Tubulin, the dimeric subunit of microtubules, is a major cell protein that is centrally involved in cell division. Tubulin is subject to specific enzymatic posttranslational modifications including cyclic tyrosine removal and addition at the COOH terminus of the α-subunit. Tubulin is normally extensively tyrosinated in cycling cells. However, we have previously shown that detyrosinated tubulin accumulates in cancer cells during tumor progression in nude mice. Tubulin detyrosination, resulting from suppression of tubulin tyrosine ligase and the resulting unbalanced activity of tubulin-carboxypeptidase, apparently represents a strong selective advantage for cancer cells. We have now analyzed the occurrence and significance of tubulin detyrosination in human breast tumors. We studied a total of 134 breast cancer tumors from patients with or without known complications over a follow-up period of 31 ± 10 months. The mean age of the patients at the time of diagnosis was 57 years. For each patient, detailed data concerning the histology and extension of the tumor were available. Tumor cells containing detyrosinated tubulin were visualized by immunohistochemical staining of paraffin-embedded tissue sections.

Cancer cells with detyrosinated tubulin were observed in 53% of the tumors and were predominant in 19.4% of the tumors. Tubulin detyrosination correlated to a high degree of significance (P < 0.001) with a high Scarf-Bloom-Richardson (SBR) grade, a known marker of tumor aggressiveness. Among SBR grade 1 tumors, 3.8% were strongly positive for tubulin detyrosination compared with 65.4% of the SBR grade 3 tumors. The SBR component showing the strongest correlation with tubulin detyrosination was the mitotic score. In the entire patient population, neither the SBR grade nor the detyrosination index had significant prognostic value (P = 0.11, P = 0.27, respectively), whereas a combined index was significantly correlated with the clinical outcome (P = 0.02). A preliminary subgroup analysis indicated that tubulin detyrosination may define high- and low- risk groups in breast cancer tumors with an SBR grade of 2. Our study shows that tubulin detyrosination is a frequent occurrence in breast cancer, easy to detect, and linked to tumor aggressiveness.

Microtubules are major cytoskeletal structures, centrally involved in the control and mechanics of cell division, being the principal components of the mitotic spindle in eukaryotic cells (1). The building block of microtubules is the αβ-tubulin heterodimer. Tubulin is subject to specific posttranslational modifications including a cycle of tyrosine removal and addition at the COOH terminus of the α subunit (2, 3). This cycle involves two enzymes, TTL4(4) and an ill-defined tubulin carboxypeptidase (5, 6), and generates two major forms of tubulin: tyrosinated tubulin (Tyr-tubulin) and Glu-tubulin. A third tubulin species (Δ2-tubulin) arises by removal of the COOH-terminal Glu residue from the α chain of Glu-tubulin (7, 8). Tyr-tubulin is the dominant tubulin species in cycling cells (9, 10). Glu-tubulin is abundant in neurons but can be present in stable microtubules of other cell types (10, 11, 12, 13, 14). Δ2-tubulin normally has high neuronal specificity (7, 8, 13, 15). However, we have previously observed an abnormal accumulation of Glu-tubulin and Δ2-tubulin in cancer cells of both fibroblastic and epithelial origin, during tumor growth in nude mice (15). This accumulation is attributable to TTL suppression and apparently represents a strong selective advantage for cancer cells (15). Here, we have tested whether accumulation of Glu- and Δ2-tubulin also occurred in human tumor cells and whether tubulin detyrosination was related to tumor severity.

Patient Population.

Entered in this study were tumor samples from 134 breast cancer patients treated at the Center Léon Bérard, between January 1996 and December 1998. The end point for clinical follow-up was December 31, 1999. All of the tumors were primary breast tumors. The study included tumors from all of the patients who had locoregional recurrence, metastasis, or death during the period of follow-up. Among these tumors, tumors <40 mm in size (n = 34) were paired with tumors of the same size from patients of the same age with no event. Tumors ≥40 mm in size (n = 18) could not be paired because of difficulties in finding subjects of the same age in the relatively small-sized population of patients with similar tumors and no event (n = 48). All of these 48 patients were included in the study. The median follow-up (from diagnosis of the tumor to December 1999) was 32.2 months (range, 12.2–48.2 months). All of the patients were seen on a regular basis, every 6 months. The principal patient characteristics are described in Table 1.

