Gene, 110 (1992) 65-70 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0378-1119/92/$05.00
65
GENE 06246
Isolation of c D N A clones encoding proteins of complex structures: analysis of the Trypanosoma brucei cytoskeleton (Recombinant DNA; immunoscreening; antibodies; ~.gtl 1 vector; antisera; immunofluorescence)
Christopher R. Birkett a*, Alberto E. Parma a*, Roger Gerke-Bonet b, Robert Woodward a and Keith Gull ~'b a Biological Laboratory, University of Kent at Canterbury, Canterbury, Kent, CT2 7NJ (U.K.); and b Department of Biochemistry and Molecular Biology. School of Biological Sciences, University of Manchester, Medical School, Manchester M13 9PT (U.K.)
Received by J.R. Kinghorn: 23 May 1991 Revised/Accepted: 14 August/20 August 1991 Received at publishers: 30 October 1991
SUMMARY
We have adapted a group of well-known procedures in order to devise a simple method that allows the isolation of specific cDNAs encoding proteins located in different regions of the Trypanosoma brucei cytoskeleton, cDNA clones were isolated by screening a Agtl 1 expression library with a polyspecific, polyclonal antiserum against a complex immunogen, in this case the complete cytoskeleton. The fusion proteins produced by the clones were then used as an affinity immunoadsorbant to select monospecific polyclonals. The monospecific antisera isolated were used as probes to identify and localize different cytoskeleton proteins by Western blotting and immunofluorescence. This method proved particularly useful for the molecular identification of minor components in a complex structure. It should prove applicable to the molecular analysis of other organelles or protein complexes.
INTRODUCTION
Most eukaryotic cells adopt different configurations and shapes during their life cycle involving the rearrangement of many subcellular components. Modulation of the shape and form of eukaryotic cells is defined by a complex internal array of filamentous structures known as the cytoskel~toti. The c~ ~oskeleton ofthc protozoan 7", bruceiis a highly Correspondence to: Dr. K. Gull, Manchester University, Department of Biochemistry and Molecular Biology, Stopford Bid., Oxford Rd., Manchester M13 9PT (U.K.) Tel. (44-61)275-5108; Fax (44-61)275-5082. * Present addresses: (C.R.B.) AFRC Institute for Animal Health, Compton Laboratory, Compton, Newbury RGI6 0NN (U.K.) Tel. (44-635)578-411; (A.E.P.) Santos Vega 224, (7000) Tandil (Argentina). Abbreviations: bp, base pair(s); cDNA, DNA complementary to RNA; EGTA, ethylene glycol-bis(p-amino-ethyl ether)-N,N,N',N'-tetra-acetic
organized array of microtubules and other filaments. It includes subpellicular microtubules, microtubule based flagellar axoneme, basal bodies, the paraflagellar rod (PFR) and flagellar attachment zones (Russell etal., 1983; Bramblett et al., 1987; Sherwin et al., 1987; Seebeck et al., 1988; Gallo and Precigout, 1988; Sherwin and Gull, 1989; Woods et al., 1989a). The dominant structure is a complex corset of subpellicular microtubules which underlies the acid; FBS, fetal bovine serum; IPTG, isopropyl-~-D-thiogalactopyranoside; kb, kilobase(s) or 1000 bp; nt, nucleotide(s); PA, polyacrylamide; PAGE, PA-gel electrophoresis; PBS, phosphate-buffered saline, 140 mM NaCI/20 mM KCI/KH2PO4/K~HPO4 pH 7.4; PFR, paraflagellax rod; pfu, plaque-forming unit(s); PMSF, phenylmethylsulfonyl fluoride; PNME buffer, 10raM Pipes/140mM NaCI/2mM MgCI2/l mM EGTA/0.5% Triton X-100 pH 6.9; S., Staphylococcus; SCC, 0.15 M NaCI/0.015 M Na3.citrate pH 7.6; SDS, sodium dodecyl sulfate; T., Trypanosoma; TBS, 10 mM Tris' HCI/140 mM NaC! pH 7.4; TNTE, 10 mM Tris. HCI/0.5 M NaCI/0.2% Triton X-100/1 mM NaN3/I mM EDTA pH 7.5.
66 plasma membrane and provides cell shape (Vickerman, 1985; Sherwin and Gull, 1989). These microtubules are cross-linked both to each other and to the plasma membrane. Many of the above features are known in great detail and very high quality descriptions exist of structures such as microtubule cross bridges and filamentous links between
A
organelles. However, there is a great paucity of information regarding the molecular identity of these important cytoskeletal components. Remodelling of the cytoskeleton during the major cellular transformations in the trypanosome life cycle will, undoubtedly, include modulations in these components. So far, we only possess molecular descriptions of the major filament forming proteins of the
B
kDa 205,..
~r~
§6,-
a
b
c
d
35"-
20"-
o a
b
Fig. 1. Characterization of anti-cytoskeleton antiserum. Polyspeeific anti-cytoskeleton antiserum was characterized by SDS-PAGE and Western blotting (panel A) and by immunofluorescence microscopy (panel B). (Panel A)Coomassie blue-stained proteins after 0.1% SDS-10% PAGE of the immunogen (a) indicate heterogeneity on the protein composition, the tubulin doublet (50 and 52 kDa) and the PFR doublet (68 and 72 kDa) are particularly abundant. A corresponding Western blot probed with anti-cytoskeleton antiserum (b) showing that some of the minor components are highly immunogenic. (Panel B) Comparison oftrypanosomes stained with pre-immune rabbit serum (a and b) and anti-cytoskeleton antiserum (e and d). Methods. Preparation of cytoskeletons and antiserum: T. brucei brucei stock 427 was grown to log phase at 27 °C in medium SDM79 (Brun and SchOnenberger, 1979). The cells were harvested and resuspended in ice cold PNME buffer. The cells were incubated for 10 rain on ice and the cytoskeleton pellet collected by centrifugation at 1500 x g, for 5 rain at 4°C. The pellet was washed once in PNME buffer by resuspension and centrifugation. The cytoskeletons were treated with 1 mg/ml of DNaseI for 15 rain at 37 ° C. The pellet of washed and nuclease treated cytoskeletons was then salt extracted by resuspending and incubating for 15 rain in 10 mM Pipes/0.6 M NaCI/2 mM MgCI2/1 mM EGTA pH 6.9. The treated material was pelleted by centrifugation at 1500 x g for 5 rain at 4°C and washed with PNME. The cytoskeletons were resuspended in sterile 0.1 M Na. phosphate (pH 7.2) and emulsified with an equal volume of complete Freund's adjuvant. This emulsion (2 ml per animal) containing cytoskeletons from 1.5 x 108 parasites was injected subcutaneously into young New Zealand white rabbits. Booster injections were given subcutaneously on days 28, 56 and 150. Serum was collected 7-14 days after each booster injection.
