IFN-γ–Regulated Toxoplasma gondii Distribution and Load in the Murine Eye

Abstract

purpose. To establish a mouse model of ocular toxoplasmosis in both wild type (WT) and immunocompromised hosts and to clarify the effects of interferon (IFN)-γ on the infectivity of Toxoplasma gondii in various parts of the eye.

methods. Susceptible WT C57BL/6, resistant WT BALB/c, and IFN-γ knockout (GKO) mice were infected with cysts of T. gondii perorally. The tissues were harvested for molecular and histopathologic studies. Analysis included a quantitative competitive polymerase chain reaction (QC-PCR) assay and reverse transcription (RT)-PCR for IFN-γ and stage conversion markers. All animals underwent ophthalmic examinations including fluorescein angiography (FA).

results. In WT C57BL/6 mice, T. gondii was detected in tissue in the following order: brain, retina, choroid, sclera, and optic nerve (ON). The highest T. gondii load was observed in the posterior retina, and was much greater than that in WT BALB/c mice. In GKO mice, disseminated infection was evident, and the T. gondii load was highest in the choroid and ON. IFN-γ mRNA expression in WT C57BL/6 mice was higher than that in WT BALB/c mice after infection. Tachyzoites existed in GKO mice, whereas bradyzoites existed in WT C57BL/6 mice. FA showed dye leakage from the retinal capillaries of GKO mice.

conclusions. The T. gondii load in the retina in the susceptible WT strain continued to increase, unlike in the resistant WT strain. IFN-γ was shown to regulate the T. gondii load and interconversion in the eye. A toxoplasmic vasculitis model was established with GKO mice and assay systems with QC-PCR and FA.

Aprotozoan parasite, Toxoplasma gondii, can cause severe, life-threatening disease, especially in immunocompromised patients, and it is an important cause of ocular disease in both immunocompetent and immunocompromised individuals, such as human immunodeficiency virus (HIV)–infected patients, ¹ patients with cancer, and organ transplant recipients. ^{2 3 4}  

Several studies ^{5 6 7} have proposed hypotheses that delineate the pathogenesis of the disease process. Despite these works, the mechanisms by which this organism causes retinochoroidal damage have remained a subject of controversy. In addition, although the interconversion between the bradyzoite and tachyzoite stages of T. gondii within the eye of the intermediate host is the critical event in the pathogenesis of ocular toxoplasmosis, the factors influencing the interconversion in the eye are unknown. 

Although many investigators ^{6 8 9 10 11} have used intraperitoneal, intraocular, or carotid artery injection of Toxoplasma parasites to infect the eye directly, with the purpose of establishing experimental models of acquired ocular toxoplasmosis, these infection routes were not natural. There is not only the disadvantage of disrupting the blood–ocular barrier, but also the possibility of causing damage to intraocular structures. Moreover, the trauma caused by the injection itself may have caused inflammation. In the present study, infection was introduced perorally (i.e., via the natural infection route), without breaching the ocular barrier, so that the ensuing disease was not complicated by inflammation or trauma to the eye. 

Because a mouse model of ocular toxoplasmosis would enhance the expeditious evaluation of novel agents, we reasoned that the quest for a satisfactory animal model of ocular toxoplasmosis must be continued. 

Interferon (IFN)-γ is the pivotal mediator inducing anti-T. gondii effector mechanisms. ^{12 13} We have reported that the destruction of the IFN-γ gene has remarkable effects on infectivity in both susceptible and resistant mice. ^{14 15} The use of IFN-γ knockout (GKO) mice for the model of immunocompromised hosts such as acquired immunodeficiency syndrome (AIDS) makes sense, considering that among HIV-infected individuals there is a decreased ability to produce IFN-γ and that this is related to the development of opportunistic infections. ¹⁶  

The purpose of the present study was to establish a mouse model of ocular toxoplasmosis in both immunocompetent wild type (WT) and immunocompromised hosts infected perorally and to clarify the effects of IFN-γ at early and subacute stages on the infectivity of T. gondii in various areas of the eye and on the interconversion of T. gondii in the retina. 

Materials and Methods

Toxoplasma gondii

Cysts of an avirulent Fukaya strain of T. gondii were obtained as previously described. ¹⁵  

Experimental Animals

Six- to 8-week-old inbred WT C57BL/6 and WT BALB/c mice were purchased (SLC, Hamamatsu, Japan). GKO mice of the same age with both C57BL/6 and BALB/c backgrounds were used. All mice had normal physical and ophthalmic examinations and had no detectable serum antibodies to T. gondii before the present infection. 

