α-Hemoglobin-stabilizing Protein: An Effective Marker for Erythroid Precursors in Bone Marrow Biopsy Specimens.

RESEARCH ARTICLE

a-Hemoglobin-stabilizing Protein: An Effective Marker for Erythroid Precursors in Bone Marrow Biopsy Specimens Hongbo Yu, MD, PhD,* Jack L. Pinkus, PhD,w and Geraldine S. Pinkus, MDw

Abstract: Accurate analysis of the erythroid lineage is essential in evaluating bone marrow biopsies and can be particularly challenging in settings of dyserythropoiesis. a-Hemoglobin-stabilizing protein (AHSP) is an erythroid-specific chaperone protein and represents a potential specific marker for erythroid elements. This study defines the immunohistochemical profile of AHSP, as compared with an established erythroid marker CD71, in 101 bone marrow biopsies including normal marrows and cases of acute pure erythroid leukemia, acute erythroid/ myeloid leukemia, other types of acute myeloid leukemia, myelodysplastic syndrome, chronic myelogenous leukemia, other types of myeloproliferative neoplasm, chronic myelomonocytic leukemia, acute lymphoblastic leukemia, plasma cell neoplasm, and metastatic carcinoma. In acute pure erythroid leukemia, blasts in 7 of 11 cases showed similar reactivity for CD71 and AHSP, whereas less extensive reactivity was observed for AHSP as compared with CD71 in the remaining 4 cases. In normal marrows and other various disorders, reactivity for AHSP was similar to CD71 and was restricted to the erythroid lineage. Mature erythrocytes were negative for AHSP as were myeloblasts, lymphoblasts, nonerythroid hematopoietic marrow elements, plasma cells, and carcinoma cells. AHSP is an effective marker for detection of normal or abnormal erythroid precursors in bone marrow biopsies and is a useful addition to an immunohistochemical panel for assessment of neoplastic cells of possible erythroid derivation. Key Words: a-hemoglobin-stabilizing protein, CD71, marker for erythroid precursors (Appl Immunohistochem Mol Morphol 2016;24:51–56)

a-H

emoglobin-stabilizing protein (AHSP) is an erythroid-specific chaperone protein that facilitates incorporation of nascent a-globin into hemoglobin.1,2 AHSP contains 102 amino acids with a molecular weight of 12 kDa and is expressed in hematoReceived for publication July 16, 2014; accepted August 11, 2014. From the *Department of Pathology, UMass Memorial Medical Center and University of Massachusetts Medical School, Worcester; and wDepartment of Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA. The authors declare no conflict of interest. Reprints: Geraldine S. Pinkus, MD, Department of Pathology, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115 (e-mail: [email protected]). Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

Appl Immunohistochem Mol Morphol

poietic tissues exclusively by cells of the erythroid lineage. It forms a stable but reversible complex with free a-globin chain, which stabilizes its structure and limits its chemical reactivity. This interaction prevents a-globin chain aggregation and precipitation in vitro and in vivo and thus maintains its availability for incorporation into hemoglobin. The interaction between AHSP and a-globin chain also prevents oxidative damage due to oxygen species production.1–3 The pattern of AHSP gene expression during human erythropoiesis indicates increasing levels of expression during cell differentiation, with the highest expression at the stage where hemoglobin synthesis is maximal.4 A reduction in expression levels occurs when the cells lose their nuclear function and hemoglobin synthesis is reduced. a-Globin gene expression in these cells occurs with the same pattern. In anucleate reticulocytes and mature red blood cells, AHSP synthesis declines, and the protein is degraded.5 The high level of AHSP within erythroid precursors makes it a potential excellent marker for detection of erythroid elements within the bone marrow. Accurate analysis of the erythroid lineage is essential in evaluating bone marrow biopsies. In paraffin-embedded tissue, particularly in dyspoietic marrows, erythroid precursors may not be easily differentiated from other types of precursors on hematoxylin and eosin (H & E)-stained and/or Giemsa-stained sections and requires immunohistochemical analysis for this distinction. Current commonly used markers for erythroid precursors include CD71, glycophorin A (also referred as CD235a), and hemoglobin. Immunostains for CD235a and hemoglobin may be difficult to interpret due to reactivity of both markers for abundant mature non-nucleated erythrocytes in the marrow.6 Our previous study,7 along with others,8 have demonstrated that CD71 (transferrin receptor) is a useful marker to identify erythroid precursors in bone marrow biopsies and is characterized by distinct membranous staining in erythroid precursors, which diminishes in intensity with maturation, with an absence of CD71 reactivity in mature red blood cells. However, CD71 is not absolutely specific for erythroid elements in hematopoietic neoplasms5; thus, a panel of markers is advised. The goal of this study is to evaluate the distribution and specificity of AHSP immunoreactivity for erythroid precursors in a variety of clinical settings in paraffin-embedded bone marrow biopsies and to compare its staining

