close

Вход

Забыли?

вход по аккаунту

?

SPLENIC MARGINAL ZONE EXPANSION IN B-CELL LYMPHOMAS OF GASTROINTESTINAL MUCOSA-ASSOCIATED LYMPHOID TISSUE (MALT) IS REACTIVE AND DOES NOT REPRESENT HOMING OF NEOPLASTIC LYMPHOCYTES

код для вставкиСкачать
JOURNAL OF
PATHOLOGY,
VOL. 179: 49-53 (1996)
SPLENIC MARGINAL ZONE EXPANSION IN B-CELL
LYMPHOMAS OF GASTROINTESTINAL
MUCOSA-ASSOCIATED LYMPHOID TISSUE (MALT) IS
REACTIVE AND DOES NOT REPRESENT HOMING OF
NEOPLASTIC LYMPHOCYTES
SUE
HARRIS, BRIDGET
S. WILKINS AND DAVID B. JONES
University Pathology, Southampton General Hospital, Southampton SO16 6 YD, U.K.
SUMMARY
It has been suggested that lymphocytes of mucosa-associated lymphoid tissue (MALT) arise from marginal zone cells and that
MALT-derived lymphomas may spread to other extra-nodal sites by homing to marginal zones in different tissues.’ Marginal zone
expansion has been observed in spleens removed during surgery for gastrointestinal MALT lymphoma, which was sufficiently extreme
in some cases to suggest neoplastic involvement. To investigate this phenomenon, polymerase chain reaction (PCR) amplification of
immunoglobulin heavy chain gene fragments was performed to demonstrate B-cell clonality in gastrointestinal MALT lymphomas and
spleens from 11 patients. Monoclonal PCR products were obtained from 9 of the 11 gastrointestinal tumours but from none of the
accompanying spleens. One additional spleen, for which the accompanying gastric lymphoma tissue was unavailable for review, yielded
a monoclonal product. However, obvious lymphoma deposits were present in this specimen, in addition to marginal zone enlargement.
It is concluded that splenic marginal zone expansion accompanying gastrointestinal MALT lymphoma is correctly interpreted as being
reactive. Splenic involvement by MALT lymphoma is uncommon and does not show preferential colonization of the marginal zone to
suggest homing of MALT-derived cells to this site.
KEY
WORDS-spleen;
marginal zone; mucosa-associated lymphoid tissue; lymphoma; polymerase chain reaction
INTRODUCTION
The concept of mucosa-associated lymphoid tissue
(MALT) has proved to be of value in the understanding
of malignant non-Hodgkin’s lymphomas (NHLs) arising
at extra-nodal sites, which show behavioural characteristics consistent with those of their presumed physiological counterparb2 Extra-nodal lymphomas are currently
subdivided for classification into those arising from
MALT sites, such as gastrointestinal tract, bronchus,
salivary gland, or thyroid gland, and those arising at
non-mucosal sites, such as brain and testis3 Lymphomas originating in MALT demonstrate an intimate
association with mucosal epithelial cells, forming
lympho-epithelial lesions by infiltration of the epit h e l i ~ mThey
. ~ tend to remain localized to the organ in
which they arise, with little spread to distant nodal sites,
unlike node-based lymphomas which often spread
extensively to other lymph nodes. When spread occurs,
lymphomas of MALT derivation may show selective
spread to other mucosal sites.4 It has been suggested
that the lymphocytes of MALT arise from marginal
zone cells and that MALT lymphomas may spread to
other extra-nodal sites by homing to marginal zones in
different tissue^.^
Addressee for correspondence: Dr B. S. Wilkins, CRC Research
Fellow in Pathology, University Department of Pathology, Level E,
South Block, Southampton General Hospital, Southampton
SO16 6YD, U.K.
CCC 0022-341 7/96/050049-05
01996 by John Wiley & Sons, Ltd.
