| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |


From the Second Department of Internal Medicine*
and First Department of Pathology,
Yamagata
University School of Medicine, Yamagata, Japan
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
GCs and FDCs in the secondary LFs have been reported to express some Ca2+-binding proteins (CBPs), including an acetic CBP, S-100 protein, a vitamin-D- dependent CBP, calbindin-D, and a Ca2+-dependent phospholipid binding protein, annexin VI.7-10 Annexin VI has a single high-affinity Ca2+-binding site and lacks the classic EF-hand Ca2+-binding sites. An EF-hand family of CBPs, calmodulin, plays roles in diverse events, including cell proliferation, smooth muscle contraction, ion channel control, and microtubular assembly.11 A calmodulin-dependent (type 2B) serine/threonine protein phosphatase, calcineurin, is also categorized as an EF-hand CBP.12 Caldesmon is a major calmodulin- and actin-binding protein, which is essential for smooth muscle and nonmuscle contraction.13 These findings suggest that CBPs may be indispensable to the correct functioning of every cell by regulating the intracellular Ca2+ concentration ([Ca2+]i). However, very few papers described details of the follicular localization of CBPs in lymphatic tissues and the inclusive CBP localizations in the secondary LFs remain to be clarified. It is well known that follicular lymphomas have FDC meshworks, and in some neoplastic lymphomas there is still a functional relationship between FDCs and neoplastic lymphoma cells similar to that observed in non-neoplastic LFs.14,15 Some authors investigated S-100 protein localization in malignant lymphoma,16-19 but there is still a mystery about follicular distribution of CBPs and functions of FDCs in follicular lymphoma.9
The aim of this study was to investigate the precise localization of six different CBPs, caldesmon, S-100 protein, calcineurin, calbindin D, calmodulin, and annexin VI, in the five zones of secondary LF, with special reference to FDCs. Furthermore, the distribution patterns of caldesmon and S-100 protein in neoplastic follicles of follicular lymphomas were determined and compared with that of the FDC marker Ki-M4p to investigate the characteristics of FDCs in neoplastic follicles.
| Materials and Methods |
|---|
|
|
|---|
Fifteen palatine tonsils from patients suffering from chronic tonsillitis, five lymph nodes showing reactive follicular hyperplasia, three appendices, and three terminal ileal specimens (including Peyer's patches) were studied. Palatine tonsils were obtained during therapeutic tonsillectomy, lymph nodes and terminal ileal tissue were obtained during right hemicolectomy for colonic cancer, and appendices were obtained during surgery for ectopic pregnancy. Tissues from 19 patients with follicular lymphoma (7 in superficial lymph nodes, 6 in stomach, and 6 in thyroid gland) were obtained during diagnostic or therapeutic surgery. Follicular lymphomas were classified and graded according to the Revised European-American Lymphoma Classification.20 Six, eight, and five cases with grades I, II, and III lymphomas, respectively, were investigated.
Ca2+-Capture Cytochemistry
Ca2+-capture cytochemistry was carried out with the method described previously by Menon et al.21 Fresh tonsils were cut into 500-µm-thick slices using a DTK-1000 microslicer (Dosaka EM, Osaka, Japan) and immersed in ice-cold 90 mmol/L potassium oxalate containing 2% glutaraldehyde and 2% formaldehyde overnight for Ca2+ localization. The samples were post-fixed with 1% osmium tetroxide containing 2% pyroantimonate for 2 hours at 4°C, rinsed in ice-cold distilled water (pH 10, adjusted with 0.1 N KOH), dehydrated in a graded series of ethanols, and routinely embedded in a low-viscosity epoxy resin.22 Thin sections containing GCs were observed using a Zeiss EM109 electron microscope (Zeiss, Oberkochen, Germany).
Antibodies
Heterologous antisera or monoclonal antibodies against the seven
CBPs were used (Table 1)
. The five
secondary LF zones were identified after hematoxylin and eosin staining
and immunostaining of CD23 (H107, mouse IgG; DAKO, Glostrup, Denmark).
