with human cytomegalovirus (HCMV) before and af- ter transplantation, kidneyspecimens were studied by in situ hybridization with SS-labeled DNAprobes rep-.
American Journal of Pathology, Vol. 132, No. 2, August 1988 Copyright 0 American Association of Pathologists
Inflammatory Cells in Transplanted Kidneys Are Infected by Human Cytomegalovirus JOHN W. GNANN,Jr., MD,*JARL AHLMEN, MD,t CHRSTIAN SVALANDER, MD,* LARS OLDING, MD,§ MICHAEL B. A. OLDSTONE, MD,* andJAY A. NELSON, PhD
From the Department ofImmunology, Research Institute of Scripps Clinic, LaJolla, California,* the Department ofNephrology, Central Hospital, Skovde,t the Department ofPathology, Sahlgren Hospital, G5teborg,t and the Department ofPathology, Karolinska Institute, Stockholm,§ Sweden
To determine which cells in kidney grafts are infected with human cytomegalovirus (HCMV) before and after transplantation, kidney specimens were studied by in situ hybridization with SS-labeled DNA probes representing HCMV immediate-early and late genes. Pretransplantation biopsies and serial posttransplantation biopsies were obtained from 7 renal grafts. All of the transplant recipients were HCMV-seronegative at the time of transplantation and all developed primary HCMV infections. HCMV nucleic acids were not detected in biopsies taken from the healthy donor kidneys before transplantation. However, biopsies taken at various interva after transplantation showed abun-
dant hybridization with HCMV immediate-early and late gene probes. Virtually all of the hybridizing cells were mononuclear inflammatory cells in the interstitial spaces ofthe kidney. Occasional hybridization was seen with renal tubular or glomerular cells. No cytomegalic cells were seen. Biopsy specimens taken after systemic anti-HCMV chemotherapy with phosphonoformate showed no uniform reduction in HCMV gene expression. These studies demonstrate that the principal HCMV-infected cells in kidneys ofrenal transplant patients with primary HCMV infections are infiltrating inflammatory cells. (Am J Pathol 1988, 132: 239-248)
HUMAN CYTOMEGALOVIRUS (HCMV) is an important cause of morbidity and mortality in patients who receive renal allografts.'" Approximately 50% ofseronegative transplant recipients develop primary HCMV infection; reactivation of endogenous latent HCMV (or re-infection) occurs in over 85% of patients who are seropositive at the time of transplantation.6 Disease caused by HCMV in this immunocompromised population ranges from a mononucleosis syndrome to severe retinitis, to fatal interstitial pneumonitis.7-9 Several key questions regarding the pathogenesis of HCMV infection in renal transplant recipients are still unanswered. The source of HCMV infection in transplant recipients remains incompletely understood. The virus may potentially be transmitted by the graft, by transfused blood, or by person-to-person contact. HCMV-seropositive transplant recipients can, in addition, reactivate endogenous latent HCMV. Studies with murine cytomegalovirus (MCMV) have shown that latent virus can be reactivated by an allogenic reaction in vi-
tro.'0 Further experiments with animal models demonstrated that latent murine or rat cytomegalovirus are transferable through kidney allografts" -13 and that latent MCMV can be reactivated from mouse kidney.14 Epidemiologic studies of human renal transplantation also suggest that the donor kidney is the most likely source of HCMV for previously seronegative recipients. An HCMV-seronegative patient who receives a kidney from a seropositive donor is at higher risk for HCMV infection (33-83% infected) than a seronegative patient who receives a graft from a seronegative donor (0-30% infected).6"5-'7 Moreover, Supported in part by USPHS grants AI-21640 (JAN) and AI-07007 and AG-04342 (MBAO). JAN is a recipient of Junior Faculty Award JFRA-1 15 from the American Cancer Society. JWG is supported by USPHS Training Grant P32
NS-07078. Accepted for publication March 31, 1988. Address reprint requests to John W. Gnann, Jr., MD, Department of Immunology - IMM6, Research Institute of Scripps Clinic, La Jolla, CA 92037.
239
240
GNANN ET AL
pairs of patients who received kidneys from the same cadaveric donor postoperatively shed HCMV strains with identical restriction enzyme analysis patterns, suggesting that the virus was transmitted with the kidneys.8,"9 This epidemiologic evidence is strong, but indirect. Most efforts to directly demonstrate HCMV in healthy renal tissue have been unsuccessful;6'2022 a single report of recovery of HCMV from healthy kidneys2' remains unconfirmed. If the renal allograft is indeed a vehicle for transmitting HCMV, the virus may be latent in renal parenchymal cells (endothelial cells, tubular epithelial cells, tissue macrophages) or in peripheral blood leukocytes trapped in the graft. Although renal transplant patients frequently shed HCMV in their urine, previous pathologic studies have not established which cells in the genito-urinary tract support virus replication. Routine light microscopy cannot clearly distinguish HCMV-infected cells unless they are end-stage cytomegalic cells with intranuclear inclusions. Indeed, histologic evidence of HCMV infection in renal grafts is often difficult to demonstrate, even in renal biopsies taken from viuric
patients.24'25 To investigate these problems further, the technique of in situ hybridization with HCMV-specific probes was used to identify virus-infected cells in biopsies from renal grafts. The study was designed to determine whether: 1) HCMV-infected cells can be detected in healthy renal tissue, 2) a predominant virus-infected cell type is definable in graft biopsies from recipients with primary HCMV infection, 3) there is sufficient infection of renal parenchymal cells to account for the diminished renal function that has been associated with HCMV infection in renal allograft recipients, and 4) anti-viral chemotherapy with phosphonoformate (PPF) alters HCMV expression in renal tissue. The principle conclusion of this study is that the predominant HCMV-infected cells in posttransplantation renal grafts are infiltrating mononuclear inflammatory cells.
