A~&&TGAGCATGATATGGAAGTTGACACTGCMTQTCT&CTACMCCCTCG 360 P K E D E H D U E V D T A M S A T T T L CMMCATTCCTTGCCTGAGCTCGAGATTTACTGCTATTTACTTGTGCTMTATTTCTCA 420 A K H S L P E L E I Y C Y L L V L I P L T T a T C A G M - T A C I L G T G A ~ T ~ G C T T ~ C C T C A ~ G T A ~ G C C C ~ G A480 I D Q K K Y S E A K A C S S A S I A R V AGMCTTaTAGACaCTGTGGAAGTTTTAGCATCCAGGCTCTACTTCTATTATTCCC 540 K N L N R R T V E V L A S R L Y P Y Y S TTAGCTATOMCTTACAGGCGATCTGGCTGAMTTCGGGGAMTCTACTGGCCTTWATA 600 L S Y E L T G D L A E I R G N L L A L H GGATTGCMCACTTCGCCATGATGAGTTOOOACA(iOAAAC 660 R I A T L R W D E L G Q E T L L I V L L L GCMTTATCTCCACTACMTTTATATGACCAGGCAGAGAMCTMGGTCMAGGCACCTC 720 R N Y L H Y N L Y D Q A E K L R S K A P GATTTGMGCCCATTCCAATCMCAGTTTTGCCGCTACTTGTTTTACCTAGGMAGATTA 780 R P E A H S N Q Q P C R Y L F Y L G K I GMCTATTCAGCTGGAGTATACTGATGCGAMGMTCATGCTCCMGCTGCCCGTMGCC 840 R T I Q L E Y T D A K E S C S K L P V S CCTGTTGCAGCTCTTGGTTTCCGAGTCCMTGCMCAAGTGIXCGGTGATAGTACGCTAC 900 P C C S S W F P S P M Q Q V G G D S T L T A T T A G G A - T C C C A ~ G C G G A C T G T T T ~ T A T ~ G M ~ T ~ G C A T T G A960 L L G E I P E R T V F U Q K G M E K A L GGCCGTACTTTWLCTTACMATGCTGTACGMTTGGGGATCTGGAGCTCTTTAGAGCTCG 1020 R P Y F D L Q M L Y E L G I W S S L E L TGGCTGACMGTTCGCTAGCACATTCACTGCAGACCGGACCCATMTCTGATTGTCAGGC 1080 V A D K F A S T F T A D R T H N L I V R TTCGTCATMTGTCATMGMCTGGGCTTCGCMCATCAGTATTTCATACTCCCGCATTT 1140 L R H R V I R T G L R N I S I S Y S R I CATTAGTGGATGTGGCCAGGMGCTGAGATTGGACTCTCCMATCCTGTTGCTGAT&TG 1200 S L V D V A R K L R L D S P N P V A D A MAGTATTGTATCCMGGUATAAGGGATGGAGCCATTGACGCCACCATCGATCATGCM 1260 E S I V S K A I R D G A I D A T I D H A A T O O G T G G A T G G T G T C A M O C ~ G A C A T T T A C T C A A C A M T G A G C C A ~ G C A G 1320 N G W U V S K E T G D I Y S T N E P Q A CTTTTMTTCCAGGATAGCGTTTTGCCTTMCATGCATMTGMGCTGTTCGTGCACTTC 1380 A F N S R I A F C L N U R R E A V R A L GATTTCCCOCAMTTCTCGCT~~GAGGAGAGAGAaCAAC 1440 R F P A N S H K D K E S A E K R R E R Q AGCAGGAGCM~GCTTOCAMGCATATAGCCGAGGAAGATGATGATGAGTTTTMGCTT 1500 Q Q E Q E L A K H I A E E D D D E F ' TCATGCACATGACTAGGTGGCATTGAGCTATGACTCTTCTACAGCTGTTGCCACAMTTT 1560 GGATCATATTATAGACMCGACAGATTCTCTTTACCTTATTTTCCTTTCAMCGMTTTG 1620 T T G T C T G O A G T A T T T W T T C C G T A T T G T T A C T ~ T ~ T G T C C G T T T ~ T C C T C C ~ A T1680 A~ TTTTAGACTTTGACATCTTTGTMCTTTGGTTATGTTTGTATGGTGAGGGGATCC 1140
CTTTTAGGTGGATGTTTACTTGAGMTTTTTCCATTCMTQGGAGMTGATGCTMATTG T C A T T A T T T W G C P
1800 1853
Figure 4. Nucleotide and deduced amino acid sequences of the carrot c D N A for 21 D7 antigen. Numbers are for nucleotides; amino acids are shown in t h e single-letter code. The star indicates the stop codon.
