| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Arizona Cancer Center and
Department of Pathology, University of Arizona, Tucson, AZ 85724
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
20% of tumors, and moderate infiltrates in 50%
(1). The cellular composition of breast tumor infiltrates
also varies among patients, and is generally heterogeneous, containing
both CD4+ and CD8+ T cells,
with fewer numbers of B cells, macrophages, and NK cells
(2). Although >70% of breast tumors contain moderate or
heavy lymphocytic infiltrates, only 24% of breast adenocarcinomas
contain a large component of tumor-infiltrating B cells
(TIL-B).3 When
present, TIL-B can comprise up to 40% of the tumor-infiltrating
lymphocyte (TIL) population (2, 3). Histologically, TIL-B
in breast tumors are arranged in aggregates (4, 5), and
are predominantly IgG+, as opposed to the low
level predominantly IgA+ population normally seen
in healthy breast tissue (6, 7, 8), or the predominantly
IgM+ population of peripheral blood
(9).
Based on serum Ab reactivity with tumor cells and Ags, patient
antitumor B cell reactions occur in >40% of breast cancer patients
(reviewed in Refs. 10 and 11). A variety of
breast tumor-associated Ags elicits naturally occurring serum Ab
responses (reviewed in Ref. 10). Breast cancer-reactive B
cells have also been identified in regional or draining lymph nodes of
breast cancer patients (for examples, see Refs. 12, 13, 14, 15).
However, little direct evidence exists that TIL-B in breast cancer are
tumor specific, although tumor cell-reactive TIL-B have been cloned
from other tumor types, including melanoma, colon carcinoma, ovarian
carcinoma, lung carcinoma, glioma, sarcoma, neuroblastoma, and
Hodgkin’s lymphoma (16, 17, 18, 19, 20, 21, 22, 23). One study found that
anti-tumor Ag Abs were produced by TIL-B in 70% of nonbreast
tumors examined (19).
The best direct evidence for breast tumor cell-reactive TIL-B comes
from a 1994 study by Katano et al. (24), in which a B cell
line established from a human breast adenocarcinoma was shown to
produce tumor cell-reactive Abs and inhibit growth of autologous tumor
cells. However, human TIL-B reactivity is not limited to tumor-specific
Ags; a study by our group found TNF-
-reactive Abs were produced by
breast cancer TIL-B (20). Furthermore, a recent study
demonstrated that Abs produced by TIL-B of typical medullary carcinoma
(TMC) of the breast specifically bind 
-actin, which occurs on the
surface of apoptotic TMC cells in vivo (25). It has also
been demonstrated that TMC TIL plasma cells are the product of
intratumoral oligoclonal proliferation and differentiation (8, 26). However, it is not clear that TMC is representative of
other more common histologic types of breast cancer, as TMC has many
unusual features, including a diagnostic plasmacytic infiltration,
favorable prognosis, and expression of HLA-DR (27, 28).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tumor tissue was obtained from three untreated women
diagnosed with primary infiltrating ductal carcinoma (IDC). The three
tumors obtained were Her-2 negative by immunohistochemistry.
Tumor-draining lymph node from one of the three patients and peripheral
blood from a healthy female donor were also acquired. Patient data are
summarized in Table I
. PBL were included
to control for poor PCR methodology that could result in repetitive
cloning of single PCR products rather than as a measure of the
peripheral repertoire, which has been extensively characterized
elsewhere (9, 29, 30, 31). Healthy breast tissue was acquired
from a reduction mammoplasty.
|
The breast cancer cell line MCF-7 was obtained from American Type Culture Collection (Manassas, VA) and is positive by flow cytometry for MUC-1, Her-2/neu and carcinoembryonic Ag (CEA), and negative for epithelial cell adhesion molecule (EpCAM). The breast cancer cell lines 3133 and 3199 were established from primary IDC of the breast by the Arizona Cancer Center Tissue Culture Core Service. The cell line 3133 is positive by flow cytometry for CEA, MUC-1, Her-2/neu, and EpCAM. The cell line 3199 is positive by flow cytometry for EpCAM, but is negative for CEA, MUC-1, and Her-2/neu (flow cytometry data; K. Trevor, unpublished data). Untransformed primary human foreskin fibroblasts were obtained from the Arizona Cancer Center Tissue Culture Core Service.