Materials and Methods Kidney Biopsy Specimens The 7 patients studied underwent renal transplantation at Sahlgren Hospital, Goteborg, Sweden. They were selected for inclusion in this in situ hybridization study on the basis of availability of adequate amounts of biopsy tissue. All were documented to be HCMVseronegative before transplantation. One patient (3) received a graft from an HCMV-seropositive livingrelated donor, and the remainder of the patients re-
AJP * August 1988
ceived cadaveric grafts. Five of the donors (Table 1) were HCMV-seropositive and 1 was seronegative (no additional specimen was available from this cadaveric donor to absolutely rule out a false-negative result). Only leukocyte-depleted packed red blood cells were used for transfusions, a practice documented to reduce the incidence of transfusion-related HCMV infection.26 All patients developed active HCMV infection (documented by seroconversion and/or positive virus cultures) within 10 weeks of transplantation (Figure 1) and were treated with PPF. Small wedge biopsies were taken from the grafts at the time of transplantation. Percutaneous needle biopsies of the graft were obtained when clinical and laboratory findings suggested acute rejection or HCMV disease. Additional biopsies were also obtained after completion of PPF therapy. Wedge biopsies were also taken from grafts removed because of irreversible rejection. The tissues were fixed in 4% paraformaldehyde (in sodium cacodylate buffer 0.1 M, pH 7.4), dehydrated in serial alcohol washes, and embedded in paraffin. Sections were examined for evidence of rejection according to criteria proposed by Porter.27 Cellular rejection was classified by light microscopy as mild, moderate, or severe. For in situ hybridization studies, 4-,u sections were mounted on acid-cleaned poly-L-lysine-coated glass slides with 0.1% Elmer's white glue (Borden Inc., Columbus, OH) and stored at 4 C. Slides were labeled with a code number and submitted for in situ hybridization in random order. Specimen identification codes were not broken until all sections had been hybridized and scored. In Situ Hybridization The in situ hybridization procedure has been described previously.28'29 Briefly, sections were deparaffinized in xylenes and hydrated in graded ethanol-water washes. The specimens were permeabilized with 0.2 N HCI (10 minutes), 1% Triton X-100 (1.5 minutes), and proteinase K (1.0 ,ug/ml for 20 minutes at 37 C). The sections were then covered with hybridization mix containing 50% formamide, 5X hybridization salts (0.9 M NaCl, 50 mM NaH2PO4, 5 mM EDTA), 5X Denhardt's solution (0.2% polyvinylpyrrolidone, 0.02% Ficoll, 0.02% bovine serum albumin), 10% dextran sulfate, 250 ,g/ml salmon sperm DNA, 125 ,g/ml HeLa cell RNA, 30 U/ml heparin, 10 mM dithiothreitol (DTT), and 1 X 108 cpm/ml of a heat-denatured 35S-labeled DNA probe. The sections were cover-slipped with Gel-Bond film (FMC Corporation, Rockland, MA), sealed with rubber cement,
CMV IN TRANSPLANTED KIDNEYS
Vol. 1 32 * No. 2
241
Table 1 -HCMV Serologic Status of Kidney Recipients and Donors
Recipient 1 2 3 4 5 6 7
Preoperative HCMV serology*
Source of
Donor HCMV
Perioperative
kidney
serology*
blood transfusions
Negative Negative Negative Negative Negative Negative Negative
Cadaver Cadaver Uving-related donor Cadaver Cadaver Cadaver Cadaver
Positive Negative Positive Positive Unknown Positive Positive
None 3 units PRBCt None 1 unit PRBC 2 units PRBC 1 unit PRBC None
Presence of HCMV-specific IgG determined by standard commercially-available ELISA and/or IFA. Unknown, no serum specimen available. t PRBC, leukocyte-depleted packed red blood cells.
