Download tubulin isotypes - Journal of Cell Science

J. Cell Sci. Suppl. 5, 243-255 (1986)
Printed in Great Britain © The Company o f Biologists Limited 1986
243
TUBULIN ISOTYPES: GENERATION OF DIVERSITY
IN CELLS AND MICROTUBULAR ORGANELLES
K . G U L L 1, P. J . H U S S E Y 1, R . S A S S E 1, A . S C H N E I D E R 2,
T . S E E B E C K 2 a n d T . S H E R W IN 1
1Biological Laboratory, University o f Kent, Canterbury, Kent, U K
2Institut fu r Allgemeine Mikrobiologie, Universität Bern, Baltzerstrasse 4,
CH-3012 Bern, Sw itzerland
SUMMARY
Diversity of tubulin isotypes is illustrated by consideration of the /3-tubulin isotypes of higher
plants and the eukaryotic microbe, Physarum polycephalum, and by the a-tubulin isotypes of the
protozoan, Trypanosoma brucei. The carrot plant expresses six, well-defined /5-tubulin isotypes
that possess characteristic two-dimensional gel coordinates. These six /3-tubulin isotypes are
differentially expressed during development of the flowering plant. In a similar manner, Physarum
expresses three separate /3-tubulin isotypes during its life cycle; of the two /31 isotypes, one is
expressed solely in the myxamoeba whilst the other is expressed both in the myxamoeba and in the
plasmodium. A further /8-tubulin isotype, /32, is expressed only in the plasmodium. In carrot and in
Physarum the generation of /3-tubulin diversity appears, in the main, to be generated by the
differential expression of a /3-tubulin multi-gene family. However, tubulin isotypes can also be
generated by post-translational modifications and T. brucei utilizes two different modifications
within one cell. First, the primary translation product, the a l -tubulin isotype, can be acetylated to
produce the a i isotype. Second, both the cx\ and oci isotypes appear to exist in both tyrosinated
and detyrosinated forms. The generation of these O'-tubulin isotypes within the same cell and their
presence in particular cellular domains, modulated throughout the cell cycle, reveals a complex
relationship between a-tubulin isotypes produced by post-translational modifications and the
dynamics of microtubule construction.
OV E R VI E W
Microtubules represent one of the most readily observed components of the
cytoskeleton and as such their structure and organization has been extensively
documented during the past 20 years using a variety of electron-microscopic
techniques. They have been described as components of the cytoplasmic architecture
of most cells as well as providing the major structural elements of mitotic and meiotic
spindles, cilia and flagella. They also occur in a host of other specialized arrange­
ments in cells as diverse as neurones and free-living flagellates. Despite the diversity
of these occurrences and structural arrangements it has become clear that the basic
biochemical unit used in their construction is a heterodimer of tubulin.
Early reports tended to emphasize the conserved nature of tubulin; however, the
introduction and application of more sophisticated techniques soon led to reports of
tubulin heterogeneity. These reports have now documented the existence of tubulin
multi-gene families in many organisms; however, the number of genes and their
arrangement in the genome varies considerably (Cowan & Dudley, 1983; Raff,
244
K. Gull and others
1984). In the yeast Schizosaccharomyces pombe two functional ar-tubulin genes have
been identified; the genes are dispersed within the genome, one having an intron and
one being intronless (Toda et al. 1984). In unicellular organisms there appear to be
no simple paradigms governing the number or arrangement of tubulin genes.
Chlamydomonas and Aspergillus both possess two a and two /3-tubulin genes
(Weatherbee & Morris, 1984), Physarum has at least four a and three /3-tubulin
D N A sequences (Schedl et al. 1984), whilst Naegleria has been reported to possess
eight cv-tubulin D N A sequences (Lai et al. 1984). In most of the above cases the
multi-tubulin gene sequences have been shown to be dispersed throughout the
organism’s genome. In a few cases, however, there is evidence of clustering of these
sequences within the genome. In trypanosomes the tubulin genes are arranged in
tightly packed clusters of tandemly repeated alternating a /¡3 pairs (Thomashow et al.
1983; Seebeck et al. 1983), whilst in Leishmania there is a cluster of tandemly
duplicated ar-tubulin genes and a completely separate cluster of tandemly repeated
/3-tubulin genes (Landfear et al. 1983).
In metazoan organisms there is also excellent evidence for the existence of tubulin
multi-gene families, although there is also some evidence for the presence of
pseudogenes in certain of the organisms studied (Lee et al. 1983). The multi-tubulin
D N A sequences observed in human, rat, mouse and chicken genomes are all
dispersed, whilst there is some clustering within the sea-urchin genome. The
molecular biology and genetics of these multi-tubulin gene families have been
excellently analysed in a recent review (Cleveland & Sullivan, 1985).
