Received: 4 January 2019
Revised: 7 August 2019
Accepted: 12 August 2019
DOI: 10.1111/cup.13565
ORIGINAL ARTICLE
Membrane attack complex (C5b-9 complex or Mac), is strongly
present in lesional skin from patients with endemic pemphigus
foliaceus in El Bagre, Colombia
Ana Maria Abreu-Velez1
| Yulieth A. Upegui-Zapata2 | Carlos A. Valencia-Yepes3 |
Eduardo Upegui-Quiceno4 | Natalia-Regina Mesa-Herrera5 |
Alejandra M. Jiménez-Echavarria4 | César D. N. Pulido6 | Bruce R. Smoller7,8 |
Michael S. Howard1
1
Georgia Dermatopathology Associates,
Atlanta, Georgia
2
School of Medicine, Medical Research Institute,
University of Antioquia, Medellin, Colombia
3
Abstract
Background: El Bagre endemic pemphigus foliaceus (El Bagre-EPF) is a new variant
of endemic pemphigus foliaceus present in the El Bagre area of Colombia, South
Department of Education, University of
Antioquia, Medellin, Colombia
America. Here, we investigate the presence of complement/C5-b9 in lesional skin of
4
patients and matched controls from the endemic area. We also aim to compare the
Programa de Estudio y Control de
Enfermedades Tropicales Group, University of
Antioquia, Medellin, Colombia
5
Biogemol Institute and Department of
Physiology and Biochemistry, School of Medicine
University of Antioquia, Medellin, Colombia
6
Department of Cardiology, CES University,
Medellin, Colombia
7
Department of Pathology and Laboratory
Medicine, University of Rochester School of
Medicine and Dentistry, Rochester, New York
8
Department of Dermatology, University of
Rochester School of Medicine and Dentistry,
Rochester, New York
patient's autoantibody levels using indirect immunofluorescent titers (IIF) and correlate with the lesional presence of complement/C5b-9.
Methods: A case-control study was carried out by testing for the presence of complement/C5b-9 in lesional skin in 43 patients affected by El Bagre-EPF, as well as
43 matched, healthy controls from the endemic area. Skin biopsies were obtained
and evaluated via hematoxylin and eosin staining, and immunohistochemistry.
Results: The presence of complement/C5b-9 was observed in all cases of the
patients affected by El Bagre-EPF and was not observed in the controls from the
endemic area (P < 0.001). The patients' autoantibody titers utilizing IIF for IgG and
IgM showed correlation between higher autoantibody titers and stronger intensity of
Correspondence
Ana Maria Abreu-Velez, MD, PhD, Georgia
Dermatopathology Associates, 1534 North
Decatur Road, NE, Suite 206, Atlanta, GA
30307-1000.
Email: abreuvelez@yahoo.com
staining with complement/C5-b9 staining (P < 0.001).
Funding information
Embassy of Japan in Colombia; Georgia
Dermatopathology Associates, Atlanta, GA;
Hospital Nuestra Señora del Carmen, El Bagre;
Mineros SA; Municipality of El Bagre,
Antioquia, Colombia
C5-B9, endemic pemphigus foliaceus in El Bagre, MAC complex
Conclusion: Patients affected by El Bagre-EPF have lesional deposition of complement/
C5b, which correlates with disease severity and previously established serologies.
KEYWORDS
Abbreviations: BMZ, basement membrane zone; CS, complement system; C5b-9 complex or
MAC, membrane attack complex; DAPI, 4',6-diamidino-2-phenylindole; DIF, IIF, direct and
indirect immunofluorescence; EPF, endemic pemphigus foliaceus; El Bagre-EPF, endemic
pemphigus foliaceus in El Bagre; FITC, fluorescein isothiocyanate; FS, fogo selvagem;
H&E, hematoxylin and eosin; ICS, intercellular stain between keratinocytes; PF, pemphigus
foliaceus; SNPs, single-nucleotide polymorphisms.