Immunohistochemistry and histology
Using routinely formalin-fixed, paraffin-embedded tissue sections, the reactivities of mAbs directed against the following Ags were studied: CD3, CD20, Ki-67, IgG, IgM, and CD27 (Ventana Medical Systems, Tucson, AZ); CD38 (VS38), CD21, IgA, and IgD (DAKO, Glostrup, Denmark). To determine the lineage of proliferating lymphocytes, sections were double labeled with CD20 and Ki-67 (B lymphocytes) or CD3 and Ki-67 (T lymphocytes). Sections of a human tonsil were placed on each slide as a positive control. Staining was performed using a Benchmark (Ventana Medical Systems) automated immunohistochemistry instrument for CD27, CD21, and Ki-67, while the embryonic stem cell automated immunostainer (Ventana Medical Systems) was used for CD20, CD3, CD38, and H chain Abs IgM, IgG, IgA, and IgD (32).
IgG H chain library preparation for repertoire analysis
Tumor and node tissue were acquired immediately postsurgery and
manually disaggregated, and lymphocytes were enriched by Ficoll
gradient centrifugation, as described (33). Parameters of
individual IgG H chain libraries are summarized in Table I
. Following
disaggregation and Ficoll gradient centrifugation, TIL were split into
two separate pools of cells and washed with PBS to remove residual mRNA
from lysed cells before RNA extraction. Tandem libraries were cloned
from the separated cells to be able to distinguish between repetitively
cloned PCR products and identical B cell progeny as from a memory B
cell response. Total RNA was extracted with TRIzol reagent
(Sigma-Aldrich, St. Louis, MO) from each of the TIL pools. Separate
reverse-transcriptase reactions were performed on the two RNA pools, as
described (34). IgG1 H chain Fd regions
(VH + CH1) were amplified
by PCR: each of seven individual 25 µl PCRs contained one consensus
degenerate VH family primer, an IgG1-specific
constant region primer (CG1z), and 3 µl cDNA. Samples were run with
one ready-to-go PCR bead (Pharmacia, Peapack, NJ). Primers have been
previously described (8). Amplifications consisted of a
4-min 94°C hot start, followed by 35 cycles, 1 min at 94°C, 2 min
at 55°C, and 3 min at 72°C, with a single terminal 10-min extension
at 72°C in a PTC-100 thermocycler (MJ Research, South San Francisco,
CA). All reactions were performed in quadruplicate, pooled, and
purified by agarose gel electrophoresis and extraction via Qiaex II
(Qiagen, Valencia, CA). PCR products were ligated into the pGEM-T
vector (Promega, Madison, WI) using 50 ng vector, 15 ng insert, and T4
DNA ligase (Promega), in triplicate ligation reactions to maximize
clone diversity.
Patient 3 TIL-B. One library was made with an IgG1-specific C region primer (CG1z), and the second library was constructed using a pan-IgG C region primer (CHIgG), as previously described (35). PCR products were cloned as described for patient 1.
After transformation of pGEM-T libraries into XL1-Blue competent cells (Invitrogen, Carlsbad, CA), random colonies were selected and grown overnight, and plasmid DNA was prepared with Wizard plasmid miniprep columns (Promega). Randomly picked clones were screened for the presence of IgG H chain insert through SacI/SacII (Promega) restriction digestion.
DNA sequencing and analysis
Clones were sequenced using standard sequencing primers (Arizona Research Laboratories, University of Arizona) and resolved using the FAKTORY program (Arizona Research Laboratories, University of Arizona). Candidate germline genes were identified via DNAPLOT (36). Percentage mutation and replacement to silent mutation ratios were calculated for V gene-encoded regions only (framework region (FR)1-complementarity-determining region 1 (CDR)1-FR2-CDR2-FR3). Taq error rates were calculated from the first H chain C region (CH1) of five randomly selected clones. To determine clonality, IgG H chain sequences with shared VDJ usage were aligned by CLUSTAL W (37). Clonality of cloned IgG H chain sequences was determined by shared VDJ germline gene usage and VDJ junctional mutation patterns, as this level of diversity is determined in the bone marrow before circulation in the periphery. Germline gene nomenclature is as per previously described (36).