The most common site of origin of MALT-derived
NHL is the stomach. This poses a challenge to the
proposed relationship between physiological MALT and
extra-nodal MALT NHL, as the normal stomach has no
organized MALT. It has been shown recently, however,
that the stomach can acquire MALT in response to
immunological stimulation, for instance as a result of
gastric colonization by Helicobacter pylori.6 Such reactive MALT is believed to be the physiological substrate
from which gastric NHL may subsequently develop.
The precise nature of the cells giving rise to MALT
lymphomas is not yet known. It has been proposed that
cells of the marginal zones found in reactive MALT
structures are the source.’ These cells are believed to be
the mucosal counterparts of marginal zone cells in the
spleen, and have been demonstrated in tonsil, stomach,
Peyer’s patches of the small intestine and mucosal
lymphoid nodules in the appendix and large bowel.4
The splenic marginal zone consists of a rim of
medium-sized lymphocytes occupying the outer edge of
B-cell nodules, forming an interface between the B-cell
compartment of the white pulp and the red pulp.’
Depending on its state of activation, the marginal zone
may be inconspicuous or may be many cells thick,
sometimes containing scattered larger blast cells. It is
believed to be important in T-independent immune
responses to carbohydrate-rich antigens, such as the
endotoxins of Gram-negative bacteria.’
As accumulations of marginal zone cells are a feature
of MALT and of splenic white pulp, but rarely of lymph
Received 14 November 1994
Accepted 8 November 1995
50
S . HARRIS ET A L
nodes, they may reflect the presence of an extra-nodal
lymphocyte recirculation pathway common to MALT
and the ~ p l e e n .Splenic
~
marginal zones are frequently
observed to be expanded in cases of gastric MALT
lymphoma; in view of the putative marginal zone cell
origin of such tumours, it would be of interest to know
whether this represented a reaction to the presence
of tumour in the stomach or direct involvement of
marginal zones by lymphoma.
We have investigated this question by using the
polymerase chain reaction (PCR) to amplify DNA from
gastrointestinal resections and splenectomy specimens
obtained from patients undergoing surgery for gastric,
small intestinal or colonic MALT lymphoma. In all but
one case, splenectomy was performed for entirely technical reasons and there was no morphological evidence
of splenic involvement by NHL. In most cases, however,
splenic marginal zones were expanded and had been
interpreted at the time of diagnosis as reactive (Fig. I).
DNA was amplified using PCR primers reactive
with variable and joining region determinants of the
immunoglobulin heavy chain gene. This technique
enabled us to establish the clonal (B lymphocyte) nature
of the gastric tumours and to determine, within the limit
of the sensitivity of the technique, the presence or
absence of a corresponding population of monoclonal
B lymphocytes within the spleens.
MATERIALS AND METHODS
Cases
Eleven cases, including nine gastric MALT lymphomas (seven low and two high grade), one high-grade
small intestinal MALT lymphoma and one high-grade
colonic MALT lymphoma, which had previously been
characterized extensively by immunophenotyping, were
selected from the diagnostic files of the Pathology
Department of Southampton General Hospital. The
tissues had been fixed in formalin and processed by
routine methods into paraffin wax. From each case one
block from the stomach/intestine and one from the
spleen were selected for PCR amplification. The spleen
only, with morphological evidence of high-grade tumour
involvement, was available from a further case in which
gastrectomy and splenectomy had been performed for
treatment of high-grade gastric NHL.
Blocks from a case of chronic lymphocytic leukaeniia
with gastric and splenic involvement were used as positive control material and a case of gastric carcinoma
with accompanying reactive spleen was used as a
negative control.
DNA extraction
Two 15pm sections were cut from each block and
placed in Eppendorf microcentrifuge tubes. To dissolve
the paraffin wax, 1 ml of xylene was added to each tube,
mixed gently and left to stand at room temperature
for 10 min. The samples were then centrifuged at
13 000 rpm in a microcentrifuge for 5 min. The xylene
was removed and replaced with 1 ml of absolute ethanol.