|
A portion of each lymphatic tissue was fixed with periodate-lysine-paraformaldehyde (PLP)23 for 6 hours at 4°C, washed with 0.01 mol/L phosphate-buffered saline (PBS, pH 7.4) containing graded concentrations of sucrose, embedded in OCT compound (Miles Laboratories, Elkhart, IN), and stored at -80°C until cryostat sectioning. Another portion of each lymphatic and follicular lymphoma tissue was fixed with 10% formalin at room temperature for 12 hours and processed routinely, and 3-µm-thick paraffin sections were cut out. Both cryostat and paraffin sections were stained immunohistochemically using the avidin-biotin-peroxidase complex (ABC) technique24 with 33' diaminobenzidine (DAB; Dojin Chemicals, Kumamoto, Japan) and then counterstained with methyl green or hematoxylin. Antigen retrieval on paraffin sections was carried out in a 0.1% trypsin solution (37°C) for 20 minutes or a 1 mmol/L EDTA/NaOH solution (pH 8.0) at 90°C for 15 minutes. Control sections were treated with PBS or nonimmune serum instead of specific antibodies. The immunostaining intensities of the CBPs in each of the five zones (MZ, OZ, ALZ, BLZ, and DZ) within the secondary LFs and those of caldesmon, S-100 protein, and Ki-M4p in neoplastic follicles of follicular lymphomas were classified into four categories: -, negative; +, weakly positive; ++, moderately positive; +++, strongly positive.
Immunoelectron Microscopy
For immunoelectron microscopy of the CBPs, tonsillar specimens were fixed with PLP for 6 hours at 4°C, cut into 60-µm-thick slices using a DTK-1000 microslicer, and subjected to indirect immunoperoxidase staining.25 The tissues used for S-100 protein immunostaining were re-fixed with 10% formalin for 10 minutes at room temperature. After immunostaining, the specimens were fixed with 1.25% glutaraldehyde and 2% osmic acid, dehydrated in a graded series of ethanols, and embedded in Epon resin, and ultrathin sections were observed using a Zeiss EM109 electron microscope.
Isolation of FDC-Associated Clusters from GCs and Immunocytochemical Staining
After tonsillar specimens were fixed with PLP for 6 hours at 4°C, 300-µm-thick slices were cut using a DTK-1000 microslicer, and the GCs were enucleated from the microslices under a stereoscope.26 To obtain FDC-associated clusters, enucleated GCs were digested with 0.05% collagenase (Sigma Chemical Co., St. Louis, MO) for 1 hour at 37°C using the method described previously26 with modifications.27 The extracts were then filtered through 90-µm nylon monofilament mesh, and cytospin slides were prepared using a Cytospin centrifuge (Shandon, Pittsburgh, PA).
FDC-associated clusters mounted on cytospin slides were subjected to double immunocytochemical staining. First, R4/23 immunostaining was carried out using the alkaline phosphatase-anti-alkaline phosphatase (APAAP) method28 with 5-bromo-4-chloro-3-indoxyl phosphatase and nitro blue tetrazolin chloride (BCIP/NBT; DAKO), which gave rise to a royal purple color, and then staining for each of the CBPs was performed using the ABC method, which yielded a brown color. The preparation subjected to S-100 protein immunostaining was re-fixed with 10% formalin for 10 minutes at room temperature before the ABC procedure. All of the tissues were counterstained with methyl green. Negative control cytospin preparations were incubated with PBS or nonimmune serum instead of an anti-CBP antibody.
In Situ Hybridization (ISH)
A 1700-base cDNA probe complementary to human calmodulin mRNA was obtained from the American Type Culture Collection (Rockville, MD)29 and labeled with biotinylated 11d-UTP using a nick translation kit (BRL, Gaithersburg, MD). For ISH, tonsillar specimens were fixed with 4% paraformaldehyde for 6 hours at 4°C, and cryostat sections were prepared, as described above, subjected to proteolytic digestion with 0.2 N HCl for 10 minutes at room temperature, followed by proteinase K (5 mg/ml) for 10 minutes at 37°C,30 and hybridized with the calmodulin probe at 42°C overnight. They were then incubated with avidin-fluorescein isothiocyanate (FITC; Boehringer Mannheim, Indianapolis, IN) for 1 hour at room temperature, and the location of the bound probe was observed under a fluorescence microscope. Positive control preparations were incubated with FITC-labeled ß-actin (DAKO) and negative controls with PBS instead of the hybridization probe for the hybridization and RNAse digestion (200 mg/ml for 1 hour at 37°C).
| Results |
|---|
|
|
|---|
Extensive Ca2+ capture was observed in the
extracellular spaces. Some follicular lymphocytes captured small
amounts of Ca2+ in their cytoplasm, whereas the majority of
FDCs exhibited marked Ca2+ capture in both their cytoplasm
and nuclei (Figure 1, A and B)
, revealing
that FDCs have a relatively high Ca2+-binding capacity.
Capture was fairly extensive at the FDC cytoplasmic peripheries, and
vascular endothelial cells captured large amounts of Ca2+
in their cytoplasm and nuclei.
|
The CBP immunohistochemical staining results for the LFs of
various lymphatic tissues, including tonsils, lymph nodes, appendices,
and Peyer's patches, are summarized in Table 2
. Caldesmon, S-100 protein, calcineurin,
calbindin-D, calmodulin, and annexin VI were positive in the LFs of
lymphatic tissues. In general, immunostaining was more intense in the
light zone (LZ) than the DZ (Figure 2, A and CF
). The BLZ contained all six CBPs, and S-100 protein was
localized in the BLZ and DZ but not in the ALZ (Figure 2B)
. The MZ
showed intense calcineurin (Figure 2C)
and annexin VI immunostaining
and weak reactions with caldesmon and calmodulin (Figure 2, E and F)
.