Induction of Toxoplasmosis in Mice

All experiments in this study were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. For quantitative competitive polymerase chain reaction (QC-PCR) analysis of the eyes and brains, WT and GKO mice were killed at 0, 6, 8, 10, 12, 21, or 28 days, and 0, 6, 10, or 12 days, respectively, after peroral infection with five cysts of the Fukaya strain administered with a syringe fitted with a 19-gauge round-ended needle on day 0 of the experiment. Blood was analyzed 8 days after infection. For reverse transcription (RT)-PCR analysis, WT and GKO mice were killed at 0, 12, or 28 days and 10 or 12 days, respectively, after peroral infection. Survival of the mice was monitored daily, and cumulative mortality was calculated. 

QC-PCR

The sequence of the T. gondii SAG1 (a major surface antigen of T. gondii) gene was based on published data. ¹⁷ Construction of competitive cDNA of SAG1 (competitor SAG1) was described previously. ¹⁸ Briefly, the cDNA fragment of SAG1 was from restricted digestion of the PCR product of the genomic SAG1 of T. gondii using the restriction enzymes (NdeI at the 451-bp site and XhoI at the 1210-bp site), and was constructed in the pET-21b(+) plasmid (Novagen, Madison, WI). SAG1 pET-21b(+) was digested with the restriction enzymes SacII and SacI at base pair positions 511 and 665. The small fragment was deleted, and the cohesive ends of the larger fragment were treated with cloned Klenow fragment (TaKaRa Co., Kyoto, Japan) to obtain blunt end insert DNA. The blunt ends of SAG1 in pET-21b(+) were ligated with T4 DNA ligase (TaKaRa Co.). The smaller molecular size product of pET-21b(+) SAG1 was confirmed to be the truncated SGA1, in which 155 base pairs (511–665) were deleted, with a DNA sequencer (model 373A; Applied Biosystems, Foster City, CA). 

Genomic (g) DNA from the cornea, iris/ciliary body, lens, retina, choroid, sclera, optic nerve (ON), brain, and blood was prepared as previously described. ¹⁴ The retina was cut along the center of the distance from the ON head and ora serrata, separating the posterior and peripheral retina, respectively. Using 1 μg of gDNA from these tissues, QC-PCR was performed to determine the distribution of T. gondii as described previously. ¹⁵ Briefly, gDNA (1 μg) extracted from these tissues was coamplified in the reaction buffer containing 10 μM bovine serum albumin with a constant concentration of truncated SAG1 DNA that competitively binds oligo primers with WT SAG1. The amplified cDNAs were electrophoretically separated on 1% agarose gels containing ethidium bromide, and the ratio to competitor (T/C) SAG1 DNA subsequently amplified was measured with a gel densitometer (IPLab; Signal Analytical Corp., Vienna, VA). The number of T. gondii was calculated as described previously. ¹⁵ Each experimental group contained three mice, and the experiments were repeated three times. Statistical analysis was performed 4 weeks and 12 days after infection in WT and GKO mice, respectively. Data were considered significant at P < 0.05 by Mann-Whitney test. 

Expression of mRNAs of SAG1, T.g.HSP30/bag1, and mIFN-γ

The expressions of SAG1 (as a marker of tachyzoites ¹⁷ ), T.g.HSP30/bag1 (as a marker of bradyzoites ¹⁹ ), and mouse (m)IFN-γ (mIFN-γ) messenger RNAs (mRNAs) were investigated by RT-PCR. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control. Four mice were used for each experimental group, and the experiment was repeated two times. 

Primers

The primers used were as follows: for SAG1, forward (5′-TCG GAT CCC CCT CTT GTT GC-3′), which corresponds to nucleotides 452 to 471 of the T. gondii SAG1 gene, and reverse (5′-CTC CAG TTT CAC GGT ACA GT-3′), which corresponds to nucleotides 1191 to 1210; for T.g.HSP30/bag1, forward (5′-GGC TCG AGA TGG CGC CGT CAG CAT-3′) and reverse (5′-GGG GAT CCC TAC TTC ACG CTG ATT TG-3′); for mIFN-γ, forward (5′-TGA ACG CTA CAC ACT GCA TCT TGG-3′) and reverse (5′-CGA CTC CTT TTC CGC TTC CTG AG-3′); for GAPDH forward (5′-ACC ACA GTC CAT GCC ATC AC-3′) and reverse (5′-TCC ACC ACC CTG TTG CTG TA-3′). 

