Chlamydophila psittaci is viable and infectious in the conjunctiva and peripheral blood of patients with ocular adnexal lymphoma: Results of a single-center prospective case–control study

Abstract

Ocular adnexal MALT lymphoma (OAML) is linked to Chlamydophila psittaci (Cp) infection. Viability and infectivity of Cp, demonstrated by growth in culture, has not been yet investigated in these patients. We conducted a single-center prospective case–control study to assess the prevalence, viability and infectivity of Cp in 20 OAML patients and 42 blood donors registered in a 6-month period. The presence of Cp in conjunctival swabs and peripheral blood mononuclear cells (PBMC) of patients and donors was assessed by TETR-PCR and in vitro cultures. From an epidemiological point of view, OAML patients often resided in rural areas, and reported a history of chronic conjunctivitis and prolonged contact with household animals (85% vs. 38% of donors; p = 0.00001). Cp was detected in lymphoma tissue in 15 (75%) patients. Cp DNA was detected in conjunctival swabs and/or PBMC from 10 (50%) patients and in PBMC from 1 (2%) donor (p = 0.01). Viability and infectivity of Cp, demonstrated by growth in culture, were confirmed in conjunctival swabs and/or PBMC from 5 (25%) patients, but not in donors (p = 0.002). This prospective study demonstrates, for the first time, that Cp present in the conjunctiva and PBMC of OAML patients is capable to grow and be isolated in cell cultures. Cp infection is common in OAML patients and exceptional in blood donors. Epidemiological data of OAML patients (prolonged contact with household animals and chronic conjunctivitis) are consistent with Cp exposure risk. © 2008 Wiley-Liss, Inc.

Ocular adnexal MALT lymphoma (OAML), an indolent malignancy that originates in the lachrymal gland, eyelids, orbital soft tissues or conjunctiva, is associated with a persistent infection of Chlamydia psittaci (Cp)1 (now reclassified as Chlamydophila psittaci,2) the etiologic agent of psittacosis, a human infection caused by exposure to infected birds.3 In OAML patients, Cp determines a local and systemic infection because its DNA can be detected both in the lymphomatous lesion of the ocular adnexae and, in near 40% of cases, in the peripheral blood mononuclear cells (PBMC).1 This putative Chlamydia–lymphoma association is also supported by objective tumor regression observed in 65% of patients with Cp-related OAML treated with the antibiotic doxycycline.4, 5

To date, the comparison of Cp infection prevalence between OAML patients and controls has been actually performed on retrospective series spanning different periods, and, importantly, methods used so far did not allow us to assess whether the presence of Cp DNA and antigens in lymphoma samples and PBMC corresponded to viable and infectious microorganisms.1 This is a relevant issue considering that the presence of viable Cp in OAML patients, demonstrated by growth in cultures, would increase the level of evidence supporting this association, fulfilling the second Koch's postulate, which states that “the pathogen must be isolated and grown in pure culture.” In the setting of bacteria-associated lymphomas, isolation in pure cultures has been demonstrated only for Helicobacter pylori,6 the bacterium related to gastric MALT lymphomas.7

Herein, we report the results of a single-center prospective case–control study focused on the prevalence of Cp infection as well as the viability and infectivity of this microorganism in conjunctival swabs and PBMC of 20 OAML patients recruited during a 6-month period. Results on OAML patients were compared to viability and infectivity rates in healthy blood donors referred to our institution during the same period. Other than to confirm our previous observations,1 this prospective study demonstrates, for the first time, that Cp present in biological samples from OAML patients is viable and infectious, i.e., it is capable to grow and be isolated in unrelated tissue cultures. These findings provide important additional insights on the genesis, biology and management of Cp-related lymphomas.

Material and methods

Study group

Twenty patients with newly diagnosed OAML (i.e., marginal zone B-cell lymphoma of MALT-type of the lachrymal gland, conjunctiva, eyelid or orbital soft tissue) and 42 healthy blood donors referred to the San Raffaele H Scientific Institute of Milan, between October 2006 and March 2007, were enrolled onto a single-center prospective case–control study. The primary endpoints of the study were to define the prevalence of Cp infection in both OAML patients and donors and to assess the viability and infectivity of Cp detected in biological samples from these patients.