Calretinin was negative in the LFs. There were no marked
differences among the four different lymphatic tissues with respect
to the immunoreactions in each secondary LF zone (Figure 2G)
.
|
|
The immunoelectron microscopic results are summarized in Table 3
. Five CBPs were observed in FDCs.
Caldesmon was limited to the peripheries of the FDC cell bodies and
long cytoplasmic processes, and it accumulated particularly in the
cytoplasm, twining around extracellular thick fibers (Figure 3A)
. The peripheral cytoplasm of FDCs
often contained stress fibers. In contrast, caldesmon was not detected
in the sagging peripheral FDC cytoplasm in contact with lymphocytes,
and the inner FDC cytoplasm either showed no or faint immunostaining.
S-100 protein was distributed evenly in the FDC cytosol (Figure 3B)
.
The FDC cytoplasm also showed diffuse calcineurin, calbindin-D (Figure 3C)
, and calmodulin immunostaining. Annexin VI was localized on the
cell membranes of GC lymphocytes but not FDCs.
|
|
To verify the immunohistochemical results, the expression of the
five CBPs except annexin VI in FDCs within isolated FDC-associated
clusters was examined. FDCs expressing the FDC marker R4/23 also
expressed caldesmon (Figure 4A)
, S-100
protein, calcineurin, calbindin-D, and calmodulin (Figure 4B)
simultaneously.
|
ISH detected calmodulin mRNA in the cytoplasm of the vast majority
of follicular cells (Figure 5)
, whereas
the LFs in negative control preparations showed no specific reactions.
|
Neoplastic follicles in grades I and II lymphomas were positive in
a lacy pattern for Ki-M4p and caldesmon (Figure 6A)
, but never contained S-100
protein-positive FDCs (Figure 6B)
, except a part of grade II neoplastic
follicles that contained only a few, small conglomerate positive
patterns (Table 4)
. In contrast, those in
grade III lymphomas were positive for Ki-M4p and S-100 protein
demonstrating a diffuse pattern often intermingled with the large
conglomerate pattern (Figure 6D
, but not for caldesmon (Figure 6C)
.
|
|
| Discussion |
|---|
|
|
|---|
The vast majority of physiological phenomena depend extensively on [Ca2+]i changes in response to various extracellular and intracellular stimuli, some of which lead to increases in [Ca2+]i, and cytosolic CBPs bind to Ca2+ at concentrations in the range of 10-5 to 10-8 mol/L.33 [Ca2+]i changes are also important in the immunological phenomena that occur in LFs. For example, the Ca2+-cAMP-dependent second messenger system appears to regulate apoptosis of GC B cells.34 Although FDCs and GCs in the secondary LFs have been found to express some CBPs, including S-100 protein, calbindin-D, and annexin VI,7-10 precise evaluation of the CBPs in their LFs, the follicular localization of other CBPs, such as caldesmon, calcineurin, and calmodulin, and Ca2+ itself in FDCs has not been reported until now.
We examined Ca2+ localization in tonsillar secondary LFs using Ca2+-capture cytochemistry and observed extensive Ca2+ capture in the extracellular spaces. Some GC lymphocytes captured relatively little Ca2+, whereas FDCs and endothelial cells captured strikingly large amounts of Ca2+ in both their cytoplasm and nuclei, suggesting that [Ca2+]i may be regulated by CBPs.
Next, the follicular localizations of six CBPs were examined immunohistochemically. Caldesmon was found throughout the LF, and its localization on FDCs was confirmed by immunocytochemical examination of isolated FDC clusters. Immunoelectron microscopy revealed that it was present characteristically on the cytoplasmic processes of FDCs that twined around extracellular thick fibers and in the peripheral cytoplasm, presumably creating a high extension. FDCs twine around the extracellular fibers and connect to neighboring FDCs via desmosome-like junctions to build up a three-dimensional meshwork structure in the secondary LFs.6 Ogata et al35 demonstrated that FDCs express fibronectin and laminin receptors that bind to the reticulin and laminin fibers in secondary LFs. Furthermore, FDCs possess actin filaments.36 These data, including our results, suggested that caldesmon on actin filaments and extracellular matrix adhesion receptors on FDCs may be the principal ways whereby FDCs twine around extracellular fibers and generate tension in the peripheral cytoplasm, contributing to the formation and maintenance of the meshwork structure.