Ophthalmic Examination Procedures

All animals underwent retinal and slit lamp examinations before infection and then every 3 days after infection. The pupils were dilated with 1 drop each of 1% tropicamide and 2.5% phenylephrine. To document the findings, we performed fluorescein angiography (FA) with a fundus camera (Canon, Tokyo, Japan) in unanesthetized mice by intraperitoneal injection of 3 μL/g of body weight of 10% sodium fluorescein (Alcon Co., Tokyo, Japan). 

Histologic Examinations

The eyes and brain samples freshly collected for histology were fixed in phosphate-buffered formaldehyde (15%), dehydrated, and embedded in paraffin. Sections were stained with hematoxylin and eosin for histologic examinations. 

Results

Mortality

GKO C57BL/6 and BALB/c mice died 11 to 12 days after infection, whereas both strains of WT mice survived for more than 1 month. 

QC-PCR Analysis

The yields of gDNA from the tissues were as follows: 1 to 2 μg gDNA from 1 mg cornea, 2 to 3 μg gDNA from 1 mg iris/ciliary body, 2 to 3 μg gDNA from 1 mg lens, 5 to 10 μg gDNA from 1 mg posterior retina, 4 to 6 μg gDNA from 1 mg peripheral retina, 1 to 2 μg gDNA from 1 mg choroid, 1 to 2 μg gDNA from 1 mg sclera, 1 to 2 μg gDNA from 1 mg ON, and 1 to 2 μg gDNA from 1 mg brain. 

The kinetics of T. gondii loads showed that in the susceptible strain WT C57BL/6, T. gondii was detected first in the brain; second in the posterior retina, peripheral retina, and choroid; and third in the sclera and ON, 10, 12, and 21 days after infection, respectively (Fig. 1A) . Although the number of protozoans in the brain, posterior, and peripheral retinas continued to increase gradually, that in choroid and sclera changed only slightly, remaining at a low level. The T. gondii load was higher in the posterior and peripheral retinas than in the other eye areas examined 4 weeks after infection, and the number of protozoans in the posterior retina exceeded that in the peripheral retina (P < 0.05). Once T. gondii in the ON was detected, the number of protozoans increased rapidly and reached almost the same level as in the peripheral retina 4 weeks after infection, despite the fact that protozoans in the ON were detected later than in the other eye parts. There were significant differences (P < 0.05) between the numbers of protozoans in the posterior retina and ON and between the posterior retina and choroid. The number of protozoans in the ON was much less than in the brain (P < 0.05). No detectable PCR product targeting the SAG1 gene was observed from the cornea, iris/ciliary body, or lens at any time during the infection. 

In the resistant strain WT BALB/c, the abundance of T. gondii in the eye parts (Fig. 1B) was much less (P < 0.05) than in WT C57BL/6 mice, and the number of protozoans grew at a slower rate in the eye parts of WT BALB/c mice than in those of WT C57BL/6 mice. T. gondii was detected in the brain and posterior and peripheral retinas 10 to 12 days after infection, with their abundance reaching a peak at 12 days and decreasing thereafter. Finally, very low numbers of protozoans were detected in the sclera, ON, and choroid 3 to 4 weeks after the infection. 

In contrast, in GKO mice from both genetic backgrounds, disseminated infection was clearly evident (Fig. 2) . The time course showed that in GKO C57BL/6 mice (Fig. 2A) , T. gondii was detected in the brain, ON, iris/ciliary body, posterior retina, peripheral retina, choroid and sclera 10 days after infection. The number of protozoans increased dramatically and continued to increase. The T. gondii number was highest in the choroid, ON, brain, and sclera 12 days after infection. The abundance of protozoans in the posterior retina was lower than in choroid, ON, and sclera and that in the iris/ciliary body was significantly lower (P < 0.05) than that in the posterior retina. The number of protozoans in the peripheral retina was less than that in the iris/ciliary body, but the difference was not significant. There was no significant difference between the number of T. gondii in the posterior and peripheral retinas. Finally, a low level of T. gondii load was observed in the cornea 12 days after infection. No detectable PCR product targeting the SAG1 gene was observed from the lens at any time during the infection. 