Tumor samples were collected at the Department of Pathology of the Center Léon Bérard. Tumors were pure histological variants of invasive breast carcinomas comprising 100 ductal, 15 lobular, and 19 other variants (including mucinous, tubular, and metaplastic). The mean largest diameter of collected tumors was 37.4 mm (range, 2–100 mm; Table 1). The histopathological type and grade of the carcinomas were evaluated according to WHO criteria and SBR grading, respectively (16). The SBR grade includes three histological parameters: tubular differentiation, nuclear pleiomorphism, and mitotic count.

Immunohistochemistry of Paraffin-embedded Tissue.

The cellular content in Glu-tubulin and in Δ2-tubulin was evaluated by immunohistochemical analysis, using specific antibodies. Tumor samples were Bouin-fixed within 2 h after surgical removal. Sections of 4 μm in thickness from paraffin-embedded blocks were deparaffinized in xylene and rehydrated in a decreasing ethanol series (100 to 50%). At this stage, tissue sections assigned to Δ2-tubulin immunohistochemistry were treated for antigen retrieval. Antigen retrieval involved treatment in a sodium citrate solution [10 mm (pH 6.0)] in a 700-W microwave oven, three times for 5 min each. The sections were left to cool down in the buffer at room temperature for about 20 min and then rinsed in PBS. All of the tissue sections were treated with 3% H2O2 for 15 min to exhaust endogenous peroxidase. The sections were then immersed in PBS containing 0.3% BSA-0.5% Triton for 6 min. After preincubation in 3% BSA in PBS for 30 min, sections were incubated overnight at 4°C with polyclonal Glu-tubulin (L3) or Δ2-tubulin (L7) antibodies. These antibodies were produced in our laboratory (13) and, beforehand, were affinity-purified against the corresponding heptapeptide. Affinity-purified L3 and L7 antibodies were used at dilutions of 1:8000 and 1:250, respectively. Sections were then rinsed 3 times in PBS-0.1% Tween and incubated, at room temperature, sequentially in biotinylated goat antirabbit secondary antibody at 1:300 dilution (30 min), and in avidin-biotin peroxidase complex (30 min; DAKO SA). Chromogenic development was obtained by adding a diaminobenzidine solution (kit DAKO SA) for 6 min. Finally, sections were slightly counterstained with Harris’ hematoxylin, dehydrated, mounted in Eukitt medium, and examined by light microscopy.

Statistics.

The association of variables was evaluated using the χ2 test, the Fisher exact test, variance analysis, or the log-rank test, depending of the nature of the variables and of the group size. The value of significance was taken as P < 0.05. Statistical analysis was carried out using Statistica software.

Analysis of Glu-tubulin in Cancer Cells.

Tissue sections stained with Glu-tubulin antibody were examined for presence of cells containing Glu-tubulin as described above. In normal areas of breast tissue sections, Glu-tubulin staining was detected in the cytoplasm of fibroblasts (Fig. 1,A) and of endothelial cells. The nuclei of myoepithelial cells were occasionally stained. This nuclear labeling may result from a cross-reaction with a nuclear protein in this cell type because tubulin is normally undetectable in cell nuclei. Most importantly, normal epithelial cells were never stained (Fig. 1,A). In contrast, Glu-tubulin was detected in transformed epithelial cells (Fig. 1, C and D). Stromal cells within the tumor were also stained. For the scoring of tumors, slides were microscopically evaluated by three observers (A. M., L. L., I. T.) without knowledge of either the prior histological results or the clinical outcome. The percentage of positively stained transformed epithelial cells in the infiltrating component were estimated by each investigator. Tumors were scored according to the percentage of positive cancer cells: grade 1, negative for Glu-tubulin staining (Fig. 1,B); grade 2, positive, with <50% of cancer cells positive for Glu tubulin (Fig. 1,C); and grade 3, strongly positive with ≥50% of cancer cells positive for Glu-tubulin (Fig. 1,D). When the scores determined by the investigators differed significantly, a consensus scoring was decided by joint examination. Nerve fibers known to exhibit high levels of Glu-and Δ2-tubulin (13) were used as internal positive controls. Negative controls were obtained by omission of the primary antibody. To test the robustness of the Glu-tubulin scoring, a sample of 34 slides was scored again by an independent observer (M. H. P.), and the scoring was compared with the consensus scoring. Scores were identical in 28 cases, showing a good reproducibility of the scoring (Kendall concordance coefficient, 0.79). With this scoring, >50% of the tumors were positive for Glu-tubulin staining (Table 1). Thus, tubulin detyrosination is a common occurrence in breast cancer cells.