Our immunohistochemical experiment demonstrated other CBPs, such as S-100 protein, calcineurin, calbindin-D, and calmodulin in LFs. The BLZ showed intense immunostaining of six CBPs. S-100 protein was not detected in the ALZ, but large amounts were present in the DZ. The presence of these proteins in FDCs was confirmed by immunocytochemistry. Immunoelectron microscopy revealed diffuse immunostaining of S-100 protein, calcineurin, calbindin-D, and calmodulin in the FDC cytoplasm. Furthermore, in our ISH study, we detected calmodulin mRNA expression throughout secondary LFs, although we were unable to identify where exactly it was localized in FDCs. S-100 protein inhibits the phosphorylation of various proteins, such as tau protein and p87 protein,37,38 calcineurin is a protein phosphatase regulated by the Ca2+-calmodulin complex,39 calbindin-D modulate [Ca2+]i,40 and calmodulin regulates [Ca2+]i by activating Ca2+,Mg2+-ATPase11 and inositol 1,4,5-trisphosphate (InsP3) 3-kinase.41 These CBPs may regulate [Ca2+]i in FDCs and exhibit diverse functions, such as protein phosphorylation in FDCs, although, at the moment, their precise roles in FDCs and follicular lymphocytes are not clear.
Annexin VI was detected immunohistochemically throughout the LF, but the highest levels were present in the MZ. Immunoelectron microscopy revealed that it was localized on the surfaces of GC lymphocytes but not FDCs. Lin et al42 suggested that annexin VI is required for budding of clathrin-coated pits, the initial step of endocytosis.
Finally, we investigated the localization patterns of caldesmon and S-100 protein and compared them with that of Ki-M4p in follicular lymphomas. All of the neoplastic follicles examined contained Ki-M4p-positive FDCs, although this was weaker in grade III follicles than in grades I and II follicles. Caldesmon was expressed clearly in a lacy pattern in grades I and II neoplastic follicles but never in grade III follicles. In contrast, S-100 protein was expressed frequently in cells with dendritic morphology (presumably FDCs) in grade III neoplastic follicles and in a part of grade II follicles, but never in grade I follicles. In the GCs of non-neoplastic follicles, S-100 protein was confined to the DZ and BLZ, whereas the LZ was richer in Ki-M4p and caldesmon than the DZ. These data suggest that the caldesmon and S-100 protein expression patterns of grades I and III follicular lymphomas differ, and furthermore, it is likely that FDC distribution pattern in the LZ within non-neoplastic follicles may be similar to that in grade I neoplastic follicles, FDC distribution pattern in the DZ and BLZ within non-neoplastic follicles may be similar to that in grade III follicles, and grade II neoplastic follicles may be intermediate between grade I and grade III follicles.
In conclusion, we have demonstrated here that 1) [Ca2+]i is considerably higher in FDCs than in their surrounding follicular lymphocytes, 2) the BLZ shows intense immunostaining of CBPs in comparison with the other four zones in LFs, 3) FDCs express five CBPs in two different patterns, 4) caldesmon expression in FDCs is related to twining around extracellular thick fibers, 5) FDC meshworks in the non-neoplastic LZ are similar to those in grade I neoplastic follicles, and FDC meshworks in the non-neoplastic DZ and BLZ are similar to those in grade III follicles.
| Acknowledgements |
|---|
| Footnotes |
|---|
Accepted for publication April 28, 1999.
| References |
|---|
|
|
|---|
proteins. Ca2+ stimulates disulfide cross-linking between the S100b (ßß) subunit and the microtubule-associated
(2) protein. Biochemistry 1988, 27:2728-2736[Medline]
This article has been cited by other articles:
![]() |
S-I Okazaki, M Yamakawa, K Maeda, N Ohta, and M Aoyagi Expression of glucocorticoid receptors in non-neoplastic lymphoid follicles and B cell type malignant lymphomas J. Clin. Pathol., April 1, 2006; 59(4): 410 - 416. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-S. Chen, E. Coustan-Smith, T. Suzuki, G. A. Neale, K. Mihara, C.-H. Pui, and D. Campana Identification of novel markers for monitoring minimal residual disease in acute lymphoblastic leukemia Blood, April 1, 2001; 97(7): 2115 - 2120. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Manuelidis, I. Zaitsev, P. Koni, Z. Yun Lu, R. A. Flavell, and W. Fritch Follicular Dendritic Cells and Dissemination of Creutzfeldt-Jakob Disease J. Virol., September 15, 2000; 74(18): 8614 - 8622. [Abstract] [Full Text] |
||||
![]() |
O. W. Kamel Unraveling the Mystery of the Lymphoid Follicle Am. J. Pathol., September 1, 1999; 155(3): 681 - 682. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |