Histologic and genetic assessment of explanted allograft valves

Histologic and Genetic Assessment of Explanted Allograf t Valves Jo& Neves, MD, Carolino Monteiro, PhD, Ricardo Santos, MSc, Ana Martins, MD, Shcia Ramos, MD, Teresa Ramos, MD, Carol Calta, MSc, Jo& Rueff, MD, and Jo50 Q. Melo, MD Institute do Cora@o, Hospital de Santa Cruz, Carnaxide, Portugal; and Departamento de Cenktica, Faculdade de Ci@ncias Medicas UNL, Lisboa, Portugal

A possible way of analyzing the immune response triggered by the allograft and the cellular viability is to compare immunocompetent and immunosuppressed patients, such as those having valve replacement and heart transplantation, respectively. These groups differ in immunosuppression therapy, preparation methods, valve hemodynamics. In the present study, we investigated polymerase chain reaction-amplified DNA flanking hypervariable (CA), regions obtained from valve leaflets taken from patients having valve replacement or heart transplantation and performed histologic analyses of the cells. In addition, we assessed an autograft valve to compare the hemodynamic effects on the cellular compo-

I t has been suggested that the long-term success of allograft heart valves directly correlates with the via-

bility of the leaflet fibroblasts [il. Therefore, much effort has been devoted to the development of sterilization and cryopreservation techniques for these valves to ensure the best possible degree of cell survival [2]. Nevertheless, degeneration remains the most frequent cause of failure [3]. One hypothesis for the mechanism of failure is that the immune response triggered by the allograft may affect cellular viability. A possible way of analyzing this conjecture is to compare immunocompetent and immunosuppressed patients, such as those having valve replacement (VR) and heart transplantation (HTx), respectively. In a recent study [4], polymerase chain reaction @‘CR)-amplified DNA flanking hypervariable (CA), regions obtained from valve leaflets taken from VR and HTx patients showed that whereas the latter contained a normal number of cells, with some cases including cells of both donor and recipient origin, the former contained very few cells of undetermined origin (Melo JQ et al, 1994, submitted for publication). The differences between these groups involve the administration of immunosuppression therapy, preparation methods, and valve hemodynamics. In the present study, we have investigated this issue further by performing parallel histologic analyses

Presented at the VI International Symposum for Cardiac Bioprostheses, Vancouver, BC, Canada, July 29-31, 1994.

Address reprint requests to Dr Melo, Department of Cardiothoracic Surgery, Hospital de Santa Cruz, Avenida Professor Reinaldo dos Santos, Carnaxide 2795 Linda-A-Velha, Portugal.

sition of the valve leaflet. We conclude that leaflet cellularity of the heart transplantation and autograft patients is superior to that of the valve replacement patients. These differences were consistent with the occurrence of an immune response in the valve replacement group, which was prevented or abrogated by immunosuppressive therapy administered to the heart transplantation group. However, it cannot be excluded that preparation procedures have a long-term effect on the extracellular matrix, leading to deterioration of cell adhesion and homing conditions.

(Ann Thorac Surg 1995;6O:SZ41-5)

of the leaflet cells and genetic determination of their origin. In addition, we assessed an autograft valve [5] to compare the hemodynamic effects on the cellular com- position of the valve leaflet.

Material and Methods

Thirteen explanted valves were retrospectively analyzed: seven VR, five HTx, and one autograft (AG). In the VR group, some of the valves were cryopreserved and the remaining were stored in an antibiotic solution until implantation. Three of the valves were from brain-dead donors, two from autopsies, and the remaining two from unknown sources. In the HTx group, all the hearts were obtained from brain-dead donors. The hearts were har- vested in sterile conditions, with no warm ischemic time, using 1 L of cold crystalloid cardioplegia, and were stored at 4°C in Ringer’s lactate. After heart transplantation, all patients received immunosuppression therapy with cy- closporine, azathioprine, and prednisone. Four patients incurred one to three treated rejection episodes. All HTx patients died with no evidence of rejection, except for 1 patient who died 1 month after surgery with signs of mild rejection. All valves from the VR and AC groups were removed during reoperations, and valves from the HTx group were collected during postmortem examination. These valves were obtained with approval from the national institutional review board. The AG valve was explanted because of major aortic regurgitation, due to the prolapse of one leaflet. The recipient, donor, and valve information for the groups is outlined in Table 1.