67
,>
Fig. 2. Localisation of antigen by imullm,,lluoresceneemicroscopy. Affinity-purified immunoglobulins were used for indirect immunofluoreseenceagainst methanol-fixed trypanosomes. Phase-contrast (a,e,e,g) and immunofluoreseeneeimage (b,d,f,h) from cells treated with lgGs selected by ).gtl I lysate (a and b), clone ).5.5 (¢ and d), clone ~.5.12 (¢ and f) and clone ).5.20 (g and h). Methods. eDNA library screening: A ).gtl 1 eDNA library constructed with mRNA poly(A) + isolated from procy¢lic T. brucei was plated at a density of 160 pfu/cm2 or I x 104 pfu per 90-ram diameter plate. To clear the rabbit sera of naturally acquired anti-E, col~ antibodies an extract of induced Y1089r cultures was prepared by heat shock lysis of the pelleted ceils. The serum was pre-adsorbed by adding 2 mg/ml of the extract to the serum. Nitrocellulose filters containing the induced proteins were obtained following the pro¢edures described by Huynh et al. (1985). The remaining protein binding capacity was blocked by pre-incubating the filters in TBS containing 20% FBS. Primary polyspecific antiserum was diluted (1:50-1:500) with TB S/20 ~o FB S/2 mg per ml E. coli-Y 1089 extract/1 mM NaN3 and incubated with the filters overnight at 4°C. Positive clones were detected by employing biotinylated S. aureus Protein A and avidin-conjugated alkaline phosphatase. Immunofluoreseence of trypanosomes: Slides were prepared by immersion in poly-L-lysine solution (100 #g/ml) and then allowed to dry without further washing. The trypanosome cells from a culture containing 5 × 106 cells/ml were harvested by eentrifugation at 1000 × g for 3 rain at 22 ° C. The pellet was resuspended in PBS. The cell suspension was spread over the surface of the poly-L-lysine coated slides and allowed to settle for 5 rain. This was monitored with a light microscope until there was an even spread of single cells bound to the slide. The slides were then immersed in 3.8% formaldehyde for 5 rain followed by an immersion in cold methanol for a further 5 rain. The cells were rehydrated by immersion in PBS for 5 min. The first antibodies (selected monospecifie polyclonals) were applied for I h at 25°C in a moist chamber. The slides were then washed with PBS, three times for 10 rain each. The second antibody was diluted 1:10 in PBS and applied under the conditions described for the first antibody and washed in PBS three times for 10 rain each.
b e c k et al., 1983; G a U o a n d Schrevel, 1985; S c h l a e p p i et al.,
o t h e r c o m p o n e n t s is n o t only h i n d e r e d by t h e very small a m o u n t s available, b u t also by the n a t u r e o f t h e proteins
1989). D i r e c t b i o c h e m i c a l identification a n d analysis o f t h e
w h i c h are often i n s o l u b l e a n d p r o n e to aggregation during
c y t o s k e l e t o n ; g-tubulin,/1-tubulin a n d P F R proteins (See-
68 purification. This has restricted the development of polyclonal antibody probes raised to individual proteins. We have recently reported an alternative approach where we have isolated a library of monoclonal antibodies by using cytoskeletons as a complex immunogen. This proved very useful for identifying protein constituents and providing cytological probes for structures such as the basal body, the flagellar attachment zone, nucleus and paraflagellar rod (Woods et al., 1989a,b). The approach was successful in many respects but monoclonal antibodies have limitations as probes for e D N A libraries since they recognize single epitopes and often recognise an epitope that is the product of post translational modifications such as the acetylation of Lys 4° in 0c-tubulin (Schneider et al., 1987). Therefore, in this paper we report a procedure whereby we prepared a complex T. brucei polyspecific, polyclonal antiserum to complete cytoskeletons and used the antibodies generated as polyspecific probes to screen a ;tgtl 1 library. Random positive clones were selected and used as immunoadsorbants to isolate monospecific sera that allowed recognition of the protein encoded by each specific e D N A and its position within the cytoskeleton. The aim of the present study was to isolate a number of e D N A clones containing sequences that code for cryptic cytoskeletal proteins and to emphasise the general applicability of the described approach to the molecular analysis of complex structures.