Volume 24, Number 1, January 2016

www.appliedimmunohist.com |

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

51


Yu et al

pattern with that for CD71. Prior studies5,7,8 have shown that CD71 is an effective, although not entirely specific, marker for erythroid precursors and is not expressed in mature erythrocytes, greatly facilitating interpretation in bone marrow biopsy specimens.

MATERIALS AND METHODS Cases were retrieved from the surgical pathology files of the Department of Pathology of UMass Memorial Medical Center, Worcester, MA, and the Department of Pathology of Brigham and Women’s Hospital, Boston, MA. This study was approved by the Institutional Review Board of University of Massachusetts Medical School and Brigham and Women’s Hospital. Representative H & E-stained and Giemsa-stained sections were reviewed to confirm the diagnoses. The neoplastic diagnoses were classified according to the WHO classification9; however, for simplicity of categorization, the cases of acute myeloid leukemia (AML) are designated by the former FrenchAmerican-British (FAB) classification10 in Table 1. Paraffin sections of 101 bone marrow biopsies fixed in formalin or Bouin’s solution were analyzed. Preliminary studies of bone marrow biopsies fixed in formalin, Bouin’s solution, B Plus fixative, or Zenker’s acetic acid solution revealed that reactivity for AHSP was preserved with all fixatives following heat-induced epitope retrieval. Specimens evaluated included 12 normal marrows, 11 cases of acute pure erythroid leukemia (FAB M6b), 20 cases of acute erythroid/myeloid leukemia (FAB M6a), 21 cases of other AML (including FAB subtypes M0-M5


and 1 case with myelodysplasia-related changes), 11 cases of myelodysplastic syndrome (MDS), 5 cases of chronic myelogenous leukemia (CML), 5 cases of other myeloproliferative neoplasm (MPN), 5 cases of chronic myelomonocytic leukemia; 2 cases of acute lymphoblastic leukemia, 5 cases of plasma cell neoplasm, and 4 cases of metastatic carcinoma. In all cases, immunoreactivity was evaluated for erythroid precursors, mature erythrocytes, blasts/tumor, and background tissue. Immunohistochemical studies were performed on 4um-thick paraffin-embedded tissue sections. Studies were performed following heat-induced epitope retrieval in a steamer (30 min, Retrieve-All-2; Covance Research Products, Dedham, MA). Slides were incubated for 1 hour at room temperature with antibodies against AHSP [rabbit polyclonal (1:100,000 dilution; Rockland Immunochemicals Inc., Gilbertsville, PA) and CD71 (clone H68.4, 1:2000 dilution; Invitrogen Corp/Life Technologies, Grand Island, NY)]. Detection was performed using the Rabbit PowerVision peroxidase system (30 min; Leica Biosystems Inc., Buffalo Grove, IL). Antibody localization was determined with a peroxidase reaction using the DAB+ system (Dako Corp., Carpinteria, CA). Sections were counterstained with hematoxylin. Pattern and intensity of staining for each antibody were evaluated.