813
Analysis of Expression at the mRNA Leve1
RNA blot analysis, performed with the DC13 cDNA as a probe, indicated the presence of the mRNA in carrot cell suspensions in the logarithmic phase of growth (data not shown). In carrot plantlets, the highest amount of the transcript corresponding to the DC13 cDNA was observed in root tips (data not shown). A periwinkle cDNA clone homologous to the DC13 cDNA was isolated from a cDNA library prepared with poly(A)+ RNA extracted from C. roseus cells. Of 2 X 105 phage screened, only two positive ones were found, both containing the same cDNA insert of 1.6 kb. Partia1 sequencing of the selected cDNA revealed about 85% homology of the predicted amino acid sequence to the carrot 21D7 antigen. The periwinkle cDNA was named CR8 (from Cutharunthus roseus) and was used for hybridization with blots of periwinkle RNA. Total RNA was isolated from asynchronously growing periwinkle cell suspensions at different days after subculturing. The results of hybridization with the probe prepared from the CR8 cDNA are shown in Figure 6. The amount of the mRNA was low during the lag phase and the early logarithmic phase and then increased to a maximum in the midlogarith&c phase, 4 d after subculturing. The level of expression began to decrease after 5 d and dropped to a minimum level by day 7. In other experiments, cell growth was arrested in the logarithmic phase by nutrient starvation, and cells were harvested when proliferative growth had ceased. The results of blot analysis of poly(A)+ RNA (Fig. 7) indicated that the amount of mRNA corresponding to the CR8 cDNA decreased
MTQDVEMKEVPAPAPSNSVT
1'
. *.*...*
1" MKQEQSARRRQHKAKF'PPGGEQEPPPPAPQDVEMKEEA4AQSGS'IQEGDGKtAATEHSQR
21' AATPSTWILEEIASLIESQAYAREVRRILRAVRLTIALRKKLNASVVNAFLNFSLVPQS
. . . **...*. .. .*... ..* * .**..*.
. ...** *.
.. .....
61" ELDTVTLEDIKEHVRQLEKAVSGKEPRF'fLRLRMLPSTSRRLNHYVLWL4VHGFFTSNN
Data base searches revealed a homologous amino acid sequence deduced from a nucleotide sequence of a mouse gene encoding tum- transplantation antigen P91A in mutant tumor cells (Lirquin et al., 1989). The degree of identity between the entire carrot and mouse polypeptides is about 45% (Fig. 5). When conservative substitutions are taken into account, the amino acid sequences show 78% homology. The most conservative are carboxy-terminal regions and the sequence composed of 39 amino acids that begins at position 202 of the 21D7 protein (82% identical and 13% similar residues). One-quarter of a11 amino acids in the latter sequence are Leus. The mouse polypeptide has the deduced molecular mass of 60 kD, and it is 30 residues longer than its carrot counterpart; however, these additional amino acids are localized in the most divergent region at the amino terminus. Computer searches revealed a possible nuclear targeting sequence of 7 amino acids (EKLRSKA) that begins at position 222 of the carrot polypeptide and shares 70% homology with the yeast histone 2B nuclear localization signal (GKKRSKA) (Moreland et al., 1987).
81' EVHARLASYLPKEDEHDMEVDTAMSATTTLAKHSLPELEIYCYLLVLIFLIDaKKYSEAK I**.* 8 121" ATRDFLLPFLEEPMDTEADLQFRPRTGKAASAPLLPEVEAYLQLLHVIFLMNSKRYEEAQ
.... * ..* . .
. . . . ..... .
**..***.. *.*..*.
.
141' ACSSASIARVKNLNRRTVEVLASRLYFWSLSYELMDLAEIRQNLLALHRIATLRHDEL
*.. ...... *.*.....*.. *.*.. **. ..*. .*. * *
*.****.i.