After gentle mixing, the samples were centrifuged at
13 000 rpm for 5 min. This step was repeated with fresh
ethanol. The ethanol was then aspirated and the sample
dried under vacuum.
DNA was extracted by incubation for 48 h at 37°C
with 400pg/ml proteinase K in 3OOpl of buffer (10 mM
Tris-HC1, pH 9.0; 50 mM KCI; 0.1 per cent Triton
X-100). Five hundred microlitres of phenol were then
added to each sample, mixed by vortex and spun at
600rpm in a microcentrifuge for 5 min. The aqueous
layer was removed to a fresh microcentrifuge tube
containing 500 p l of chloroform-iso-amyl alcohol (24:l),
mixed well and spun as above. Finally, the aqueous layer
was transferred to a fresh microcentrifuge tube and the
DNA precipitated overnight by adding 1/10 volume of
3 M sodium acetate (pH 5.2) and 2 volumes of cold
absolute ethanol.
Samples were then spun at 14000rpm at 4°C for
30 min. The ethanol was aspirated and the samples were
dried under vacuum. Each sample was re-dissolved in
50 pl of filtered-sterilized distilled water.
Primers
The primers used were consensus sequence primers for
the third framework regions at the 3’ end of the variable
segment (FR3A), and LJH for the joining segment of the
human immunoglobulin heavy chain gene (IgH), as
described by Wan el ul.,“’ with sequences as folIows:
Variable region:
FR3A
5‘-ACA CGG C(C/T)(G/C) TGT ATT ACT GT-3’.
Joining region:
LJH
5’-TGA GGA GAC GGT GAC C-3’
PCR-amplified fragments generated were between 70
and 120 base pairs in length.
P C R amplijication
Two rounds of amplification were performed using
a Hybaid thermal cycler. Appropriate positive and
negative controls were included in each amplification
reaction.
The first-round reaction mixture for each sample
consisted of 1 pl of target DNA; 250 ng of each primer
FR3A and LJH; 200 ,UM each of deoxyadenosine, deoxycytidine, deoxyguanosine and deoxythymidine triphosphate (dNTPs); 3.5 mM magnesium chloride; 1.0 units of
Taq DNA polymerase (Promega) in PCR buffer (10 mM
Tris-HC1, pH 9; 50 mM KCl, 0-1 per cent Triton X-100);
and sterile water to a total volume of 50pl. Each
reaction mixture was covered with mineral oil (Sigma).
The first round of PCR consisted of an initial denaturation step at 95°C for 6 min, followed by 30 cycles of
denaturation at 94°C for 45 s, annealing at 55°C for 45 s,
and extension at 72°C for 45 s. This was followed by a
final extension at 72°C for 5 min. Amplified products
SPLENIC MARGINAL ZONE IN MALT LYMPHOMA
51
Fig. l-(A) Reactive splenic white pulp from a case of gastric MALT NHL. Iinniunostained using DBA44, an antibody which preferentially stains
mantle zone B-cells and a population of red pulp B - c e l l~ . Note
'~
the wide (unstained) marginal zones. (B) Marginal zone expansion in a case of
gastric MALT NHL, considered to be equivocal on morphological grounds (H&E stained section). The gastric tumour was found to be clonal
following PCR with IgH primers, but the spleen yielded a polyclonal PCR product
from this single-round PCR were not always visible on a
polyacrylamide gel.
To increase the concentration of product, 1 pl of the
first-round product was used to set up a second PCR.
The second PCR consisted of 20 cycles of amplification,
performed as for the first round, using l p l of the
first-round product as the target DNA.
Following amplification, 15pl of each sample was run
on a 10 per cent polyacrylamide gel at 250 V for 2 h,
stained with ethidium bromide and photographed under
ultraviolet light.