positive clones representing poorly immunogenic proteins. From the positive clones obtained after secondary and tertiary screenings twelve were randomly selected as a working group. These clones were used to purify monospecific
kDa 205- i
•
116" 845845"
A 25"
i~i~, ~
RESULTS AND DISCUSSION
a (a) Characterization of polyclonal antisera A number of polyspecific anti-cytoskeleton antisera raised against 1". brucei cytoskeletons in individual rabbits had similar, but not identical, reactive profiles when analyzed by Western blotting (Towbin et al., 1979; Burnette, 1981), against the immunogen protein mixture. A consistent feature of such blots is that often minor proteins in the cytoskeleton complex were highly immunogenic whilst the abundant cytoskeletal tubulins were, by comparison, rather poor immunogens (Fig. 1A). This effect is apparent with the PFR components and many of the high molecular weight proteins. When the same antiserum was used for immunofluorescence studies it was revealed that the flagellum, the cell body and the residual nuclear envelope were major targets for the anti-cytoskeleton antiserum. (b) Cytoskeleton clones and monospecific polyclonal isolation The unfractionated and pre-adsorbed anti-cytoskeleton antiserum was used at relatively low dilutions of 1:501:500 during screening in order to maximize the number of
r~
5.5
5.12
5.15
5.20
Fig. 3. Proteins of T. bruceicytoskeleton were size fractionated by 0.1 SDS-10% PAGE, electroblottedonto nitrocellulosesupports and probed with total anti-cytoskeletonantiserum (a) or the affinityselected fractions obtained with the ~5.5, ,1.5.12,~5.15 and ~$.20 fusion proteins. Monaspecific polyclonais selected with clone ).5.5 revealed a band of approx. 120 kDa. The IGgs selected by clones ).5.12 and ).5.15 showed a differential bindingto the PFR proteins. The antibodies selected by clone ).5.20 revealed a doublet of bands of 165 kDa and 180 kDa, respectively.The arrowheads indicate the bands obtained with each monospecific polyclonal. Methods. Fusion-protein affinity selection: To obtain monospecific polyclonalsfrom isolated clones a lawn of E. coli Y 1090r was inoculated with enough phage to produce a confluentplate. Just as the plaques first appeared the expression of the fusion protein was indaced by overlaying with a dry nitrocellulosefilter previouslysoaked in 10 mM IPTG. The plates were incubated overnightat 42 °C and the filters pre-incubated in TBS/20% FBS for 30 rain. Polyspecificantiserum was diluted 1:101:50 with TBS/20% FBS/2mg per ml ).gtll-lysogen extract/0.5mM PMSF/I mM NaN~ and incubated with the membrane containing the fusion protein for 1 h at room temperature. Non-specific antibodies were removed by a series of washes (5 x 5 rain) with TNTE, followed by 3 x 5 min 2 M urea/0.1 M glycine/l% Triton X-100 washes with a final series of 3 × 5 rain washes with 50 mM Na. acetate (pH 5.0). The specific antibodyfractionwas then eluted with 3 ml per filterof 0.2 M glycine•HC! (pH 2.0) which was neutralizedwith 1 M "Iris.The eluted antibodieswere dialysed against TBS and concentrated by ultrafiltration.
69 polyclonals from the polyvalent anti-cytoskeleton serum as described in Fig. 3 legend. The different clones selected yielded enough immunoglobulin to perform indirect immunofluorescence assays and Western blots. The pattern of immunofluorescence varied from clone to clone and in some cases specific structures could be identified (Fig. 2). Whilst the use of affinity selected antibodies gave us the location of individual cytoskeleton proteins we were able to use Western blotting to establish the individual biocht~mical identity of gene products locating to the same cytoskeletal area. We have chosen to illustrate this procedure with four clones shown in Figs. 2 and 3. Two ofthese positive clones, ).5.12 and ).5.15, selected immunoglobulins that detected the paraflagellar rod in immunofluorescence and both of the paraltagellar rod proteins (PFRI and PFR2) by Westem blotting (Figs. 2 and 3). Interestingly, ).5.12 selected a monospecific antiserum that gave the same level of reaction with both PFR1 and PFR2 proteins, whilst clone ).5.15 selected an immunoglobulin fraction that consistently gave a better reaction with PFR2 than wifi] PFR1 (Fig. 3). The small amount of polyspecific antibody obtained by this
I~XLm X
m m
X
X
X
Ill m III
method may explain the ability to see differential binding specificity. Other clones, such as ).5.5 affinity selected an immunoglobulin fraction that recognized proteins in the cytoskeleton of the main cell body but when used for Western blotting recognized diffe~.!ent proteins. Clone ).5.5 selected antiserum when used in Western blotting reveals the gene product to be a polypeptide of around 120 kDa. Other clones recognized specific structures such as the flagellum, but with a rather different pattern to that described earlier for the ).5.12 and )~5.15. For instance, clone 3.5.20 antiserum reveals a rat~er punctate pattern along the flagellum and in Western b~lotting detects a doublet of proteins of around 165 and Ig0 kDa (Figs. 2 and 3). (c) Genomic anallysis of specific cDNA clones Genomic analysis ofclones ).5.5, ).5.12, ).5.15 and ).5.20 was performed by using the inserts as probes for nitrocellulose filters containing single and double digested 2". brucei gen,~,~,i~: DNA. Clone ).5.5 contains an insert of approx. 452 ~:~ ~nd when it is purified and used as a probe for a Southern blot it is clear that the homologous sequences in
mLX X X X m o . u o t.)
mX a .Xx m Q . (J 0
X
:)
t.)
n_ n w O X
kb 21.2J"
..o
t
e
e" f
W °
i
Q
5,h,.
4.~
I
4.2" 3.5,-
i
2.o..
i
1.9"
I
~
/
1.7D-
Q o.9..
Q
Q
O
O,8--
qlD
A
B
C
D
Fig. 4. Southern blots of singly and doubly digested T. brucei genomic DNA probed with the inserts isolated from clones ~.5.5 (panel A), ~.5.12 (panel B), 45.15 (panel C) and ~.5.20 (panel D). The digested genomic DNA was fractionated on a 1% agarose gel followed by transfer to nitrocellulose. The nitrocellulose filters containing the size-fractionated DNA were hybridized for 16 h in the presence of 50% formamide at 42°C as described in Sambrook et al. (1989). The inserts were labelled by random priming (Feinberg and Vogelstein, 1983) using [ ~)2P] dCTp (sp.act. > 3000 Ci/mmol) to a specific activity of 1-4 × l09 cpm/#g. The hybridized filters were washed 3 x 15 min with 0.1% SDS/0.1 x SSC at 65 ° C. Hybridization signals were detected by autoradiography at -80°C on KONICA X-ray film with an mtensifying screen. B, 8amHI; C, Clal; E, EcoRl; P, Pstl; Pv, Pvul; S, Sail; X, XbaI.
