The Royal Veterinary and Agricultural University, Copenhagen, Denmark; ... of Veterinary Pathology, Biilowsvej 13, DK-1870 Frederiksberg C, The Royal ...
Mycopathologia 112:11-18, 1990. 9 1990 Kluwer Academic Publishers. Printed in Belgium.
A murine model for the study of the impact of inoculation on the foeto-placental unit
Aspergillusfumigatus
Henrik Elvang Jensen 1, & Jann Hau 2
I National Veterinary Laboratory, Bfilowsvej 27, P.O. Box 373, DK-1503 Copenhagen V, Denmark; 2Department of Veterinary Pathology, Laboratory Animal Unit, Biilowsvej 13, DK-1870 Frederiksberg C, The Royal Veterinary and Agricultural University, Copenhagen, Denmark; *Present address. Department of Veterinary Pathology, Biilowsvej 13, DK-1870 Frederiksberg C, The Royal Veterinary and Agricultural University, Copenhagen, Denmark Received 31 May 1989; accepted 8 January 1990
Key words: Aspergillusfumigatus, aspergillosis, mycotic abortion, murine model, foeto-placental unit
Abstract Pregnant female BALB/c mice were inoculated intravenously with Aspergillus fumigatus conidia in different concentrations (1 x 101-1 x 108) on day 10 of pregnancy. Pregnancy was confirmed by the presence ofmurine alpha-fetoprotein (m-AFP) in the maternal circulation. Conidia inoculation of 1 x 103 and greater significantly (p < 0.0001) resulted in abortion or placental lesions. Uterine cell infiltrations were recognized in 75 ~o of the mice which aborted following the inoculation of 1 x 103 conidia or more. In the placental lesions of mice inoculated with 0.5-1 x 108 conidia, hyphae, haemorrhage, thrombosis, necrosis, calcification, and inflammation were regular findings. Histopathology of other maternal organs revealed that extrauterine organ lesions were closely correlated with the dose of conidia inoculated, and the following pattern of reaction was found: liver > lung > kidney > brain > heart.
Introduction Mycotic placentitis and abortion are of great importance in cattle production. Since 1920, when the first case was reported [ 13], there has been a steady increase in reports on this subject. In Denmark mycotic abortion accounts for approximately 14~o of all bovine abortions [9] and comparable figures have been reported from other countries [7, 10, 15, 16]. Aspergillus fumigatus has been reported to be responsible for approximately 75 ~o of the cases of mycotic abortion, with most of the remaining 25 ~o being caused by zygomycetes and Candida spp. [6]. Experimental inoculation (vena jugu-
laris) of pregnant cows and ewes with A. fumigatus conidia has demonstrated that the placental tissue is highly susceptible to mycotic infection subsequently resulting in abortion [2, 6]. In both species an apparent resistance to infection in extrauterine maternal tissue has been found, and when present, the lesions are predominantly restricted to the liver in cows [6], and to the heart in ewes [2]. The pathogenicity of Aspergillus spp. and Candida spp. against embryonic murine tissues has been demonstrated when injected i.v. into pregnant mice [ 12]. Prior to the inoculation with a mixture ofAspergillus spp. and Candida spp. the mice had been treated with ACTH.
12 The aim of the present study was to evaluate if a murine model is suitable to study the pathogenesis of fungal abortion in the cow and other mammalian species.
Table I. Findings in the placenta and uterus of female BALB/c mice inoculated intravenously with Aspergillusfumigatus conidia in different concentrations (1 x 101-1 x 108) on day 10 of pregnancy. Placenta
Uterus
Materials and methods Group
Dose
Animals and pregnancy diagnosis Inbred primipregnant female BALB/cABom mice were used. The mice (25-30 g) were fed a commercial pelleted feed (Altromin, Brogaarden, Denmark) ad libitum and had free access to drinking water (tap water). Conception time was determinated by the presence of a vaginal plug, and was designated day 1 of pregnancy. On day 10 of pregnancy, prior to the experimental inoculation, a venous blood sample was collected from the tall vein for pregnancy diagnosis. This was determined by the presence of murine alpha-fetoprotein (m-AFP). The serum content of m-AFP was measured by rocket immunoelectrophoresis as previously described [4]. Only mice which were m-AFP positive on day 10 of pregnancy were used in the experiment.