In many cases where multiple tubulin DNA sequences have been observed in an
organism’s genome it has subsequently been shown that at least some (in the case of
mammals), and more often many, of these D N A sequences do represent functional
tubulin genes. In some cases these genes can be shown to be different and yet to code
for the same tubulin polypeptide. For instance, in Chlamydomonas there are two
different /3-tubulin genes expressed, yet both genes encode an identical tubulin
polypeptide (Youngblom et al. 1984). In chicken there may be around seven to nine
different /3-tubulin genes; D N A sequencing and other techniques have shown that at
least five chicken jS-tubulin genes encode different authentic polypeptides (Cleveland
& Sullivan, 1985). In some cases a link has been made between individual tubulin
genes and particular tubulin isotypes that can be resolved and recognized by two­
dimensional gel electrophoresis. In Aspergillus nidulans there are two genes for
ar-tubulin, tub A and tubB, and two genes for /3-tubulin,, ben A and tubC. The tub A
gene codes for two proteins, a l -tubulin and cr3-tubulin, whilst the tubB gene codes
for <*2-tubulin. The benA gene codes for two /3-tubulin isotypes, /31-tubulin and
/32-tubulin, whilst the tubC gene encodes a third /3-tubulin isotype, /33-tubulin
(May et al. 1985). Similar relationships between an individual tubulin gene and a
particular, identifiable tubulin isotype have also been established in Drosophila (Raff
& Fuller, 1984). These relationships are particularly useful in permitting links to be
more easily established between investigations of the cell and molecular biology of
the tubulin multi-gene family.
Tubulin isotypes
245
In this paper we will discuss the initial evidence that suggests that higher plants
also contain multiple tubulin isotypes, and that these tubulin isotypes are differ­
entially expressed within different parts of the mature plant. We will then discuss the
generation and usage of different tubulin isotypes in two other organisms. First, the
generation of multiple /3-tubulin isotypes as the products of separate genes in
Physarum polycephalum, and second the generation of a--tubulin isotypes via two
separate post-translational modifications in Trypanosoma brucei.
H I G H E R P L A N T T U B U L I N S : AN E X A M P L E OF T H E D I F F E R E N T I A L
E X P R E S S I O N OF M U L T I P L E / 3 - T U B U L I N I S O T Y P E S
Considering the large amount of information pertaining to tubulin heterogeneity
in animals and eukaryotic microbes it is surprising that little detail exists regarding
the position in higher plants. We have initiated a study of this problem by using an
immunological approach to identify and characterize possible tubulin isotypes
amongst plant polypeptides separated by two-dimensional polyacrylamide gel elec­
trophoresis. In early experiments root tip cell lysates of Phaseolus vulgaris were
analysed by one-dimensional polyacrylamide gel electrophoresis (Hussey & Gull,
1985). The ar-tubulin and /3-tubulin polypeptides were then identified by Western
blotting of the separated polypeptides onto nitrocellulose paper and probing with a
panel of well-characterized monoclonal antibodies that recognize evolutionarily
conserved epitopes. These initial experiments showed that the plant ar-tubulin
migrated ahead of the plant /3-tubulin on Laemmli sodium dodecyl sulphatepolyacrylamide gels. This is the reverse migration to that characteristic of animal cell
tubulins, where the /8-tubulin is the faster-migrating species. This a//3 inversion of
the tubulins was first described in Physarum (Clayton et al. 1980) and has now been
reported in Dictyostelium, Paramecium, Tetrahymena and Crithidia. The two­
dimensional gel analysis of Phaseolus polypeptides revealed that both the a and
/3-tubulins of this organism could be separated into four discrete isotypes. We have
recently concentrated our studies of plant tubulins onDaucus carota, the carrot, and
have used monoclonal antibodies and immunoblotting protocols to reveal the
presence of a complex family of /3-tubulin isotypes within this plant. We have been
able to detect six well-defined /3-tubulin isotypes that are differentially expressed
within various parts of the carrot plant. The /35 isotype is present in the vegetative
phase of the flowering plant; it is expressed in the stem, the midrib and the leaf
lamina. However, there is a marked increase in its relative abundance from stem to
midrib, until it is found as the dominant /3-tubulin isotype in the leaf lamina. All of
the organs of the floret except pollen possess /SI, /32 and /33-tubulin isotypes.
However, in the stamen, where pollen is developing, a /34-tubulin isotype is detected
that increases in relative abundance in the mature stamen. In the pollen the /34tubulin is the major /3-tubulin isotype. The /36-tubulin isotype is only expressed in
seedlings.
There is, as yet, little detailed evidence available to link this complex expression of
/3-tubulin isotypes with the presence of a multi-tubulin family. However, preliminary
246
K. Gull and others
evidence from in vitro translation of mRNA preparations from various tissues and
organs of the plant suggests that the /3-tubulin heterogeneity is a reflection of the
presence of multiple /3-tubulin mRNA species, and Southern blot analysis using a
heterologous j3-tubulin cD N A probe has revealed the presence of multiple restriction
fragments.