© 2019 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
J Cutan Pathol. 2019;1–5.
wileyonlinelibrary.com/journal/cup
1
2
ABREU-VELEZ ET AL.
conjugated monoclonal antibodies to human total IgG or IgG4, as
1 | I N T RO D UC T I O N
described elsewhere9-11; (d) and test positive by immunoblotting for
From the first documentation of the autoimmune nature of pemphi-
reactivity against desmoglein 1 (Dsg1),9,10 (e) and for plakin molecules
gus (including pemphigus foliaceus [PF] and its variants including
including envoplakin, periplakin, and/or desmoplakins I-II. Additional
endemic pemphigus foliaceus [EPF]), the role of autoantibodies and
criteria included (f) that each patient's serum must immunoprecipitate
the complement system (CS) has been recognized as pathogenic in
a concanavalin A affinity-purified bovine tryptic 45 kDa fragment of
1-6
Multiple authors have described
Dsg1,12 and (g) each patient's serum had to yield a positive result
the association of the CS and its fixation by pemphigus antibodies,
using an ELISA test when screening for autoantibodies to PF anti-
documenting the role that the CS plays in pemphigus.1-6 In pemphi-
gens.13 In this study, one biopsy specimen was taken in each case; half
gus, as well as in Brazilian EPF (fogo selvagem), granular comp-
to be used for H&E and immunohistochemistry (IHC) stains, and half
lement/C3 deposits along intercellular spaces of the epidermis (ICS)
for immunofluorescence examination.
lesional skin and in cell cultures.
combined with IgG autoantibodies were documented as a diagnostic
hallmark.7,8 Here, we aim to determine the presence of the membrane
2.1 | H&E and IHC
attack complex/MAC (complement/C5b-9 complex), in lesional skin
using a case-control study in patients affected by a new variant of
We performed H&E and IHC stains as previously described.14 All sam-
endemic pemphigus foliaceus (El Bagre-EPF) in El Bagre, Colombia,
ples were run with positive and negative controls. The interpretation
South America. This variant of EPF, in contradistinction to fogo
of IHC was based on overall staining intensity15: intensity (0: no
selvagem, affects men more than women, and is not present in people
immunoexpression; 1+: weak immunoexpression; 2+: moderate immu-
younger than 20 years; further, in about one third of the patients,
noexpression; and 3+: strong immunoexpression first developed by
there are autoantibodies directed not only to the skin but to cell junc-
McCarty et al, and termed an H&E score. For our IHC staining we uti-
9-15
lized a Leica (Buffalo Grove, IL) staining system. Specifically, for pri-
The study rationale was based on the fact that we recently discovered
mary staining we utilized a Bond Max platform autostainer with bond
new intra-cytoplasmic and intranuclear autoantigens. Such data imply
polymer refined detection DS9800, a horseradish peroxidase linker
the opening of the cell membrane, such as may happen through the
polymer, and DAB chromogen (brown staining). Positive and negative
ring structure formed by a complement/C9 pore in the membrane, all-
controls were consistently included for study. For IHC, we utilized
owing free diffusion of molecules into and out of the cells.
antibodies for complement/C5b-9 at the dilution of 1:50, without
tions in multiple organs of the body, causing systemic anomalies.
antigen retrieval (Dako; Agilent Technologies, Santa Clara, CA).
2 | MATERIALS AND METHODS
2.2 | Statement on ethics
A case-control study was done by searching for the existence of
complement/C5-b9 in lesional skin in patients, and controls appropriately matched from the endemic area. We tested 43 patients
affected by El Bagre-EPF and 43 controls matched by age, sex,
demographics, comorbidities, work activities, weight, exposure to
chemicals, socioeconomic status and income, and food intake. About
one-third of El Bagre-EPF patients had a Senear-Usher-like disease,
as documented previously.9,10 The 43 controls from the endemic
A human quality assurance review board approved the studies at the
Hospital Nuestra Señora del Carmen in El Bagre and all participants
provided signed consent. The studies have been approved by the
appropriate institutional and/or national research ethics committees,
and performed in accordance with the ethical standards established in
the 1964 Declaration of Helsinki and its later amendments, or comparable ethical standards.
area were healthy individuals without skin disease. As internal controls, we also tested skin samples from healthy patient breast reduc-
3 | RESULTS
tions, five patients with cutaneous lupus, five with bullous
pemphigoid (BP), five with dermatitis herpetiformis (DH), and five
Figure 1A shows a typical clinical hyperpigmented form in a patient
with pemphigus vulgaris (PV).