Other sections from the same tumor blocks as those used for Glu-tubulin scoring were stained with Δ2-tubulin antibody. These tumor sections were analyzed without knowledge of the Glu-tubulin score. Staining of normal cell types was almost identical to the staining obtained with Glu-tubulin antibody, except that the nuclear staining of myoepithelial cells was almost constant, whereas fibroblasts were only exceptionally stained (Fig. 1,E). Normal epithelial cells were never stained (Fig. 1,E). In contrast, Δ2-tubulin could be detected in cancer cells (Fig. 1,F). The scoring was done as described for Glu-tubulin. Over 20% of the tumors scored positive for Δ2-tubulin staining (Table 1). Ninety % (28 of 31) of the Δ2-tubulin-positive tumors were also Glu-tubulin positive. There was a strong overall correlation between positivity for Δ2-tubulin and positivity for Glu-tubulin (P < 0.00001).

Relationship between Tubulin Detyrosination and Markers of Tumor Severity.

Tubulin detyrosination could occur at random among breast tumors or could be related to tumor severity, as assessed by clinical and cytological markers. To test which of these possibilities was correct we analyzed the relationship between the tumor detyrosination score and known markers of tumor severity. Major prognostic factors in breast tumors include patient age, tumor size and axillary lymph node involvement (17). Cytological markers include the steroid receptor status and the so-called SBR grade. In our population, detyrosination was unrelated to age, tumor size, axillary node involvement and receptor status (Table 2). In contrast, there was a highly significant association between positivity for Glu-tubulin and high SBR grade (Table 2). The SBR grade has three components, scoring tubular differentiation, nuclear pleiomorphism, and the proportion of mitotic cells, respectively (16). Nuclear pleiomorphism was not significantly related to detyrosination (Table 2). Interestingly, detyrosination was strongly and specifically correlated with the mitotic score. In strongly detyrosinated tumors, the mitotic score was high (Table 2). The score of tubular differentiation also differed among detyrosination classes, but this was probably the result of statistical fluctuations: the score was lower in the grade 2 Glu-tubulin-positive tumors than in the two other groups (Table 2) and did not differ significantly between Glu-tubulin-negative tumors (grade 1) and strongly positive (grade 3) tumors (P = 0.32, NS). In contrast, the SBR grade and the mitotic score both differed significantly between the two extreme classes of negative and strongly positive (P < 0.001 and P < 0.01, respectively).

Tubulin Detyrosination and Clinical Outcome.