G E N E R A T I O N OF 0 - T U B U L I N I S O T Y P E S AS T H E P R O D U C T S OF S E P A R A T E
G E N E S IN P. P O L Y C E P H A L U M
The slime mould P. polycephalum expresses two distinct isotypes of /3-tubulin,
separable by two-dimensional gel electrophoresis. These two electrophoretically
separable /3-tubulin isotypes, /3l-tubulin and /32-tubulin, are differentially expressed
within the life cycle of this eukaryotic microbe. The ¡31-tubulin is the only /3-tubulin
isotype expressed within the myxamoeba, whilst the macroscopic, syncytial plas­
modium expresses both the /31-tubulin and the /32-tubulin isotypes. The tubulins
from Physarum have been characterized as components of microtubules purified
from both the myxamoeba and plasmodium by cycles of assembly/disassembly
in vitro. Also, they have been extensively characterized by peptide mapping, reaction
with well-characterized monoclonal antibodies and by in vitro translations of specific
m RNAs selected by hybridization to cloned tubulin D N A sequences (Burland et al.
1983; Roobol et al. 1984).
Recently, Schedl et al. (1984) have provided an understanding of the number and
arrangement of /3-tubulin D N A sequences within the Physarum genome. These
workers were able to use restriction-fragment length polymorphisms to analyse the
multiple D N A sequences detected on Southern blots of Physarum D NA using
heterologous /3-tubulin-specific DNA probes. This mapping approach has revealed
the presence of at least three /3-tubulin loci, bet A , betB and betC] no linkage was
detected between any of these three loci. What relationship exists between these
three /3-tubulin loci and the two electrophoretically defined /3-tubulin isotypes, /31
and /32-tubulin?
The ¡31-tubulin isotype
The /31-tubulin is expressed in both myxamoebae and the plasmodium; however, a
variety of types of evidence suggests that this apparently simple pattern of expression
belies a more complex regulatory phenomenon. We have previously shown that
neither growth of Physarum myxamoebae nor the in vitro assembly of Physarum
myxamoebal tubulin is sensitive to the classical anti-microtubule agent, colchicine.
However, both growth of the cells and polymerization of Physarum tubulin are
extremely sensitive to inhibition by members of the benzimidazole carbamate group
of drugs (Quinlan et al. 1981). Burland et al. (1984) have isolated and characterized a
number of mutants that show resistance to the anti-microtubule fungicide MBC
(methyl benzimidazole-2-yl-carbamate). Analogy with other systems suggested that
at least some of these mutants would occur within the structural genes for /3-tubulin.
Burland and his colleagues were able to define two unlinked resistance loci, benA and
Tubulin isotypes
247
benD, that appeared to cosegregate with some of the previously mentioned ¡3-tubulin
restriction fragments in meiotic analysis. Mutations in ben A cosegregate with the
bet A locus and mutations in benD cosegregate with the betB locus. Thus, the general
conclusion of this study is the suggestion that ben A and benD each define a different
/3-tubulin structural gene. More direct evidence to confirm this proposal for benD
has come from the observation of Burland et al. that one mutation in benD,
benDUQ, results in the production of a novel, electrophoretically altered /3-tubulin
polypeptide in the myxamoebae. Myxamoebae carrying the 6ewD210 mutation also
express a normal wild-type /81-tubulin isotype (presumably the product of the benA
gene). The assumption from this work is that there are likely to be two /3-tubulin
genes that are expressed in the Physarum myxamoeba and that they encode indi­
vidual polypeptides that are either identical or are so similar that they colocalize in
the same two-dimensional gel spot. Confirmation of the second of these alternatives
comes from the actual amino acid sequence of the myxamoebal )3-tubulins. Physarum
myxamoebal tubulin has been purified in this laboratory (Clayton, Dawson & Gull)
and the amino acid sequence of both the a and /? subunits has been determined by
Singhofer-Wowra and Little in Heidelberg. During the sequencing of the /3-tubulin a
very clear heterogeneity was detected at position number 283 in the polypeptide
chain. The heterogeneity involved the detection of a double signal (alanine and
serine) at this position. No other heterogeneities were detected during the se­
quencing of nearly the entire polypeptide chain, so providing direct evidence for the
normal expression of two very similar /3-tubulin isotypes in the Physarum myx­
amoeba and that these two isotypes are the products of different genes, presumably
benA and benD.
The benA and benD genes exhibit differential expression during the life cycle of
Physarum. The altered, electrophoretically novel ¿>ewD210 /3-tubulin can be easily
located on two-dimensional gels of polypeptides from the plasmodial form of this
mutant. However, in plasmodia of such mutants Burland et al. were able to show
that there was no wild-type /3l-tubulin isotype. Thus, the benD gene appears to be
expressed in both the myxamoeba and the plasmodium, whilst the benA gene
appears to be expressed only in the myxamoebal stage of the life cycle. The initial
view of the Physarum /31-tubulin isotype was that it was expressed in both the
myxamoebae and the plasmodium. We can now see that a more complex scenario
actually pertains: there are two /31-tubulin isotypes expressed in the myxamoeba,
differing from each other in only one known amino acid and colocalizing to the same
two-dimensional gel spot. They are encoded by two separate genes, only one of
which is expressed in the plasmodium.
The ¡32-tubulin isotype
The /32-tubulin isotype is expressed only in the plasmodium. This /32-tubulin
isotype is found when RNA extracted from the plasmodium is translated in vitro and,
thus, the assumption is that this /32-tubulin isotype is also a direct gene product,
presumably from the betC locus. Since this /3-tubulin gene(s) is expressed only in the
plasmodium (a syncytium) it is not amenable to study by mutant selection. However,
248
K. Gull and others
a3
i
m
cel
1
Fig. 1. The tubulin isotypes of T. brucei separated by two-dimensional gel electro­
phoresis and detected by immunoblotting using monoclonal antibodies to a and /3-tubulin
(a mixture of DM 1A and K M X ). The blot shows the relationship between the a 1, ah and
/3-tubulin species.
when this gene is cloned it should be immediately recognizable by virtue of the
unique nature of the region encoding the C-terminal sequence of the /32 polypeptide
(G ull et al. 1985; Birkett et al. 1985).
G E N E R A T I O N O F a r - T U B U L I N I S O T Y P E S AS T H E P R O D U C T S OF P O S T - T R A N S ­
L A T I O N A L M O D I F I C A T I O N S I N T. B R U C E I
T he T. brucei genome contains a w’ell-characterized tubulin multi-gene family of
around 10 or and 10 /3 genes that comprise a clustered array of alternating a and ¡5
genes (Im boden et al. 1986). There is no evidence for the presence of heterogeneity
within these gene clusters that might result in the production of multiple tubulin
isotypes. However, our studies of the tubulin polypeptides of this organism clearly
indicate that it does use two post-translational modification mechanisms in order to
generate distinct isotypes of or-tubulin.
Acetylated oc-tubulin
T he effect of the first of these two post-translational modification systems is seen
when total protein from procyclic forms of trypanosomes is analysed by two­
dimensional gel electrophoresis and immunoblotted with well-characterized anti-artubulin monoclonal antibodies. T h is procedure detects the presence of two clearly
separated O'-tubulin isotypes (F ig. 1). In accordance with the nomenclature applied
previously to the tubulins of Chlamydomonas, we have termed the more basic of the
two proteins, al-tu bu lin , and the apparently higher molecular weight, less-basic
isotype, ar3-tubulin. A large quantitative difference is seen between the two
or-tubulin isotypes, the a3 isotype being the most abundant. In order to assess the
probable relationship between these two isotypes we have selected ar-tubulin m RN A
by preparative hybridization and have analysed this ar-tubulin m RN A by translation
Tubulin isotypes
r
249
-■>
S*Owr«~x
\
*
I
2
jt
Fig. 2. Transmission electron micrograph (TEM ) showing the cytoskeletal, sub-pellicu­
lar microtubules of T. brucei. The structure in the flagellum near to the axoneme is the
paraflagellar rod.
in vitro. Two-dimensional gel analysis of the translation products revealed only a
single ar-tubulin electrophoretic species, the ad-tubulin isotype. T hus, arl-tubulin
presum ably represents the primary transcription product, whilst a^-tubulin is most
probably a modified derivative of arl-tubulin. We have now been able to show that
this is indeed the case and that the post-translational modification that produces the
a^-tubulin isotype is an acétylation. T he same post-translational modification has
been described previously in Chlamydomonas by Rosenbaum’s group (M cKeithan
et al. 1983; L ’Hernault & Rosenbaum, 1983, 1985).
Trypanosom e cells contain a precisely arranged microtubule cytoskeleton and we
have determined the distribution of the a l and tv3-tubulin isotypes within the
microtubular organelles of the interphase cell. T he cytoplasmic pool of soluble
tubulin contains almost exclusively the a \-tubulin isotype. In contrast, the mem­
brane associated sub-pellicular microtubules (Figs 2, 3) contain both isotypes,
a 3 -tubulin being the major isotype; whilst the flagellar axonemal microtubules are
almost completely devoid of the a l -tubulin isotype. T his pattern of generation
of a distinct ar-tubulin isotype via acétylation, a post-translational modification,
is very similar to the patterns discovered in other eukaryotic microbes such
as Chlamydomonas reinhardtii, Polytomella agilis (M cKeithan et al. 1983) and
Crithidia fasciculata (Russell et al. 1984; Russell & Gull, 1984). The role of the
acetylated <x3-tubulin isotype is still rather unclear. The post-translational modifi­
cation is reversible in that the modification occurs either just before or just after
the tubulin is deposited in the flagellar axonemal microtubules ( L ’Hernault &
K. Gull and others
250
Rosenbaum , 1983; Russell & G ull, 1984) and is deacetylated upon resorption of the
flagellum ( L ’Hernault & Rosenbaum, 1985). Our studies with T. brucei, together
with those recently reported by Piperno & Fuller (1985), argue that acétylation is not
3A
I
I;* 4 *
B
Fig. 3. T EM of negatively stained cytoskeletons of T. brucei showing a cell at an early
stage in the cell cycle (A), and one at a later stage with one mature and one daughter
flagellum (B). T he arrangement of the sub-pellicular microtubules is seen in both cells.