Sensitivity studies
To investigate the sensitivity of our PCR method for
detecting clonal rearrangements from DNA extracted
from fixed paraffin-embedded material, the B-cell line
Nam 6" was diluted in a reactive lymphocyte population separated from fresh blood. While keeping the
total number of cells constant (2 x lo6), dilutions of
Nam 6 cells between 1 and 100 per cent were prepared.
Each dilution was fixed in 10 per cent buffered
formalin for 24 h before being centrifuged and the
formalin removed. The resulting pellet was then processed overnight through graded alcohols, cleared in
xylene, and finally impregnated with paraffin wax using
a Shandon Hypercentre. The wax-impregnated cell
pellets were then carefully transferred to microcentrifuge tubes and the DNA was extracted as described
previously.
Each sample was amplified using the FR3A/LJH
primers as described previously.
RESULTS
Spleen pathology
None of the 12 spleens from patients with gastrointestinal lymphoma was significantly enlarged, weighing
between 50 and 180 g. In the case with obvious morphological involvement by lymphoma, several nodules of
white tumour tissue were present within the paren-
chyma, the largest measuring 4 cm in diameter. In some
of the other spleens, fine miliary nodularity due to
prominence of white pulp structures throughout the
parenchyma was noted at macroscopic examination, but
most were described as having normal macroscopic
appearances.
Histologically, 9 of the 12 spleens showed striking
marginal zone expansion of white pulp nodules. This
reflected increased numbers of morphologically normal
marginal zone lymphocytes with occasional blast cells,
as seen in reactive splenic tissue, e.g., in some cases of
idiopathic thrombocytopenic purpura. The unusual feature in our cases was the marked degree of this marginal
zone expansion (Fig. 1). In each case, it had been
concluded at the time of diagnosis that the appearances
were reactive, but in three cases the possibility of
lymphomatous involvement had been suggested.
In two cases, the spleens appeared entirely normal,
with non-reactive white pulp nodules. The final case was
the case with obvious lymphomatous involvement. This
showed high-grade non-Hodgkin's lymphoma in the
large nodules and reactive marginal zone expansion of
intervening white pulp nodules. There was no evidence
of association of the tumour nodules with pre-existing
marginal zones.
Immunostaining showed evidence of light chain
restriction (kappa) only in the morphologically obvious
high-grade lymphoma deposits. Marginal zones in all
cases showed mixed kappa and lambda light chain
expression.
PCR analysis
All samples of stomach, intestine, and spleen yielded
amplifiable DNA. Nine of the 11 gastrointestinal
samples yielded clonal products with the FR3A/LJH
primers (lanes 5, 7, and 9 in Fig. 2); some cases showed
rearrangement of both IgH alleles (lanes 7 and 11).
Eleven out of 12 spleen specimens produced polyclonal
smears (lanes 6, 8, and 10 in Fig. 2). The spleen from
which a clonal product was obtained was the single case
which showed morphological evidence of high-grade
52
S. HARRIS ET A L
Fig. 2-Polyacrylamide gel showing FR3AILJH PCR products. Lane
O=DNA standard pBR322 Hae 111 digest; lanes 1 and 2=positive
control case (CLL), Gut (G) and spleen (S); lanes 3 and 4=negative
control case (adenocarcinoma). gut (G) and spleen (S); lanes 5 , 7, and
9=clonal gastrointestinal samples; lanes 6 , 8, and lO=polyclonal
spleen specimens corresponding to lanes 5 , 7, and 9; lane 11 =clonal
product from the spleen with morphological evidence of NHL
involvement
tumour NHL involvement (lane 11 in Fig. 2). Two
gastric cases, one high-grade and one low-grade MALT
lymphoma, produced polyclonal smears following PCR.
In another case, gastrectomy and splenectomy were
performed after chemotherapy, the initial diagnosis
having been made from gastric biopsies. Histology
showed no evidence of residual tumour in the stomach
or of splenic involvement, but PCR-amplified D N A
from the stomach produced a weak clonal band.