Demographic, epidemiological and clinical data, mostly regarding prolonged and/or occupational contact with household animals and previous diseases, were collected from patients and donors by means of an interview. Patients with OAML were assessed by conventional staging procedures (physical examination, ultrasonography of the neck, total-body computerized tomography scan, gastroscopy and bilateral bone marrow biopsy). Written informed consent was obtained from each patient and donor. This study conformed to the tenets of the Declaration of Helsinki and was approved by the IRB of the San Raffaele H Scientific Institute of Milan, Italy.

Samples collection and DNA extraction

Lymphoma tissue, conjunctival swabs and blood samples were obtained from each OAML patient; blood samples were drawn from healthy donors.

Five micrometer thick-sections were cut from formalin-fixed, paraffin embedded lymphoma tissue specimens from the 20 OAML patients. Specimens were deparaffinized and DNA was extracted by using the QIAamp Tissue Kit (Qiagen, Hilden, Germany) and assessed by polymerase chain reaction (PCR) (see later).

Two conjunctival swabs were collected from each OAML patient within 1 month from histological diagnosis and before any treatment: 1 swab was quickly forwarded to the Centro di Riferimento Oncologico of Aviano for PCR analyses, and 1 swab was placed in a saccarose-phosphate-glutamate transport medium and quickly forwarded to the Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna of Pavia for Cp cultures. Researchers from both laboratories worked independently and blindly, and results were matched only at the time of data analysis.

Two samples of peripheral blood from each OAML patient and healthy donor were collected in ethylene-diamine-tetra-acetic acid containing tubes. One sample per individual was forwarded without delay to the above-mentioned laboratories and immediately processed.

DNA extraction from PBMC samples was carried out following phenol/chloroform extraction technique after proteinase K digestion. DNA was extracted from infected cultures with a commercial DNA extraction kit (QIAamp Tissue kit, Qiagen, Hilden, Germany).

Cultures

PBMC were recovered with Histopaque®-1077 (Sigma), washed, resuspended in PBS and inoculated onto J774A.1 cell monolayers grown in RPMI medium supplemented with foetal calf serum (10%), glutamine (200 mM) and glucose (20%) in shell vials with coverslips. Standard techniques for culturing chlamydiae were employed. In particular, the isolation medium was supplemented with cycloheximide (1 μg/ml) and the inoculums were centrifuged at 1,500g for 1 hr at 37°C onto the cell monolayer. After incubation at 37°C for 72–96 hr, coverslips were collected, fixed with methanol, stained with an antichlamydia monoclonal antibody (Merifluor Chlamydia, Meridian Bioscience) and examined for the presence of chlamydial inclusions under UV illumination (Fig. 1). At least 5 culture passages were carried out for each sample.

Figure 1. Chlamydial inclusions in J774A.1 cells. Vitality and infectivity of Cp in a PBMC sample from an OAML patient was confirmed by its grow in in vitro culture. Immunostaining with a fluorescein-isothiocyanate-labelled anti-chlamydia monoclonal antibody and examination under UV illumination showed the presence of multiple intracellular chlamydial infectious elementary bodies (arrows). The identity of the inclusions was confirmed on DNA extracted from infected cultures by a specific PCR assay.8

The identity of the inclusions was confirmed on DNA extracted from infected cultures by a PCR assay targeting a region of the 16S ribosomal gene of chlamydiae.8 Identity of amplified products of the expected size (∼269 bp) was assessed by an RFLP assay using the enzyme MseI. The band patterns were compared with the patterns of the different chlamydial species in our database. Specificity of the amplified 16S fragments was confirmed by direct sequencing and comparison with sequences available in GenBank.

TETR-PCR

The presence of Cp DNA in diagnostic lymphoma samples, patient's PBMC and conjunctival swabs and donors' PBMC was investigated using a TETR-PCR, as previously reported.1, 9 Amplified DNA is the end of the 16S rRNA gene and the beginning of the 16S-23S spacer region in the ribosomal genes.9 The primers pairs specific for Chlamydia trachomatis and Chlamydia pneumoniae are located entirely in the 16S rRNA gene; for Cp, 1 primer is located in the 16S rRNA gene and 1 primer in the 16S-23S spacer region. DNA equivalent to <1 I.F.U. per PCR tube of Chlamydia trachomatis L2, Chlamydia pneumoniae TW-183 or Chlamydia psittaci ORNI were included as positive controls. Blank reactions filled with 50 μl of PCR mixture were interspersed every 3–4 samples to monitor possible cross-contamination.

An additional PCR protocol allowing the amplification of the heat shock protein 60 (hsp60) gene was also used to confirm the presence of Cp DNA.10

PCR products were visualized by gel electrophoresis. Amplicon fragment sizes were quantified by image analysis (Image Station 440CF, NEN Life Science Products, Boston, MA). Specificity of the amplicons was confirmed by direct sequencing of both sense and antisense strands of purified PCR products using an ABI PRISM 310 Genetic Analyzer (Perkin Elmer, Foster City, CA). Sequence specificity was assessed by BLAST search (http://www.ncbi.nlm.nih.gov/blast) and heterogeneity of Chlamydiae sequences across the different samples was investigated by aligning them with the MultAlin software (http://prodes.toulouse.inra.fr/multalin/multalin.html).

Statistical methods

Clinical and epidemiological characteristics of patients and donors and distribution of Cp infection were compared using Fisher exact test. Independent associations among studied variables and Cp infection (outcome variable) were assessed by logistic regression. All statistical tests were two-sided, with a significant “p” value of 0.05. Analyses were carried out using the Statistica 4.0 statistical package for Windows (Statsoft, Tulsa, OK).

Results

Study group

Patients' and donors' characteristics are reported in Table I. Fourteen OAML patients had single-lesion disease (stage IE), while no patient had elevated lactate dehydrogenase serum levels or systemic symptoms. Compared to OAML patients, donors were more commonly young males living in urban areas. OAML patients reported a history of chronic conjunctivitis and prolonged contact with household animals more commonly than controls. Seventeen (85%) OAML patients and 16 (38%) donors had prolonged contact with animals (p = 0.00001); occupational exposure to animals was reported in 4 cases. Prolonged contact with cats was reported by 14 (70%) patients and 14 (33%) donors (p = 0.006); prolonged contact with birds was reported in 6 (30%) and 2 (5%) cases, respectively (p = 0.004). The period of contact was significantly longer among OAML patients (Table I).

Table I. Patients and Donors Characteristics

Variables	OAML patients	Blood donors	p
1 This subgroup includes patients with bilateral involvement of the orbits (n = 3), regional lymphadenopathies (n = 2) and other extranodal sites of disease (n = 1). 2 Dogs, fishes, chickens, horses and other household animals. 3 Comparison of contact with cats on the entire series (70% of OAML patients and 33% of donors). 4 Comparison of contact with birds on the entire series (30% of OAML patients and 5% of donors).
Number of cases	20	42
Median age (range)	60 (25–86)	41 (20–57)	0.00001
Male/Female ratio	8/12 (0.66)	29/13 (2.23)	0.03
Site of disease
Lachrymal gland	5 (25%)	–	–
Conjunctiva	9 (45%)	–	–
Orbital soft tissue	6 (30%)	–	–
Stage > I1	6 (30%)	–	–
Chronic conjunctivitis	10 (50%)	2 (5%)	0.00001
Residency in rural areas	11 (55%)	6 (14%)	0.0008
Prolonged contact with
cats	11 (55%)	14 (33%)
cats and birds	2 (10%)	0 (0%)
cats, birds and others2	1 (5%)	0 (0%)	0.0063
birds	1 (5%)	0 (0%)
birds and others	2 (10%)	2 (5%)	0.0044
Occupational contact with animals	3 (15%)	1 (2%)	0.09
Median duration of contact (range; ys.)	37 (2–65)	11 (1–40)	0.00001
Hepatitis C virus infection	4 (20%)	–