Tubulin isotypes
251
a specific marker for flagellar tubulin. Rather, the production of tubulin isotypes by
acetylation may provide a marker for stable microtubules. This is suggested by the
very different ratios of art to « 3 -tubulin in the various trypanosomal microtubule
types. The cytoplasmic pool of soluble tubulin contains very little or no a3-tubulin
isotype, while the very stable microtubules of the flagellum axoneme contain almost
exclusively ar3-tubulin. The microtubules of the membrane-associated sub-pellicular
array, which are of intermediate stability, are also intermediate in their content of
« 3 -tubulin. It is clear that possession of the acetylated O'-tubulin is not restricted to
the microtubules of the flagellum and so appears not to be linked with the production
of doublet (axonemal) or triplet (basal body) microtubules. The suggestion of a link
with the more stable microtubules in a cell involves an experimental, operational
definition of microtubule stability (resistance to drugs, etc.). There is, at present, no
direct evidence to suggest a causal relationship between acetylation of O-tubulin and
the production of stable microtubules. The true function of a-tubulin acetylation in
the cell may well have nothing to do with inherent microtubule stability itself, but
may just correlate with this observed property of these particular subsets of micro­
tubules. However, it is clear that this particular modification occurs in many cells
and, moreover, has an intimate association with the dynamics of microtubule
polymerization in the cell.
Tyrosinated oc-tubulin
Barra et al. (1973) first reported the post-translational addition of tyrosine to a
brain protein that was subsequently identified as o-tubulin. This modification of
what is now known to be the C-terminus of ar-tubulin has been shown to occur in a
number of cells, but is not the primary post-translational event in this complex
scenario. Cloning and sequencing of otubulin genes and cDNAs has revealed that
most O'-tubulin polypeptides are translated with a tyrosine as their C-terminal amino
acid (Tyr-tubulin; Cleveland & Sullivan, 1985). In vivo the initial post-translational
modification is the removal of this tyrosine by a specific carboxy peptidase (Agarana
et al. 1978), so exposing the penultimate glutamic acid residue (Glu-tubulin).
It appears likely that this reaction occurs preferentially, whilst the ar-tubulin is
in a microtubule (Thompson, 1982; Kumar & Flavin, 1981). Such detyrosinated
O'-tubulin can then act as a substrate for a cytoplasmic tubulin tyrosine ligase, which
restores a tyrosine residue to the C-terminus of the a -tubulin polypeptide (Raybin &
Flavin, 1975, 1977; Thompson, 1982; Flavin & Murofushi, 1984). Recently, the
presence of this tt-tubulin modification cycle has been demonstrated in T. brucei
(Stieger et al. 1984) by in vivo labelling with [3H]tyrosine under conditions of
stringent inhibition of protein synthesis. We have recently extended this initial
observation by studying the pattern of radiolabelled products from such an experi­
ment using two-dimensional gel electrophoresis. In the presence of protein synthesis
inhibitors the incorporation of [3H]tyrosine into tubulin reaches a plateau after 2h.
When a lysate from cells that have been labelled under these conditions is analysed by
two-dimensional gel electrophoresis and fluorography, it is clear that radioactive
tyrosine is incorporated exclusively into O'-tubulin. Both oc\ and O'3-tubulins (see
252
K. Gull and others
above) are labelled. Knowledge of the distribution of these tubulin isotypes in
T. brucei cells and the results of various kinetics experiments suggest that it is the
a l -tubulin isotype that is the true substrate for the tubulin tyrosine ligase. The
soluble a \-tubulin then being incorporated into a cellular microtubule during the
course of the experiment (being acetylated in the process) and so appearing as the
a3 electromorph on two-dimensional gels. The specificity of the in vivo labelling
experiments for O'-tubulin is not a trivial result since the T. brucei /3-tubulin gene is
unusual in that it also encodes a tyrosine as the C-terminal amino acid of the
/3-tubulin polypeptide. Thus, the absence of labelling of the /3-tubulin during these
in vivo experiments shows that the tubulin tyrosine ligase-catalysed modification in
T. brucei is, as with other organisms, restricted to or-tubulin. Unlike the acetylation
described earlier, the detyrosination-tyrosination cycle does not result in shifts of
two-dimensional gel coordinates that correlate with precursor-product species.