70 the trypanosome DNA are contained in single fragments when a variety of enzymes are used (Fig. 4A) indicating that the isolated clone may represent a single copy sequence. Clones ~.5.12 and ~.5.15 contain inserts of 388 bp and 2.2 kb, respectively. The purified inserts were used sequentially as probes for a Southern blot containing singly and doubly digested genomic DNA. The Southern blot patterns obtained were not identical and, taken with the above described Western blotting differences, may indicate some heterogeneity of the genes encoding PFR proteins within the T. brucei genome (Figs. 3, 4B and 4C). Clone 45.20 contains an insert of 1.8 kb, which produced a different pattern to the ones obtained with 45.12 and 45.15 when used as a probe for Southern-blot analysis. This confirms the protein data showing that although the ~.5.20 gene product locates to the flagellum it is distinct from the PFR proteins (Fig. 4D). This isolation and analysis of a range of eDNA probes to cytoskeletal proteins illustrates a simple approach to the identification of minor components of complex cellular structures. It is particularly applicable to cellular systems that lack the power of mutational genetics and show a high degree of structural organisation.
(d) Conclusions (1) We developed a screening system that allows the identification of clones encoding rare proteins of the T. brucei cytoskeleton. This method not only eliminates the need to purify a specific protein before raising antisera but also allows the isolation of cDNAs coding for a large range of proteins with differential immunogenicities. (2) The use of the polyspecific, pol:,clonai antisera eliminates the need to characterize and isolate specific antibodies in order to screen a eDNA library. (3) The system is sensitive enough that it allows the identification of cDNAs encoding closely related proteins. It also permits the physical localization of the protein under study to particular areas of the cell. (4) The method only requires that the target protein can be isolated as part of a complex mixture of proteins and should be a useful approach to the molecular analysis of complex organelles.
ACKNOWLEDGEMENTS
This work was funded by grant GB427057 CB from the Medical Research Council and received additional financial support from the United Nations Development Program/World Bank/World Health Organization special programme for Research and Training in Tropical Diseases. A.E. Parma held a fellowship from CONICET (Argentina).
REFERENCES Bramblett, G.T., Chang, S. and Flavin, M.: Periodic crosslinking of microtubules by cytoplasmic microtubule-associated and microtubulecorset proteins from a trypanosomatid. Prec. Natl. Acad. Sci. USA 84 (1987) 3259-3263. Brun, R. and SchOnenberger, M.: Cultivation and in vitro cloning of procyclic culture forms of Trypanosoraa brucei in a semi-defined medium. Acta Tropica 36 (1979) 289-292. Burnette, W.N.: Western blotting: electrophoretic transfer of proteins from sodium dodecyl sulfate polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated Protein A. Anal. Biochem. 112 (1981) 195-203. Feinberg, A.P. and Vogelstein, B.: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132 (1983) 6-13. Guile, J.M. and Precigout, E.: Tubulin expression in trypanosomes. Biol. Cell 64 (1988) 137-143. Gallo, J.M. and Schrevel, J.: Homologies between paraflagellar rod proteins from trypanosomes and euglenoids revealed by monoclonal antibody. Ear. J. Cell. Biol. 36 (1985) 163-168. Huynh, T.V., Joung, R.A. and Davies, R.W.: Construction and screening cDNA libraries in ,~gtl0 and ~gtl 1. In: GIover, D.M. (Ed.), DNA cloning. A practical approach. IRL Oxford, 1985, pp. 49-78. Russell, D.G., Newsam, R.J., Palmer, G.C.N. and Gull, K.: Structural and biochemical characterization of the paraflagellar rod in Trypanosoma brucei. Eur. J. Cell. Biol. 30 (1983) 137-143. Sambrook, 3., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. Sehlaeppi, K., Deflorin, 3. and Seebeck, T.: The major component of the paraflagellar rod of Trypanosoma brucei is a helical protein that is encoded by two identical, tandembly linked genes. J. Cell. Biol. 109 (1989) 1695-1709. Schneider, A., Sherwin, T., Sasse, R., Russell, D.G., Gull, K. and Seebeck, T.: Subpellicular and fiagellar microtubules of Trypanosoma bnlcei brucei coi~tain the same ~ tubulin isoforms. J. Cell Biol. 104 (1987) 431-438. Seebeck, T., Whittaker, P.A., Imboden, M.A., Hardman, N. and Braun, R.: Tubulin genes of Trypanosoma brucei: a tightly clustered family of alternating genes. Prec. Natl. Acad. Sci. USA 80 (1983) 4634-4638. Seebeck, T.H., Schneider, A., Kueng, V., Schlaeppi, K. and HemphiU, A.: The cytoskeleton of Trypanosoma brucei - the beauty of simplicity. Protoplasma 145 (1988) 188-194. Sherwin, T. and Gull, K.: Direct visualisation of detyrosination along single microtubules reveals novel mechanisms of microtubule assembly during cytoskeletal duplication in Trypanosoma brucei. Cell 57 (1989) 211-222. Sherwin, T., Schneider, A., Sasse, R., Seebeck, T. and Gull, K.: Distinct localization and cell cycle dependence of COOH terminally tyrosinated alpl~a tubulin in the microtubules of Trypanosomabrucei. J. Cell. Biol. 104 (1987) 439-446. Towbin, H., Staehelin, T. and Gordon, J.: Electrophoretie transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Prec. Natl. Acad. Sci. USA 76 (1979) 43504354. Woods, A., Sherwin, T., Sasse, R., MacRae~ T.H., Baines, A.J. and Gull, K.: Definition of individual components within the cytoskeleton of Trypanosoraa braceiby a library of monoclonal antibodies. J. Cell Sci. 93 (1989a) 491-500. Woods, A., Baines, A.J. and Gull, K.: Evidence for a Mr 88000 glycoprotein with a transmembrane association to a unique flagellum attachment region in Trypanosoma bracei. J. Cell Sci. 93 (1989b) 501508.