and incubated in a humidified chamber at 37 C for 18 hours. After hybridization, coverslips were removed, and the slides were washed twice in 2X SSC (0.3 M NaCl, 0.03 M sodium citrate for 15 minutes at 21 C), twice in 0. 1 X SSC (15 minutes at 21 C), once in 0.1 X SSC (10 minutes at 60 C), once in 0.1 X SSC (15 minutes at 37 C), and once in 2X SSC (10 minutes at 21 C). The slides were then dehydrated in graded ethanol-water washes, air-dried, and dipped in NTB-2 photographic emulsion (Eastman Kodak Co., Rochester, NY). After 4 days of exposure at 4 C, the slides were developed with Kodak D- 19, fixed, counterstained with Gill-2 hematoxylin, and cover-slipped. The slides were examined by light microscopy for characteristic silver grain formation. The relative frequency of hybridizing cells in a section was visually scored on a semi-quantitative scale: 0 (no hybridizing cells), 1 + (rare hybridizing cells), 2+ (0.01-0.1% of inflammatory cells in a section showed hybridization), 3+ (0.1-1% of inflammatory cells in a section showed hybridization), 4+ (more than 1% of inflammatory cells in a section showed hybridization).
protein and a related 71 kd phosphoprotein.35 The IE and late probes do not contain sequences homologous with human DNA and do not cross-hybridize.33'36 Probe DNA was subcloned from genomic viral DNA cosmids37 and nick-translated with deoxyadenosine 5'-(a [35S]thio)triphosphate and deoxycytidine 5'-(a[35S]thio)triphosphate.38 A negative control probe was prepared by nick-translating pUC18 that contained no viral DNA insert. Specific activity of these probes was approximately 1 X 108 cpm/,ug of DNA. Probes were stored in 10 mM DTT at -70 C for up to 2 weeks.
HCMV Probes HCMV genomic DNA probes encoding immediate-early (IE) and late genes were used in this study (Figure 2). HCMV IE genes are transcribed prior to viral protein synthesis in restricted areas of the genome.30'3' These genes are expressed in permissive and non-permissive infections.28'32 The IE gene probe used in this study was the 10 kb EcoRI J fragment from HCMV strain AD 169, a region that encodes at least five IE messages and is abundantly transcribed during the IE phase of replication.33'34 Late genes encode viral structural proteins and are transcribed after the onset of viral DNA replication.30'3' The late gene probe, also derived from strain AD 169, was a 6.6 kb BamHI R fragment that detects 4.0 kb and 1.9 kb late mRNAs coding for the 65 kd phosphorylated matrix
Results
HCMV Serology and Cultures Sera were assayed for HCMV IgG and IgM antibodies by standard immunofluorescence and enzymelinked immunosorbent assays. Urine specimens for HCMV culture were processed in standard fashion and inoculated onto human foreskin fibroblast cells, which were monitored for the appearance of characteristic viral cytopathic effect.
HCMV nucleic acids were not detected in any biopsies taken from healthy donor kidneys prior to implantation (Table 2). No hybridization with HCMV IE or late gene probes was seen with these specimens (Figure 3A). In contrast, renal biopsies taken after transplantation during episodes of active HCMV infection and/ or acute rejection showed abundant hybridization with HCMV probes. The majority of the hybridizing cells were mononuclear inflammatory cells in the interstitial spaces of the kidney (Figure 3B, C, D). In the sections with the most extensive hybridization, approximately 5% of infiltrating mononuclear cells were positive for HCMV. In other sections, only 0.01-1.0% of inflammatory cells were positive for HCMV. The
242
GNANN ET AL
AJP * August 1988
j,Jon '85
jDoc| l 'g t Febj1
|
Mellitus
Prednisolone
.
Biopsy'
July
June
Kidney disease
May
p
Biopsy'
Transplant'
'84
Apr
CMV 1gM +, leulkopenia
Fever,
May
Mar Dled
Begin PPf Interstitial pneumonia
2 yoM c
Biopsy', graft removed
Biopsy Biopsy'
Transplant'
43yoF Nov '84
Aug
Sept
Prednisolone
I
CMV
+
1gM
Nov
Tt
Bgh Fever,
Oct
j
Cyclosporlne
Repeat PPF
CMV urine culture
+
abrl. LFTs, interstitial pneumonia Transplant'
Blopsy', graft removed
Biopsy'
June May '83 33 yoM Poiycystic Kidney disease Cyclosporhne A
July
J
Sept
Aug
j
Oct
Nov
--Hi
I
t
Prednisolone
I
t
Begin CMV Igo + PPFC CMV tgM
Fever,,
*bnl LFTt, Interstitial pneumonia
4
D
Nov 84
35yoM
Cyclosporkse
Feb
Jan
ec
Diabetes
Meliltus
BIopsy' Graft removed
Biopsy' Biopsy
Transplant'
Prednlsolone
M
May
Apr
_
t
T
AzathoprIneFever,
A
DIed
Begin
abnl. LFTs
PPF
CMV IgM
+
CMV urIne culture +
Biopsy' Biopsy' Biopsy
Transplant' May
Apr 84
5
20yoF
|
I
June | I
uly
Cyclosporlne A
Prednlsolone
Azathloprlne
ATG
Graft removed | Aug
Sept
II ~ ~I
Figure 1-The clinical courses of the 7 renal transplant patients included in this study are shown on these time lines. Biopsies examined by in situ hybridization are marked with an asterisk. yo, years old; F, female; M, male; PPF, phosphonoformate; CMV, human cytomegalovirus; abnl. LFTs, abnormal liver function tests; GN, giomerulonephritis; ATG, antithymocyte globulin. Positive tests are designated by +.