Construction and panning of phage display library
Protocols for phage display library construction, panning, and
analysis are as previously described (34). Briefly, a
150-mg sample of tumor from patient 1 was homogenized with mortar and
pestle in liquid nitrogen, and RNA extracted using TRIzol reagent
(Sigma-Aldrich). RNA was further purified by lithium chloride
extraction, and a total of 92 µg purified RNA was obtained. Of this,
20 µg RNA was used in a reverse-transcriptase reaction using the
Superscript II First Strand system (Invitrogen). cDNA was treated with
RNaseH (Invitrogen) and used directly in PCR amplification of H and L
chain V regions, unrestricted for isotype. A Fab (H chain V region plus
CH1 and L chain VL plus
CL) library of 1 x
108 Fab clones was cloned in the pCOMBX phage
display vector (gift of C. Barbas, The Scripps Research Institute, La
Jolla, CA). The phage-displayed library was panned with the MCF-7
breast cancer cell line in six sequential rounds of panning. In brief,
4 x 1012 phage were added to 4 x
106 allogeneic breast cancer cells (MCF-7),
incubated for 30 min at room temperature, centrifuged, and washed five
times with PBS to remove unbound phage. Cell-bound phage was recovered
by trypsin digestion (Invitrogen) and used to reinfect
Escherichia coli strain XL1-B (New England Biolabs,
Cambridge, MA). After overnight growth, phage was isolated by
polyethylene glycol precipitation, and the cycle was repeated.
The phage library was also panned with autologous tumor tissue lysate
(soluble protein), with negative selection on healthy breast tissue
lysate. Lysates were prepared by homogenization of snap-frozen tissue
in liquid nitrogen, repeated freeze-thaw cycles, and extraction
overnight at 4°C in PBS containing 1% Nonidet P-40 (Sigma-Aldrich),
10 µg/ml aprotinin (Sigma-Aldrich), and 1 mM PMSF (Sigma-Aldrich).
ELISA plates (Costar, Garden Grove, CA) were coated overnight at 4°C
with 20 µg protein/well. Four wells were coated with tumor tissue
lysate, and four wells with healthy breast tissue lysate. Plates were
then blocked for 1 h with 5% BSA, and 1 x
1012 phage were added to each tumor tissue lysate
well. Plates were incubated at 37°C for 2 h. Wells were washed
with PBS + 0.1% Tween 20 (Sigma-Aldrich) to remove unbound phage, with
increasing numbers of washes in successive rounds of panning. Phage
were then eluted by the addition of glycine elution buffer (0.1 M
glycine-HCl, pH 2.2), and neutralized with 3 M Tris. To reduce
nonspecific binding, eluted phage were then added to healthy breast
tissue lysate-coated wells for 30 min, and unbound phage was recovered
in panning rounds 3, 4, and 5. Phage was rescued by transfection of
E. coli strain XLB-1 (Invitrogen). After overnight growth,
phage was isolated by polyethylene glycol precipitation, and the cycle
was repeated.
Analysis of phage-displayed Ab pools by flow cytometry
Following selection of the phage Fab library on MCF-7 cells, phage pools were assessed by flow cytometry, as described (34). Cell lines MCF-7, 3133, 3199, and foreskin fibroblasts were analyzed. Cells were incubated with preselection phage display library, postselection phage display library, or antitetanus toxoid phage-displayed Fab as a negative control (tetanus toxoid, provided by C. Barbas, The Scripps Research Institute) (38). Fab reactivity was determined by flow cytometry using anti-M13 mouse mAb (Pharmacia) as secondary Ab, and FITC-labeled goat anti-mouse F(ab')2-specific Ab (Jackson ImmunoResearch Laboratories, West Grove, PA) as tertiary Ab, as described (34). Ten thousand cells in the gated (live) population were counted per tube.