affected by El Bagre-EPF. Figure 1B shows a typical H&E stained
Skin biopsy specimens from the patients and controls were evalu-
biopsy, showing small intra-corneal blisters, as well as acantholysis in
ated by hematoxylin and eosin (H&E), direct and indirect immunofluo-
the epidermis. In Figures 1C,D, we show some of the patterns of
rescence (DIF, IIF), and immunoblotting, immunoprecipitation, and
positivity seen by IHC staining using the complement/C5b-9
enzyme-linked immunosorbent assay (ELISA). Only patients meeting
antibody.
the following diagnostic criteria for El Bagre-EPF were included; spe-
The patients with a larger percentage of compromised skin
cifically, they had to: (a) display clinical and epidemiological features
(according to the Lund and Browder skin burn area chart)16 showed
described for the disease; (b) live in the endemic area;9,10 (c) have
higher titers of serum autoantibodies, as well as stronger reaction to the
serum displaying intercellular staining (ICS) between epidermal
complement/C5-b9 antibody. The data parallel the clinical forms of pem-
keratinocytes and the basement membrane zone (BMZ) of the skin
phigus (according to Viera's clinical classification for fogo selvagem).17
demonstrated via either DIF or IIF using fluorescein isothiocyanate
(a) The clinically inactive form (no apparent clinical compromise, but still
3
ABREU-VELEZ ET AL.
F I G U R E 1 A, A clinically
hyperpigmented form in a patient
affected by El Bagre-EPF with
hyperkeratotic plaques on the back (white
arrow). B, An hematoxylin and eosin
(H&E) stain of a patient affected by El
Bagre-EPF showing the epidermal edema
(black arrow) and a small subcorneal
blister (red arrow; ×400). C, IHC stain
using antibody to C5-b9 shows positive
staining (indicated in fuschia) in the
corneal layer (red arrow; +++), BMZ
(green arrow; +++), and the extracellular
matrix (black arrow; ++; ×100). D, AntiC5-b9 staining at higher magnification
highlighting positive staining in the
corneal layer (black arrow) and the BMZ
(green arrow; ×400). E, Positive antiC5-b9 staining on extracellular matrix
fibers and/or mesenchymal-endothelial
cell junctions (black arrows; ++),
sebaceous gland basement membrane
(red arrow; ++), and eccrine glands (blue
arrow; +++) in El Bagre-EPF. F, A
representative section of normal skin
showing negative anti-C5-b9
immunohistochemical staining (×200)
positive via our DIF with ICS [positive staining +] and serum antibody
antibody (as well as presence in specific parts of the epidermis, the der-
titers 1:20 to 1:40. (b) The localized, prurigoid, and hyperpigmented
mis, and the skin appendages with the intensity of positivity).
forms (9%-24% of the skin involved; our positive staining for
complement/C5-b9 was ++, present in the locations presented in
Table 1). (c) The Senear-Usher and generalized forms including
erythrodermic, bullous exfoliative, and hyperkeratotic (from 24%-100%
skin involvement; our complement/C5-b9 staining was +++, and our
serum titers of autoantibodies ran between 1:320 and 1:640).
3.1 | Assessing the autoantibody levels in relation to
disease severity and therapeutic response in El BagreEPF pemphigus patients
The more active clinical cases displayed not only intercellular staining
In Table 1, we show the results of our IHC staining in El Bagre-EPF,
between the keratinocytes (ICS), but also to the BMZ, to dermal neu-
with controls from the endemic area, as well as from patients with other
rovascular bundles, to dermal skin appendages and their cell junctions,
autoimmune skin diseases. The table displays the localization of immuno-
to neuroreceptors, and to mesenchymal-endothelial dermal cell
reactivity, and the strength of the staining using the complement/C5b-9
junctions.
4
ABREU-VELEZ ET AL.