181" K I S D D L n O g I S T P N H R A L D L V ~ ~ ~ E F L D K L D ~ S F L ~ L R T A T L ~ A D 201' WETLLNLLLRNYLHYNLYDPAEgLRSKAPRFULRSKAPRFEAliSN~FCRYLFYLGKIRTIQLNTDA
..........................
8;.
*. ..... .**.* *.*..**.**..*
241" OOATLLMLLRNnfflSLYDOAEHLVSKSM"PEOAN"EWARnWTaRIRA1PLEYSEA 261' K E S - C S K L P V S P - C C S S W F P S P M ~ f f i D S T L L L G E I P E R ~ F M ~ ~ E ~ L ~ Y F D ~
. *. ..
. * .....
.*.**.**.*
*.* .......*
8.
*
301" ~~NALRgAPQHTAVGFKQTWiKLLIWELLLGEIPDRLQFRQPSLKRSLMPYFLLTQ 319' L Y E L G I W S S L E L V A D E F A S T F T A D R T H N L I V R L R H N V I R M K
* .... * *.*...* ,* *..**.**.**.*.**.*
**. .****** .*...*
361" AVRTGNLAKFNQVLWFGEKFQTDGTYTLIIRLRHNVIKMVIU4ISLSYSRISLADIAOK
379' LRLDSPNPVADAESIVSRAIRDOAIDATIDATID~~SKETQDIYSTNEP~FNSRIAF **S *****.a**
......................
..+***.
***
**. **.*
421" LQLDSPE---DAEFIVAKAIRDGVIEASI~EKG~QSKEMIDIYSTREPQLAFH~ISF
439' CLNMHNEAVRALRFPANS-HKDKESAEKRRE~EQELAKHIAEEDDDEF
.*.*.***..* .** ****.****.**. *.**..**.*.*.*
**..I*
478" CLDIIUWSVKAMRFPPKSYNKDLESAEEREEREWLEFAKEMAEDDDDSFP
The alignment between the deduced amino acid sequences of carrot 21D7 antigen (upper line) and mouse antigen P91A (lower line). The single-letter code for amino acids is used. Asterisks and dots show identical and similar amino acids, respecFigure 5.
tively. T h e possible nuclear targeting sequence, which has 70% homology to the nuclear localization signal of the yeast histone 2 6 (Moreland et al., 1987), is underlined.
Smith et al.
814
0 1 2 3 4
567
Plant Physiol. Vol. 101, 1993
8 days
-A
0
2
5
8
11 14 17 20 24 hours
12 40
I ••^
a> "5
20
10
3
5
E12
°L 0
2
4
6
8
Time of Culture (day)
SI60 -2,40
Figure 6. Expression of the gene for 21D7 antigen in growing periwinkle (C. roseus) cell suspensions. Total RNA was extracted and analyzed by northern blot hybridization (6 Mg/lane) with a probe prepared from CR8 cDNA, the C. roseus homolog of DC13 cDNA.
dramatically in comparison with that of control cells in the logarithmic phase. A recently established system (Nishida et al., 1992) in which cell division in periwinkle suspension cultures was partially synchronized by withdrawal and subsequent addition of auxin was also used to study the expression of the gene corresponding to CR8 cDNA. Stationary-phase cells of the strain TN21 were transferred to fresh medium free of 2,4-D and cell division ceased. After 2 d, readdition of 2,4-D to the medium rapidly induced the progression through the cell cycle and resulted in a single peak of DNA replication from 9 to 14 h. Consequent cell division occurred from 17 to 20 h with about 75% synchrony (Fig. 8).
Figure 7. Decrease in mRNA corresponding to the periwinkle gene for 21D7 antigen in growth-arrested cells. Lane 1, Control cells in the logarithmic phase of growth; lanes 2 and 4, cells in the logarithmic phase were transferred to the medium without nitrogen or Sue, respectively, and cultivated for 2 d; lanes 3, 5, and 6, 7-d-old cells were transferred to the medium with only 1/50 of the usual content of total nitrogen (lane 3) or reduced content of Sue (lane 5, 0.5%; lane 6, 1%) and cultivated for 5 d. Under the described conditions, the proliferation of cells ceased completely, although the cells remained viable. Poly(A)+ RNA was extracted, electrophoresed (0.1 Mg/lane), and hybridized with a probe prepared from CR8 cDNA.