In GKO BALB/c mice (Fig. 2B) , the kinetics patterns of the abundance of protozoans in the various parts of the eye, except the iris/ciliary body, were similar to those in GKO C57BL/6 mice. It should be noted that in GKO BALB/c mice, the number of protozoans in the iris/ciliary body was much greater than in the posterior or peripheral retina, whereas in GKO C57BL/6 mice the number of protozoans in the iris/ciliary body was much less than in the posterior retina. 

The number of protozoans in the eye parts of GKO C57BL/6 mice, except the iris/ciliary body and cornea, was significantly greater (P < 0.05) than in GKO BALB/c mice. 

In WT mice, intensities of the SAG1 band in blood were very faint compared with the competitor bands (Fig. 3) . In contrast, in GKO mice, these band intensities were the opposite, which means that the abundance of T. gondii in the blood of GKO mice was much greater than that in WT mice at day 8 after infection. 

Kinetics of Expression of mIFN-γ

mIFN-γ mRNA was detected in small amounts in WT C57BL/6 before infection, but then increased in amount (Fig. 4A) after infection. In contrast, the amount of mIFN-γ mRNA in BALB/c mice before infection was very small, increased moderately at 12 days after infection, and then maintained at this level. 

Kinetics of Expression of SAG1 and T.g.HSP30/bag1 mRNAs

In the retina of both WT mice, the SAG1 mRNA expression was low at day 12 after infection and still lower at 4 weeks after infection (Fig. 4B) . On the contrary, T.g.HSP30/bag1 mRNA was detected at a moderate level 12 days after infection in the retina of WT C57BL/6 mice and at a higher level 4 weeks after infection, whereas it was not detected in the retina of BALB/c mice at any time point examined. In both GKO mice, however, the levels of SAG1 mRNA expression was low 10 days after infection and showed a rapid increase at 12 days, whereas the levels of T.g.HSP30/bag1 mRNA expression were very low compared with those of WT C57BL/6 mice. 

Histopathology

GKO mice showed mild inflammation 12 days after infection in the eyes and brains, including sludging or congestion of blood in the retinal vessels (Fig. 5) , whereas there was no evidence of inflammation in WT mice. 

Fluorescein Angiography

Dye leakage from retinal capillaries was observed in GKO C57BL/6 mice (Fig. 6) and GKO BALB/c mice (data not shown) by FA, but this was not a characteristic finding in WT mice. 

Discussion

Recently, a few experimental models of acquired ocular toxoplasmosis have been developed in mice, hamsters, and cats. ^{6 7 8 9 10 11} The animals are inoculated systemically via the intraperitoneal ^{9 10 11} or peroral route ⁷ with bradyzoites, vascularly via the carotid artery, ⁶ or intracamerally ⁸ with tachyzoites. In this study, we established a mouse model of ocular toxoplasmosis by a physiological infection route—that is, by peroral infection with cysts in both immunocompetent and immunocompromised hosts. 

It was noted that there was a difference in the number of protozoans in the various parts of the eyes between WT C57BL/6 (Fig. 1A) and WT BALB/c (Fig. 1B) mice. Our results indicate that an immunogenetic predisposition may exist for developing ocular toxoplasmosis. Williams et al. ²⁰ reported that susceptibility to T. gondii in mice is affected by several genes including one within the major histocompatibility complex. Furthermore, Meenken et al. ²¹ reported the possible association between the severity of ocular involvement of congenital toxoplasmosis and the presence of HLA-Bw62 antigen. 

There have been numerous reports concerning the critical roles of IFN-γ in the modulation and inflammation of T. gondii infection, ^{11 22 23} in which immunocompetent WT mice and immunocompromised hosts, such as nude mice ²⁴ or cytokine-knockout mice, have been used. ²⁵ In the present study, expression of mIFN-γ mRNA occurred at higher levels in the retinas of susceptible WT C57BL/6 mice than in those of resistant WT BALB/c mice. In our model, the high level of IFN-γ may have exerted a protective effect against T. gondii infection and may have abrogated ocular inflammation. Shen et al. ¹⁰ reported that there was an earlier and subjectively greater expression of IFN-γ in the brain and eye of B6MRL/lpr and B6MRL/gld mice than in WT, although no significant difference in the degree of ocular inflammation and apoptosis was detected between them. Gazzinelli et al. ¹¹ reported that treatment of mice with monoclonal antibody against IFN-γ resulted in a marked increase of ocular lesions and the severity of inflammatory response, and that a high expression level of IFN-γ mRNA was detected by RT-PCR in the eyes of C57BL/6 mice. It has been shown that, in response to various IFN-γ inducers including T. gondii infection, C57BL/6 mice produce generally high levels of IFN-γ, whereas BALB/c mice produce low levels. ²⁶ Regarding the serum levels of IFN-γ in C57BL/6 and BALB/c mice after T. gondii infection, there are a few conflicting opinions. ^{26 27} Shirahata et al. ²⁶ reported that C57BL/6 mice are the best producers of IFN-γ, whereas BALB/c mice are consistently poor producers, and the former showed a significant prolongation of mean survival time after intraperitoneal infection with tachyzoites of the relatively avirulent strain S-273 of T. gondii, compared with that of BALB/c mice. They concluded that there was no direct correlation between the susceptibility to T. gondii and the levels of serum IFN. On the contrary, Liesenfeld et al. ²⁷ showed that serum levels of IFN-γ of C57BL/6 mice (susceptible) were lower at 7 days after peroral infection with T. gondii (when they had developed necrosis) than those of BALB/c mice (resistant) that did not have necrosis develop in the ilea and survived. The detailed mechanisms of IFN-γ–mediated genetic control of the susceptibility to T. gondii infection remain to be elucidated. 