A total of 46 patients had complications related to breast cancer (recurrence and metastasis, metastasis, death) within the period of observation. The patient selection process (see “Materials and Methods”) equalized major prognostic factors such as age and tumor size among patients with or without complications. Therefore, in the studied population, neither age nor tumor size was related to the clinical outcome (P = 0.67 and P = 0.32, respectively). The relationship between the clinical outcome and both the detyrosination and SBR classes is shown in Table 3. Neither the detyrosination index nor the SBR index had significant prognostic value in our study (P = 0.27 and P = 0.11, respectively), although in both cases, a trend toward accumulation of patients with complications in the highest-grade class was observed. Could the detyrosination score be used to improve the prognostic value of the SBR grade? It is the clinician’s experience that the SBR classification is nonambiguous for the extreme classes (1 and 3) but that the class 2 tumors (which are the most numerous) pose a problem, some of them being almost class 1, and others almost class 3. We tried to use the detyrosination score to reclassify the SBR2 tumors: SBR2 tumors were considered as SBR3 if they were strongly positive for Glu-tubulin (grade 3). This defined a corrected SBR index, designated SBRc. The SBRc correlated significantly with the occurrence of complications (P = 0.02, Table 3). The apparently improved prognostic value of SBRc compared with SBR was explained by subgroup analysis that showed that among SBR grade 2 patients, cancer complications occurred in 5 of 8 patients with a detyrosination index of 3 and in only 15 of 58 patients with lower detyrosination grades (P < 0.05, Fisher’s exact test).

The present study shows that tubulin detyrosination is a specific and common occurrence in human breast primary tumor cells. Glu-tubulin was absent in the epithelial cells of normal breast tissue but was observed in over 50% of carcinomatous tumor cells. Δ2-tubulin accumulation was also frequent, although of somewhat more restricted occurrence than Glu-tubulin accumulation. In cultured cycling cells, Glu-tubulin levels can vary to some extent as a function of the cell cycle (10, 18, 19), whereas Δ2-tubulin has only been observed after TTL suppression (15). Therefore, in principle, Δ2-tubulin accumulation is a more specific marker of TTL suppression than Glu-tubulin accumulation. However, in the present study, Glu- and Δ2-tubulin signals were very strongly correlated, which indicated that both Δ2-tubulin accumulation and Glu-tubulin accumulation resulted from TTL suppression. In cultured cells with suppressed TTL activity, Glu-tubulin is much more abundant than Δ2-tubulin (8, 15). Therefore, Glu-tubulin is an easier marker of TTL loss to detect than is Δ2-tubulin. This probably accounts for the excess number of Glu-tubulin-positive tumors compared with the number of Δ2-tubulin-positive tumors observed in this study. We have found Glu-tubulin scoring of tumors to be easy and reproducible among observers. This scoring should, therefore, be easy to perform in routine clinical practice and offers a unique possibility to detect the loss of a putative tumor suppressor gene (TTL) by the accumulation of an abnormal variant of a major cell protein (Glu-tubulin).

Our study shows that tubulin detyrosination does not occur at random among breast tumors. Instead, tubulin detyrosination is more frequent in tumors with a high SBR grade, a known marker of tumor severity (16). The detyrosination grade was apparently unrelated to the differentiation status of the tumor, as assessed by the tumor histomorphology and steroid receptor status, whereas it strongly correlated with the mitotic score. These results agree with previous studies, which indicated that the tyrosination cycle is not a differentiation marker (2, 3, 20) and that its inhibition in cancer cells somehow favors tumor growth (15).

Is tubulin detyrosination a clinically useful marker of tumor prognosis? Our preliminary data indicate that tubulin detyrosination may define high- and low-risk groups in breast cancer patients, which represent the majority of breast tumor patients, with an SBR grade of 2. A simple combination of the SBR and of the detyrosination grades apparently yielded an index with improved prognostic value. These results are encouraging but are based on the analysis of groups of small size and obviously need to be confirmed by studies of much larger patient populations. We believe that the ease of the Glu-tubulin scoring and its potential usefulness justify its evaluation in such studies. The ubiquity of the tubulin tyrosination cycle (2, 3, 20) suggests that TTL elimination may occur in several types of cancers. In a preliminary limited survey of colon and lung cancers, we have, indeed, observed Glu-tubulin and Δ2-tubulin positive tumors. Therefore, assaying tubulin detyrosination may be of clinical interest in epithelial cancers in general.

Fig. 1.