Oct
CMV
Begin PPF CMV leukopenla, 1gM
IgG
Fever,
+
abnl. LFTs
6
28yoF Nephritis
Transplant' Biopsy' Oct Sept s8j1 I I I I
Biopsy' Nov I
Azathloprine
Predrdsolone
t
T
Fever |CV
Dec I
Jan 86 I
'
1gM
+
CMV IgG + I
Transplant'
7 7
57o
Nephrltls
Biopsy'
I Nov i Azathloprlne Prednhiolone
Cyclosporine A
-
ATG
Dec +1 I I
Biopsy', graft removed Jan '86 Feb
t
Begin
Ii I
~~~~~~~~~~~culture+
Mar
Apr
I
t
CMV IgG
t CMV urine
CMV urine culture +, abnl. LFTs. Begin PPF.
Oct 85
Mar
Feb I
+
IPPFI Fever, CMV tgM + leukopenia CMV urine culture +
frequency of hybridizing cells did not correlate with the severity of cellular rejection (Table 2). The distribution of hybridizing cells in the sections was not uniform. Rather, hybridizing cells tended to occur in focal clusters. In biopsies that sampled both renal cortex and medulla, hybridizing cells were more frequently found in the cortex. Hybridization was occasionally seen with tubular epithelial cells (Figure 4A) and rarely with cells in the glomerular tuft (Figure 4B). Approximately 800 glomeruli were present in posttransplantation biopsy sections studied with HCMV probes, but only 3 glomeruli demonstrated clear hybridization. The possibility that the hybridiz-
ing cell in the glomerular tuft was actually an inflammatory cell could not be ruled out. A few examples of hybridization with apparent vascular endothelial cells were noted, but precise identification of the cell type was difficult. The specificity of the HCMV probes was demonstrated by the absence of hybridization when the labeled plasmid was used as a probe (Figure 3F). No typical cytomegalic cells were noted by light microscopy. Efforts at precise identification of infected cells by double-labeling the same specimen with immunocytochemical techniques plus in situ hybridization failed because of poor antigen preservation in the for-
CMV IN TRANSPLANTED KIDNEYS
Vol. 132 * No. 2
243
HCMV Strain AD169 DNA Probes for In Situ Hybridization 1.95Kb IE RNA 1.41Kb IE RNA 1.70Kb IE RNA 2.25Kb IE RNA 1.70Kb IE RNA
1.9Kb L RNA 4.0Kb L RNA
D I - 33Kb Banm Hi R, 6.6Kb
A Figure 2-A physical map for the EcoRI fragments of HCMV DNA (strain AD169) is shown with the relative positions of the DNA fragments used for in situ hybridization and the transcripts they encode. The immediate-early probe (EcoRI J fragment) and the late probe (BamHl R fragment) were subcloned from genomic viral DNA cosmids.
4
J
| ,7K -
7mfig iol
'h
III K IcIlSIOIbM E IMIP IVI
I
1 IA
II
I
so
100
I
I
I
I
I
0.0
0.1
0.2
0.3
0.4
I
D
I
H
8
I~~ 150 Kllobases I
c
v
|RIIde- J IftHqXITT
I
0.6 Map Unit 0.5
C
I LrIFl EEco RI
~~~~~ I I
200
235
I
I
I
I
0.7
0.8
0.9
1.0
Probes used to detect different kinetic classes of HCMV transcripts
malin-fixed, paraffin-embedded tissues. The authors plan additional prospective studies with the doublelabeling techniques to identify the infected cells in fresh biopsy material. Biopsies obtained after PPF therapy still contained HCMV-positive cells (Figure 3E). While the frequency of hybridizing cells in posttherapy biopsies was lower in some cases (eg, Table 2, patient 4), this finding was not consistent. These findings indicate that PPF therapy failed to eliminate HCMV from the kidney, despite favorable clinical responses. The in situ hybridization protocol was modified to identify the predominant nucleic acid species hybridizing with the IE probe. Heating the specimen at 90 C for 3 minutes before adding the probe mixture caused no obvious increase in the amount of hybridization observed. However, pretreatment of the section with ribonuclease (100 ,ug/ml at 37 C for 30 minutes) resulted in a marked reduction (although not elimination) of hybridization. These data indicate that hybridization occurred primarily with viral mRNA rather than with viral genomic DNA. Discussion In situ hybridization experiments clearly demonstrated that mononuclear inflammatory cells are the predominant HCMV-infected cells in renal grafts during the first 3 months after transplantation. The subpopulation of mononuclear cells that hybridized with the HCMV probes has not yet been identified precisely, but the round cell infiltrate in renal grafts has been shown previously to consist primarily of T lymphocytes and monocytes/macrophages.39-4' The relative importance of HCMV
infection and host-vs.-graft response as stimuli for the influx of inflammatory cells seen in these biopsies is not known. Table 2-In Situ Hybridization of Biopsies from Transplanted Kidneys Using HCMV-Specific Probes
Recipient 1
Probe
Pretransplant
Posttransplant pre-PPF*
Posttransplant post-PPF
CMV-IE CMV-Late Control
Ot 0
+++ +++
+ +++
0
0 (mild)
0 (moderate)
++++ ++++
+++ ++
0 (mild)
0 (mild)
++++ ++
+ +++
0 (mild)
0 (severe) 0
(normal4t 2
3
CMV-IE CMV-Late Control
CMV-IE CMV-Late Control
0 0 0 (normal) 0 0 0
(normal) 4
5
6
7
CMV-IE CMV-Late Control CMV-IE CMV-Late Control CMV-IE CMV-Late Control
CMV-IE CMV-Late Control
0 0 0 (normal) 0 0 0 (normal) 0 0 0
(normal) 0 0 0 (normal)
++++ +++
+
0 (moderate) 0 0 0 (mild)
0 (severe)
+ +
+ +
0 (mild)
0 (mild)
+++ +
+++ ++
0 (mild)
0 (moderate)
0 (severe) + ++
Phosphonoformate therapy. t Relative frequency of hybridizing cells (approximate percentage of infiltrating cells showing hybridization): + (1.0%).