Cp detection

Cp DNA was detected in lymphoma samples in 15 (75%) patients. As reported in Table II, TETR-PCR detected the presence of Cp DNA in biological samples of 10 (50%) OAML patients: in conjunctival swabs of 5 (25%) patients, in PBMC samples of 3 (15%), in both samples of 2 patients (10%); all these patients had a Cp-positive lymphoma. Only the PBMC sample of 1 (2%) donor resulted positive in TETR-PCR (p = 0.01; Odds ratio: 13.7). PCR for the Cp hsp60 gene confirmed positive results obtained with TETR-PCR in all assessed cases (data not shown).

Table II. Chlamydophila psittaci Status in OAML Patients and Donors

Variables	OAML patients	Blood donors	p1
1 Detection of Cp DNA by TETR-PCR on biological samples. PBMC, peripheral blood mononuclear cells. 2 Cp isolated from cultures and confirmed by DNA sequencing. 3 Comparison of PBMC positivity between patients and donors (Fisher exact test).
Number of cases	20	42
Cp DNA (TETR-PCR) in2
Conjunctival swab	5 (25%)	Not assessed
PBMC	3 (15%)	1 (2%)
PBMC + conjunctival swab	2 (10%)	Not assessed	0.01
Cp (cultures) in3
Conjunctival swab	0 (0%)	Not assessed
PBMC	3 (15%)	0 (0%)
PBMC + conjunctival swab	2 (10%)	Not assessed	0.002

Cp was isolated (in vitro cultures) from 5 (25%) OAML patients (in PBMC of 3 patients and in PBMC and conjunctival swabs of other 2 patients) (Fig. 1), but not from donors (p = 0.002; Odds ratio: infinity) (Table II). All patients with positive cultures had a Cp-positive lymphoma.

Results obtained with TETR-PCR and culture tests were not identical, but highly concordant, with some differences between types of specimens (Table III). Concordant results were obtained in 65% of conjunctival swabs, with a higher sensitivity for TETR-PCR. Concordance rate for PBMC samples was of 70%, with similar sensitivities for both tests.

Table III. Chlamydophila psittaci Status in Biological Samples from OAML Patients According to the Used Methods

Variables	Negative culture tests	Positive culture tests	p1
1 Fisher exact test.
Conjunctival swab
Positive TETR-PCR (n = 7)	6 (86%)	1 (14%)
Negative TETR-PCR (n = 13)	12 (92%)	1 (8%)	0.99
PBMC
Positive TETR-PCR (n = 5)	3 (60%)	2 (40%)
Negative TETR-PCR (n = 15)	12 (80%)	3 (20%)	0.56

In OAML patients, the prevalence of Cp infection in lymphomatous tissue, conjunctival swabs and PBMC was not significantly correlated to age, stage of disease, history of chronic conjunctivitis, residence area or prolonged contact with cats or birds. Importantly, Cp was detected in OAML tissue samples in 9 of the 12 (75%) patients ≤60 years and in 6 of the 8 (75%) patients older than 60 years (p = 1.00). Results obtained with TETR-PCR and culture tests according to the primary site of lymphoma are reported in Table IV; Cp detection rates in conjunctival swabs and PBMC were identical in conjunctival and intraorbital lymphomas. Four patients had concomitant hepatitis C virus infection; all of them were females, older than 80 years in 3 cases, and invariably affected by Cp-positive lymphomas. TETR-PCR and culture tests detected the presence of Cp in conjunctival swab, PBMC and both in 1 patient each. These features could suggest a higher Cp detection rate in HCV-positive patients, but with a distribution between conjunctival and PBMC samples similar to that observed for the entire series.