The positive control case (chronic lymphocytic
leukaemia involving stomach and spleen) yielded clonal
bands from both sites (lanes 1 and 2 in Fig. 2). The
negative control case (adenocarcinoma) produced polyclonal smears from stomach and spleen (lanes 3 and 4 in
Fig. 2).
Sensitivity analysis
When amplified with the FR3A/LJH primers, using
DNA extracted from fixed paraffin-embedded cells,
clonal products could be amplified when the Nam 6
population comprised at least 5 per cent of the reactive
B-cell population. Similar sensitivity has been reported
with DNA extracted from unfixed tissue^.'^"^
DISCUSSION
Our study has shown evidence of a monoclonal B-cell
population in only 1 of 12 spleens removed during
surgery for primary gastric or intestinal MALT lymphomas. In nine cases, a monoclonal IgH rearrangement
was detected in the accompanying gastrichntestinal
specimens. These findings suggest that the spleen is a
rare site of involvement by gastrointestinal lymphoma.
The expanded splenic marginal zones seen in most of
our cases were confirmed as being polyclonal and, hence,
presumably reactive. In the single case where a clonal
IgH band was generated from a spleen, there was
obvious morphological evidence of involvement of the
organ by high-grade lymphoma. Unfortunately, the
gastrectomy specimen from this case (dating from 1975)
could not be traced for histological review and inclusion
in the study, so the diagnosis of a lymphoma of MALT
origin may be questioned.
Spencer et al.’ reported a case of small intestinal
MALT lymphoma in which cells bearing an idiotypic
determinant of the primary tumour were found within
gastric mucosa, in splenic marginal zones and red pulp.
This finding was interpreted as indicating migration of
tumour cells via an extra-nodal lymphoid circulatory
pathway which included these sites. Our results suggest
that if such traffic occurs, it is beneath the level of
detection of our PCR method.
In one case, diagnosis of low-grade gastric MALT
lymphoma was made on the basis of a small biopsy, in
which there was insufficient tissue for DNA extraction.
Subsequent gastrectomy (with splenectomy) was performed due to persistent ulceration following chemotherapy. Histological examination of the gastrectomy
specimen revealed no evidence of residual lymphoma,
confirmed upon review, but a clonal IgH rearrangement
was detected by PCR, indicating the high sensitivity of
our PCR method.
It remains possible that DNA from the various subpopulations of splenic B lymphocytes is diRerentially
amplified by our PCR method. Thus, IgH DNA from
marginal zone cells might be underrepresented in the
PCR product and the sensitivity of the method for
detecting those cells might be less than the apparent
sensitivity as tested. Application of microdisse~tion,~~
to harvest marginal zone cells from histological sections,
or other techniques for enrichment of marginal zone
lymphocyte DNA in the PCR reaction mixture would be
required to exclude this possibility.
All of the tissue blocks yielded amplifiable DNA, but
gastric blocks from two cases, one of which had been
diagnosed as high-grade and the other as low-grade
MALT lymphoma, produced polyclonal smears rather
than clonal bands. It has been reported previously that
the framework 3 primer FR3A will amplify only 77 per
cent of MALT lymphoma^,'^ possibly because these
tumours are more prone to somatic mutations than are
follicular lymphomas. Consequently PCR failure may
result from mis-priming by the FR3A primer.
In conclusion, with the reservations concerning sensitivity and selectivity discussed above, our results do not
support the hypothesis that a shared recirculation pathway exists between the splenic marginal zones in gastric
or intestinal MALT. If such a pathway does exist, our
results suggest that neoplastic mucosal lymphocytes
traffic through it in small numbers, if at all. It seems to
be correct to interpret the striking marginal zone expansion seen in spleens from patients with primary gastric/
intestinal B-cell lymphoma as being reactive in nature.
We have also not found evidence to suggest that any
lymphoid cell population from the spleen, including
marginal zone cells, seeds MALT after undergoing
neoplastic change to give rise to MALT lymphomas.