0
5
10
15
20
25
Time of Culture (hour)
Figure 8. Expression of the periwinkle gene for 21D7 antigen during auxin-stimulated growth of suspension cultures. C. roseus cells were arrested at the d phase by auxin starvation. Addition of 4.4 x 10~6 M 2,4-D led to the progression of the cell cycle. -A, Cells were harvested after 2 d of cultivation without auxin. Cells at 0 h were collected after 15 min of incubation with the auxin; other points were harvested at the times indicated above the blot. Total RNA (6 fjg of each sample) was analyzed by northern hybridization with the CR8 cDNA. Incorporation of [3H]thymidine into the DNA fraction (middle panel) and the number of cells during treatment (lower panel) are shown.
In this partially synchronized system, a low level of expression of the gene corresponding to the CR8 cDNA was detected in arrested cells after incubation in auxin-free medium for 2 d (Fig. 8, —A). After readdition of auxin, the level of expression immediately increased at least 3-fold (0 h; the cells were harvested after a 15-min incubation with auxin) and remained high throughout the S phase during the next 14 h. After 14 h, the amount of mRNA transcript rapidly decreased below the level observed in arrested cells and remained at a low level until the end of mitosis. When cDNA homologs of rRNA and the extensin gene were used as probes for RNA hybridization, constitutive patterns of expression were observed in auxin-synchronized cells (data not shown). DISCUSSION
Using Mab 21D7, we have isolated and characterized a carrot cDNA for the 21D7 antigen. A polyclonal antiserum against a fusion protein specified by the cDNA was used to show that the isolated clone corresponded to the gene of the 21D7 antigen. The results of immunoblotting and immunoprecipitation demonstrated that the polyclonal antiserum and the monoclonal antibody recognized the same carrot antigen. Additional evidence was found through the use of peptide mapping experiments with V8 protease, which cleaved pro-
A cDNA for Plant Cell Division-Associated Antigen 21 D7 teins after the acidic residues aspartic or glutamic acid. The patterns of partially digested polypeptides recognized by the two different antibodies were very similar, indicating that polypeptide fragments appeared as a result of cleavage of the same protein with identical positions of accessible acidic residues. Sequencing of carrot cDNA for the 21D7 antigen showed that it contained an ORF encoding 488 amino acids with a high Leu content. The apparent molecular mass of 21D7 antigen detennined by immunoblotting is 10 kD smaller than the deduced mass of the polypeptide, probably due to some posttranslational proteolytic cleavage. A similar amino acid sequence derived from the mouse gene for the tum- transplantation antigen P91A (Lirquin et al., 1989) was found in data bases. The overall homology between the putative sequence of carrot antigen 21D7 and mouse antigen P91A is about 45%, and it is up to 70 to 80% in conservativeregions in the middle of the polypeptides and in the carboxy termini. Thus, we conclude that the 21D7 protein should be a plant counterpart of the mouse antigen. The similar sizes of the deduced amino acid sequences and the same sizes (2.2 kb) of the corresponding mRNAs also support our conclusion. The high similarity between the carrot and mouse polypeptides suggests that their function might be well conserved evolutionarily. The monoclonal antibody 21D7 raised against the carrot protein did not react with mouse proteins (Smith et al., 1988). Nothing is known about the position of the corresponding epitope in the carrot protein. However, monoclonal antibodies typically recognize small epitopes composed of a few amino acids (Harlow and Lane, 1988); therefore, this epitope could be nonconservative or hidden in the structure of the mouse protein. Mouse antigen P9 1A is a representative of the wide group of TSTAs of experimental tumors. Expression of TSTA elicits a T-cell-mediated immune rejection response in syngeneic animals, so cell lines producing TSTA fail to form tumors in vivo and have been named “tum-” cells (Boon, 1983). Usually, a tum+ cell line, which can form progressive tumors in animals after mutagen treatment in vitro, generates stable tum- mutant cell clones. There is evidence that mutations of many unrelated genes could result in the