1
2
1
2
3
4
3
4
a
Control
4
- (9) -(12)
A -(14)
b
1 x 10 ~
4
-(10) (10)
-(7) (7)
-
-
c
1 • 102
4
-(10) -(11)
A -(9)
-
I
d
1 x 103
4
A A
A A
I I
I -
e
1X
10 4
4
A A
A -(9)
I
I -
f
1 x 105
3
A A
A
I I
I
g
1•
10 6
4
A A
A A
I I
-
h
1•
10 7
2
A
A
I
I
1
N + H(7)
N
1"
I + H(4)
-
3*
I+H(5) I + H(IO)
i 0.5 x 10 s
Experimental inoculations A. fumigatus (American Type Culture Collection, Rockville, USA, strain no. 42202) conidia were harvested from cultures incubated on CzapekDox agar for 7 days at 37 ~C. The conidia were recovered by filtration through sterilized gauze and suspended in sterile 0.9% saline with 0.1% Tween 80 (Sigma P 1754) and counted in a hemocytometer according to Emmons et aL [3 ]. On day 10 of pregnancy 9 groups (group b-j) of mice, 2 - 4 each, were inoculated (i.v.) with 0.5 ml of conidia suspension, corresponding to 1 x 101 to 1 x 108 per injection (Tablel). A control group (group a) was injected (i.v.) with 0.5ml sterile 0.9% saline containing 0.1% Tween 80. Except for mice that either succumbed or were killed by cervical dislocation when severely ill, the mice were euthanized by an overdose of pentobarbitone (i.p.) on day 19 of pregnancy (Table 1).
Number of mice
j
lxl0 s
N+H(4) I
-
Mice inoculated with 1 x 101-0.5 x 108 were killed on day 19 of pregnancy (9 days post inoculation), whereas mice (*) inoculated with 0.5-1 x 108 either succumbed or were killed severely ill on day 12 of pregnancy (2 days post inoculation). Abbreviations: - : normal. I: infiltration (PMN cells). H: hyphae present. N: necrosis. A: abortion. (No.): number of foetuses/placentae.
Post-mortem procedures A complete necropsy was performed on each animal which died or was sacrificed, and the number of foetuses and/or placentae in the uterus was recorded. Lung, heart, kidney, liver, brain and uterus with contents, were removed and fixed in toto in 10% buffered formalin. The tissues were embedded in paraffin, sectioned at 5/~m and
13 stained with haematoxylin and eosin (HE), periodic acid-Schiff(PAS), Mallory's phospho tungstic acid-haematoxylin method (PTAH), and Grocoifs methenamine-silver method (GMS). The uterine contents and histopathological findings were recorded individually.
1 x 103 conidia or more. However, one mouse in group e, which was inoculated with 1 x 10 4 conidia neither aborted nor had placental lesions. The post-mortem status of pregnancy and histopathological findings are listed in Table 1. The main necropsy findings were as follows:
Statistics
Uterus
The significance of the results was evaluated by Fischer's exact test.
The uterine findings in mice which had aborted consisted of submucosal oedema and polymorphonuclear (PMN) cell infiltration. In some of the animals small amounts of necrotic debris and P M N cells were found in the lumen of the uterus (Fig. 1). Uterine infiltrations were found in 75 ~o of the mice which had aborted following the inoculation of 1 x 103 conidia or more (Table 1). In mice which had not aborted only a minute infiltration of P M N cells in the uterine wall was seen.
Results
Experimental animals Inoculation of A. fumigatus conidia significantly (p < 0.0001) resulted in abortion or placental lesions in the pregnant mice inoculated with
Fig. 1. Uterusfrom a mouse which aborted due to inoculation (i.v.) of 1 x 103Aspergillusfumigatusconidia. In the lumen necrotic debris and polymorphonuclear cells are seen. (HE, Obj. x 10). Inset (Obj. x 25).
14 Placenta
Placental material was found only in mice which had not aborted. In the placental discs of mice inoculated with 0.5-1 x 108 conidia the following features were regular findings: thrombosis, haemorrhage, necrosis, P M N cell infiltration, hyphal growth, and dystrophic calcification. The thrombosed vessels were constricted to the maternal blood sinuses located at the base of the placental disc, whereas no thrombosis was seen in the labyrintic plexus (Fig. 2). In the extrafoetal membranes substantial tissue necrosis and detachment from the uterine wall was observed in all mice inoculated with 0.5-1 x 108 conidia. In the space between the uterine wall and the detached necrotic membranes PMN cells and hyphal growth could be demonstrated (Fig. 3). In mice inoculated with 1 x 10~-1 x 102 conidia and in one mouse inocu-
lated with 1 x 10 4 (no. 4, group e) no alterations were found in the placental discs or the extrafoetal membranes.
Foetus
Hyphal growth was found in the foetuses of two animals (no. 1 in group i and j). In both mice the infection was accompanied by necrosis, and in one (no. 1, group i) by dystrophic calcification, too. No inflammatory reactions were observed in any of the foetuses.