Table IV. Chlamydophila psittaci Status in Biological Samples from OAML Patients According to the Primary Site of Disease

	Conjunctival lymphoma	Orbit lymphoma1	p2
1 This subgroup includes lymphomas arising in the lachrymal gland and orbit soft tissue. 2 Fisher exact test.
Number of patients	9 (45%)	11 (55%)
Conjunctival swabs
Positive TETR-PCR	3 (33%)	4 (36%)	0.53
Positive cultures	0 (0%)	2 (18%)	0.16
Positive ≥1 method	3 (33%)	5 (45%)	0.37
PBMC
Positive TETR-PCR	2 (22%)	3 (27%)	0.49
Positive cultures	2 (22%)	3 (27%)	0.49
Positive ≥1 method	3 (33%)	5 (45%)	0.37

Treatment and Cp eradication

Therapeutic management of OAML patients was heterogeneous: 14 patients were treated with doxycycline, 3 received chemotherapy, 1 was irradiated and 2 patients were carefully observed. The median follow-up for OAML patients was 8 months.

Eight of the 10 OAML patients with Cp infection detected by TETR-PCR in PBMC (n = 2), conjunctival swabs (n = 4) or both (n = 2) were treated with doxycycline. One month after doxycycline treatment was concluded, PBMC and conjunctival swabs were collected again and assessed with TETR-PCR to analyze eradicating activity, as previously reported.4, 5 Eradication rate was high: Cp DNA was found in post-antibiotic PBMC samples of a single patient and was no longer detectable in post-antibiotic conjunctival swabs. Treatment heterogeneity and short follow-up do not allow us to formulate reliable conclusions on therapeutic activity and efficacy.

Discussion

This prospective case–control study provides further original evidence on the association between Cp and OAML and expands our knowledge on the biology of this microorganism in lymphoma patients. This study demonstrates, for the first time, that Cp is viable and infectious in OAML patients as unequivocally demonstrated by the ability of this microorganism to grow in pure cultures, a presence that was confirmed by molecular techniques. Second, our results confirm that Cp is present at the time of tumor diagnosis in conjunctival swabs and/or PBMC of half of OAML patients, whereas it is almost invariably absent in donors. Third, this study delineates the epidemiological profile of OAML patients, who are more commonly resident in rural areas, with a history of chronic conjunctivitis and prolonged contact with household animals, particularly cats and birds. These figures are consistently with an increased risk of Cp exposure.

A further added value of this study is given by confirmation of a strict association between Cp infection and OAML in a prospective series enrolled over a period of 6 months. This excludes potential selection biases like those hypothesized for our previous retrospective series, which included patients and controls collected in a 15-year period.1 As confirmatory data, Cp prevalence in lymphoma tissue was similar both in the present prospective study (75%) and in the previous retrospective study (80%).1 Like for most lymphoma studies using donors as controls, registered patients and donors were not matched for age, gender and geographical area. In fact, OAML patients came from 3 different Italian regions, and half of them were older than 60 years, while donors came only from Lombardy. Discrepancy in age could represent a major weakness in this study considering that a longer exposure to environmental chlamydiae may be associated with higher infection rates. Actually, this was demonstrated only for C. pneumoniae but not for Cp.11 In fact, Cp seropositivity in elderly population (>65 years) of Western countries is very rare (3%).11 Even if serological tests and PCR are not comparable methods, the significantly higher Cp prevalence observed in OAML patients seems to suggest that this association is not simply due to differences in the cumulative bacterial exposure. This is confirmed by the fact that Cp prevalence was equal in OAML patients younger and older than 60 years. Discrepancies in geographical areas between registered patients and donors could also be important considering that chlamydial diffusion in the general population varies among different regions and is related to the prevalence of Cp infection in wild animals, which is affected by the transport of this micro-organism in migrant birds.12 Notwithstanding, Cp positivity in the current series of OAML patients was homogeneous among subgroups from different regions (data not shown), excluding bona fide a related bias.