appearance of TSTA and produce the tum- phenotype (Boon and Van Pel, 1989). Studies of P91A antigen expression transfer by gene fragments led to a hypothesis that immunogenicity was not related to the entire polypeptide encoded by the gene of P91A antigen (Boon and Van Fel, 1989). Lirquin et al. (1989) have demonstrated that peptides causing immune rejection might be generated by the local transcription and translation of small subgenic regions of the mutant gene, rather than the degradation of the large protein product. Therefore, there is no information about the possible function and localization of the entire P91A gene product in mouse cells (Lirquin et al., 1989). Before the present study, no similarities had been observed between the sequence of P9 1A and other sequences in existing data bases. The nuclear localization of the plant 21D7 antigen (Smith et al., 1988)led us to undertake a computer search for nucleic acid-binding and nuclear localization motifs in both the plant
81 5
and mammalian polypeptides. Analysis of the mouse protein sequence yielded no significant similarity with known motifs, but the plant protein had a sequence with 70% homology to the nuclear localization signal from the yeast histone 2B (Moreland et al., 1987). However, the degree of homology between the mouse sequence and the yeast motif is only 30%, despite the location of the corresponding sequence in the evolutionarily conserved region. Recently, a new hypothesis on the possible structure of nuclear targeting sequences emerged from studies of many diverse yeast and mammalian nuclear localization signals (Dingwall and Laskey, 1991). These “bipartite”sequences are usually composed of two small basic clusters separated by a spacer of 10 amino acids, but some spacers could be shorter or longer (Garcia-Bustos et al., 1991). The sequences of both carrot and mouse proteins contain severa1short basic clusters, but the spacers between them are longer than 10 amino acids, so caution is required in the interpretation of their similarity to these motifs. Moreover, some nuclear proteins do not possess their own nuclear localization signals and enter the nucleus via cotransport with another protein (Silver, 1991); thus, we cannot exclude this possibility in the case of the 21D7 antigen. Experimental systems to examine nuclear localization sequences of plant proteins have been developed only recently, so the structure of plant signals is understood poorly (Camngton et al., 1991). So, in our opinion, further studies are necessary to define a nuclear targeting sequence within the carrot 21D7 polypeptide. Analysis of amount of the mRNA corresponding to the mouse gene of tum- transplantation antigen P91A demonstrated that it was expressed at the same level in both original and mutant tumor cell lines cultivated in vitro. In mouse liver and spleen cells, it was expressed at significantly lower levels (Lirquin et al., 1989). These observations are consistent with correlations between 21D7 gene expression and cell proliferation in plants. Many genes associated with cell division and progression of the cell cycle are highly conserved among eukaryotes (reviewed in Jacobs, 1992). For example, protein kinase ~34‘~‘’from alfalfa has 64% similarity to the mammalian enzyme (Hirt et al., 1991), and plant cyclins have approximately 47% homology to their Xenopus counterparts (Hemerly et al., 1992). Expression of the plant gene for 21D7 antigen at the mRNA level supports the conclusions of previous studies indicating that this gene is closely associated with cell proliferation. In carrot seedlings, the high level of expression observed in root tips is similar to that of other cell division-related genes, e.g. the periwinkle gene for proliferating-cell nuclear antigen (Kodama et al., 1991b) or the S-phase-specific periwinkle gene cycO7 (Ito et al., 1991). Accumulation of mRNA corresponding to cDNA for the 21D7 antigen was observed in carrot and periwinkle cellsuspension cultures in the mid and late logarithmic phase. The decrease in the mRNA content in the stationary phase was associated with the decrease in cell proliferation. When cell division was arrested by nutrient starvation, the amount of 21D7 mRNA also dropped significantly. Partially synchronized periwinkle cell suspensions were used to study the dependence between expression of the
Smith et al.