Liver
In all mice inoculated with 1 x 106-1 x 108 conidia inflammatory reactions were found, whereas lower doses only provoked reactions in
Fig. 2. Placental disc from a mouse euthanized 2 days after inoculation (i.v.) of 1 x 108 Aspergillusfumigatus conidia. Thrombosis is situated in the blood sinuses at the base of the placental disc (hollow arrows), whereas thrombosis in the labyrintic plexus (L) is absent. Arrow: haemorrhage. F: foetus. (PTAH, Obj. x 2.5).
15 some (25-75~o) of the mice. The inflammatory reaction consisted of mononuclear cell infiltration with granuloma formation. In some of the granulomas a few P M N cells were seen. Only at the highest dose (1 x 108) hepatocyte necrosis was found. Hyphal growth was never observed in the liver.
nuclear and P M N cell infiltration. In only one animal (no. 2, group j) hyphal growth was found disseminated in vessels (both arteries and veins) and lung tissue. In the vessels and tissues hyphae were accompanied by thrombosis and necrosis respectively. Polymorphonuclear cell infiltration was found both in the necrotic tissue and thrombosed vessels containing hyphae.
Lung Kidney At doses from 1 • 10 7 to 1 X 108 conidia all mice showed inflammatory reactions, whereas doses between 1 x 103 and 1 x 10 6 induced reactions in 25-50 ~o of the animals only. Lower doses did not give reactions. The inflammatory reaction consisted of interstitial accumulation of histiocytes, granuloma formation with occasional development of multinucleated giant cells, and perivascular mono-
Renal lesions were only found in mice inoculated with 1 x 10 6 conidia or more. In two mice only focal inflammatory reactions without hyphal growth were seen. In one (no. 4, group g) the reaction consisted of interstitial accumulation of histiocytes, whereas in the other (no. 2, group j) tubular epithelial necrosis and P M N cell infiltration was found. In the remaining mice the hyphal
Fig. 3. Uterusfrom a mouse euthanized 2 days after inoculation (i.v.) of 1 x 108Aspergillusfumigatusconidia. Betweenthe uterine wall (UW) and the detached necrotic extrafoetal membrane (EM) hyphae and polymorphonuclear cells are seen. (GMS, Obj. x 25).
16 growth was accompanied by a massive P M N cell infiltration and necrosis. Hyphal growth was mainly restricted to the papilla (necrosis and abscess formation) and the pelvic cavity (purulent pyelitis). Furthermore, in mice inoculated with 1 x 10 7 conidia or more, intratubular P M N cellcasts, interstitial accumulations of P M N cells, fibroblasts and histiocytes were regular findings.
Brain
Cerebral lesions were only found in mice inoculated with 1 x 10 7 conidia or more. In mice without hyphal growth (dose: 1 x 107) the lesions consisted of hyperaemia, perivascular infiltrations of mononuclear cells and abscess formation. In animals with hyphal growth (mice inoculated with 0.5-1 x 108 conidia) thrombosis, necrosis, and P M N cell infiltrations were dis-
seminated. The latter were found both perivascularly and disseminated in the brain tissue.
Heart
Inflammatory reactions were only found in the hearts of mice inoculated with 1 x 108 conidia. The reaction consisted of necrosis accompanied by P M N cell infiltration and perivascular histiocyte accumulation. No hyphae were found in the hearts of the test animals.
Control animals
In the extrauterine maternal tissues no lesions were found following injection of 0.9~o saline containing 0.1 ~o Tween 80. One mouse (no. 2) in the control group aborted. In the uterus of this
Fig. 4. Placental disc from a mouse (control group) inoculated (i.v.) with 0.5 ml sterile 0.9% saline containing 0.1% Tween 80. Focal necrosis (N) in the periphery of the placental disc (PD) and the degenerated Reicher's membrane is seen. (PAS, Obj. x 10).
17 animal only a negligible number of PMN cells were found in the wall, whereas no exudation to the lumen had occurred. In the animals which had not aborted focal necrosis was observed in the periphery of the placental discs (Fig. 4). These necrotic areas were never associated with thrombosis, cell infiltration nor calcification. Alterations in the extrafoetal membranes were never observed in the control mice.