IFN-γ also plays a role in interconversion of T. gondii between bradyzoites and tachyzoites. ^{28 29 30} The present study confirms that T. gondii in the retina of WT C57BL/6 formed bradyzoites in the retina 4 weeks after infection, whereas in both strains of GKO mice T. gondii formed tachyzoites by 12 days after infection. 

In our study, differences were noted not only in the number of protozoans but also in the distribution of parasites in the various eye parts between WT (Fig. 1) and GKO (Fig. 2) mice. It is well known that ocular toxoplasmosis in immunocompromised hosts differs distinctly, both clinically and histopathologically, from the disease in immunocompetent patients. ³¹  

Our WT C57BL/6 mice data showing the amount of T. gondii in the retina are consistent with the clinical findings that the macula or the posterior pole is more easily affected than the peripheral retina. ³² The features appearing to distinguish ocular toxoplasmosis in immunocompromised patients from that in immunocompetent individuals include multiple active lesions, massive necrosis in all areas of the retina, a greater number of organisms in retinal lesions, a greater size of retinochoroiditis, ^{1 33 34 35} and the occasional spread of organisms to uveal tissue. ³⁶ Our present data (Fig. 2) in GKO mice strongly confirm these clinical data, although the precise reasons for the particular parasite distribution in the retina need further study. 

The present study clearly showed that the uveal tract, especially the anterior uvea, was somewhat difficult to infect with T. gondii in the immunocompetent host, as previously described. ³⁷ There is some argument about the attribution of severe granulomatous iridocyclitis to T. gondii infection of the anterior segment. The anterior segment may show acute granulomatous uveitis at the reactivation of the disease, ³² possibly on the basis of a reaction to exposed antigens in the retina, but without apparent reactivation and proliferation of parasites. 

The present data confirmed the clinical findings of HIV-associated ocular toxoplasmosis in which a prominent inflammatory reaction in the vitreous body and anterior chamber has been described. ^{1 33 38 39} The reason for the different clinical and histopathological findings in the uveal tract between immunocompetent and immunocompromised hosts may be that the uveal tract is the highly vascularized middle layer of the eye, and the immunocompromised host, such as the GKO mice, has severe parasitemia, ⁴⁰ as reported previously ^{14 41} and as shown in the present study. 

Parasites are present in the ON of some patients, ^{33 42} and it has been suggested that T. gondii can reach the retina through the ON from intracranial sites of infection. ^{43 44} Such transmission would explain the frequent parapapillary location of lesions. The presence of parasites in the ON may simply reflect the hematogenous spread through the ocular branches of the ophthalmic artery, and ocular toxoplasmosis may develop without evidence of intracranial disease. ³¹ On the contrary, in both strains of GKO mice the levels of T. gondii loads in the ON were much higher than those in the posterior and peripheral retinas and in the brain (Fig. 2) . Our present study confirms a report ⁴⁵ describing toxoplasmic papillitis as the initial manifestation of AIDS. 

The abundance of protozoans in the sclera is quite different in WT and GKO mice. Although toxoplasmosis is the most common infectious cause of posterior intraocular inflammation, it is rarely described in association with scleritis. The sclera overlying a focus of toxoplasmic retinochoroiditis, however, can develop clinically apparent scleritis. ⁴⁶ Although the sclera is relatively avascular, it is a highly porous structure and is next to the choroid, where a high level of T. gondii load was detected in GKO mice. Thus, scleritis caused by toxoplasmosis may be more common, especially in immunocompromised patients than was previously thought. 