Analysis of phage-displayed Ab pools by ELISA
Following selection of the phage Fab library on autologous tumor tissue lysate (soluble protein), phage pools were used in ELISAs against autologous tumor tissue lysate and healthy breast tissue lysate, as described (34). Plates were coated overnight with an excess of either autologous tumor tissue lysate or healthy breast tissue lysate (>20 µg) in 25 µl at 4°C, and blocked with 5% BSA in PBS. Phage (50 µl) that had been selected on autologous tumor tissue lysate were added to each well, incubated 2 h at 37°C, and washed 10 times with H2O, and HRP-conjugated anti-M13 Ab (Pharmacia) was added. Following a 1-h incubation at 37°C, plates were washed 10 times with H2O, and Fab binding was detected with ABTS substrate (Roche Molecular Biochemicals, Indianapolis, IN).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The three cases of IDC of the breast were reviewed for TIL
using H&E staining and immunohistochemical markers of B lymphocytes, T
lymphocytes, plasma cells, and follicular dendritic cells. The tumors
of patients 1 and 2 were largely composed of fibrous stroma, with
interspersed islands of malignant cells (Fig. 1, A and D). The
tumor of patient 3 consisted of large tumor nests interconnected by
bands of fibrous stromal tissue (Fig. 1G). Although TIL were
found scattered throughout the stroma and interspersed between tumor
cells in all three tumors, most lymphocytes clustered in dense
aggregates. In many cases, the aggregates occurred in stromal
areas immediately adjoining tumor nests. This is seen most clearly in
the tumor of patient 3, which had both a higher relative composition of
malignant cells as opposed to acellular stroma, and a denser
lymphocytic infiltrate. Lymphocytes also occurred adjacent to areas of
morphologically nonmalignant tissue within the tumor, but were not
observed outside the tumor margins.
|
, B
and E). In contrast, the heavier infiltrate of patient 3 was
composed of 30–40% B cells (Fig. 1H). B lymphocytes
occurred almost exclusively adjacent to tumor nests on the periphery of
the main tumor mass in aggregates reminiscent of lymph node germinal
centers. B lymphocytes were negative for the memory phenotype marker
CD27 (data not shown). Although small numbers of patient 3 germinal
center B cells stained for IgD or IgM, high levels of background
staining for IgA, IgM, and IgG from serum Abs rendered the isotype of
most B cells difficult to interpret (data not shown).
Aggregates of T lymphocytes were observed in the tumors of all three
patients (Fig. 1, C, F, and I), with
individual infiltrating CD3+ T lymphocytes
observed between malignant cells within tumor nests. In patient 3, T
lymphocytes surrounded B lymphocyte germinal centers as occurs in a
germinal center light zone. Patient 3 germinal centers also contained
interdigitating CD21+ follicular dendritic cells
in the B cell area (Fig. 1J). CD21+
cells were not observed outside of B cell germinal centers. Plasma
cells (CD38) were rare, frequently IgA+ (as
normally seen in healthy breast tissue) (6), and occurred
randomly in relation to other lymphocytes (data not shown).
Morphologically, TIL were small with a mature chromatin pattern.
Lymphocyte mitoses were extremely infrequent. Low B lymphocyte
proliferation was confirmed by low staining for the proliferation
marker Ki-67 (Fig. 1, E and K). Although adjacent
tumor cells and same-slide tonsil controls were
Ki-67+, <2% of lymphoid cells were positive
(Fig. 1L).
Calculation of PCR Taq polymerase error rate
To calculate the background level of PCR-induced mutations, five
IgG H chain clones were randomly selected and compared with the human
germline IgG1 CH1 region sequence
(36). Only one error was detected in a total of 1485 bp,
for a total error rate of 0.67 Taq-induced mutations per
1000 bp. In a 296-bp V gene-encoded region, this would result in 0.2
Taq-induced errors, which is negligible for the purposes of
this study. Previous studies have demonstrated comparable levels of
PCR-generated error in Ab amplifications (39, 40).
Oligoclonal expansion of TIL-B
To determine whether the aggregates of B cells observed in tumors
were the result of random recruitment from the periphery or the
proliferation of TIL, IgG H chain libraries were generated by RT-PCR
and random clones were sequenced from each of the three IDC tumors, a
tumor-draining lymph node from patient 3, and PBL from a healthy donor.