t Extent of cellular rejection (normal, mild, moderate, severe).27
AJP * August 1988
GNANN ET AL
244
i
4'
:.. *
:
*,,
'!
x !
.
...-3
. _-w a - *. 9
, *
Vol. 132 * No. 2
CMV IN TRANSPLANTED KIDNEYS
245
Figure 3-Photomicrographs of in situ hybridization of kidney biopsies using HCMV-specific probes. The histologic diagnosis given is based on examination A-Pretransof the entire section; all diagnostic features may not be present in the photomicrograph (original magnifications are shown in parentheses). plantation biopsy (patient 1), IE probe; normal histology; no hybridization. (X200) B-Posttransplantation biopsy (patient 1), IE probe; mild cellular rejection, C-Posttransplantation biopsy (patient 1), IE probe; moderate cellular rejection, no inclusions; hybridizing no inclusions; many hybridizing cells. (X200) cells are marked by dark grains. (x400) D-Posttransplantation biopsy (patient 1), late probe; mild cellular rejection, no inclusions; note absence of hybridizing cells in glomerulus. (X400) E-Postphosphonoformate therapy biopsy (patient 7), late probe; moderate cellular rejection with fibrosis, no F-Posttransplantation biopsy (patient 1), pUC1 8 control probe; moderate cellular inclusions; hybridization with mononuclear inflammatory cells. (X600) rejection, no inclusions; no hybridization. (X200)
Transplant recipients may shed large amounts of HCMV in their urine for months or years without evidence of viremia or HCMV-induced disease. Therefore, some cell type(s) in the urinary tract must be capable of supporting HCMV replication. HCMV clearly can infect renal tubular epithelial cells under some circumstances (eg, cytomegalic inclusion disease of the newborn25'42) and HCMV infection of human mesangial cells has been demonstrated in vitro. However, in the studies reported here, infection of renal parenchymal cells was detected infrequently. The authors speculate, therefore, that activated mononuclear cells may serve as host cells for HCMV replication in renal grafts. All biopsies that were positive for HCMV hybridized with both the IE and late gene probes. Since late genes are not transcribed in the HCMV replication cycle until after the onset of viral DNA synthesis, this hybridization with the late probe indicates that these leukocytes may be permissively infected. Peripheral blood mononuclear cells (PBMNs) from renal transplant recipients with active HCMV infections also hy-
bridize with both IE and late gene probes. In contrast, PBMNs from healthy HCMV-seropositive individuals hybridized with the IE gene probe28 but rarely hybridized with the late gene probe (Gnann J, Schrier RD, Truax AB, Oldstone MBA, Nelson JA, manuscript in preparation). Previous studies have shown that HCMV can nonpermissively infect mononuclear leukocytes in vitro32 and in vivo.28 Others found that T lymphocytes stimulated in an allogenic mixed lymphocyte reaction can be permissively infected by HCMV in vitro." The data presented in this study demonstrate activation of viral replication in vivo that proceeds at least as far as late gene expression in mononuclear cells ofimmunocompromised renal transplant patients. On the basis of these data at least 2 hypotheses can be proposed for the pathogenesis of primary HCMV infection in transplant patients. First, virus carried in a small number of cells in the graft may be reactivated by a combination of allogenic stimulation'0'45 and therapeutic immunosuppression' and begin to replicate. Activated host lymphocytes infiltrating the kid-
Figure 4-Photomicrographs of in situ hybnidization of kidney biopsies using HCMV-specific probes. The histologic diagnosis given is based on examination A-Posttransof the entire section; all diagnostic features may not be present in the photomicrograph (original magnifications are shown in parentheses). plantation biopsy (patient 1), IE probe; mild cellular rejection, some tubular cells with nucle polymorphism; note hybridization (arrow) with probable tubular epithelial cells. (X400) B-Postphosphonoformate therapy biopsy (patlent 2), IE probe; minimal evidence of rejection; note hybridization with cell in glomerular tuft. (X400)
246
AJP * August 1988
GNANN ET AL