65
GENE 06246
Isolation of c D N A clones encoding proteins of complex structures: analysis of the Trypanosoma brucei cytoskeleton (Recombinant DNA; immunoscreening; antibodies; ~.gtl 1 vector; antisera; immunofluorescence)
Christopher R. Birkett a*, Alberto E. Parma a*, Roger Gerke-Bonet b, Robert Woodward a and Keith Gull ~'b a Biological Laboratory, University of Kent at Canterbury, Canterbury, Kent, CT2 7NJ (U.K.); and b Department of Biochemistry and Molecular Biology. School of Biological Sciences, University of Manchester, Medical School, Manchester M13 9PT (U.K.)
Received by J.R. Kinghorn: 23 May 1991 Revised/Accepted: 14 August/20 August 1991 Received at publishers: 30 October 1991
SUMMARY
We have adapted a group of well-known procedures in order to devise a simple method that allows the isolation of specific cDNAs encoding proteins located in different regions of the Trypanosoma brucei cytoskeleton, cDNA clones were isolated by screening a Agtl 1 expression library with a polyspecific, polyclonal antiserum against a complex immunogen, in this case the complete cytoskeleton. The fusion proteins produced by the clones were then used as an affinity immunoadsorbant to select monospecific polyclonals. The monospecific antisera isolated were used as probes to identify and localize different cytoskeleton proteins by Western blotting and immunofluorescence. This method proved particularly useful for the molecular identification of minor components in a complex structure. It should prove applicable to the molecular analysis of other organelles or protein complexes.
INTRODUCTION
Most eukaryotic cells adopt different configurations and shapes during their life cycle involving the rearrangement of many subcellular components. Modulation of the shape and form of eukaryotic cells is defined by a complex internal array of filamentous structures known as the cytoskel~toti. The c~ ~oskeleton ofthc protozoan 7", bruceiis a highly Correspondence to: Dr. K. Gull, Manchester University, Department of Biochemistry and Molecular Biology, Stopford Bid., Oxford Rd., Manchester M13 9PT (U.K.) Tel. (44-61)275-5108; Fax (44-61)275-5082. * Present addresses: (C.R.B.) AFRC Institute for Animal Health, Compton Laboratory, Compton, Newbury RGI6 0NN (U.K.) Tel. (44-635)578-411; (A.E.P.) Santos Vega 224, (7000) Tandil (Argentina). Abbreviations: bp, base pair(s); cDNA, DNA complementary to RNA; EGTA, ethylene glycol-bis(p-amino-ethyl ether)-N,N,N',N'-tetra-acetic
organized array of microtubules and other filaments. It includes subpellicular microtubules, microtubule based flagellar axoneme, basal bodies, the paraflagellar rod (PFR) and flagellar attachment zones (Russell etal., 1983; Bramblett et al., 1987; Sherwin et al., 1987; Seebeck et al., 1988; Gallo and Precigout, 1988; Sherwin and Gull, 1989; Woods et al., 1989a). The dominant structure is a complex corset of subpellicular microtubules which underlies the acid; FBS, fetal bovine serum; IPTG, isopropyl-~-D-thiogalactopyranoside; kb, kilobase(s) or 1000 bp; nt, nucleotide(s); PA, polyacrylamide; PAGE, PA-gel electrophoresis; PBS, phosphate-buffered saline, 140 mM NaCI/20 mM KCI/KH2PO4/K~HPO4 pH 7.4; PFR, paraflagellax rod; pfu, plaque-forming unit(s); PMSF, phenylmethylsulfonyl fluoride; PNME buffer, 10raM Pipes/140mM NaCI/2mM MgCI2/l mM EGTA/0.5% Triton X-100 pH 6.9; S., Staphylococcus; SCC, 0.15 M NaCI/0.015 M Na3.citrate pH 7.6; SDS, sodium dodecyl sulfate; T., Trypanosoma; TBS, 10 mM Tris' HCI/140 mM NaC! pH 7.4; TNTE, 10 mM Tris. HCI/0.5 M NaCI/0.2% Triton X-100/1 mM NaN3/I mM EDTA pH 7.5.
66 plasma membrane and provides cell shape (Vickerman, 1985; Sherwin and Gull, 1989). These microtubules are cross-linked both to each other and to the plasma membrane. Many of the above features are known in great detail and very high quality descriptions exist of structures such as microtubule cross bridges and filamentous links between
A
organelles. However, there is a great paucity of information regarding the molecular identity of these important cytoskeletal components. Remodelling of the cytoskeleton during the major cellular transformations in the trypanosome life cycle will, undoubtedly, include modulations in these components. So far, we only possess molecular descriptions of the major filament forming proteins of the
B
kDa 205,..
~r~
§6,-
a
b
c
d
35"-
20"-
o a
b
Fig. 1. Characterization of anti-cytoskeleton antiserum. Polyspeeific anti-cytoskeleton antiserum was characterized by SDS-PAGE and Western blotting (panel A) and by immunofluorescence microscopy (panel B). (Panel A)Coomassie blue-stained proteins after 0.1% SDS-10% PAGE of the immunogen (a) indicate heterogeneity on the protein composition, the tubulin doublet (50 and 52 kDa) and the PFR doublet (68 and 72 kDa) are particularly abundant. A corresponding Western blot probed with anti-cytoskeleton antiserum (b) showing that some of the minor components are highly immunogenic. (Panel B) Comparison oftrypanosomes stained with pre-immune rabbit serum (a and b) and anti-cytoskeleton antiserum (e and d). Methods. Preparation of cytoskeletons and antiserum: T. brucei brucei stock 427 was grown to log phase at 27 °C in medium SDM79 (Brun and SchOnenberger, 1979). The cells were harvested and resuspended in ice cold PNME buffer. The cells were incubated for 10 rain on ice and the cytoskeleton pellet collected by centrifugation at 1500 x g, for 5 rain at 4°C. The pellet was washed once in PNME buffer by resuspension and centrifugation. The cytoskeletons were treated with 1 mg/ml of DNaseI for 15 rain at 37 ° C. The pellet of washed and nuclease treated cytoskeletons was then salt extracted by resuspending and incubating for 15 rain in 10 mM Pipes/0.6 M NaCI/2 mM MgCI2/1 mM EGTA pH 6.9. The treated material was pelleted by centrifugation at 1500 x g for 5 rain at 4°C and washed with PNME. The cytoskeletons were resuspended in sterile 0.1 M Na. phosphate (pH 7.2) and emulsified with an equal volume of complete Freund's adjuvant. This emulsion (2 ml per animal) containing cytoskeletons from 1.5 x 108 parasites was injected subcutaneously into young New Zealand white rabbits. Booster injections were given subcutaneously on days 28, 56 and 150. Serum was collected 7-14 days after each booster injection.
