immune system activation associated with a naturally occurring infection in xenopus laevis

Developmental and Comparative Immunology, Vol. 16, pp. 453-462, 1992 0145-305X/92 $5.00 + .00 Printed in the USA. All rights reserved. Copyright © 1992 Pergamon Press Ltd.

IMMUNE SYSTEM ACTIVATION ASSOCIATED WITH A NATURALLY OCCURRING INFECTION IN Xenopus laevis

Laura Haynes,* Fiona A. Harding,I- Anne D. Koniski,* and Nicholas Cohen*

*Department of Microbiology and immunology, University of Rochester Medical Center, Rochester, NY 14642 and -i-Department of Immunology, University of California, Berkeley, CA 94720

(Submitted October 1991; Accepted January 1992)

OAbstract--Immune system activation corre- lated with a naturally occurring infection has been found in the South African clawed frog Xenopus laevis. The microorganism thought to be the cause of this infection is coccobacilloid and approximately 1 Ixm in diameter. Since this microorganism does not grow on conven- tional bacterial media and it has been observed intracellularly, it may be an obligate intracellular bacterium. It has been found in Xenopus peripheral blood and in highly vascularized organs such as the spleen and liver. Splenomeg- aly is the only pathology thus far described for infected frogs; infection is not associated with increased morbidity or mortality. This infection has been found in all outbred frogs examined in shipments from one South African and three separate North American vendors, and has been transmitted to animals bred and raised in our laboratory. This infection has a profound effect on the immune system of Xe- nopus. Significant numbers of splenocytes from infected individuals exhibit morphology commonly associated with activated T lymphocytes. There is constitutive production of T-cell growth factor (TCGF) and both IgM and IgY. Freshly harvested splenocytes from infected animals proliferate in response to a TCGF- containing supernatant, indicating that they express receptors for TCGF, a trait exclusively exhibited by activated lymphocytes. These splenocytes also show an increase in the activation marker recognized by the monoclonal antibody FJ17.

r~Keywords--Xenopus; Immune system activation; Microbial infection; TCGF; IL-2.

Address correspondence to Laura Haynes.

Introduction

The South African clawed flog Xeno- pus laevis has one of the best character- ized immune systems among the non- mammalian vertebrates. It displays poly- morph ic c lass I and class II m a jo r histocompatibility complex (MHC) anti- gens, which are involved in mixed lymphocyte responses, T-cell cytotoxicity, restricted T-cell proliferation, allograft rejection, and T-B-ce l l cooperat ion [(1); Harding et al. submitted]. It also pro- duces three immunoglobulin isotypes: IgM, IgY (IgG-like), and IgX (found in the gut) (1). Additionally, interleukin 1 (IL-1)- and interleukin 2 (IL-2)-like cytokines have been described (2,3). Xeno- pus T-cell growth factor (TCGF), an IL- 2-like cytokine, is produced by T cells upon stimulation with antigen or mitogen. Like mammalian IL-2, it induces the proliferat ion of lymphoblasts , but has no effect on fleshly harvested resting splenocytes (3).

Although much is known about the Xenopus immune system, there is very little informat ion regarding responses to naturally occurring pathogens. This paper describes what appears to be the a c t i v a t i o n o f the X e n o p u s i m m u n e system by a naturally occurring intracellular microorganism. It also serves as a word of caution to those investigators s tudying the immune sys t em of this species.

453

454 L. Haynes et at.

Materials and Methods

Animals

Large adult female Xenopus laevis were purchased from Xenopus 1 (Ann Arbor, MI), Connecticut Valley Biologi- cal (Southampton, MA), NASCO (Fort Atkinson, WI), or the African Xenopus Facility (Noordhoek, South Africa). J strain Xenopus (1) were bred and main- tained in our laboratory. Assays involving splenocytes from uninfected frogs were performed prior to March 1990, since all animals in our colony exhibit this infection after this date.

Microscopy

Freshly harvested live splenocytes from infected frogs were observed with a Nikon Diaphot-TMD inverted microscope equipped with a hookup to a vid- eocassette recorder (VCR) and a video terminal. Photographs of the video terminal were taken with a 35-mm camera. Photomicrographs of freshly harvested living splenocytes were also taken with a Leitz SM microscope equipped with a Polaroid camera. Adherent splenocytes from infected frogs were obtained by in- cubating fresh splenocytes in complete medium with 1% FBS (see below) for 1 h in a 6-well plate (Costar, Cambridge, MA) with a sterile coverglass in each well. The coverglass was then rinsed with amphibian PBS (APBS), mounted on a slide, and observed with a Zeiss Ax- ioskop equipped with a 35-ram camera.