A summary of the data is found in Table II.
|
). Similar degrees of
clonality were observed in IgG H chains from patients 2 and 3 tumors.
Of the 31 IgG1 H chains sequenced from patient 2, 68% belonged to one
of five clonal groups. Of the 58 IgG H chain sequences examined from
patient 3, 62% belonged to one of the 11 identified clonal groups
(Fig. 2).
|
, demonstrates definitively that B cell
proliferation rather than PCR artifact produced the clonal groups.
Somatic hypermutations were also far in excess of the calculated 0.2
PCR-induced mutations per V gene-encoded region. However, a small
number of clones were identical, with no unique somatic mutations in
comparison with each other, as might result from one or more cell
divisions during which no somatic hypermutations were acquired. Most
identical clones were verified as derived from identical B cell clones
rather than PCR artifact, having been cloned from independent TIL
pools. For example, the identical clones 1–18 and 2–38 were derived
from libraries 1 and 2, respectively (Fig. 3A).
|
), derived from a progenitor B cell using
the germline V gene 4-39*06, because mutations were not linear. This
may be explained by a phenomenon described by McHeyzer-Williams et al.
(41), in which chimeric molecules produced during PCR
confuse the linearity of descent for clones. One IgG H chain library for patient 3 was produced with an IgG1-specific primer, and the second library with a pan-IgG-specific primer. All sequences were of the IgG1 isotype regardless of primer used, suggesting that the IgG1 isotype dominates the IgG TIL-B pool.
Only two reiterated sequences of 26 occurred in the tumor-draining
lymph node sequences, for a total of 7.7% clonality (Fig. 2). No
overlap existed between the sequences isolated from patient 3 TIL and
patient 3 tumor-draining lymph node to suggest node/tumor trafficking
(with the caveat that a larger sample size might be required to detect
intersection). No repeated clones were observed in PBL sequences,
consistent with expected low levels (<1/20,000) of reiterated clones
for healthy donor peripheral blood (29) (Fig. 2).
Germline VH gene usage
VH germline gene family usage by TIL-B was
analyzed for evidence of biased usage that might suggest epitope
selection (Fig. 4). Library repertoires
generated by this and similar methods have generally matched those
generated by single cell RT-PCR (31, 8). The most common
VH germline genes used by TIL-B belonged to the
VH3 and VH4 families. In
addition, VH1 gene segments were used by 10–35%
of TIL-B IgG and tumor-draining node IgG H chains, while no
VH1 occurred in the PBMC repertoire.
|
TIL-B IgG H chain V gene sequences were compared with progenitor
germline V genes to determine levels and patterns of somatic
hypermutation. Somatic mutations clustered in the CDRs, consistent with
affinity maturation (Fig. 5). Mutation
levels and patterns were similar to the tumor-draining lymph
node-derived sequences and known affinity-matured Ab sequences
(reviewed in Ref. 30), suggesting a functional ectopic
germinal center reaction in the breast tumors. TIL-B IgG H chain CDR1
and CDR2 regions were mutated on average between 8 and 13%,
respectively, in comparison with germline V genes. In contrast, CDR1
and CDR2 regions of IgG1 H chains derived from PBL were on average
mutated only 0.3% and 0.4% from germline. Most, if not all, of the
observed somatic hypermutations in TIL clonal groups could be accounted
for by mutations accumulated during intratumoral proliferation, as
shown in Fig. 2. The replacement mutation pattern of some Abs was
nonrandom, as calculated by the method of Lossos et al.