RESULTS The results of immunohistochemical studies for AHSP and CD71 are summarized in Table 1. For each

TABLE 1. Comparison of Immunohistochemical Staining Patterns for AHSP and CD71 in 101 Normal and Neoplastic Bone Marrow Biopsies AHSP Diagnosis

CD71

No. Cases

Erythroids

RBCs

Background

Erythroids

RBCs

Background

Normal

12

MDS AML FAB M0 FAB M1 FAB M2 FAB M3 FAB M4 FAB M5 FAB M6a FAB M6b AML with MDS-related changes CML Other MPNw CMML ALL

11

12 Erythroids 11

0 RBCs 0

0 Blasts 0

12 Erythroids 11

0 RBCs 0

0 Blasts 0

4 9 2 1 2 2 20 11 1 5 5 5 2 Erythroids 5 4

0 0 0 0 0 0 0 0 0 0 0 0 0 RBCs 0 0

0 0 0 0 0 0 0 11* 0 0 0 0 0 Tumor 0 0

4 9 2 1 2 2 20 11 1 5 5 5 2 Erythroids 5 4

0 0 0 0 0 0 0 0 0 0 0 0 0 RBCs 0 0

0 0 0 0 0 0 0 11 0 0 0 0 0 Tumor 0 4

Plasma cell neoplasm Metastatic carcinoma

4 9 2 1 2 2 20 11 1 5 5 5 2 5 4

For each marker, the number of cases exhibiting relevant immunoreactivity is indicated. Cases of AML are listed according to their FAB designation. *Only a subset of neoplastic erythroblasts was reactive in four cases. wMegakaryocytes in 2 cases of primary myelofibrosis were negative for AHSP but focally positive for CD71. ALL indicates acute lymphoblastic leukemia; AML, acute myeloid leukemia; CML, chronic myelogenous leukemia; CMML, chronic myelomonocytic leukemia; erythroids, erythroid precursors; FAB, French-American-British; MDS, myelodysplastic syndrome; MPN, myeloproliferative neoplasm; RBCs, mature erythrocytes.

52 | www.appliedimmunohist.com

Copyright

r

2015 Wolters Kluwer Health, Inc. All rights reserved.




case, studies for AHSP and CD71 were performed simultaneously. In normal bone marrows, reactivity for AHSP was specific and restricted to the erythroid lineage and showed a cytoplasmic staining pattern. Mature erythrocytes were negative for AHSP (Fig. 1B). The highest expression level for CD71 was observed in early erythroid precursors and the lowest level in late normoblasts (Fig. 1C). Nearly all mature erythrocytes were negative for AHSP and CD71, greatly facilitating interpretation. The staining pattern for CD71 is predominantly membranous, whereas the staining pattern for AHSP is mainly cytoplasmic. Both antibodies were nonreactive in cells of myeloid lineage, megakaryocytes, and other marrow elements and both show no significant background staining. Staining patterns for AHSP and CD71 were evaluated in cases of acute erythroid leukemia (Fig. 2). Both markers highlighted erythroid precursors in all cases. In acute pure erythroid leukemia (FAB M 6b), 7 of 11 cases showed similar reactivity for CD71 and AHSP (Figs. 2A– C), whereas less extensive reactivity was observed for AHSP as compared with CD71 in the remaining 4 cases (Figs. 2D–F). However, both markers clearly demonstrated that the blasts were of erythroid lineage. In 20 cases of acute erythroid/myeloid leukemia (FAB M 6a), AHSP showed similar reactivity to CD71 (Figs. 2G–I). In other FAB subtypes of AML (Figs. 3A–C), MDS (Figs. 3D–F), CML, other MPN, chronic myelomonocytic leukemia, acute lymphoblastic leukemia, plasma cell neoplasm and metastatic carcinoma, AHSP showed similar reactivity to CD71, and served as an effective marker for detection of normal or abnormal erythroid precursors. Myeloblasts, lymphoblasts, and nonerythroid marrow elements, as well as plasma cells and carcinoma cells, were negative for AHSP. Studies for AHSP showed less background staining in comparison with CD71 in these FAB subtypes of AML cases. Megakaryocytes in 2 cases of primary myelofibrosis were reactive for CD71 but not for AHSP (Figs. 4A–C). No significant reactivity for AHSP was seen in metastatic carcinoma cells, whereas focal cytoplasmic staining for