Production of PHA-induced Supernatants (PHA SN)

PHA SNs containing T-cell growth factor (TCGF) were generated as previously described (3). Briefly, 5 × 106 Xe- nopus splenocytes/mL were incubated in complete medium [Leibovitz's L-15 me-

dium (Gibco, Grand Island, NY) ad- justed to amphibian osmolarity (220 mOsm) and supplemented with 1.25 × 10 -5 M HEPES buffer (Gibco), 100 U/mL penicillin, 100 ~g/mL streptomy- cin (Gibco), 1 x l0 -2 M NaCHO 3, 5 x 10 5 M 2-mercaptoethanol (Sigma)] with 0.25% bovine serum albumin (BSA) and I ~g/mL PHA-P (Sigma, St. Louis, MO). The 24-h and 48-h supernatants were collected and the mitogen was removed from the supernatant by absorption with chicken erythrocytes. The resulting supernatant was precipitated with satu- rated ammonium sulfate (SAS), dialyzed (Spectra/Por membrane, M r cutoff 6000- 8000) with APBS, and sterile filtered before use.

Control (medium alone) and PHA SNs were also generated with splenocytes from three infected J strain Xeno- pus. Twenty-four-h SNs were collected, treated to remove mitogen as above, concentrated with Centricon Microcon- centrators (Amicon, Beverly, MA) with a cutoff of 10,000 Mr, and sterile filtered before use.

Assay for Response to PHA SN

The PHA SN (or control SN) was assayed for activity on fresh splenocytes or thymic blasts (see below) in a 3-day 3H- thymidine incorporation assay. One hundred p~L of the supernatant was incubated with 1 x 105 splenocytes in 100 ~L c o m p l e t e m e d i u m wi th 1% hea t - inactivated fetal bovine serum (FBS, Hyclone, Logan, UT) in 96-well round bottom plates (Costar). All experiments were performed with the same lot of FBS. After 48 h 1 ~Ci/well 3H-thymidine (Amersham, Arlington Heights, IL) was added. The cultures were harvested 24 h later and processed for liquid scintilla- tion spectrometry. All cultures were plated in triplicate and the data are pre- sented as the mean counts per minute (cpm) _+ SE.

Xenopus immune system activation 455

Production of Thymic Blasts

Thymuses were aseptically removed from young adults (6 months to 1 year of age) and cultured in complete medium with 2 ixg/mL PHA-P and 10% SAS PHA SN for the indicated number of days. The blasts were harvested by centrifuga- tion over Histopaque ~ = 1.077 (Sigma) and washed twice in complete medium with 1% FBS. They were then assayed for their response to SNs as detailed above.

AR film) at -70°C until the desired in- tensity was achieved.

Monoclonal Antibodies

FJ17 is an IgM mouse monoclonal antibody that recognizes an unknown determinant on activated Xenopus T lymphocytes (M. Flajnik, personal communication). 6.16 is an IgG mouse anti- Xenopus IgM monoclonal antibody (5) and XYll .2 .2 is an IgG mouse anti- Xenopus IgY monoclonal antibody (6).

Production of Supernatants From Metabollically Labelled Cells

Splenocytes were washed in sterile APBS (SAPBS) and resuspended to 5 × 10 6 splenocytes/mL in RPMI without methionine (RPMI Select-amine Kit, Gibco) at amphibian osmolarity. The cells were incubated for 1 h at 27°C in complete methioine-free medium (supplemented with I00 U/mL penicillin, 100 p.g/mL s t r ep tomyc in , 1 × 1 0 - 2 M

NaHCO3, 5 x 10 -5 M 2-mercaptoethanol, 5 ixg/mL insulin, and 5 ixg/mL trans- ferrin) at 26°C and then put in culture at 5 x 106 splenocytes/mL in complete methionine-free medium with 0.5 mCi 35S- methionine/2 mL with or without 1 to 2 p.g/mL PHA. Supernatants were collected after 36 h and concentrated using Centricon Microconcentrators with a cutoff of 10 ,000 M r.