(42), suggesting affinity maturation (Table II and Fig. 6).
|
|
To determine whether TIL-B Abs were reactive with tumor cells, a
phage-displayed Fab library was generated from patient 1 TIL by the
methods of Barbas et al. (34). The Fab library was panned
with the allogeneic breast cancer cell line MCF-7. Binding of
phage-displayed Fab to MCF-7 cells was measured by flow cytometry. Mean
fluorescent intensity increased 2.7-fold from 13 to 36 after only four
rounds of panning with MCF-7 cells, and 24-fold to a mean fluorescent
intensity of 312 after six rounds of panning (Fig. 7). The MCF-7-selected Ab pool bound the
3199 breast cancer cell line with 8-fold greater intensity in
comparison with the unselected Ab pool, and the 3133 breast cancer cell
line with 2-fold greater intensity. In contrast, binding of TIL-B Abs
to human primary fibroblasts decreased 50% after panning with
MCF-7.
|
7-fold increase in binding to
autologous lysate was observed, as measured by ELISA against autologous
tumor tissue lysate protein (Fig. 8). An
equal enrichment for binding to healthy breast tissue lysate was also
observed in the tumor tissue lysate-selected Abs.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). However, unlike germinal centers from true lymphoid tissue,
germinal centers contained no dark zone of Ki-67-positive proliferating
B cells (Fig. 1). Mantle zones and CD38+ plasma
cells were also absent. Despite low Ki-67 staining of TIL-B, intratumoral oligoclonal expansion of TIL-B was established by the presence of clonal groups derived from common progenitor B cells in the three breast tumors examined, regardless of histologic TIL-B density. The seemingly contradictory molecular and histological proliferation data may reflect a low B lymphocyte proliferative rate over a period of time, or alternatively may be a relic of a previous period of proliferative activity and subsequent quiescence. Between 44 and 68% of IgG H chain sequences from TIL-B belonged to clonal groups, while only 7% of tumor-draining lymph node sequences and 0% of PBL sequences were clonal, consistent with the large repertoires of those populations.
TIL-B IgG H chain mutation levels, patterns, and germline gene usage
suggest that TIL-B undergo affinity maturation intratumorally,
presenting the possibility of production of high-affinity
anti-tumor Ag Abs. However, this conclusion stems from indirect
evidence of affinity maturation, which can only be resolved through Ag
affinity studies. TIL-B IgG H chains contained somatic mutations that
clustered in the Ag-contacting CDRs, as is observed in affinity-matured
Abs (reviewed in Refs. 30 and 43), and as was
also seen in tumor-draining lymph node, but not peripheral blood
IgG (Fig. 5). As calculated by the polynomial algorithm of Lossos et
al. (42), replacement and silent mutations occurred
nonrandomly in some TIL-derived Ig. A modest bias in usage of
individual germline genes was consistent with epitope selection, with
use of germline genes 1–18, 3–30, 4–39, and 4–61 by all three TIL-B
repertoires (Fig. 4). Of these genes, only 3–30 is normally
overrepresented in the peripheral repertoire of young or elderly adults
(31, 44).
TIL-B-derived Fabs were reactive with the allogeneic breast cancer cell
line MCF-7, indicating that TIL-B proliferate in response to tumor Ag
rather than nonspecific inflammatory or cytokine signals (Fig. 7).
Although Fabs were selected for binding to MCF-7, binding of these Abs
to other breast cancer cell lines demonstrated the presence of reactive
epitope(s) common to these cell lines. Although the TIL-B Fab pool
selected against MCF-7 cell surface bound breast cancer cell lines
preferentially in comparison with nonmalignant primary fibroblasts, the
determination of true specificity awaits further characterization.
Panning of the Fab library on autologous tumor tissue lysate (soluble
protein) yielded Fabs with equal reactivity for soluble lysates from
tumor and healthy breast tissue (Fig. 8). This indicates that at least
some TIL-B produce Abs reactive with Ags shared by breast tumor and
healthy breast.
Our study suggests that while TIL-B undergo tumor Ag-driven expansion in intratumoral follicles, deletion of autoreactive B cells may be deficient. Because few proteins expressed by tumor cells are truly tumor specific, the majority of Abs produced will likely be directed against autoantigens, and only a small percentage against tumor-specific Ags. Isolation of tumor-specific Abs from TIL will thus require judicious technique. We speculate that the lack of negative selection in intratumoral germinal centers may be the source of autoreactive breast tumor-associated serum Abs described by other groups, some of which are associated with pathologic autoimmune states (45, 46). Future investigations are needed to investigate the identities of Ags reactive with TIL-B Abs and implications of ectopic germinal centers in breast cancer.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Julia A. Coronella, Arizona Cancer Center, University of Arizona, 1515 North Campbell Avenue, Tucson, AZ 85724-5024. E-mail address: jcoronella{at}azcc.arizona.edu
3 Abbreviations used in this paper: TIL-B, tumor-infiltrating B cell; CDR, complementarity-determining region; CEA, carcinoembryonic Ag; EpCAM, epithelial cell adhesion molecule; FR, framework region; IDC, infiltrating ductal carcinoma; TIL, tumor-infiltrating lymphocyte; TMC, typical medullary carcinoma.