0 1995 by The Society of Thoracic Surgeons 0003-4975/95/$9.50 0003-4975(95)00274-O

S142 CARDIAC BIOPROSTHESES NEVES ET AL Ann Thorac Surg HISTOLOGIC/GENETIC ALLOGRAFT ASSESSMENT 1995;60:5141-5

Table 1. Clinical Summa y for the Donors and Recipients

Case

Implant Age (Y)

D R

Sex Blood Group Cold Ischemic Cryopreservation Follow-Up

D R D R Time (h) (4 ho)

Vl v2 v3 v4 V.5 V6 v7 Tl T2 T3 T4 T5 Al

23 72 M M A- Ai 21 62 M F A+ Ai 40 64 M M A+ A+ 30 44 M F A+ AA 46 17 M M A+ o-

. 51 F NA A- . 31 M NA A- 29 53 M M O- o+ 28 53 M M A+ A+ 19 42 F M A- A+ 25 60 M M A+ A -- 23 61 M M O- o+ . . 64 F Ai

48 51 2 24 19 6 24 30 2

48 307 12 624 0 73 NA 0 239 NA 0 230

2.0 1

2.2 . 2 2.8 5 1.5 . 46 2.0 . . . 60 0.1 . 9

A = autograft; D = donor; F = female; M = male; NA = not available; R = recipient; T = heart transplantation group; v= valve replacement group.

Two leaflets of each explanted aortic valve were used for histologic analysis. These were fixed with buffered formaldehyde, embedded in paraffin, and stained with hematoxylin-eosin and elastic van Gieson for routine histologic evaluations, including determination of endothelial cell content. An immunohistochemical study was performed to evaluate the lymphocyte population among the cells with the avidin-biotin immunoperoxide method using monoclonal antibodies LCA (CD45), L26 (CD20), and UCHL, (CD45RO) (Dako A/S, Glostrup, Denmark). Labeling with the monoclonal antibodies was done ac- cording to the manufacturer’s recommendations.

The third leaflet of each valve was used for genetic evaluation. This material was presented in paraffin- embedded blocks in 10 cases and as fresh tissue in three cases. These specimens were divided into fragments, one to four for the VR group and four to 21 for the HTx group, depending on the size and quality of the sample. Blood samples or small portions of native aortic tissue were collected from the recipient as a source of constitutional DNA for comparative studies. From solid tissue, DNA was obtained by the standard proteinase K method followed by phenol-chloroform extraction, whereas the guanidine-HCI method was used for blood [6]. We amplified DNA by PCR using primers of which one was fluorescently labeled. Oligonucleotide primers flanking the hypervariable (CA),, repeat regions [7-91 were syn- thesized using a 381 DNA synthesizer (ABI, Foster City, CA). A fluorescent dye (5-carboxy-fluorescein [FAM] or 2’,7’-dimethoxy-4’,5’-diclhoro-6-carboxy-fluorescein UOE]; ABI, Foster City, CA) was attached to the 5’ end of the PCR primers using aminolink 2 (ABI), and the primers were purified by high-performance liquid chroma- tography. The PCR amplifications were performed in a model 480 thermocycler (Perkin Elmer, Vaterstetten, Germany), and the obtained products were subjected to electrophoresis in an automated DNA sequencer (model 373A; ABI). The size of the (CA),-containing amplimers

was determined by referencing them to the internal lane size standards using GENESCAN 672 software [lo, 111.

Results

Histologic studies indicated that valve leaflets in the VR group lost their normal structure. There was evidence of hyalinization of the matrix. Four of these cases (V3, V5, V6, and V7) had focal calcification, in one case as early as 2 months after implantation. A small number of fibroblasts was found only in cases Vl, V2, and V3, and in these cases there was evidence of focal polymorphonuclear infiltration due to nonspecific acute endocarditis. Endothelial cells were not found in any of the VR cases (Fig 1A). In cases V2, V4, and V6, we observed lymphocytes, which were identified as T cells (Figs lB, 1C). One of these, case V4, was the only one in which cells from both donor and recipient were identified by their specific genetic profiles. In the HTx group, the histologic results were normal (Fig 1D). There was a slight decrease in the number of fibroblasts at the free edge of the leaflets (Fig 1E). The endothelium was focally maintained, and none of the leaflets showed inflammatory infiltration. The AG valve demonstrated conserved structure and cellularity, including endothelium, without leukocyte infiltration (Fig 1F).