ney as part of the host-vs.-graft response may then be infected by HCMV. These infected mononuclear cells could disseminate HCMV to distant target organs (eg, retina, lung). Alternatively, a small number of HCMV-infected donor lymphocytes or monocytes carried in the graft may become activated and begin to express virus. These donor leukocytes could move into the bloodstream and infect circulating host PBMNs that may later enter the graft as part of the rejection response. Indeed, the 2 hypotheses are not mutually exclusive and both could be operative. The focal clustering of infected inflammatory cells that was observed, however, lends support to the first hypothesis. Patient 2 is unique among this group of patients because she received a graft from an HCMV seronegative donor. Nonetheless, patient 2 did develop HCMV infection about I10 weeks after surgery, presumably acquired from transfused blood. Interestingly, this was the longest transplantation-to-infection interval seen in this small group of patients. In addition, the HCMV-specific hybridization with the biopsies from patient 2 was the most intense seen in all the specimens studied, although the pattern and distribution of hybridizing cells was similar to that observed in other kidneys. These findings may relate to differences in inoculum size and/or route oftransmission. If HCMV is indeed transferred via the graft, why were infected cells not detected by in situ hybridization in kidney tissue from HCMV-seropositive donors? One possible explanation is that latently infected cells do not express sufficient amounts of IE mRNA to be detected by this system. The in situ hybridization technique is clearly sensitive enough to detect nonpermissively infected lymphocytes, however.28 An alternative explanation is that HCMV is present only in a small number of cells in a kidney and could be missed by sampling error with the small biopsies obtained. The 7 patients studied were treated with PPF, an anti-HCMV drug (still investigational in the United States) that inhibits viral DNA polymerase.47 Even after clinically successful therapy with PPF,48 HCMV nucleic acids were still readily detectable in renal tissue. Thus, although PPF may have inhibited HCMV replication, virus was still present after completion of a course of therapy. Several investigators have reported a positive correlation between HCMV infection and allograft rejection.4954 A recent multicenter study indicated that the adverse effect of HCMV infection on allograft survival was especially marked in patients who received cadaveric kidneys, developed primary HCMV infections, and were treated with anti-thymocyte preparations.54
However, the confounding contributions of multiple variables have thus far made it difficult to establish firmly a direct cause-and-effect relationship between HCMV infection and allograft rejection. Although the reason that HCMV infection is associated with a higher risk of allograft rejection remains unclear, direct renal damage caused by extensive HCMV infection of tubular or glomerular cells is ruled out by the data that show minimal parenchymal cell infection.
References 1. Peterson PK, Balfour HH, Marker SC, Fryd DS, Howard RJ, Simmon RL: Cytomegalovirus disease in renal allograft recipients: A prospective study of the clinical features, risk factors and impact on renal transplanta-
tion. Medicine 1980,59:283-300 2. Glenn J: Cytomegalovirus infections following renal transplantation. Rev Infect Dis 1981, 3:1151-1178 3. Marker SC, Howard RJ, Simmons RL, Kalis JM, Connelly DP, Najarian JS, Balfour HH: Cytomegalovirus infection: A quantitative prospective study of three hundred twenty consecutive renal transplants. Surgery 1981, 89:660-671 4. Betts RF: Cytomegalovirus infection in transplant patients. Prog Med Virol 1982,28:44-64 5. Peterson P, Ferguson R, Fryd D, Balfour H, Rynasiewillz J, Simmons R: Infectious diseases in hospitalized renal transplant recipients: A prospective study of a complex and evolving problem. Medicine 1982, 61: 360-372 6. Ho M: Cytomegalovirus: Biology and infection. New York, Plenum Medical Book Company, 1982, pp 171203 7. Betts RF, Freeman RB, Douglas RG, Talley TE: Clinical manifestations of renal allograft derived primary cytomegalovirus infection. Am J Dis Child 1977, 759763 8. Rubin RH, Cosimi AB, Tolkoff-Rubin NE, Russell PS, Hirsch MS: Infectious disease syndromes attributable to cytomegalovirus and their significance among renal transplant recipients. Transplantation 1977, 24:458464 9. Suwansirikul S, Rao N, Dowling JN, Ho M: Primary and secondary cytomegalovirus infection: Clinical manifestations after renal transplantation. Arch Intern Med 1977, 137:1026-1030 10. Olding LB, Jensen FC, Oldstone MBA: Pathogenesis of cytomegalovirus infection. I. Activation of virus from bone marrow-derived lymphocytes by in vitro allogenic reaction. J Exp Med 1975, 141:561-572 11. Hamilton JD, Seaworth BJ: Transmission of latent cytomegalovirus in a murine kidney tissue transplantation model. Transplantation 1985, 39:290-296 12. Klotman ME, Starnes D, Hamilton JD: The source of murine cytomegalovirus in mice receiving kidney allografts. J Infect Dis 1985, 152:1192-1196 13. Bruning JH, Bruggeman CA, Van Boven CPA, Van Breda Vriesman PJC: Passive transfer of cytomegalovi-
CMV IN TRANSPLANTED KIDNEYS
Vol. 132 * No. 2
14. 15.