In this study, biological sampling from patients and donors was performed with identical protocols and, more importantly, complementary molecular and microbiological methods were concurrently applied to the same samples at different institutions, and results were matched only at the time of data analysis. Concordance rates between TETR-PCR and culture tests were high, strengthening the role of Cp detection in PBMC as a reliable parameter of bacterial eradication in OAML patients treated with doxycycline.4, 5 The higher sensitivity of TETR-PCR in conjunctival swabs is in line with recently reported discrepancies among methods used for chlamydia detection.13 These discrepancies may reflect the presence of persistently infected cells. In vivo, chlamydiae can enter a persistent phase in response to various micro-environmental conditions. In this state, it would be detected by TETR-PCR but not by cultures due to the very limited, or virtually absent, production of infectious elementary bodies.14, 15 Consequently, these findings seem to strengthen the role of Cp detection in PBMCs by TETR-PCR as a reliable method to screen disseminated infection and assess Cp eradication in OAML patients treated with doxycycline.

This study demonstrates, for the first time, that Cp isolated from conjunctival swabs and PBMC of OAML patients is viable and infectious, confirming that, at the time of lymphoma diagnosis, this microorganism is responsible for both a local and a systemic infection in many OAML patients. Viability and infectivity of Cp in biological samples of OAML patients were demonstrated by the successful in vitro growth and isolation of this microorganism, which is further substantiated by previously reported data on Cp-eradication and lymphoma regression after doxycycline treatment4 and by the direct visualization of intact Chlamydia elementary bodies within OAML-associated macrophages/monocytes by electron microscopy (Ponzoni et al., personal communication). The demonstration that Cp can be isolated and grown in vitro from biological samples of OAML patients improves the evidence level supporting this bacteria–lymphoma association and fulfills the second Koch's postulate. These postulates are still a useful benchmark in judging whether there is a cause-and-effect relationship between an infectious agent and a clinical disease. Before this study, Helicobacter pylori were the unique bacterium with lymphomagenetic potential isolated in pure cultures from lymphoma patients.6

Epidemiological characteristics of OAML patients in this study reflect an exposure background that strongly supports the relationship between chlamydial infection and these lymphomas. These patients are more commonly females living in rural areas, with a history of chronic conjunctivitis and prolonged contact with household animals. These observations deserve to be further investigated since such animals, mostly cats, canaries and parrots are well-known carriers of Cp.12 The occurrence of high rates of Cp infection in these animals16 raises the possibility that human Cp infections, like chronic conjunctivitis,17 may be under-estimated. This infection can be asymptomatic both in pets and humans, and a prolonged contact with an apparently healthy animal shedding infectious chlamydiae can result in repeated infection cycles in humans, with the development of 1 or more lymphomas in some cases.18 These lymphomagenetic potentials may preferentially occur in organs considered as “first barriers” to exposure to air-transported antigens.18 In effect, transmission from animals to humans mainly occurs through aerosols of fecal or feather dust where infectious Cp may remain viable for several months.19

In conclusion, this prospective case–control study demonstrated, for the first time, that Cp is viable and infectious in conjunctival swabs and/or PBMC in a significant proportion of OAML patients. These patients more frequently report prolonged contact with household animals and chronic conjunctivitis, which suggests an increased risk of Cp exposure. Finally, this study demonstrated that asymptomatic Cp infection is extremely rare in blood donors, which indicates a low risk of infection with this microorganism via blood transfusion and underscores the specific relationship between Cp and OAML.

Acknowledgements

This paper is dedicated to the memory of Mr. Sergio Vigo (National Reference Laboratory for Animal Chlamydioses, Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna, Pavia, Italy), whose technical contribution for the isolation of the microorganism in cell culture was critical.

Contribution: A.J.M.F., R.D., M.P., C.D., S.R., and S.M. designed the research, analyzed data and wrote the paper; G.P.D., L.M. and A.G.R. interviewed patients and collected biological samples; R.D., E.P., M.G.C., and L.P. designed and performed PCR analyses; N.V. and S.M. designed and performed culture tests and made the figures.