Received for publication March 20, 2002. Accepted for publication May 31, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
genes with no or little somatic mutation throughout life. Eur. J. Immunol. 23:3272.[Medline]
2 microglobulin expression in medullary and atypical medullary carcinoma of the breast: histopathologically similar but biologically distinct entities. J. Clin. Pathol. 53:286. IV spectrin in paraneoplastic lower motor neuron syndrome. Proc. Natl. Acad. Sci. USA 98:6945.This article has been cited by other articles:
|
|
 |
|
  S. N. Willis, S. S. Mallozzi, S. J. Rodig, K. M. Cronk, S. L. McArdel, T. Caron, G. S. Pinkus, L. Lovato, K. L. Shampain, D. E. Anderson, et al. The Microenvironment of Germ Cell Tumors Harbors a Prominent Antigen-Driven Humoral Response J. Immunol., March 1, 2009; 182(5): 3310 - 3317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Coppola and J. J. Mule Ectopic Lymph Nodes Within Human Solid Tumors J. Clin. Oncol., September 20, 2008; 26(27): 4369 - 4370. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schmidt, D. Bohm, C. von Torne, E. Steiner, A. Puhl, H. Pilch, H.-A. Lehr, J. G. Hengstler, H. Kolbl, and M. Gehrmann The Humoral Immune System Has a Key Prognostic Impact in Node-Negative Breast Cancer Cancer Res., July 1, 2008; 68(13): 5405 - 5413. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Yang, G. C. Waldbieser, and C. J. Lobb The Nucleotide Targets of Somatic Mutation and the Role of Selection in Immunoglobulin Heavy Chains of a Teleost Fish J. Immunol., February 1, 2006; 176(3): 1655 - 1667. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Kotlan, P. Simsa, J.-L. Teillaud, W. H. Fridman, J. Toth, M. McKnight, and M. C. Glassy Novel Ganglioside Antigen Identified by B Cells in Human Medullary Breast Carcinomas: The Proof of Principle Concerning the Tumor-Infiltrating B Lymphocytes J. Immunol., August 15, 2005; 175(4): 2278 - 2285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-J. Kim, T. Kammertoens, M. Janke, O. Schmetzer, Z. Qin, C. Berek, and T. Blankenstein Establishment of Early Lymphoid Organ Infrastructure in Transplanted Tumors Mediated by Local Production of Lymphotoxin {alpha} and in the Combined Absence of Functional B and T Cells J. Immunol., April 1, 2004; 172(7): 4037 - 4047. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sapino, P. Cassoni, E. Ferrero, M. Bongiovanni, L. Righi, N. Fortunati, P. Crafa, R. Chiarle, and G. Bussolati Estrogen Receptor {alpha} Is a Novel Marker Expressed by Follicular Dendritic Cells in Lymph Nodes and Tumor-Associated Lymphoid Infiltrates Am. J. Pathol., October 1, 2003; 163(4): 1313 - 1320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Nzula, J. J. Going, and D. I. Stott Antigen-driven Clonal Proliferation, Somatic Hypermutation, and Selection of B Lymphocytes Infiltrating Human Ductal Breast Carcinomas Cancer Res., June 15, 2003; 63(12): 3275 - 3280. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Deduced intermediates; numbers inside circles
indicate sequenced clones; numbers next to arrows indicate numbers of
mutations in comparison with germline at a given branching.
A, Patient 1, clonal group 1. Clones 1–35 and 2–26
contained no V gene mutations in comparison with germline. Clones 2–38
and 1–18 shared a single common mutation from germline.
B and C, Patient 2, clonal groups 3, 5.
D and E, Patient 3 clonal groups 4,
8.