Immunohistochemical staining of Glu- and Δ2-tubulin in breast tumor tissue. Paraffin-embedded breast tissue sections were stained as described in “Materials and Methods” for Glu-tubulin (A–D) or Δ2-tubulin (E and F). Sections were counterstained with Harris’s hematoxylin. Examples of immunostaining in normal areas of breast tissue are shown for Glu-tubulin (A) and for Δ2-tubulin (E), respectively. Nuclei are stained in blue by hematoxylin, and Glu- or Δ2-tubulin is stained in brown. Examples of tumors scored as negative (grade 1), positive (grade 2), or strongly positive (grade 3) for Glu-tubulin staining are shown in B, C, and D, respectively. A tumor strongly positive for Δ2-tubulin is shown in F.

Fig. 1.

Immunohistochemical staining of Glu- and Δ2-tubulin in breast tumor tissue. Paraffin-embedded breast tissue sections were stained as described in “Materials and Methods” for Glu-tubulin (A–D) or Δ2-tubulin (E and F). Sections were counterstained with Harris’s hematoxylin. Examples of immunostaining in normal areas of breast tissue are shown for Glu-tubulin (A) and for Δ2-tubulin (E), respectively. Nuclei are stained in blue by hematoxylin, and Glu- or Δ2-tubulin is stained in brown. Examples of tumors scored as negative (grade 1), positive (grade 2), or strongly positive (grade 3) for Glu-tubulin staining are shown in B, C, and D, respectively. A tumor strongly positive for Δ2-tubulin is shown in F.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported in part by a grant from La Ligue to D. J. and by a United States Department of Defense Grant BCRP DAMD 17-00-1-0618 (to R. L. M.).

4

The abbreviations used are: TTL, tubulin tyrosine ligase; Glu-tubulin, detyrosinated tubulin; Δ2-tubulin, α tubulin lacking both tyrosine and glutamic acid from its COOH terminus; SBR, Scarff-Bloom-Richardson; NS, not significant.

Table 1

Patient characteristics

No. 134 
Age (yr)  
 Median 55.5 
 Range 25.8–85.8 
Tumor-Node-Metastasis (TNM)  
 Tumor size  
  <20 mm 30 (22.4%) 
  20–40 mm 37 (27.6%) 
  ≥40 mm 67 (50.0%) 
 Node involvement  
  N0 47 (35.1%) 
  N1 87 (64.9%) 
 Metastasis  
  M0 120 (89.5%) 
  M1 14 (10.4%) 
Receptor status  
 ERa+PgR+ 77 (57.5%) 
 ER+PgR− 12 (9.0%) 
 ER−PgR+ 6 (4.5%) 
 ER−PgR− 24 (17.9%) 
 Noninterpretable 15 (11.2%) 
SBR Grade  
 1 16 (11.9%) 
 2 66 (49.3%) 
 3 52 (38.8%) 
Glu-tubulin grade  
 1, negative 63 (47%) 
 2, positive 45 (33.6%) 
 3, strongly positive 26 (19.4%) 
Δ2-tubulin grade  
 1, negative 103 (77.0%) 
 2, positive 27 (20.0%) 
 3, strongly positive 4 (3.0%) 
Clinical evolutionb  
 No complication 82 (61.2%) 
 Metastasis 32 (23.9%) 
 Death related to cancer 14 (10.4%) 
 Death not related to cancer 5 (3.7%) 
 Unknown 1 (0.7%) 
No. 134 
Age (yr)  
 Median 55.5 
 Range 25.8–85.8 
Tumor-Node-Metastasis (TNM)  
 Tumor size  
  <20 mm 30 (22.4%) 
  20–40 mm 37 (27.6%) 
  ≥40 mm 67 (50.0%) 
 Node involvement  
  N0 47 (35.1%) 
  N1 87 (64.9%) 
 Metastasis  
  M0 120 (89.5%) 
  M1 14 (10.4%) 
Receptor status  
 ERa+PgR+ 77 (57.5%) 
 ER+PgR− 12 (9.0%) 
 ER−PgR+ 6 (4.5%) 
 ER−PgR− 24 (17.9%) 
 Noninterpretable 15 (11.2%) 
SBR Grade  
 1 16 (11.9%) 
 2 66 (49.3%) 
 3 52 (38.8%) 
Glu-tubulin grade  
 1, negative 63 (47%) 
 2, positive 45 (33.6%) 
 3, strongly positive 26 (19.4%) 
Δ2-tubulin grade  
 1, negative 103 (77.0%) 
 2, positive 27 (20.0%) 
 3, strongly positive 4 (3.0%) 
Clinical evolutionb  
 No complication 82 (61.2%) 
 Metastasis 32 (23.9%) 
 Death related to cancer 14 (10.4%) 
 Death not related to cancer 5 (3.7%) 
 Unknown 1 (0.7%) 
a