T A B L E 1 IHC staining showing the number of positive stain in the skin and its appendages sites, correlation with skin severity affecting
percentage of skin accordingly with the scale use for burns, IIF serologic titers of autoantibodies in this study using C5-B9 antibody
Number of El
Bagre-EPF cases
Strength
of staining
Positivity to the skin
sites and its appendices
Controls from the
endemic area
Lupus
BP
DH
PV
5/43
(+++)
Intracorneal stain
0/43
0/5
0/5
4/5
0/5
43/43
(+++)
Intercellular stain between
the keratinocytes
0/43
0/5
0/5
0/5
5/5
25/43
(++)
BMZ
0/43
5/5
5/5
Below the sub
epidermal blister
mainly in the
papillary dermis (5/5)
0/5
43/43
(++)
Neurovascular bundles
0/43
3/5
3/5
3/4
3/5
23/43
(++)
Sebaceous gland basal membrane
0/43
3/5
0/5
0/5
0/5
31/43
(++)
Sweat glands and their ductus
0/43
3/5
3/5
0/5
3/5
32/43
(++)
Mesenchymal-endothelial cell
junctions in the dermal
connective tissue
0/43
0/5
4/5
0/5
0/5
34/43
(+++)
Dermal spindled cells
0/43
0/5
2/5
0/5
0/5
Abbreviations: BMZ, basement membrane zone; BP, bullous pemphigoid; DH, dermatitis herpetiformis; PV, pemphigus vulgaris.
4 | DISCUSSION
pathway, with the contribution of mannose-binding lectin (MBL and
MBL-associated proteases).19-22 All three pathways lead to the activation
We previously reported that in El Bagre-EPF the immune response in
of the complement C5, C6, C7, C8, and C9 proteins, resulting in the
acute cases begins with IgG autoantibodies. However, all cases progress
sequential assemblies of C5b-7, C5b-8, and C5b-9 on the target cell. This
toward chronicity, and often show relapsing episodes. We documented
process results in cell membrane pore formation.19-22 The mechanism of
that the immune response eventually becomes polyclonal, displaying
pore formation contrasts with pore formation via proteins such as perforin
autoreactivity directed toward epidermal keratinocytic intercellular junc-
and the cholesterol dependent cytolysins, where it is believed that
tions as well as the dermal-epidermal BMZ, and the BMZs of dermal skin
pre-pore assembly happens prior to the concurrent release of the trans-
appendages. In addition, autoreactivity is noted against neurovascular
membrane regions.22 The activation of complement and the subsequent
supply cell junctions, and against multiple skin neural receptors.18 Other
formation of complement/C5b-9 open channels on cell membranes,
immunoglobulins and complement components such as IgM, IgE, IgD,
usually leading to cell death. However, when the number of channels
complement/C1q, complement/C3c, complement/C4, and kappa and
assembled on the surface of nucleated cells is limited, sublytic comple-
lambda light chains are seen, in addition to the previously reported IgG.7
ment C5b-9 can induce cell cycle progression by activating signal trans-
Often weak ICS staining with IgG is observed in chronic cases. In addition
duction pathways and transcription factors, and inhibiting apoptosis.22
to this staining, other immunoglobulins and complement as well as
Recently, patients affected by PF including EPF, were evaluated for
deposits of fibrinogen and albumin are also seen targeting unique mole-
992 single-nucleotide polymorphisms (SNPs) of 44 CS genes, genotyped
cules in other parts of the skin. As reported previously, in about one-third
through microarray hybridization in 229 PF patients, and 194 controls.23
of the El Bagre-EPF patients the immune response is directed not only to
After excluding SNPs with minor allele frequency they found some gene
the skin, but also to several cell-junction and neural receptors in most
expression (<1%), out of a Hardy-Weinberg equilibrium in controls or in
organs and tissues.18 The patients with a more extended percentage of
strong linkage disequilibrium (r2 ≥ 0.8); 201 SNPs remained for logistic
compromised skin (according to Lund and Browder skin burn surface
regression. The authors also reported increased serum levels of comple-
16
scale)
show higher titers of serum autoantibodies as well as stronger
reactivity to the complement/C5-b9 antibody. This is indicative of epi-
ment C3, and C5-b9 deposition in lesions observed more frequently in
the PF and EPF patients in comparison with the normal controls.23
18
Of interest is the correlation we found between positivity of other
when the initial antibodies may damage the cell plasma membrane
autoimmune blistering and skin diseases such lupus, BP, DH, and PV
exposing internal cell molecules to the immune system.19
and the positivity of the staining for complement C5-b9 that perfectly
tope spreading and/or exposure to new autoantigens as shown by us
The CS contains multiple plasma- and membrane-bound proteins and
matched what we had reported on IHC staining with other immuno-
receptors working in the vanguard of host defense, killing pathogens,
globulins and complement.24 Indeed, our findings suggest that in
opsonizing and changing cells and microorganisms, and linking innate
these other autoimmune skin diseases, the immune response may also
19-22
immunity to adaptive immune responses.