16. 17.
18.
rus by cardiac and renal organ transplants in a rat model. Transplantation 1986,41:695-698 Porter KR, Starnes DM, Hamilton JD: Reactivation of latent murine cytomegalovirus from kidney. Kidney Int 1985, 28:922-925 Ho M, Suwansirikul S, Dowling JN, Youngblood LA, Armstrong JA: The transplanted kidney as a source of cytomegalovirus infection. N Engl J Med 1975, 293: 1109-1112 Betts RF, Freeman RB, Douglas RG Jr, Talley TE, Rundell B: Transmission of cytomegalovirus infection with renal allograft. Kidney Int 1975, 8:387-394 Smiley ML, Wlodavar CG, Grossman RA, Barker CF, PerloffLJ, Tustin NB, Starr SE, Plotkin SA, Friedman HM: The role of pretransplant immunity in protection from cytomegalovirus disease following renal transplantation. Transplantation 1985, 40:157-161 Chou S: Acquisition of donor strains of cytomegalovirus by renal-transplant recipients. New Engl J Med 314:
1418-1423 19. Wertheim P, Buurman C, Geelen J, Noordaa JVD: Transmission of cytomegalovirus by renal allograft demonstrated by restriction enzyme analysis (letter). Lancet 1983, 1:980-981 20. Nankervis GA: Comments on CMV infections in renal transplant patients. Yale J Biol Med 1976,49:27-28 21. Balfour HH Jr, Slade MS, Kalis JM, Howard RJ, Simmons RL, Najarian JS: Viral infections in renal transplant donors and their recipients: A prospective study. Surgery 1977, 81:487-492 22. Naraqi S, Jackson GG, Jonasson 0, Rubenis M: Search for latent cytomegalovirus in renal allografts. Infect Immun 1978, 19:699-703 23. Orsi EV, Howard JL, Baturay N, Ende N, Ribot S, Eslami H: High incidence of virus isolation from donor and recipient tissues associated with renal transplantation. Nature 1978, 272:372-373 24. Zollinger HU, Mihatsch MJ: Light, electron and immunofluorescent microscopy and clinical aspects, Renal Pathology in Biopsy. New York, Springer-Verlag, 1978, pp 613-614 25. Herrera GA, Alexander RW, Cooley CF, Luke RG, Kelly DR, Curtis JJ, Gockerman JP: Cytomegalovirus glomerulopathy: A controversial lesion. Kidney International 1986, 29:725-733 26. Lang DJ, Ebert PA, Rodgers BM, Boggess HP, Rixse RS: Reduction of post perfusion CMV infections following the use of leukocyte depleted blood. Transplantation 1977, 17:391-395 27. Porter KA: Renal transplantation, Pathology of the Kidney, 2nd ed. Edited by RH Heptinstall. Boston, Little-Brown, 1974, pp 977-1042 28. Schrier RD, Nelson JA, Oldstone MBA: Detection of human cytomegalovirus in peripheral blood lymphocytes in a natural infection. Science 1985, 230:10481051 29. Wiley CA, Schrier RD, Denaro FJ, Nelson JA, Lampert PW, Oldstone MBA: Localization of cytomegalovirus proteins and genome during fulminant central nervous system infection in an AIDS patient. J Neuropath Exp Neurol 1986, 45:127-139 30. Demarchi JM: Human cytomegalovirus DNA: Restriction enzyme cleavage maps and map locations for im-
31.
32.
33.
34.
35.
36.