816
21D7 gene a t the mRNA level and progression through the cell cycle. Although mRNA w a s detectable in auxin-starved cells, which were arrested i n the G1 phase, transcription appeared to be highly auxin-inducible. Immediately after the addition of 2,4-D, t h e amount of the transcript increased severa1 times. Similar rapid induction of transcription by auxin was observed with t h e par gene, which w a s shown to be expressed abundantly i n response to auxin prior to the initiation of DNA synthesis i n tobacco mesophyll protoplasts (Takahashi et al., 1991). The amount of mRNA corresponding to the gene of 21D7 antigen remained high throughout the S phase, then rapidly declined a t the end of DNA replication and w a s low until the end of mitosis. In the same system, similar decreases i n the amount of transcript a t t h e end of D N A replication were reported for the S-phase-specific genes cycO7 (Ito et al., 1991) a n d proliferating-cell nuclear antigen (Kodama e t al., 1991b). Periodic transcription of genes encoding b o t h structural a n d regulatory products was found to be important for the proper execution of events i n complex cell cycles (reviewed i n McKinney and Heintz, 1991). Some of these so-called cell cycle-regulated genes, e.g. histone genes, are expressed during the S phase exclusively (McKinney a n d Heintz, 1991). It is likely that t h e expression of the gene for t h e 21D7 antigen is also regulated a t the transcriptional level during t h e progression of t h e cell cycle and that t h e half-life of its mRNA is relatively short. We cannot attribute t h e gene for 21D7 antigen to the group of S-phase-specific genes because there are detectable amounts of mRNA in arrested cells and in cells during the G1 phase. However, the cell cycle-dependent appearance a n d disappearance of mRNA suggests a possible link between the function of the gene a n d progression of the cell cycle. The evolutionarily conserved sequence of t h e animal and plant proteins supports t h e speculation that these proteins might play an important role i n t h e cell proliferation. ACKNOWLEDCMENTS
We are deeply grateful to Drs. J.A. Smith and Z.R. Sung for providing Mab 21D7 and to Dr. M. Matsumoto for the carrot root cDNA expression library. We thank Drs. R. Phillips and R.W. Joy IV for critica1 reading of the manuscript. Received July 27, 1992; accepted October 29, 1992. Copyright Clearance Center: 0032-0889/93/lOl/O809/09. The nucleotide sequence of the cDNA for 21D7 antigen presented in this paper will appear in the DDBJ,EMBL, and GenBank nucleotide sequence data bases with the accession number D13434. LITERATURE CITED
Amino S, Fujimura T, Komamine A (1983) Synchrony induced by double phosphate starvation in suspension culture of Catharanthus roseus. Physiol Plant 5 9 393-396 Boon T (1983) Antigenic tumor cell variants obtained with mutagens. Adv Cancer Res 3 9 121-151 Boon T, Van Pel A (1989) T cell-recognized antigenic peptides derived from the cellular genome are not protein degradation products but can be generated directly by transcription and translation of short subgenic regions. A hypothesis. Immunogenetics 2 9 75-79 Carrington JC, Freed DD, Leinicke AJ (1991) Bipartite signal se-
Plant Physiol. Vol. 101, 1993
quence mediates nuclear translocation of the plant potyviral NLa protein. Plant Cell 3 953-962 Carro11 SB, Laughon A (1987) Production and purification of polyclonal antibodies to the foreign segment of beta-galactosidase fusion protein. In DM Glover, ed, DNA Cloning, Vol3. IRL Press, Oxford, Washington, pp 89-111 Cleveland DW, Fischer SG, Kirschner MW, Laemmli UK (1977) Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel electrophoresis. J Biol Chem 252 1102-1106 Dingwall C, Laskey RA (1991) Nuclear targeting sequences-a consensus? Trends Biochem Sci 16: 478-482 Dudits D, Bogre L, Gyorgyev J (1991) Molecular and cellular approaches to the analysis of plant embryo development from somatic cells in vitro. J Cell Sci 9 9 475-484 Fujimura T, Komamine A (1979) Synchronization of somatic embryogenesis in a carrot cell suspension culture. Plant Physiol 6 4 162-1 64 Garcia-Bustos J, Heitman J, Hall MN (1991) Nuclear protein localization. Biochim Biophys Acta 1071: 83-101 Harlow E, Lane D (1988) Antibodies. A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Hemerly A, Bergounioux C, Van