ER, estrogen receptor; PgR, progesterone receptor.

b

For patients with complications, the most severe complication is taken into account. Node involvement and metastatic expansion were rated according to the TNM classification of the International Union against Cancer.

Table 2

Relationship between tubulin detyrosination and clinical/biological characteristics of primary breast cancer

Tubulin detyrosination gradeP              a
1 (negative) (n = 63)2 (positive) (n = 45)3 (strongly positive) (n = 26)
Age     
 < 45 yr 13 (20.6%) 11 (24.4%) 4 (15.4%)  
 45–65 yr 31 (49.2%) 21 (46.7%) 10 (38.5%) NS 
 ≥ 65 yr 19 (30.2%) 13 (28.9%) 12 (46.1%)  
Tumor size     
 < 20 mm 17 (27.0%) 10 (22.2%) 3 (11.5%)  
 20–40 mm 18 (28.6%) 10 (22.2%) 9 (34.6%) NS 
 ≥ 40 mm 28 (44.4%) 25 (55.6%) 14 (53.8%)  
Axillary node involvement     
 N0 21 (33.3%) 16 (35.6%) 10 (38.5%)  
 N1 42 (66.7%) 29 (64.4%) 16 (61.5%) NS 
Receptor status     
 ERb+PgR+ 39 (61.9%) 25 (55.6%) 13 (50.0%)  
 ER+PgR− 7 (11.1%) 4 (8.8%) 1 (3.8%)  
 ER−PgR+ 3 (4.8%) 3 (6.7%) 0 (0.0%) NS 
 ER−PgR− 10 (15.9%) 5 (11.1%) 9 (34.6%)  
Non determined 4 (6.3%) 8 (17.8%) 3 (11.5%)  
SBR grade     
 1 11 (17.5%) 4 (8.9%) 1 (3.8%)  
 2 37 (58.7%) 21 (46.7%) 8 (30.8%) <0.001 
 3 15 (23.8%) 20 (44.4%) 17 (65.4%)  
Differentiation score     
 1 3 (4.8%) 3 (6.7%) 0 (0.0%)  
 2 15 (23.8%) 17 (37.8%) 4 (15.4%) <0.05c 
 3 45 (71.4%) 25 (55.6%) 22 (84.6%)  
Nuclear pleiomorphism score     
 1 3 (4.8%) 3 (6.7%) 1 (3.8%)  
 2 39 (61.9%) 17 (37.8%) 11 (4.2%) NS 
 3 21 (33.3%) 25 (55.6%) 14 (53.8%)  
Mitotic score     
 1 35 (55.6%) 18 (40.0%) 8 (30.8%)  
 2 14 (22.2%) 10 (22.2%) 5 (19.2%) <0.05 
 3 14 (22.2%) 17 (37.8%) 13 (50.0%)  
Tubulin detyrosination gradeP              a
1 (negative) (n = 63)2 (positive) (n = 45)3 (strongly positive) (n = 26)
Age     
 < 45 yr 13 (20.6%) 11 (24.4%) 4 (15.4%)  
 45–65 yr 31 (49.2%) 21 (46.7%) 10 (38.5%) NS 
 ≥ 65 yr 19 (30.2%) 13 (28.9%) 12 (46.1%)  
Tumor size     
 < 20 mm 17 (27.0%) 10 (22.2%) 3 (11.5%)  
 20–40 mm 18 (28.6%) 10 (22.2%) 9 (34.6%) NS 
 ≥ 40 mm 28 (44.4%) 25 (55.6%) 14 (53.8%)  
Axillary node involvement     
 N0 21 (33.3%) 16 (35.6%) 10 (38.5%)  
 N1 42 (66.7%) 29 (64.4%) 16 (61.5%) NS 
Receptor status     
 ERb+PgR+ 39 (61.9%) 25 (55.6%) 13 (50.0%)  
 ER+PgR− 7 (11.1%) 4 (8.8%) 1 (3.8%)  
 ER−PgR+ 3 (4.8%) 3 (6.7%) 0 (0.0%) NS 
 ER−PgR− 10 (15.9%) 5 (11.1%) 9 (34.6%)  
Non determined 4 (6.3%) 8 (17.8%) 3 (11.5%)  
SBR grade     
 1 11 (17.5%) 4 (8.9%) 1 (3.8%)  
 2 37 (58.7%) 21 (46.7%) 8 (30.8%) <0.001 
 3 15 (23.8%) 20 (44.4%) 17 (65.4%)  
Differentiation score     
 1 3 (4.8%) 3 (6.7%) 0 (0.0%)  
 2 15 (23.8%) 17 (37.8%) 4 (15.4%) <0.05c 
 3 45 (71.4%) 25 (55.6%) 22 (84.6%)  
Nuclear pleiomorphism score     
 1 3 (4.8%) 3 (6.7%) 1 (3.8%)  
 2 39 (61.9%) 17 (37.8%) 11 (4.2%) NS 
 3 21 (33.3%) 25 (55.6%) 14 (53.8%)  
Mitotic score     
 1 35 (55.6%) 18 (40.0%) 8 (30.8%)  
 2 14 (22.2%) 10 (22.2%) 5 (19.2%) <0.05 
 3 14 (22.2%) 17 (37.8%) 13 (50.0%)  
a