The CS is stimulated in
three ways: via the classical pathway, which comprises the complement
be more complex than the classically described and currently accepted
mechanisms.
proteins C1, C2, C3, and C4; the alternative pathway, with the involve-
In this study, we confirm previous records that determined that
ment of complement C3 and protein factors B, D, and P; and the lectin
late components of the complement cascade such as C5-b9 are seen
5
ABREU-VELEZ ET AL.
in skin biopsies from PV and PF.25 In El Bagre-EPF, we have noted
intracellular and nuclear antigens that are not yet characterized. We
hypothesize that this may be the result of the attachment of C5-b9 to
the plasma membranes, exposing antigens by either antigen spreading
epitopes and or possibly by neoantigen formation during cell membrane damage. We believe this process exposes cell membrane, cytoplasmic and nuclear antigens resulting in the targeting of normally
protected antigens by dysregulated (intolerant) lymphocytes.
5 | CO NC LUSIO N
This study suggests that C5-b9 deposition measured by IHC staining
may be a biomarker for more intense disease activity both clinically
and immunologically in patients affected by El Bagre-EPF, and possibly in other autoimmune skin blistering diseases.
CONF LICT OF IN TE RE ST
None.
ORCID
Ana Maria Abreu-Velez
https://orcid.org/0000-0002-7692-4133
RE FE R ENC E S
1. Cormane RH, Chorzelski TP. Bound complement in the epidermis of
patients with pemphigus vulgaris. Dermatologica. 1967;134(6):
463-466.
2. Cormane RH, Chorzelski T. Localization of complement in epithelium
by means of immunofluorescence in patients with pemphigus. (preliminary report). Przegl Dermatol. 1967;54(4):432-436.
3. Kawana S, Janson M, Jordon RE. Complement fixation by pemphigus
antibody. I. in vitro fixation to organ and tissue culture skin. J Invest
Dermatol. 1984;82(5):506-510.
4. Kawana S, Geoghegan WD, Jordon R. Complement fixation by pemphigus antibody. III. Altered epidermal cell membrane integrity mediated by pemphigus antibody and complement. J Invest Dermatol.
1986;86(1):29-33.
5. Krasny SA, Kumar V, Tarfare N, Beutner EH, Chorzelski TP. Intercellular complement C3 binding in normal skin of patients without
pemphigus. J Am Acad Dermatol. 1986;14(1):52-58.
6. Xia P, Jordon RE, Geoghegan W. Complement fixation by pemphigus
antibody. V. Assembly of the membrane attack complex on cultured
human keratinocytes. J Clin Invest. 1988;82(6):1939-1947.
7. dos Reis VM, Cucé LC, Rivitti EA. Anatomopathology and direct and indirect
immunofluorescence of lesions of endemic pemphigus foliaceus resistant
to corticoid therapy. Rev Inst Med Trop Sao Paulo. 1991;33(2):97-103.
8. Messias-Reason I, Bosco DG, Nisihara RM, Jakobsen LH, PetzlErler ML, Jensenius JC. Circulating levels of mannan-binding lectin
(MBL) and MBL-associated serine protease 2 in endemic pemphigus
foliaceus. Clin Exp Dermatol. 2008;33(4):495-497.
9. Abreu Velez AM, Hashimoto T, Bollag W, et al. A unique form of
endemic pemphigus in northern Colombia. J Am Acad Dermatol. 2003;
49(4):599-608.
10. Abreu Velez AM, Beutner EH, Montoya F, et al. Analyses of autoantigens in a new form of endemic pemphigus foliaceus in Colombia.
J Am Acad Dermatol. 2003;49(4):609-614.