Montagu M, Inze D, Ferreira P (1992) Genes regulating the plant cell cycle: isolation of a mitoticlike cyclin from Arabidopsis thaliana. Proc Natl Acad Sci USA 8 9 3295-3299 Hirt H, Pay A, Gyorgyev J, Bako L, Nemeth K, Borge L, Schweyen RJ, Heberle-Bors E, Dudits D (1991) Complementation of a yeast cell cycle mutant by an alfalfa cDNA encoding a protein kinase homologous to p34cd'2.Proc Natl Acad Sci USA 8 8 1636-1640 Huynh TV, Young RA, Davis RW (1986) Construction and screening of cDNA libraries in lambda gtlO and lambda g t l l . In DM Glover, ed, DNA Cloning, Vol 1. IRL Press, Oxford, Washington, pp 49-78 Ito M, Kodama H, Komamine A (1991) Identification of a nove1 Sphase-specific gene during the cell cycle in synchronous cultures of Catharanthus roseus cells. Plant J 1: 141-148 Jacobs T (1992) Control of the cell cycle. Dev Bioll53: 1-15 Kodama H, Ito M, Hattori T, Nakamura K, Komamine A (1991a) Isolation of genes that are preferentially expressed at the G1/S boundary during the cell cycle in synchronized cultures of Catharanthus roseus cells. Plant Physiol95 406-411 Kodama H, Ito M, Suzuka I, Komamine A (1991b) Molecular cloning of the gene for plant proliferating-cellnuclear antigen and expression of this gene during the cell cycle in synchronized cultures of Catharanthus roseus cells. Eur J Biochem 197: 495-503 Kodama H, Kawakami N, Watanabe A, Komamine A (1989) Phasespecific RNAs during the cell cycle in synchronous cultures of Catharanthus roseus cells. Plant Physiol89: 910-917 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685 Lehrach H, Diamond D, Wosney JM, Boedtker H (1977) RNA molecular weight determination by gel electrophoresis under denaturing conditions, a critical reexamination. Biochemistry 1 6 4743-4751 Lirquin C, Van Pel A, Mariame B, De Plaen E, SzicoraJ-P,Janssens C, Reddenhase MJ, Lejeune J, Boon T (1989) Structure of the gene of tum- transplantation antigen P91A: the mutated exon encodes a peptide recognized with Ld by cytolytic T cells. Cell58 293-303 Luna EJ, Kidd GH, Branton D (1979) Identification by peptide analysis of the spectrin-binding protein in human erythrocytes. J Biol Chem 254 2526-2532 McKinney JD, Heintz N (1991) Transcriptional regulation in the eucaryotic cell cycle. Trends Biochem Sci 16: 430-435 Moreland RB, Langevin GL, Singer RH, Garcea RL, Hereford LM (1987) Amino acid sequences that determine the nuclear localization of yeast histon 2B. Mo1 Cell Biol7: 4048-4057 Nishida T, Ohníshi N, Kodama H, Komamine A (1992) Establishment of synchrony by starvation and readdition of auxin in suspension cultures of Catharanthus roseus cells. Plant Cell Tissue Organ Cult 28: 37-43 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. A
A cDNA for Plant Cell Division-Associated Antigen 21 D7 Laboratory Manual, Ed 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Sanger F, Nicklen S, Couson AR (1977) DNA sequencing with chain-tenninating inhibitors. Proc Natl Acad Sci USA 7 4 5463-5467
Schmidt GW, Bartlett SG, Grossman AR, Cashmore AR, Chua N (1981) Biosynthetic pathway of two polypeptide subunits of the light-harvesting chlorophyll a/b protein complex. J Cell Biol 91: 468-478
Silver PA (1991) How proteins enter the nucleus. Cell 6 4 489-497 Smith JA, Choi JH, Krauss MR, Karu AE, Sung ZR (1985) Monoclonal antibodies that recognize proteins unique to somatic embryos of Daucus carota. In M Terzi, L Pitto, ZR Sung, eds, Somatic Embryogenesis of Carrots. IPRA, Rome, pp 86-94 Smith JA, Krauss MR, Borkird C, Sung ZR (1988) A nuclear protein associated with cell divisions in plants. Planta 174: 462-472
81 7
Sung ZR, Feinberg A, Chorneau R, Borkird C, Furner I, Smith JA, Terzi M, LoSchiavo F, Guiliano G, Pitto L, Nutti-Ronchi V (1984) Developmental biology of embryogenesis from carrot cultures. Plant Mo1 Biol Rep 2: 3-14 Takahashi Y, Kusaba M, Hiraoka Y, Nagata T (1991) Characterization of the auxin-regulated par gene from tobacco mesophyll protoplasts. Plant J 1: 327-332 Towbin H, Staehelin T, Gordon J (1979) Electrophoretictransfer of proteins from polyacrylamide gels to nikocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76: 4350-4354
Watson CJ, Jackson JF (1985) An altemative procedure for the synthesis of double-stranded cDNA for cloning in phage and plasmid vectors. In DM Glover, ed, Molecular Cloning, Vol 1. IRL Press, Oxford, Washington, pp 79-88