a-Hemoglobin-stabilizing Protein

CD71 was observed in all 4 cases of metastatic carcinoma (Figs. 4D–E). Cases in the study were fixed either in formalin or Bouin’s solution, both of which were acceptable for AHSP studies. In formalin-fixed bone marrows, the intensity, specificity, and pattern of AHSP expression were comparable to those in the Bouin’s fixed material, including an absence of reactivity for mature erythrocytes. Thus, AHSP staining is a robust, specific marker for erythroid precursors in both Bouin’s and formalin-fixed bone marrow biopsies as well as in B Plus and Zenker’s acetic acid fixed specimens (personal observation).

DISCUSSION In this study, we demonstrate that AHSP is a highly effective marker for highlighting erythroid precursors in bone marrow biopsies. Immunoreactivity for AHSP is characterized by a distinct cytoplasmic staining pattern, which is easily recognizable and its expression is restricted to cells of erythroid lineage within bone marrow biopsies. Myeloblasts, lymphoblasts, and nonerythroid marrow elements, as well as plasma cells, carcinoma cells, and mature erythrocytes were negative for AHSP, greatly facilitating interpretation. However, in 4 of 11 cases of acute pure erythroid leukemia, AHSP was less sensitive than CD71 in characterizing the blast forms, although a subset of the blasts was AHSP positive. In cases of metastatic carcinoma, neoplastic cells were AHSP-negative but showed focal reactivity for CD71. In cases of primary myelofibrosis, megakaryocytes were AHSP-negative but were reactive for CD71; however, this pattern would likely not create a problem in interpretation of cell type. Thus, AHSP is highly specific for detection of cells of erythroid derivation, although less sensitive than CD71 in detecting an erythroid derivation for neoplastic erythroid elements. However, CD71, although more sensitive, is not entirely specific. The use of a panel of markers should obviate any problems in accurate interpretation. Although an aspirate count may be preferred over immunohistochemical staining of biopsies for determining

FIGURE 1. Immunoreactivity for a-hemoglobin-stabilizing protein (AHSP) and CD71 in a normal bone marrow biopsy. (A) Normal bone marrow biopsy with maturing trilineage hematopoiesis (H&E, 400). (B) Clusters of erythroid precursors show reactivity for AHSP with a cytoplasmic staining pattern (immunoperoxidase, 400; inset, 1000). (C) CD71 immunoreactivity shows a membranous and cytoplasmic pattern in erythroid precursors (immunoperoxidase, 400; inset, 1000). The vast majority of mature erythrocytes are non-reactive for both AHSP and CD71 (B and C). H&E indicates hematoxylin and eosin. Copyright

r




53


Yu et al


FIGURE 2. Comparison of a-hemoglobin-stabilizing protein (AHSP) and CD71 studies in acute erythroid leukemia. (A) Bone marrow biopsy with acute pure erythroid leukemia (H&E, 400). In this case, reactivity for AHSP (B) is similar to that noted for CD71 (C). In 4 of 11 cases, as noted in this case (D), only a subset of neoplastic cells was reactive for AHSP (E) as compared with CD71 (F) (B, C, E, and F, immunoperoxidase, 400; D, H&E, 400). (G) Bone marrow biopsy with acute erythroid/myeloid leukemia (H&E, 400). Studies for AHSP (H) and CD71 (I) show similar reactivity for erythroid precursors (H and I, immunoperoxidase, 400). H&E indicates hematoxylin and eosin.