The VR group was assessed genetically with markers D5S82, D5S299, and MCC. Results for cases V4 through V7 using marker D5S299 are shown in Table 2. Amplifi- cation yielding amplimers of three sizes and thus conclu- sive results of mixed cell origin was found only for case V4, and donor cells but not recipient cells were found for cases VS and V7. The remaining cases did not have more than two sizes of amplimers, or control DNA was not available (case V3), thus making results up to now inconclusive. The presence of donor cells in case V5 was inferred from the results obtained with marker D5S299: a 158-base pair sized amplimer was obtained from the frag-

Ann Thorac Sq 1995;60:!341-5

CARDIAC BIOPROSTHESES NEVESETAL 5143 HISTOLOGIC/GENETIC ALLOGMFT ASSESSMENT

Fig 1. (A) Sumple V4 showing an absence uf fibroblasts and endothelinl cells and hyalinization of the connective tissue. (BJ Sample V4 showing lymphocytes. CC) Staining of T cells for case V4 using anti-CD45RO monoclonul antibody. (0) Sample T5 showing normal histologic results. (EJ T5 showing a slight decrease in the number of fibroblnsts at the free edge and maintained elastic fibers. (FJ Sample Al showing con-

served structure and cellularity including endothelium, without leukocyte infiltration. (A, B, D, E, F: elastic van Gieson and hematoxylin and eosin, X 100 before 45% reduction; C: peroxidase staining, X 1,000 before 45% reduction.)

ments, whereas amplification of the recipient control yielded PCR products of 178 and 184 base pairs. For the HTx group, we obtained three different-sized amplimers in three cases (Tl, T3, and T5), indicating the presence of two cell populations. In case Tl, a lo- to 50-fold diIference in the area of one amplimer as compared with the other suggests that one of the populations was homozygous and the other

heterozygous [X2]. Table 3 summarizes the histologic and genetic results.

Comment

Tissue quality depends upon the integrity of the extracellular matrix and consequently the fibroblasts, which

s144 CARDIAC BIOPROSTHESES NEVES ET AL HISTOLOGIC/GENETIC ALLOGRAFT ASSESSMENT

Ann Thorac Surg 1995;60%41-5

Table 2. Genetic Results in the Vulve Replacement Group

With Marker D5S299, Indicating Polymerase Chain Reaction Amplification Area for Each Size Amplimev

Amplimer Size

Fragment 158 160 178 182 184

V4 A V4 B V5 A V5 B V5 control V6 A V6 control V7 A v7 c v7 control

.

. 1,785

.

4,551 2,086

11,837

1,979 13,641

. .

.

v = valve replacement group.

770 1,873 3,591 6,949

. 59,879 63,170

1,199 848 36,648 . 29,734

... ... ...

... ... ... 6,061 ... ...

are responsible for continuous degradation and synthesis of its components. Based on this concept, we have initi- ated studies to investigate the factors that affect cell survival in allograft heart valve replacement, which in- clude immunosuppression therapy, preparation methods, and valve hemodynamics. Results obtained from study of the AG valve showed preserved connective tissue and maintained cellularity, as opposed to the findings for the VR valves; therefore, the hemodynamic factor is an unlikely mechanism affecting cell survival because both the AG and the VR valves were implanted using similar surgical techniques.

Determination of cell origin can be used to assess the overall condition of the matrix as well as the immune response induced by the allogeneic tissue. In the absence

of an immune response, the matrix should be maintained with cells of donor origin and eventually be repopulated by recipient cells. In contrast, lymphocyte infiltrates, destruction of donor cells, and matrix damage occur in the presence of an immune response. Once altered, the matrix will not support cell adhesion and the tissue will become progressively acellular.