53
SPLENIC MARGINAL ZONE IN MALT LYMPHOMA
REFERENCES
I . lsaacson PG, Spencer J. Low grade B cell lymphoma of gut-associated
lymphoid tissue (GALT): a model of the structure and migration pathways
of the B cell component of normal human GALT. Digestion 1990; 46
(Suppl 2): 274279.
2. Harris N. Extranodal lymphomas and mucosa-associated lymphoid tissue
(MALT): a unifying concept. Am J Surg Pathol 1991; 15: 879-884.
3. Harris NL, Jaffe ES, Stein H, ef a/. A revised European-American
classification of lymphoid naoplasms; a proposal from the International
Lymphoma Study Group. Blood 1994; 84: 1361-1392.
4. Isaacson, PG, Spencer J. Malignant lymphoma of mucosa-associated
lymphoid tissue. Histopathology 1987; 11: 445462.
5 . Spencer J, Diss TC, Isaacson PG. A study of the properties o f a low-grade
mucosal B cell lymphoma using a monoclonal antibody specific for tumour
immunoglobulin. J Pathol 1990; 160 231-238.
6. Wotherspoon AC, Ortiz-Hidalgo C, Falzon M, Isaacson PG. Helicobacter
pylori-associated gastritis and primary B cell gastric lymphoma. Lancet
1991; 338 1175-1176.
7. Harris NL. Low-grade B-cell lymphoma of mucosa-associated lymphoid
tissue and monocytoid B-cell lymphoma. Related entities that are distinct
from other low-grade B-cell lymphomas. Arch Pathol Lab Med 1993; 117:
771-775.
8. Van Krieken JH, Te Velde J. Normal histology of the human spleen. Am J
Surg Pathol 1988; 1 2 777-785.
9. Buryn G, Zeyers B, Van Furth R. Mechanisms of host defence
against infection with Streptococcus pneumonia. CIin Znfect Dis 1992; 14
251-262.
10. Wan JH, Trainor KJ, Brisco MJ, Morley AA. Monoclonality in B cell
lymphoma detected in paraffin wax embedded sections using the polymerase
chain reaction. J Clin Pathol 1992; 43: 888-890.
11. Klein G, Bombos L, Gothoskar B. Sensitivity of Epstein Barr virus (EBV)
producer and non-producer human lymphoblastoid cell lines to superinfection with EB-virus. Int J Cancer 1972; 1 0 4457.
12. Trainor K, Brisco M, Storey C. Monoclonality in B lymphoproliferative
disease, detected by PCR. Blood 1990; 7 5 2220-2222.
13. Wan JH, Sykes PJ, Orell SR, Morley AA. Rapid method for detecting
monoclonality in B cell lymphoma in lymph node aspirates using PCR.
J Clin Pathol 1992; 45: 420423.
14. Pan LX, Diss TC, Peng HZ, Isaacson PG. Clonality analysis of defined B
cell populations in archival tissue sections using microdissection and PCR.
Histopathology 1994; 2 4 323-327.
15. Diss TC, Peng H, Wotherspoon C, Isaacson PG, Pan L. Detection of
monoclonalitv in low grade B cell lvmohomas using the uolvmerase chain
reaction is dependent on primer selection and lymphoma type. J Pathol
1993; 169 291-295.
16. Treasure J, Wilkins BS, Delsol G, Jones DB. Immunomorphometric
analysis of normal and reactive splenic white pulp using a panel of
monoclonal antibodies. J Pathol 1993; 170 357A.
I
,
I
-
1~
Документ
Категория
Без категории
Просмотров
0
Размер файла
587 Кб
Теги
malti, doesn, associates, lymphomas, marginal, represented, tissue, mucosal, reactive, cells, expansion, neoplastic, gastrointestinal, lymphoid, zone, homing, lymphocytes, splenic
1/--страниц
Пожаловаться на содержимое документа