However, it is clear that this a-tubulin terminal tyrosine cycle does operate within
the T. brucei cell to produce or-tubulin isotypes with distinct cellular localizations.
Previous work with polyclonal antibodies raised against synthetic peptides rep­
resenting the tyrosinated and detyrosinated C-terminus of cv-tubulin have led to the
general conclusion that these two ar-tubulin isotypes may be differentially distributed
amongst the individual microtubules of interphase and mitotic cells (Gundersen
et al. 1984; Gundersen & Bulinski, 1986). In our studies we have used a monoclonal
antibody (Y L l/2 ) that is specific for tyrosinated tubulin (Kilmartin et al. 1982;
Wehland et al. 1984) and have used immunofluorescence microscopy to reveal the
changes in distribution of Tyr-tubulin during the T. brucei cell cycle. A distinct
advantage of using T. brucei in these studies is that the position of a particular
individual cell in the cell cycle can be estimated with reasonable accuracy, since there
are particular structural landmarks that occur with distinct cell cycle timings. These
landmarks include cell size and shape, the presence or absence of a daughter
flagellum, the length of any such flagellum (Fig. 3), the position of the basal bodies
and the position (and segregation) of nuclear and kinetoplast D NA as visualized by
the intercalating dye, DAPI (4'6,diamidino-2-phenylindole). When trypanosome
cells are viewed by immunofluorescence microscopy using an anti-/3-tubulin mono­
clonal antibody, or an anti-or-tubulin monoclonal antibody whose epitope is not
subject to post-translational modification, then the cell body is seen to be intensely
fluorescent due to the massive numbers and homogeneous distribution of the subpellicular micro tubules. The flagellum is seen as a wavy line attached to the side of
the cell body. However, immunofluorescent staining with the Y L l/2 antibody (Tyrtubulin-specific) reveals a completely different pattern that is modulated throughout
the cell cycle. The markers outlined above permit changes in the cell-cycle-related
staining to be deduced from populations of asynchronous cells. Cells at the start of
the cell cycle exhibit bright fluorescence at the posterior third of the cell; the
flagellum is not stained. As the cell cycle progresses, a short daughter flagellum forms
on the new basal body and this new flagellum stains very brightly with Y L l/2 . This
bright fluorescence of the daughter flagellum is maintained as it continues to
elongate, as is this generally brighter fluorescence of the posterior portion of the cell.
Tubulin isotypes
253
This pattern of fluorescence then changes, concomitant with separation of the
kinetoplast D N A (as seen by DAPI double staining) and the separation of basal
bodies. At this time the intensity of staining of the posterior third of the cell is
reduced and the very bright staining of the daughter flagellum is lost almost
completely. This point at which the daughter flagellum loses its ability to be
recognized by the Y L l/ 2 antibody correlates well with the point at which it has
grown to its full length. Interestingly, the basal bodies of the trypanosome cell stain
brightly with the Y L l/2 antibody at all times during the cell cycle.
The staining pattern of the T. brucei flagellum permits a modulation of the
tyrosination cycle to be observed in microtubules of known lineage and strongly
suggests that the tyrosinated state of a-tubulin is a marker of newly formed
microtubules. According to this view, the pool of soluble tubulin consists of the
primary translation products containing the C-terminal tyrosine coded for by the
mRNA, as well as the ‘recycled’ Glu-tubulin to which a new terminal tyrosine has
been added by the tubulin tyrosine ligase. Tyr-tubulin is the species that actually
participates in the polymerization into a microtubule, and once incorporated into the
polymer it can be detyrosinated by the action of the microtubule-associated tubulin
carboxy peptidase. The amount of Glu-tubulin that accumulates in a particular
microtubule is dependent upon the residence time of subunits in that microtubule
and such ‘old’ microtubules can be expected to possess an elevated proportion of
Glu-tubulin.
The actual function of this unique detyrosination-tyrosination cycle as well as the
O’-tubulin acetylation, described earlier, remains poorly understood. However, our
studies of these two post-translational modifications in T. brucei have revealed how
each modification can produce particular tubulin isotypes whose existence and
distribution are intimately linked to the formation of the precisely ordered micro­
tubule cytoskeleton of this organism. We feel that the cytoskeleton of this simple, yet
important, microorganism represents a highly suitable system for further investi­
gations of tubulin isotype diversity and function.
We thank John Kilmartin for generous gifts of monoclonal antibody.
The work described in this paper was supported by grants to K .G . from the Science and
Engineering Research Council, the Medical Research Council and the WHO/World Bank
Special Programme for Research and Training in Tropical Diseases. The carrot tubulin project
was supported by an SE R C CA SE award to K .G . and to D r C. W. Lloyd, Norwich.