the percentage of erythroid elements, immunostains on bone marrow biopsies may be necessary if adequate aspirate smears are not available, in cases with marked dyserythropoiesis, and for more accurate quantitation of the erythroid population. Several markers can be used to demonstrate erythroid lineage in paraffin sections, including AHSP, CD71, CD235a, hemoglobin, and spectrin. In the current study, we demonstrate that AHSP, similar to CD71, is also an effective marker for erythroid precursors. A previous study has demonstrated that CD71 is a superior marker as compared with CD235a and hemoglobin, particularly for ease in assessment, given its negativity in mature erythrocytes.7 Further, an earlier study showed that spectrin is also superior to CD235a for detection of erythroid precursors.11 Spectrin was found to be most highly expressed in early erythroid precursors (cytoplasmic), with decreasing expression in maturing forms (membranous), including mature erythrocytes. As mature


erythrocytes do express spectrin, it is most likely that immunohistochemical studies for spectrin will have a high background as compared with AHSP and CD71. Our results show AHSP is an effective marker of erythroid precursors in normal bone marrow biopsies and in erythroid blasts in acute erythroid leukemia. These findings are similar to those in a recent study from Raesset al.5 However, some of our results differ from their findings. In the prior study, CD71 positivity was noted for myeloblasts in AML cases and lymphoblasts in acute lymphoblastic leukemia cases. In the present study, reactivity for CD71 was limited to the erythroid lineage and did not show significant reactivity either in myeloblasts or lymphoblasts. In addition, both AHSP and CD71 reactivity were seen for megakaryocytes in primary myelofibrosis in their study. In our study, cytoplasmic staining for CD71 was noted in megakaryocytes in 2 cases of primary myelofibrosis, but no reactivity for AHSP in Copyright

r





a-Hemoglobin-stabilizing Protein

FIGURE 3. a-hemoglobin-stabilizing protein (AHSP) and CD71 studies in other types of AML and MDS. (A) Bone marrow biopsy with acute myeloid leukemia without maturation (H&E, 400). Studies for AHSP (B) and CD71 (C) are both negative in myeloblasts. Rare erythroid precursors are reactive for both markers. (B and C, immunoperoxidase, 400). (D) Bone marrow biopsy with MDS, refractory anemia with excess blasts-2 (H&E, 400). Studies for AHSP (E) and CD71 (F) highlight dysplastic erythroid precursors but are negative in myeloblasts (E and F, immunoperoxidase, 400). H&E indicates hematoxylin and eosin.

FIGURE 4. a-hemoglobin-stabilizing protein (AHSP) and CD71 studies in primary myelofibrosis and metastatic carcinoma. (A) Bone marrow biopsy with primary myelofibrosis (H&E, 400). Megakaryocytes are negative for AHSP (B), but show cytoplasmic staining for CD71 (C) (B and C, immunoperoxidase, 400; insets, 600). (D) Bone marrow biopsy with metastatic carcinoma (H&E, 400). Study for AHSP (E) is negative in carcinoma cells, but weak cytoplasmic reactivity for CD71 (F) is seen in tumor cells (E and F, immunoperoxidase, 400; insets, 600). H&E indicates hematoxylin and eosin. Copyright

r




55


Yu et al

megakaryocytes was detected in those cases. The prior study did provide evidence that CD71 is not absolutely lineage specific, although highly selective for erythroid precursors. It should be noted that different retrieval and detection systems were used in these studies and may have contributed to the variation in staining patterns. In our cases of acute pure erythroid leukemia, 7 of 11 showed similar reactivity for CD71 and AHSP for the neoplastic cells, whereas more extensive reactivity was observed for CD71 as compared with AHSP in the remaining 4 cases. This finding underscores the necessity of using multiple markers such as AHSP, CD71, CD235a, and hemoglobin for confirmation of erythroid lineage when analyzing a neoplastic process of uncertain lineage involving the bone marrow. In summary, AHSP is a highly effective marker for detection of cells of erythroid lineage in paraffin sections of bone marrow biopsies and is useful for delineation of an erythroid derivation for blast forms in cases of acute leukemia. REFERENCES 1. Kihm AJ, Kong Y, Hong W, et al. An abundant erythroid protein that stabilizes free alpha-haemoglobin. Nature. 2002;417: 758–763. 2. Favero ME, Costa FF. Alpha-hemaglobin-stabilizing protein: an erythroid molecular chaperone. Biochem Res Int. 2011;2011:373859.