Discussion The present study demonstrates for the first time that abortion can be induced in pregnant female mice by i.v. inoculation of A.fumigatus conidia. Foetal death in mice is generally associated with resorption of the placental and foetal tissue but, in the present study, abortion is believed to have taken place, since no sign of foeto-placental material could be demonstrated in the animals with empty uteri. The actual mechanism resulting in abortion following injection of low doses of conidia (1 x 103-1 x 107) cannot be determined by the present study, but the fact that hyphae and thrombosis, necrosis, inflammation, and calcification were regular findings in the placentae of mice which were sacrificed 2 days post inoculation suggests a pathogenesis compatible to that previously found in cattle and ewes [2, 6]. However, the placenta in ewes seems to be more selectively infected after i.v. inoculation of A. fumigatus conidia [2] in comparison to the present murine model, where other organs also were affected frequently. Focal necrosis in the periphery of the placental disc is natural in the late state of pregnancy in mice, and was observed in the placental discs obtained from normal non-inoculated BALB/c female mice on day 19 of pregnancy, too. In the placentae the thrombosed vessels were predominantly situated at the base of the placental disc, whereas no thrombosis was seen in the labyrintic plexus (Fig. 2). This finding agrees well with our recent demonstration of the localization of plasmin/plasminogen present as a rim on the
trophoblastic microvilli in the human placenta [8 ]. In view of our previous finding of pregnancy zone protein (PZP) on the trophoblastic surface in both human [1] and murine [1, 5] placentae, and the affinity between plasmin/plasminogen and PZP [ 11], it is tempting to speculate that immobilized plasmin on the trophoblastic surface prevented intravascular coagulation in the labyrintic plexus in the present experiment (Fig. 2). Thrombosis was not prevented in the basal trophoblast free maternal blood sinuses in accordance with our localization studies of plasmin [8]. The nature of the reactions in the extrauterine tissues is in accordance with that previously described in non-pregnant female NMRI-mice [ 14]. Comparison of the distribution pattern of lesions in the extrauterine tissues to other studies may be of limited relevance since both the strain/stock of mouse and the Aspergillus strain are of equally decisive importance for the manifestation and progression of the infection [ 14]. In conclusion, the mouse seems applicable as an animal model for the study of A. fumigatus infections, and further studies aimed at elucidating the abortive mechanisms are in progress.
References 1. Chemnitz J, Hau J, Svendsen P, Folkersen J, Westergaard JG, Christensen BC. Immunohistochemical demonstration of human and murine pregnancy-associated serum proteins in maternal and placental tissue. Bibl Anat 1982; 22: 87-92. 2. Cysewski SJ & Pier AC. Mycotic abortion in ewes produced by Aspergillusfumigatus: Pathologic changes. Am J Vet Res 1968; 29: 1135-51. 3. Emmons CW, Binford CH, Utz JP & Kwon-Chung KJ. Medical mycology. (3. ed.) Lea & Febiger, Philadelphia, 1977. 4. Hau J, Svendsen P, Teisner B, Thomsen Pedersen G & Kristiansen B. Correlation between fetal weight and maternal serum levels of murine alpha-foetoprotein and quantification of four molecular forms. Biol Reprod 1981; 24: 683-89. 5. Hau J, Teisner B & Svendsen B. Placental affinity for pregnancy-associated murine protein-1 examined by crossed affino immunoelectrophoresis. Electrophoresis 1983; 4: 369-70. 6. Hill MWM, Whiteman CE, Benjamin MM & Ball L.
18
7.
8.
9. 10. 11.
12.
Pathogenesis of experimental bovine mycotic placentitis produced by Aspergillus fumigatus. Vet Path 1971; 8: 175-92. Hugh-Jones ME & Austwick PKC. Epidemiological studies in bovine mycotic abortion. Vet Rec 1967; 81: 273-76. Jensen HE, Poulsen OM & Hau J. Localization of plasmin on the surface of the human troph0blast. Biomed Biochim Acta 1989; 48: 437-40. Krogh HV. Bovine mycotic abortion in Denmark. Nord Vet Med 1985; 37: 27-33. McCausland IP, Slee KJ & Hirst FS. Mycotic abortion in cattle. Aust Vet J 1987; 64: 129-32. Poulsen OM & Hau J. Interaction between pregnancy zone protein and plasmin. Arch Gynaecol Obstet 1988; 243: 157-64. Samborski Z, Szumigalska D, Madej JA, Sielicka B & Michalski Z. Fizjologia i patologia rozrodu oraz sztuczne unasienianie. Med Wet 1983; 39: 195-98.
13. Smith T. Mycosis of the bovine fetal membranes due to a mould of the genus Mucor. J Exp Med 1920; 31 : 115-22. 14. Walzl HL, Ackerbauer H, Meingassner JG & Mieth H. Histopathology of organ lesions in mice after an intravenous or intratracheal or intrarenal infection with Aspergillusfumigatus. Mykosen 1986; 30: 10-18. 15. Weikl A. Mycotic abortion in cattle. Vet Med Rev 1965; 2: 71-80. 16. Williams BM, Shreeve BJ, Hebert CN & Swire PW. Bovine mycotic abortion: Some epidemiological aspects. Vet Rec 1977; 100: 382-85.
Address for correspondence: Dr. H.E. Jensen Department of Veterinary Pathology Btilowsvej 13, DK-1870 Frederiksberg C The Royal Veterinary and Agricultural University Copenhagen, Denmark