REFERENCES
C. E ., B a r r a , H. S. & C a p u t t o , R. (1978). Release of [14C]tyrosine from tubulinyl[14C]-tyrosine by brain extract. Separation of a carboxy peptidase from tubulin tyrosine ligase.
Molec. Cell. Biochem. 19, 17-22.
B a r r a , H. S ., R o d r i q u e z , J . A ., A r c e , C . A . & C a p u t t o , R . (1973). A soluble preparation from
rat brain that incorporates into its own proteins [ 14C ] -arginine by a ribonuclease-sensitive system
and [ 14C ] -tyrosine by a ribonuclease-insensitive system. J . Neurochem. 20, 97-108.
BlRKETT, C. R., F o s t e r , K . E. & G u l l , K . (1985). Evolution and patterns of expression of the
Physarum multi-tubulin family analysed by the use of monoclonal antibodies. In Molecular
Genetics o f the Filamentous Fungi (ed. W. Timberlake). New York: A . R. Liss.
A garan a,
254
K. Gull and others
B u r l a n d , T . G ., G u l l , K ., S c h e d l , T ., B o ston , R. S . & D ove , W. F. (1983). Cell type-
dependent expression of tubulins in Physarum. J . Cell Biol. 97, 1852-1859.
(1984). Genetic analysis of resistance to
benzimidazoles in P hysarum : Differential expression of ft tubulin genes. Genetics 108, 123-141.
C layto n , L ., Q u in l a n , R . A, R o obol , A ., P o gso n , C. I. & G u l l , K . (1980). A comparison of
tubulins from mammalian brain and Physarum polycephalum using S D S polyacrylamide gel
electrophoresis and peptide mapping. F E B S Lett. 115, 301-305.
C l e v e l a n d , D. W. & S u lliv a n , K . F. (1985). Molecular biology and genetics of tubulin. A. Rev.
Biochem. 54, 331-365.
C o w an , N. J. & D u d l e y , L . (1983). Tubulin isotypes and the multigene tubulin families. Int.
R ev. Cytol. 85, 147-173.
F la v in , M . & M u r o fu sh i , H. (1984). Meth. Enzym . 106, 223-237.
G u l l , K ., B irkett , C. R., B l in d t , A. R ., D e e , J ., F oster , K . E. & P a u l , E. C. A. (1985).
Expression of a multi-tubulin family and the in vivo assembly of microtubular organelles in
Physarum polycephalum. In Microtubules and Microtubule Inhibitors (ed. M. De Brabander &
J . De Mey). Amsterdam: Elsevier.
G u n d e r se n , G . G . & B u l in sk i , J. C. (1986). Distribution of tyrosinated and nontyrosinated
a'-tubulin during mitosis, jf. Cell Biol. 102, 1118-1126.
G u n d e r se n , G . G ., K a l n o sk i , M. H. & B u l in sk i , J. C. (1984). Distinct populations of
mictrotubules: Tyrosinated and Nontyrosinated alpha tubulin are distributed differently
in vivo. Cell 38, 779—789.
H u s s e y , P. J. & G u l l , K . (1985). Multiple isotypes of a and /3 tubulin in the plant Phaseolus
vulgaris. F E B S L ett. 181, 113-118.
I m bo d en , M ., B l u m , B ., D e L a n g e , T ., B r a u n , R. & S e e b ec k , T . (1986). T u bu lin mRNAs of
Trypanosoma brucei. jf. molec. Biol. 188, 393-402.
KlLMARTIN, J . V., W r i g h t , B. & M i l s t e i n , C. (1982). Rat monoclonal antitubulin antibodies
derived using a new non-secreting rat cell line. jf. Cell Biol. 93, 576-582.
K u m a r , N. & F l a v i n , M. (1981). Preferential action of a brain detyrosinating carboxypeptidase on
polymerised tubulin, jf. biol. Chem. 256, 7678-7686.
L a i , E. Y ., R em ill a r d , S. P. & F u lto n , C. (1984). Tubulin and actin; Yin-Yang gene expression
during Naegleria differentiation. In Molecular Biology of the Cytoskeleton (ed. G. G. Borisy,
D . W. Cleveland & D . B. Murphy). New York: Cold Spring Harbor Laboratory Press.
L a n d fe a r , S. M ., M c M ahon -Pratt , D . & W irth , D. F. (1983). Tandem arrangement of
tubulin genes in the protozoan parasite Leishmania enriettii. Molec. Cell. Biol. 3 , 1070-1076.
L e e , M. G . S ., L e w is , S. A., Wil d e , C. D. & C ow an , N. J. (1983). Evolutionary history of a
multi-gene family: an expressed human tubulin gene and three processed pseudogenes. Cell
33, 477-487.
L ’H e r n a u l t , S . W. & R o sen ba u m , J. L . (1983). Chlamydomonas alpha tubulin is post-translationally modified in the flagella during flagellar assembly, jf. Cell Biol. 97, 256-263.