In this study, FA clearly demonstrated retinal vasculitis in GKO mice (Fig. 6) . Davidson et al. ⁶ observed retinal vasculitis and perivasculitis with late hyperfluorescence adjacent to focal segments of affected vessels in a feline model of ocular toxoplasmosis after carotid artery inoculation with tachyzoites. Holland et al. ⁴⁷ reported retinal vasculitis without focal necrotizing retinochoroiditis with acquired systemic toxoplasmosis, concluding that retinal vasculitis may be the only ophthalmic disorder during the early stages of a newly acquired T. gondii infection. A perivascular distribution of organisms appears to be a frequent feature of disease in immunosuppressed patients. ^{33 36} In addition, a great number of parasites were detected in the blood of GKO mice, and sludging of blood in the retinal vessels, which has become an important pathogenetic feature in infectious diseases, ⁴⁸ was observed in our study. These findings strongly suggest that T. gondii can reach the eye through the blood stream but may be arrested temporarily in terminal capillary loops of the retina. ³¹ Parasites are then released after intracellular reproduction and bursting of the cell. ³⁶ Vasculitis may develop in response to reactions between circulating antibodies and local T. gondii antigens. ⁴⁹  

Future studies of toxoplasmic retinochoroiditis will surely contribute much to our understanding of the regulation of immune response and inflammation, and they may help to clarify some of the pathologic processes involved in toxoplasmic retinochoroiditis. 

 Supported by Grants 12671697, 13670242, and 12557025 for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, Japan.

 Submitted for publication February 12, 2003; revised May 22, 2003; accepted June 2, 2003.

 Disclosure: K. Norose, None; H.-S. Mun, None; F. Aosai, None; M. Chen, None; L.-X. Piao, None; M. Kobayashi, None; Y. Iwakura, None; A. Yano, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Kinetics of T. gondii abundance in tissues of (A) WT C57BL/6 and (B) WT BALB/c mice. Brain (♦), ON (⋄), cornea (□), iris/ciliary body (▪), lens (+), peripheral retina (○), posterior retina (•), sclera (▵), and choroid (▴). Data are expressed as the mean number of parasites/mg of specimen DNA ± SD. The experiments were performed three times, with similar results.

Figure 2.

 

Kinetics of T. gondii abundance in (A) GKO C57BL/6 and (B) GKO BALB/c mice. Symbols and description of data are as in Figure 1 .

Figure 3.

 

Analysis of T. gondii abundance in the blood using QC-PCR. Lane 1: WT C57BL/6 mouse; lane 2: WT BALB/c mouse; lane 3: GKO C57BL/6 mouse; lane 4: GKO BALB/c mouse. Blood samples were obtained 8 days after infection. Top band (molecular size 759 bp) shows the SAG1; bottom band (molecular size 605 bp) shows competitor SAG1.

Figure 4.

 

Expression of mIFN-γ, SAG1, and T.g.HSP30/bag1 mRNAs during T. gondii infection. (A) mIFN-γ mRNA. Lanes 1, 3, and 5: WT C57BL/6 mice before infection, 12 days, and 4 weeks after infection, respectively; lanes 2, 4, and 6: WT BALB/c mice before infection, 12 days, and 4 weeks after infection, respectively. Top: mIFN-γ RT-PCR products; bottom: GAPDH RT-PCR products. (B), SAG1 and T.g.HSP30/bag1 mRNAs. Lanes 1 and 3: WT C57BL/6 mice 12 days and 4 weeks after infection, respectively; lanes 2 and 4: WT BALB/c mice 12 days and 4 weeks after infection, respectively; lanes 5 and 7: GKO C57BL/6 mice 10 and 12 days after infection, respectively; lanes 6 and 8: GKO BALB/c mice 10 and 12 days after infection, respectively. Top: SAG1 RT-PCR products; middle: T.g.HSP30/bag1 RT-PCR products; bottom: GAPDH RT-PCR products.

Figure 5.

 

Histologic sections from GKO C57BL/6 mice. (A) Before infection; (B) day 12 after infection. Arrow: region of sludging or congestion of blood in retinal vessels. Magnification, ×400.

Figure 6.

 

Fluorescein angiogram in GKO C57BL/6 mouse. (A) Before infection; (B) day 11 after infection. Arrows: regions of retinal capillary leakage before and after infection.