67
,>
Fig. 2. Localisation of antigen by imullm,,lluoresceneemicroscopy. Affinity-purified immunoglobulins were used for indirect immunofluoreseenceagainst methanol-fixed trypanosomes. Phase-contrast (a,e,e,g) and immunofluoreseeneeimage (b,d,f,h) from cells treated with lgGs selected by ).gtl I lysate (a and b), clone ).5.5 (¢ and d), clone ~.5.12 (¢ and f) and clone ).5.20 (g and h). Methods. eDNA library screening: A ).gtl 1 eDNA library constructed with mRNA poly(A) + isolated from procy¢lic T. brucei was plated at a density of 160 pfu/cm2 or I x 104 pfu per 90-ram diameter plate. To clear the rabbit sera of naturally acquired anti-E, col~ antibodies an extract of induced Y1089r cultures was prepared by heat shock lysis of the pelleted ceils. The serum was pre-adsorbed by adding 2 mg/ml of the extract to the serum. Nitrocellulose filters containing the induced proteins were obtained following the pro¢edures described by Huynh et al. (1985). The remaining protein binding capacity was blocked by pre-incubating the filters in TBS containing 20% FBS. Primary polyspecific antiserum was diluted (1:50-1:500) with TB S/20 ~o FB S/2 mg per ml E. coli-Y 1089 extract/1 mM NaN3 and incubated with the filters overnight at 4°C. Positive clones were detected by employing biotinylated S. aureus Protein A and avidin-conjugated alkaline phosphatase. Immunofluoreseence of trypanosomes: Slides were prepared by immersion in poly-L-lysine solution (100 #g/ml) and then allowed to dry without further washing. The trypanosome cells from a culture containing 5 × 106 cells/ml were harvested by eentrifugation at 1000 × g for 3 rain at 22 ° C. The pellet was resuspended in PBS. The cell suspension was spread over the surface of the poly-L-lysine coated slides and allowed to settle for 5 rain. This was monitored with a light microscope until there was an even spread of single cells bound to the slide. The slides were then immersed in 3.8% formaldehyde for 5 rain followed by an immersion in cold methanol for a further 5 rain. The cells were rehydrated by immersion in PBS for 5 min. The first antibodies (selected monospecifie polyclonals) were applied for I h at 25°C in a moist chamber. The slides were then washed with PBS, three times for 10 rain each. The second antibody was diluted 1:10 in PBS and applied under the conditions described for the first antibody and washed in PBS three times for 10 rain each.
b e c k et al., 1983; G a U o a n d Schrevel, 1985; S c h l a e p p i et al.,
o t h e r c o m p o n e n t s is n o t only h i n d e r e d by t h e very small a m o u n t s available, b u t also by the n a t u r e o f t h e proteins
1989). D i r e c t b i o c h e m i c a l identification a n d analysis o f t h e
w h i c h are often i n s o l u b l e a n d p r o n e to aggregation during
c y t o s k e l e t o n ; g-tubulin,/1-tubulin a n d P F R proteins (See-
68 purification. This has restricted the development of polyclonal antibody probes raised to individual proteins. We have recently reported an alternative approach where we have isolated a library of monoclonal antibodies by using cytoskeletons as a complex immunogen. This proved very useful for identifying protein constituents and providing cytological probes for structures such as the basal body, the flagellar attachment zone, nucleus and paraflagellar rod (Woods et al., 1989a,b). The approach was successful in many respects but monoclonal antibodies have limitations as probes for e D N A libraries since they recognize single epitopes and often recognise an epitope that is the product of post translational modifications such as the acetylation of Lys 4° in 0c-tubulin (Schneider et al., 1987). Therefore, in this paper we report a procedure whereby we prepared a complex T. brucei polyspecific, polyclonal antiserum to complete cytoskeletons and used the antibodies generated as polyspecific probes to screen a ;tgtl 1 library. Random positive clones were selected and used as immunoadsorbants to isolate monospecific sera that allowed recognition of the protein encoded by each specific e D N A and its position within the cytoskeleton. The aim of the present study was to isolate a number of e D N A clones containing sequences that code for cryptic cytoskeletal proteins and to emphasise the general applicability of the described approach to the molecular analysis of complex structures.