L ’H e r n a u l t , S. W. & R o sen ba u m , J. L . (1985). Chlamydomonas a'-tubulin is post-translationally
modified by acetylation on the e-amino group of a lysine. Biochemistry 24, 473-478.
M a y , G . S ., G a m bin o , J ., Wea th er bee , J. A. & M orris , N. R. (1985). Identification and
functional analysis of /3 tubulin genes by site specific integrative transformation in Aspergillus
nidulans. jf. Cell Biol. 101, 712-719.
M c K eith a n , T . W ., L efebv r e , P. A., S ilflo w , C. D. & R o sen ba u m , J. L . (1983). M ultiple
form s of tubulin in Polytomella and Chlamydomonas: evidence for a precursor of flagellar a
tu b u lin .^ . Cell Biol. 96, 1056-1063.
P ipern o , G . & F u l l e r , M. T . (1985). Monoclonal antibodies specific for an acetylated form of a
tubulin recognise the antigen in cilia and flagella from a variety of organisms. X Cell Biol. 101,
2085-2094.
Q u in l a n , R . A., R o obol , A., P ogson , C. I. & G u l l , K . (1981). A correlation between in vivo
and in vitro effects of the microtubule inhibitors colchicine, parbendazole and nocodazole on
myxamoebae of Physarum polycephalum. J. gen. Microbiol. 122, 1-6.
R a f f , E. C. (1984). Genetics of microtubule systems, jf. Cell Biol. 98, 1-10.
R a f f , E. C. & F u l l e r , M. T . (1984). Genetic analysis of microtubule function in Drosophila. In
Molecular Biology o f the Cytoskeleton (ed. G. G. Borisy, D. W. Cleveland & D . B. Murphy).
New York: Cold Spring Harbor Laboratory Press.
B u r la n d , T . G ., S c h e d l, T ., G u l l , K . & D o v e , W . F .
Tubulin isotypes
255
R a y b in , D . & F la v in , M . (1975). An enzyme tyrosylating alpha tubulin and its role in microtubule
assembly. Biochem. biophys. Res. Commun. 60, 1384-1390.
A ., W i l c o x , M ., P a u l , E. C. A. & G u l l , K . (1984). Identification of tubulin isoforms in
the plasmodium of Ph.ysa.rum polycephalum by in vitro microtubule assembly. Eur. J . Cell Biol.
33, 24-28.
R u s s e l l , D . G . & G u l l , K . (1984). Flagellar regeneration of the trypanosome Crithidia
fasciculata involves post-translational modification of cytoplasmic alpha tubulin. Molec. Cell.
Biol. 4, 1182-1185.
R u s s e l l , D . G . , M i l l e r , D . & G u l l , K . (1984). Tubulin heterogeneity in the trypanosome
Crithidia fasciculata. Molec. Cell. Biol. 4 , 779-790.
S c h e d l , T ., O w e n s , J ., D o v e , W . F. & B u r l a n d , T . G. (1984). Genetics of the tubulin gene
families of Physarum. Genetics 108, 143-164.
S e e b e c k , T ., W h i t t a c k e r , P. A ., I m o b o d e n , M., H a r d m a n , N. & B r a u n , R. (1983). Tubulin
genes of Trypanosoma brucei\ A tightly clustered family of alternating genes. Proc. natn. Acad.
Sci. 80, 4634-4638.
S t i e g e r , J., W y l e r , T . & S e e b e c k , T . (1984). Partial purification and characterisation of
microtubular protein from Trypanosoma brucei.J. biol. Chem. 259, 4596-4602.
T h o m a s h o w , L . S ., M i l h a u s e n , M ., R u t t e r , W. J. & A g a b i a n , N . (1983). Tubulin genes are
tandemly linked and clustered in the genome of Trypanosoma brucei. Cell 32, 35-43.
T h o m p s o n , W. C. (1982). The cyclic tyrosination/detyrosination of alpha tubulin. Meth. Cell Biol.
24, 235-255.
R o o b o l,
T o d a , T ., A d a ch i , Y ., H iraoka , Y. & Y an a gida , M . (1984). Identification of the pleiotropic cell
division gene NDA2 as one of the two different or tubulin genes in Schizosaccharomyces pombe.
Cell 37, 233-242.
J. A. & M o r r i s , N. R. (1984). Aspergillus contains multiple tubulin genes. J . biol.
Chem. 259, 15 452-15 459.
W e h l a n d , J., S c h r o d e r , H. C. & W e b e r , K . (1984). Amino acid sequence requirements in
the epitope recognised by the a'-tubulin specific monoclonal antibody Y L l/2 . EM BO J . 3 ,
1295-1300.
Y o u n g b l o m , H ., S c h l o s s , J. A. & S i l f l o w , C. D. (1984). The two /3 tubulin genes of
Chlamydomonas reinhardtii code for identical proteins. Molec. Cell. Biol. 4, 2686-2696.
W eath e rb ee,