Ps correspond to one-way ANOVA considering SBR grade and scores as quantitative variables.

b

ER, estrogen receptor; PgR, progesterone receptor.

c

For interpretation of this result, see Text.

Table 3

Relationship between tubulin detyrosination, SBR, SBRc, and cancer- related complications

P correspond to Pearson χ2. Log-rank tests gave similar results.
 Occurrence of cancer-related complications  
Tubulin detyrosination grade   
 1 22/63 (34.9%)  
 2 12/45 (26.7%) P = 0.27 
 3 12/26 (46.2%)  
SBR grade   
 1 3/16 (18.8%)  
 2 20/66 (30.3%) P = 0.11 
 3 23/52 (44.2%)  
SBRc   
 1 3/16 (18.8%)  
 2 15/58 (25.9%) P = 0.02 
 3 28/60 (46.7%)  
P correspond to Pearson χ2. Log-rank tests gave similar results.
 Occurrence of cancer-related complications  
Tubulin detyrosination grade   
 1 22/63 (34.9%)  
 2 12/45 (26.7%) P = 0.27 
 3 12/26 (46.2%)  
SBR grade   
 1 3/16 (18.8%)  
 2 20/66 (30.3%) P = 0.11 
 3 23/52 (44.2%)  
SBRc   
 1 3/16 (18.8%)  
 2 15/58 (25.9%) P = 0.02 
 3 28/60 (46.7%)  
1
Jordan M. A., Wilson L. Microtubules and actin filaments: dynamic targets for cancer chemotherapy.
Curr. Opin. Cell Biol.
,
10
:
123
-130,  
1998
.
2
Barra H. S., Arce C. A., Argaraña C. E. Posttranslational tyrosination/detyrosination of tubulin.
Mol. Neurobiol.
,
2
:
133
-153,  
1988
.
3
MacRae T. H. Tubulin post-translational modification-enzymes and their mechanisms of action.
Eur. J. Biochem.
,
244
:
265
-278,  
1997
.
4
Ersfeld K., Wehland J., Plessman U., Dodemont H., Gerke V., Weber K. Characterization of the tubulin-tyrosine ligase.
J. Cell Biol.
,
120
:
725
-732,  
1993
.
5
Hallak M. E., Rodriguez J. A., Barra H. S., Caputto R. Release of tyrosine from tyrosinated tubulin. Some common factors that affect this process and the assembly of tubulin.
FEBS Lett.
,
73
:
147
-150,  
1977
.
6
Webster D. R. Tubulinyl-Tyr carboxypeptidase Barrett A. J. Rawlings N. D. Woessner J. F. eds. .
Handbook of Proteolytic Enzymes
,
:
1565
-1567, Academic Press London  
1998
.
7
Paturle-Lafanechèere L., Eddé B., Denoulet P., Van Dorsselaer A., Mazarguil H., Le Caer J. P., Wehland J., Job D. Characterization of a major brain tubulin variant which cannot be tyrosinated.
Biochemistry
,
30
:
10523
-10528,  
1991
.
8
Lafanechèere L., Job D. The third tubulin pool.
Neurochem. Res.
,
25
:
11
-18,  
2000
.
9