positive clones representing poorly immunogenic proteins. From the positive clones obtained after secondary and tertiary screenings twelve were randomly selected as a working group. These clones were used to purify monospecific
kDa 205- i
•
116" 845845"
A 25"
i~i~, ~
RESULTS AND DISCUSSION
a (a) Characterization of polyclonal antisera A number of polyspecific anti-cytoskeleton antisera raised against 1". brucei cytoskeletons in individual rabbits had similar, but not identical, reactive profiles when analyzed by Western blotting (Towbin et al., 1979; Burnette, 1981), against the immunogen protein mixture. A consistent feature of such blots is that often minor proteins in the cytoskeleton complex were highly immunogenic whilst the abundant cytoskeletal tubulins were, by comparison, rather poor immunogens (Fig. 1A). This effect is apparent with the PFR components and many of the high molecular weight proteins. When the same antiserum was used for immunofluorescence studies it was revealed that the flagellum, the cell body and the residual nuclear envelope were major targets for the anti-cytoskeleton antiserum. (b) Cytoskeleton clones and monospecific polyclonal isolation The unfractionated and pre-adsorbed anti-cytoskeleton antiserum was used at relatively low dilutions of 1:501:500 during screening in order to maximize the number of
r~
5.5
5.12
5.15
5.20
Fig. 3. Proteins of T. bruceicytoskeleton were size fractionated by 0.1 SDS-10% PAGE, electroblottedonto nitrocellulosesupports and probed with total anti-cytoskeletonantiserum (a) or the affinityselected fractions obtained with the ~5.5, ,1.5.12,~5.15 and ~$.20 fusion proteins. Monaspecific polyclonais selected with clone ).5.5 revealed a band of approx. 120 kDa. The IGgs selected by clones ).5.12 and ).5.15 showed a differential bindingto the PFR proteins. The antibodies selected by clone ).5.20 revealed a doublet of bands of 165 kDa and 180 kDa, respectively.The arrowheads indicate the bands obtained with each monospecific polyclonal. Methods. Fusion-protein affinity selection: To obtain monospecific polyclonalsfrom isolated clones a lawn of E. coli Y 1090r was inoculated with enough phage to produce a confluentplate. Just as the plaques first appeared the expression of the fusion protein was indaced by overlaying with a dry nitrocellulosefilter previouslysoaked in 10 mM IPTG. The plates were incubated overnightat 42 °C and the filters pre-incubated in TBS/20% FBS for 30 rain. Polyspecificantiserum was diluted 1:101:50 with TBS/20% FBS/2mg per ml ).gtll-lysogen extract/0.5mM PMSF/I mM NaN~ and incubated with the membrane containing the fusion protein for 1 h at room temperature. Non-specific antibodies were removed by a series of washes (5 x 5 rain) with TNTE, followed by 3 x 5 min 2 M urea/0.1 M glycine/l% Triton X-100 washes with a final series of 3 × 5 rain washes with 50 mM Na. acetate (pH 5.0). The specific antibodyfractionwas then eluted with 3 ml per filterof 0.2 M glycine•HC! (pH 2.0) which was neutralizedwith 1 M "Iris.The eluted antibodieswere dialysed against TBS and concentrated by ultrafiltration.
69 polyclonals from the polyvalent anti-cytoskeleton serum as described in Fig. 3 legend. The different clones selected yielded enough immunoglobulin to perform indirect immunofluorescence assays and Western blots. The pattern of immunofluorescence varied from clone to clone and in some cases specific structures could be identified (Fig. 2). Whilst the use of affinity selected antibodies gave us the location of individual cytoskeleton proteins we were able to use Western blotting to establish the individual biocht~mical identity of gene products locating to the same cytoskeletal area. We have chosen to illustrate this procedure with four clones shown in Figs. 2 and 3. Two ofthese positive clones, ).5.12 and ).5.15, selected immunoglobulins that detected the paraflagellar rod in immunofluorescence and both of the paraltagellar rod proteins (PFRI and PFR2) by Westem blotting (Figs. 2 and 3). Interestingly, ).5.12 selected a monospecific antiserum that gave the same level of reaction with both PFR1 and PFR2 proteins, whilst clone ).5.15 selected an immunoglobulin fraction that consistently gave a better reaction with PFR2 than wifi] PFR1 (Fig. 3). The small amount of polyspecific antibody obtained by this
I~XLm X
m m
X
X
X
Ill m III
method may explain the ability to see differential binding specificity. Other clones, such as ).5.5 affinity selected an immunoglobulin fraction that recognized proteins in the cytoskeleton of the main cell body but when used for Western blotting recognized diffe~.!ent proteins. Clone ).5.5 selected antiserum when used in Western blotting reveals the gene product to be a polypeptide of around 120 kDa. Other clones recognized specific structures such as the flagellum, but with a rather different pattern to that described earlier for the ).5.12 and )~5.15. For instance, clone 3.5.20 antiserum reveals a rat~er punctate pattern along the flagellum and in Western b~lotting detects a doublet of proteins of around 165 and Ig0 kDa (Figs. 2 and 3). (c) Genomic anallysis of specific cDNA clones Genomic analysis ofclones ).5.5, ).5.12, ).5.15 and ).5.20 was performed by using the inserts as probes for nitrocellulose filters containing single and double digested 2". brucei gen,~,~,i~: DNA. Clone ).5.5 contains an insert of approx. 452 ~:~ ~nd when it is purified and used as a probe for a Southern blot it is clear that the homologous sequences in
mLX X X X m o . u o t.)
mX a .Xx m Q . (J 0
X
:)
t.)
n_ n w O X
kb 21.2J"
..o
t
e
e" f
W °
i
Q
5,h,.
4.~
I
4.2" 3.5,-
i
2.o..
i
1.9"
I
~
/
1.7D-
Q o.9..
Q
Q
O
O,8--
qlD
A
B
C
D
Fig. 4. Southern blots of singly and doubly digested T. brucei genomic DNA probed with the inserts isolated from clones ~.5.5 (panel A), ~.5.12 (panel B), 45.15 (panel C) and ~.5.20 (panel D). The digested genomic DNA was fractionated on a 1% agarose gel followed by transfer to nitrocellulose. The nitrocellulose filters containing the size-fractionated DNA were hybridized for 16 h in the presence of 50% formamide at 42°C as described in Sambrook et al. (1989). The inserts were labelled by random priming (Feinberg and Vogelstein, 1983) using [ ~)2P] dCTp (sp.act. > 3000 Ci/mmol) to a specific activity of 1-4 × l09 cpm/#g. The hybridized filters were washed 3 x 15 min with 0.1% SDS/0.1 x SSC at 65 ° C. Hybridization signals were detected by autoradiography at -80°C on KONICA X-ray film with an mtensifying screen. B, 8amHI; C, Clal; E, EcoRl; P, Pstl; Pv, Pvul; S, Sail; X, XbaI.