247
mediate-early, early, and late RNAs. Virology 1981, 114:23-38 Wathen MW, Stinski MF: Temporal patterns ofhuman cytomegalovirus transcription: Mapping the viral RNAs synthesized at immediate-ealy, early, and late times after infection. J Virol 1982,41:462-477 Rice GPA, Schrier RD, Oldstone MBA: Cytomegalovirus infects human lymphocytes and monocytes: Virus expression is restricted to immediate-early gene products. Proc Natl Acad Sci 1984,81:6134-6138 Jahn G, Knust E, Schmolla H, Sarre T, Nelson JA, McDougall JK, Fleckenstein B: Predominant immediateearly transcripts of human cytomegalovirus AD 169. J Virol 1984,49:363-370 Stenberg RM, Witte PR, Stinski MF: Multiple spliced and unspliced transcripts from human cytomegalovirus immediate-early region 2 and evidence for a common initiation site within immediate-early region 1. J Virol 1985, 56:665-675 Ruger B, Klages S, Walla B, Albrecht J, Fleckenstein B, Tomlinson P, Barrell B: Primary structure and transcription of the genes coding for the two virion phosphoproteins pp65 and pp71 of human cytomegalovirus. J Virol 1987, 61:446-453 Nowak B, Gmeiner A, Sarnow P, Levine AJ, Fleckenstein B: Physical mapping of human cytomegalovirus genes: Identification of DNA sequences coding for a virion phosphoprotein of 71 kDa and a viral 65-kDa poly-
peptide. Virology 1984,134:91-102 37. Fleckenstein B, Muller I, Collins J: Cloning of the complete human cytomegalovirus genome in cosmids. Gene 1982, 18:39-46 38. Rigby PWJ, Dieckmann M, Rhodes C, Berg P: Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J Mol Biol 1977, 113:237-251 39. Hancock WW, Thomson NN, Atkins RC: Composition of interstitial cellular infiltrate identified by monoclonal antibodies in renal biopsies of rejecting human renal allografts. Transplantation 1983, 35:458-463 40. Hall BM, Bishop GA, Farnsworth A, Duggin GG, Harvath JS, Sheil AGR, Tiller DJ: Identification of the cellular subpopulations infiltrating rejecting cadaver renal allografts: Preponderance of the T4 subset of T cells. Transplantation 1984, 37:564-570 41. McWinnie DL, Thompson JF, Taylor HM, Chapman JR, Bolton EM, Carter NP, Wood RFM, Morris PJ: Morphometric analysis of cellular infiltration assessed by monoclonal antibody labeling in sequential human renal allograft biopsies. Transplantation 1986, 42:352358 42. Fetterman GH, Sherman FE, Fabizio NS, Studnicki FM: Generalized cytomegalic inclusion disease of the newborn: Localization ofinclusions in the kidney. Arch Pathol 1968, 86:86-94 43. Heieren MA, Van der Wonde FJ, Balfour HH: Cytomegalovirus replicates efficiently in human kidney mesangial cells. Proc Natl Acad Sci USA 1988, 85:16421646 44. Braun RW, Reiser HC: Replication of human cytomeg-
248
45.
46. 47.
48.
49.
50.
51.
GNANN ET AL
alovirus in human peripheral blood T cells. J Virol 1986,60:29-36 Wu BC, Dowling JN, Armstrong JA, Ho M: Enhancement of mouse cytomegalovirus infection during hostversus-graft reaction. Science 1975, 190:56-58 Jordan MC, Shanley JD, Stevens JG: Immunosuppression reactivates and disseminates latent murine cytomegalovirus. J Gen Virol 1977, 37:419-423 Helgstrand E, Eriksson B, Johansson NG, Lannero B, Larsson A, Misiorny A, Noren JO, Sjoberg B, Stenberg K, Stening G, Stridh S, Oberg B, Alenius S, Philipson L: Trisodium phosphonoformate: A new antiviral compound. Science 1978,201:819-821 Ahlmen J, Wijnveen A-C, Brynger H, Lycke E: Clinical experiences with phosphonoformate (Foscarnet): Treatment of viral diseases following renal transplantation. Scand J Urol Nephrol 1985, 92(suppl):41-44 Lopez C, Simmons RL, Mauer SM, Najarian JS, Good RA, Gentry S: Association of renal allograft rejection with virus infections. Am J Med 1974, 56:280-289 Simmons RL, Lopez C, Balfour H, Kalis J, Rattazzi LC, Najarian JS: Cytomegalovirus: Clinical virological correlations in renal transplant recipients. Ann Surg 1974, 180:623-634 Cappel R, Hestermans 0, Toussaint C, Vereerstraeten P, van Beers D, deBraekeller J, Schoutens S: Cytomega-
AJP * August 1988
52.
53.
54.
55.
lovirus infection and graft survival in renal graft recipients. Arch Virol 1978, 56:149-152 May AG, Betts RF, Freeman RB, Andrus CH: An analysis of cytomegalovirus infection and HLA antigen matching on the outcome of renal transplantation. Ann Surg 1978, 187:110-117 Pass RF, Whitley RJ, Diethelm AG, Whelchel JD, Reynolds DW, Alford CA Jr: Outcome of renal transplantation in patients with primary cytomegalovirus infection. Transpl Proc 1979, 11:1288-1290 Van Es A, Baldwin WM, Oljans PJ, Tanke HJ, Ploem JS, Van Es LA: Expression of HLA-DR on T lymphocytes following renal transplantation, and association with graft rejection episodes and cytomegalovirus infection. Transplantation 1984, 37:65-69 Rubin RH, Tolkoff-Rubin NE, Oliver D, Rota TR, Hamilton J, Betts RF, Pass RF, Hillis W, Szmuness W, Farrell ML, Hirsch MS: Multicenter seroepidemiologic study of the impact of cytomegalovirus infection on renal transplantation. Transplantation 1985,40:243-249
Acknowledgment This is Publication Number 5093-IMM from the Department of Immunology, Scripps Clinic and Research Foundation, La Jolla, CA, 92037.