3. Kong Y, Zhou S, Kihm AJ, et al. Loss of alpha-hemoglobinstabilizing protein impairs erythropoiesis and exacerbates betathalassemia. J Clin Invest. 2004;114:1457–1466. 4. dos Santos CO, Duarte AS, Saad ST, et al. Expression of alphahemoglobin stabilizing protein gene during human erythropoiesis. Exp Hematol. 2004;32:157–162. 5. Raess PW, Paessler ME, Bagg A, et al. a-Hemoglobin-stabilizing protein is a sensitive and specific marker of erythroid precursors. Am J Surg Pathol. 2012;36:1538–1547. 6. Pinkus GS, Said JW. Intracellular hemoglobin-a specific marker for erythroid cells in paraffin sections. An immunoperoxidase study of normal, megaloblastic, and dysplastic erythropoiesis, including erythroleukemia and other myeloproliferative disorders. Am J Pathol. 1981;102:308–313. 7. Marsee DK, Pinkus GS, Yu H. CD71 (transferrin receptor): an effective marker for erythroid precursors in bone marrow biopsy specimens. Am J Clin Pathol. 2010;134:429–435. 8. Dong HY, Wilkes S, Yang H. CD71 is selectively and ubiquitously expressed at high levels in erythroid precursors of all maturation stages: a comparative immunochemical study with glycophorin A and hemoglobin A. Am J Surg Pathol. 2011;35:723–732. 9. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: IARC Press; 2008. 10. Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol. 1982;51:189–199. 11. Sadahira Y, Kanzaki S, Wada H, et al. Immunohistochemical identification of erythroid precursors in paraffin embedded bone marrow sections: spectrin is a superior marker to glycophorin. J Clin Pathol. 1999;52:919–921.

Copyright

r



Effective identification and localization of immature precursors in bone marrow biopsy.

Bone marrow necrosis: ten-year retrospective review of bone marrow biopsy specimens.

Dysmegakaryopoiesis in myelodysplastic syndromes (MDS): an immunomorphometric study of bone marrow trephine biopsy specimens.

Alveolar rhabdomyosarcoma infiltrating bone marrow at presentation: the value to diagnosis of bone marrow trephine biopsy specimens.

Early reconstitution of haematopoiesis after allogeneic bone marrow transplantation: a prospective histopathological study of bone marrow biopsy specimens.

Routine histological compared with immunohistological examination of bone marrow trephine biopsy specimens in disseminated neuroblastoma.

Characterization of tissue findings in bone marrow biopsy specimens with small monoclonal B-cell populations.

Immunohistochemical detection of neuroblastoma in frozen sections of bone marrow trephine biopsy specimens.

Arsenic Primes Human Bone Marrow CD34+ Cells for Erythroid Differentiation.

Proposal of an appropriate decalcification method of bone marrow biopsy specimens in the era of expanding genetic molecular study.

Identification of cytomegalovirus in bone marrow biopsy.

Bone marrow biopsy in non-Hodgkin's lymphoma.

Micronuclei in bronchial biopsy specimens from heavy smokers: characterization of an intermediate marker of lung carcinogenesis.

Thioredoxin-interacting protein regulates the differentiation of murine erythroid precursors.

Schistosomal eggs identified on bone marrow biopsy.

Pain and anxiety during bone marrow biopsy.

Letter: Bone-marrow aspiration and biopsy.

An evaluation of cellularity in various types of bone marrow specimens.

Bone marrow biopsy in children: a study of 111 patients.

Differentiation of functionally active mouse T lymphocytes from functionally inactive bone marrow precursors. IV. Recovery of T-cell function from bone marrow precursors in a histo-incompatible environment.

Bone marrow trephine biopsy involvement by lymphoma: review of histopathological features in 511 specimens and correlation with diagnostic biopsy, aspirate and peripheral blood findings.

A marker chromosome in post-transplant bone marrow.

Prostatic acid phosphatase by radioimmunoassay tumor marker in bone marrow.

Adequacy of powered vs manual bone marrow biopsy specimens: a retrospective review of sequential marrow aspirates and biopsies in 68 patients.