Genetic assessment allows us to determine the cellular origin by observing the number of different-sized amplimers obtained by PCR amplification of dispersed (CA),, hypervariable regions. The occurrence of three or four sized amplimers precisely indicates a dual (recipient and donor) cell origin. One amplimer can indicate a single homozygous origin or a mixed homozygous origin with the same sized amplimer. Two sized amplimers can indicate a single heterozygous, a dual homozygous, or a mixed heterozygous origin with the same sized amplimer, or a mixed heterozygous and homozygous origin with the same sized amplimer. Definite confirmation can be made only by using recipient and donor controls or by increasing the number of markers.

Histologically, three major differences were found between the VR and the HTx groups. The most striking was the number of fibroblasts, which was close to normal in the HTx group but strongly reduced in the VR group. A few fibroblasts were found in valves explanted after 2 to 6 months of implantation (Vl, V2, and V3). The endothelium was maintained in the HTx group but absent in the VR group, and finally, hyalinization of the matrix was observed in the VR group, sometimes with focal calcification. Genetic profiles indicated donor cells in three valves after 1,6, and 19 years of implantation. These cells were most likely fibroblasts; however, they were very scarce and were not identified histologically. This differ- ence in genetic fibroblast identification with histologic findings can be explained in that PCR amplification is a

Table 3. Summary of Histologic and Genetic Results for the Explanted Allograft Valves

Fibroblast Location”

Case Free Edge Base Endothelium Leukocytes

Vl + + NO P v2 t + No P, T cells v3 0 + NO P v4 0 0 No T cells v5 0 0 No No V6 0 0 NO T cells v7 0 0 No No Tl f-+ ++ Yes No T2 +++ ++ Yes No T3 +++ ++ Yes No T4 -++ ++ Yes No T5 -++ ++ Yes No Al -++ ++- Yes NO

s Fibroblast scale: 0 = acellular; i = few; +-I = moderate;

Number of Amplimers

Donor Recipient D5S82 D5S299 MCC D5S346

Cells Cells F C F C F C F C

? 7 7 Yes Yes 7 Yes Yes Yes Yes Yes YCS Yes

? ? ? Yes No ? NO Yes No Yes No Yes Yes

2 2 0 0 2 2 ND ND 1 1 0 0 1 1 ND ND 1 NC 1 NC 1 NC ND ND

2 NC 3 NC 3 NC ND ND 1 2 1 2 1 1 ND ND 2 2 2 2 2 2 ND ND 2 1 1 2 1 1 ND ND 3 2 2 2 2 2 ND ND

ND ND ND ND 2 2 1 2

ND ND ND ND ND ND 3 0 1 2 ND ND ND ND ND ND

ND ND 3 1 ND ND 2 2 . . . . . . .

+ * + = “(mm,

c = control; F = fragment; replacement group.

NC = n,, control; ND = not done; I’ = polymorphonuclear; T = heart transplantation group; v = valve

Ann Thorac Surg CARDIAC BIOPROSTHESES NEVES ET AL 5145 1995;6O:S711-5 HISTOLOGIC/GENETIC ALLOGRAFT ASSESSMENT

more sensitive technique and larger fragments are used for DNA extraction. Focal polymorphonuclear infiltration was observed in cases Vl, V2, and V3, which was due to acute endocarditis. Although genetic evaluation of cell origin for these cases using three markers was inconclusive, we can deduce from the histologic findings of polymorphonuclear infiltrates and T cells (V2 and V6) that recipient cells were present. I f equal-sized amplimers for each marker are found in the fragments and the control, a minimum of five markers are necessary to determine the origin of the amplified DNA. At present, this analysis is underway. Because of the short implantation time, it is most likely that the fibroblasts in cases Vl and V3 were from the donor. Case V4 was the only occurrence in the VR group in which cells of both donor and recipient origin were found, but the presence of T cells raises the question of whether the cells of recipient origin included fibroblasts as well.

The presence of T cells indicates that there is an immune response against the allograft. It is of interest to note that in cases V2 and V4, in which T cells were identified, the donor was male and the recipient female. This finding suggests that an immune response against the male histocompati- bility antigen H-Y occurred in these patients [13].

A dual cell origin was found in three cases in the HTx group, and the remaining cases were found to contain cells of only donor origin. Morphologic studies indicated that the number of endothelial cells was negligible as compared with the number of fibroblasts [12].