Wehland J., Willingham M. C., Sandoval I. V. A rat monoclonal antibody reacting specifically with the tyrosylated form of α-tubulin. I. Biochemical characterization, effects on microtubule polymerization and organization in vivo.
J. Cell Biol.
,
97
:
1467
-1475,  
1983
.
10
Gundersen G. G., Kalnoski M. H., Bulinski J. C. Distinct populations of microtubules: tyrosinated and nontyrosinated α tubulin are distributed differently in vivo.
Cell
,
38
:
779
-789,  
1984
.
11
Wehland J., Weber K. Turnover of the carboxy-terminal of α-tubulin and means of reaching elevated levels of detyrosination in living cells.
J. Cell Sci.
,
88
:
185
-203,  
1987
.
12
Kreis T. E. Microtubules containing detyrosinated tubulin are less dynamic.
EMBO J.
,
6
:
2597
-2606,  
1987
.
13
Paturle-Lafanechèere L., Manier M., Trigault N., Pirollet F., Mazarguil H., Job D. Accumulation of Δ2-tubulin, a major tubulin variant that cannot be tyrosinated, in neuronal tissues and in stable microtubule assemblies.
J. Cell Sci.
,
107
:
1529
-1543,  
1994
.
14
Guillaud L., Bosc C., Fourest-Lieuvin A., Denarier E., Pirollet F., Lafanechèere L., Job D. STOP proteins are responsible for the high degree of microtubule stabilization observed in neuronal cells.
J. Cell Biol.
,
142
:
167
-179,  
1998
.
15
Lafanechèere L., Courtay-Cahen C., Kawakami T., Jacrot M., Rüdiger M., Wehland J., Job D., Margolis R. L. Suppression of tubulin tyrosine ligase during tumor growth.
J. Cell Sci.
,
111
:
171
-181,  
1998
.
16
Elston C. W., Ellis I. O. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up.
Histopathology
,
19
:
403
-410,  
1991
.
17
Fisher E. R., Costantino J., Fisher B., Redmond C. Pathologic findings from the National Surgical Adjuvant Breast Project (protocol 4). Discriminants for 15 years survival.
Cancer (Phila.)
,
71
:
2141
-2150,  
1993
.
18
Geuens G., Gundersen G. G., Nuydens R., Cornelissen F., Bulinski J. C., DeBrabander M. Ultrastructural colocalization of tyrosinated and detyrosinated αtubulin in interphase and mitotic cells.
J. Cell Biol.
,
103
:
1883
-1893,  
1986
.
19
Bré M. H., Pepperkok R., Kreis T. E., Karsenti E. Cellular interactions and tubulin detyrosination in fibroblastic and epithelial cells.
Biol. Cell.
,
71
:
149
-160,  
1991
.
20
Preston S. F., Deanin G. G., Hanson R. K., Gordon M. W. The phylogenetic distribution of tubulin:tyrosine ligase.
J. Mol. Evol.
,
13
:
23
-44,  
1979
.