70 the trypanosome DNA are contained in single fragments when a variety of enzymes are used (Fig. 4A) indicating that the isolated clone may represent a single copy sequence. Clones ~.5.12 and ~.5.15 contain inserts of 388 bp and 2.2 kb, respectively. The purified inserts were used sequentially as probes for a Southern blot containing singly and doubly digested genomic DNA. The Southern blot patterns obtained were not identical and, taken with the above described Western blotting differences, may indicate some heterogeneity of the genes encoding PFR proteins within the T. brucei genome (Figs. 3, 4B and 4C). Clone 45.20 contains an insert of 1.8 kb, which produced a different pattern to the ones obtained with 45.12 and 45.15 when used as a probe for Southern-blot analysis. This confirms the protein data showing that although the ~.5.20 gene product locates to the flagellum it is distinct from the PFR proteins (Fig. 4D). This isolation and analysis of a range of eDNA probes to cytoskeletal proteins illustrates a simple approach to the identification of minor components of complex cellular structures. It is particularly applicable to cellular systems that lack the power of mutational genetics and show a high degree of structural organisation.
(d) Conclusions (1) We developed a screening system that allows the identification of clones encoding rare proteins of the T. brucei cytoskeleton. This method not only eliminates the need to purify a specific protein before raising antisera but also allows the isolation of cDNAs coding for a large range of proteins with differential immunogenicities. (2) The use of the polyspecific, pol:,clonai antisera eliminates the need to characterize and isolate specific antibodies in order to screen a eDNA library. (3) The system is sensitive enough that it allows the identification of cDNAs encoding closely related proteins. It also permits the physical localization of the protein under study to particular areas of the cell. (4) The method only requires that the target protein can be isolated as part of a complex mixture of proteins and should be a useful approach to the molecular analysis of complex organelles.
ACKNOWLEDGEMENTS
This work was funded by grant GB427057 CB from the Medical Research Council and received additional financial support from the United Nations Development Program/World Bank/World Health Organization special programme for Research and Training in Tropical Diseases. A.E. Parma held a fellowship from CONICET (Argentina).
REFERENCES Bramblett, G.T., Chang, S. and Flavin, M.: Periodic crosslinking of microtubules by cytoplasmic microtubule-associated and microtubulecorset proteins from a trypanosomatid. Prec. Natl. Acad. Sci. USA 84 (1987) 3259-3263. Brun, R. and SchOnenberger, M.: Cultivation and in vitro cloning of procyclic culture forms of Trypanosoraa brucei in a semi-defined medium. Acta Tropica 36 (1979) 289-292. Burnette, W.N.: Western blotting: electrophoretic transfer of proteins from sodium dodecyl sulfate polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated Protein A. Anal. Biochem. 112 (1981) 195-203. Feinberg, A.P. and Vogelstein, B.: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132 (1983) 6-13. Guile, J.M. and Precigout, E.: Tubulin expression in trypanosomes. Biol. Cell 64 (1988) 137-143. Gallo, J.M. and Schrevel, J.: Homologies between paraflagellar rod proteins from trypanosomes and euglenoids revealed by monoclonal antibody. Ear. J. Cell. Biol. 36 (1985) 163-168. Huynh, T.V., Joung, R.A. and Davies, R.W.: Construction and screening cDNA libraries in ,~gtl0 and ~gtl 1. In: GIover, D.M. (Ed.), DNA cloning. A practical approach. IRL Oxford, 1985, pp. 49-78. Russell, D.G., Newsam, R.J., Palmer, G.C.N. and Gull, K.: Structural and biochemical characterization of the paraflagellar rod in Trypanosoma brucei. Eur. J. Cell. Biol. 30 (1983) 137-143. Sambrook, 3., Fritsch, E.F. and Maniatis, T.: Molecular Cloning. A Laboratory Manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. Sehlaeppi, K., Deflorin, 3. and Seebeck, T.: The major component of the paraflagellar rod of Trypanosoma brucei is a helical protein that is encoded by two identical, tandembly linked genes. J. Cell. Biol. 109 (1989) 1695-1709. Schneider, A., Sherwin, T., Sasse, R., Russell, D.G., Gull, K. and Seebeck, T.: Subpellicular and fiagellar microtubules of Trypanosoma bnlcei brucei coi~tain the same ~ tubulin isoforms. J. Cell Biol. 104 (1987) 431-438. Seebeck, T., Whittaker, P.A., Imboden, M.A., Hardman, N. and Braun, R.: Tubulin genes of Trypanosoma brucei: a tightly clustered family of alternating genes. Prec. Natl. Acad. Sci. USA 80 (1983) 4634-4638. Seebeck, T.H., Schneider, A., Kueng, V., Schlaeppi, K. and HemphiU, A.: The cytoskeleton of Trypanosoma brucei - the beauty of simplicity. Protoplasma 145 (1988) 188-194. Sherwin, T. and Gull, K.: Direct visualisation of detyrosination along single microtubules reveals novel mechanisms of microtubule assembly during cytoskeletal duplication in Trypanosoma brucei. Cell 57 (1989) 211-222. Sherwin, T., Schneider, A., Sasse, R., Seebeck, T. and Gull, K.: Distinct localization and cell cycle dependence of COOH terminally tyrosinated alpl~a tubulin in the microtubules of Trypanosomabrucei. J. Cell. Biol. 104 (1987) 439-446. Towbin, H., Staehelin, T. and Gordon, J.: Electrophoretie transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Prec. Natl. Acad. Sci. USA 76 (1979) 43504354. Woods, A., Sherwin, T., Sasse, R., MacRae~ T.H., Baines, A.J. and Gull, K.: Definition of individual components within the cytoskeleton of Trypanosoraa braceiby a library of monoclonal antibodies. J. Cell Sci. 93 (1989a) 491-500. Woods, A., Baines, A.J. and Gull, K.: Evidence for a Mr 88000 glycoprotein with a transmembrane association to a unique flagellum attachment region in Trypanosoma bracei. J. Cell Sci. 93 (1989b) 501508.