11. Hisamatsu Y, Abreu Velez AM, Amagai M, Ogawa MM, Kanzaki T,
Hashimoto T. Comparative study of autoantigen profile between Colombian and Brazilian types of endemic pemphigus foliaceus by various biochemical and molecular biological techniques. J Dermatol Sci. 2003;32(1):
33-41.
12. Abreu Velez AM, Javier Patiño P, Montoya F, et al. The tryptic cleavage product of the mature form of the bovine desmoglein
1 ectodomain is one of the antigen moieties immunoprecipitated by
all sera from symptomatic patients affected by a new variant of
endemic pemphigus. Eur J Dermatol. 2003;13(4):359-366.
13. Abreu-Velez AM, Yepes MM, Patiño PJ, et al. A cost-effective, sensitive and specific enzyme linked immunosorbent assay useful for
detecting a heterogeneous antibody population in sera from people
suffering a new variant of endemic pemphigus. Arch Dermatol Res.
2004;295(10):434-441.
14. Abreu-Velez AM, Yepes-Naranjo MM, Avila IC, et al. Tissue inhibitor
of metalloproteinase 1, matrix metalloproteinase 9, αlpha-1 antitrypsin, metallothionein and urokinase type plasminogen activator
receptor in skin biopsies from patients affected by autoimmune blistering diseases. Our Dermatol Online. 2013;4(3):275-280.
15. Vijayashree R, Aruthra P, Ramesh Rao KA. Comparison of manual and
automated methods of quantitation of oestrogen/progesterone
receptor expression in breast carcinoma. J Clin Diagn Res. 2015;9(3):
EC01-EC05.
16. Lund CC, Browder NC. The estimation of areas of burns. Surg Gynecol
Obstet. 1944;79:352-358.
17. Viera JP. Consirecoes sobre o penfigo foliaceo no Brasil. Sao Paulo [Different
aspects of Brazilian pemphigus foliaceus]. Sao Paulo:EmpresaGraficada
RevistadosTribunais; 1948:220.
18. Abreu-Velez AM, Howard MS, Yi H, Florez-Vargas AA. Patients affected
by a new variant of endemic pemphigus foliaceus have autoantibodies
colocalizing with MYZAP, p0071, desmoplakins 1-2 and ARVCF, causing
renal damage. Clin Exp Dermatol. 2018;43(6):692-702.
19. Muller-Eberhard HJ. Molecular organization and function of the complement system. Annu Rev Biochem. 1988;57:321-347.
20. Shin ML, Rus HG, Niculescu FI. Membranes attack by complement:
assembly and biology of the terminal complement complexes. In:
Lee A, ed. Biomembranes. JAI Press; 1996:123-149.
21. Spicer BA, Law RHP, Caradoc-Davies TT, et al. The first transmembrane region of complement component-9 acts as a brake on its selfassembly. Nat Commun. 2018;9(1):3266.
22. Tegla CA, Cudrici C, Patel S, et al. Membrane attack by complement:
the assembly and biology of terminal complement complexes.
Immunol Res. 2011;51(1):45-60.
23. Bumiller-Bini V, Cipolla GA, de Almeida RC, Petzl-Erler ML,
Augusto DG, Boldt ABW. Sparking fire under the skin? Answers from
the association of complement genes with pemphigus foliaceus. Front
Immunol. 2018;9:695.
24. Abreu-Velez AM, Googe PB Jr, Howard MS. Immunohistochemistry
versus immunofluorescence in the diagnosis of autoimmune blistering
diseases. Our Dermatol Online. 2013;4(suppl. 3):585-595.
25. Kawana S, Geoghegan WD, Jordon RE, Nishiyama S. Deposition of
the membrane attack complex of complement in pemphigus vulgaris
and pemphigus foliaceus skin. J Invest Dermatol. 1989;92(4):588-592.
How to cite this article: Abreu-Velez AM, Upegui-Zapata YA,
Valencia-Yepes CA, et al. Membrane attack complex (C5b-9
complex or Mac), is strongly present in lesional skin from
patients with endemic pemphigus foliaceus in El Bagre,
Colombia. J Cutan Pathol. 2019;1–5. https://doi.org/10.1111/
cup.13565