From the results obtained, we conclude that leaflet cellularity of HTx and AG patients is superior to that of VR patients. These differences were consistent with the occurrence of an immune response in the VR group, which was prevented or abrogated by immunosuppressive therapy administered to the HTx group. However, it cannot be excluded that preparation procedures, including ischemic time, antibiotic treatment, and cryopreservation, have a long-term effect on the extracellular matrix, leading to deterioration of cell adhesion and homing conditions.

This research was funded bv the Programa CIENCIA/JNICT, Funda+o Calouste Gulbengian, and The Portuguese Health Ministry. We also acknowledge support from the LusoAmerican Foundation and CMDTIUNL. Ricardo Santos is a recipient of a BIG scholarship (CIENCIAIJNICT).

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

O’Brien MF, Johnston N, Stafford G, et al. A study of the cells in the explanted viable cryopreserved allograft valve. J Cardiac Surg 1988;3(Suppl):279-87. McNally RT, Brockbank KG. Issues surrounding the preser- vation of viable allograft heart valves. J Med Eng Tech 1992;16:34-8. Angel1 W, Oury JH, Lamberti JJ, Koaiol J. Durability of the viable aortic allograft. J Thorac Cardiovasc Surg 1989;98: 48-56. Saiki RK, Gelford DH, Stoffel S, et al. Primer-direct enzy- matic amplification of DNA with a thermostable DNA polymerase. Science 1988;239:487-91. Ross DN. Pulmonary valve autotransplantation (the Ross operation). J Cardiac Surg 1988;3(Suppl):313-9. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning-a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1989. Breukel C, Tops C, van Leeuwen C, et al. CA repeat polymorphism at the D5S82 locus, proximal to adenomatous polyposis coli (AX). Nucleic Acids Res 1991;19:5804. Van Leeuwen C, Tops C, Breukel C, van der Klift H, Fodde R, Khan PM. CA repeat polymorphism at the D5S299 locus linked to adenomatous polyposis coli (APC). Nucleic Acids Res 1991;19:5805. Van Leeuwen C, Tops C, Breukel C, et al. CA repeat polymorphism within the MCC (mutated in colorectal can- cer) gene. Nucleic Acids Res 1991;19:5805. Mayrand PE, Corcoran KP, Ziegle JS, Robertson JM, Hoff LB, Kronick MN. The use of fluorescence detection and internal lane standards to size PCR products automatically. Appl Theoret Electrophoresis 1992;3:1-11. Ziegle JS, Su Y, Corcoran KP, et al. Application of automated DNA sizing technology for genotyping microsatellite loci. Genomics 1992;14:1026-31, Me10 JQ, Monteiro C, Neves J, et al. The allograft valve in heart transplantation and valve replacement. Genetic assessment of the origin of the cells using DNA profiles. J Thorac Cardiovasc Surg 1995;109;218-23. Simpson E, Gordon RD. Responsiveness to H-Y antigen, Ir gene complementation and target cell specificity. Immunol Rev 1977;35:59-75.

DISCUSSION

DR ALAIN F. CARPENTIER (Moderator) (Paris, France): Thank you for this very interesting presentation. 1 have one question. 1s the immune response more important when you have viable cells or when you do not have viable cells? Were you able to make these correlations?

DR MELO: We have been involved in these studies for the last 5 years. Immune response is more important when you have

viable cells. When we started the cryopreservation protocol we thought that viability was a very important issue, but we do not believe that any more.

SIR MAGDI H. YACOUB (London, England): This is a very interesting study. I am sure that PCR is a very good way of DNA fingerprinting; however, did 1 understand you to say that the presence of DNA implied tiable cells? I thought that was

implied in your message, but you could have DNA material from dead tissues; as a matter of fact, the same techniques with PCR were used to extract DNA from ancient Egyptian mummies, so it could be dead and still have DNA.

DR MELO: I was very cautious in my conclusion, because we could not show cells on our explanted homografts but still by PCR we could prove that DNA was present.

DR NELSON GIRALDO (Medellin, Columbia): Do you have any data on cell viability of the homograft before implantation?

DR MELO: Yes, we have a very careful protocol evaluating cryopreserved homografts, and we could demonstrate by cell cultures 60% cell viability before implantation.

Histologic and genetic assessment of explanted allograft valves

Documents

Transcript of Histologic and genetic assessment of explanted allograft valves