SDS-Polyacrylamide Gel Electrophoresis (SDS-PAGE)

Twelve percent SDS-reducing gels were prepared, run as described (4) in a vertical gel slab unit (Hoefer Scientific, San Francisco, CA), and stained with 0.25% Coomassie blue. To prepare auto- radiographs, gels were soaked for 1 h in Autofluor (National Diagnostics, Man- ville, N J) and dried. The dried gels were exposed to X-ray film (Kodak X-Omat

Flow Cytometer Analysis

Splenocytes were centrifuged over Histopaque 0 = 1.119 (Sigma), washed twice in complete medium plus 10% FBS, and cultured with 1 Ixg/mL PHA for the indicated period of time. Cells were then washed twice in APBS with 1% BSA and 0.1% NaN 3 (ABN). One hundred IXL of hybridoma supernatant containing the primary monoclonal antibody FJ17 in 0.1% NaN 3 or 100 IXL ABN (negative control) was added to 1 × 10 6

cells. Samples were incubated for 30 min on ice, centrifuged, and 50 IXL of a 1:20 dilution of the secondary antibody, anti- mouse IgM (ix-chain specific)-FITC con- jugate (Sigma) was added. Samples were incubated for an additional 30 min on ice and then washed twice in ABN and resuspended in 1 mL ABN. Analysis was carried out on an EPICS Profile Ana- lyzer (Coulter, Hialeah, FL). The percent-positive population was calculated by subtracting the percentage of cells stained with the secondary antibody alone from the percentage of cells stained with the primary plus secondary antibodies.

Results

Microorganism

Figure 1A is a photograph of splenocytes taken from an infected frog. The

456 L. Haynes et al.

A

Figure 1. Splenocytes from infected Xenopus. (A) Photograph of freshly harvested splenocytes from an infected frog taken from a VCR tape that was recorded through a 60x objective and further amplified. Arrows indicate microorganism; E = erythrocyte; L - lymphocyte. (B) Photomicrograph of infected adherent splenocytes (400x). Arrows indicate microorganism; L - nonlymphoid leu- kocyte; E = erythrocyte.

microorganism is coccobacilloid and is a p p r o x i m a t e l y 1 ixm in d iamete r . It seems to have a dense outer layer that makes it appear refractile when observed by light microscopy. It is found intracellularly only in nonlymphoid leukocytes (Fig. 1B) and exhibits motility when it is ext racel lu lar . The mic roorgan i sm appears to fill the entire cytoplasmic space in these cells, suggesting that it is divid- ing within the cell. It has also been observed in peripheral blood, liver, and tes- tes. Addit ional ly , this mic roorgan ism cannot be cultured on nutrient agar, nu-

trient broth, blood agar, or chocola te agar, suggesting that it may be an obligate intracellular parasite.

Pathology

This m i c r o o r g a n i s m was found in flogs from four vendors and was appar- ently spread to animals bred and raised in our facility. At this time, all f logs that we have examined have been infected. Infected flogs do not exhibit increased mortality or alterations in behavior. The


only internal pathology noted thus far is splenomegaly, which is most likely re- lated to the in vivo activation of the splenocytes. Freshly harvested splenocytes from infected animals (Fig. 2) are en- larged with pseudopodia or are banana- shaped, morphology that is associated with activated lymphocytes (7).

Constitutive Production of Xenopus TCGF

TCGF is secreted by mitogen- or antigen-stimulated Xenopus T lymphocytes and not by unstimulated cells. It induces the proliferation of lymphoblasts but has no effect on unstimulated splenocytes (3). In order to determine if lymphocytes from infected frogs are secreting this cytokine, 24-h control (medium alone) and PHA SNs were simultaneously prepared. J strain Xenopus from our inbred colony were used to eliminate any poten- tial allogeneic responses. After the SNs were treated to remove mitogen (3), they

were assayed for their ability to induce proliferation of 4-day-old thymic blasts (Fig. 3). Both the control and PHA SNs can induce proliferation, although the control SN contains about a third of the activity found in the PHA SN. This indi- cates that the splenocytes from infected animals are const i tut ively secret ing TCGF and that this secretion can be increased by stimulation with mitogen.

Constitutive Production of Cytokines and Immunoglobulin

Figure 4 shows an autoradiograph of control (no mitogen) and PHA-induced 35S-methionine-labelled supernatants run on 12% reducing SDS gels. Superna- tants from an uninfected frog are shown in lanes A (control SN) and B (PHA SN). They show that upon mitogen stimulation there is increased secretion of a 16,000 M r protein. Since only the PHA SN, and not the control SN, from uninfected frogs exhibits biological activity

Figure 2. Activated splenocytes from infected Xenopus. Arrows indicate freshly harvested lymphocytes exhibiting lymphoblastoid morphology (1040×).


8OOO " - - ' T " -

6000

O .

4000

2000

MEDIUMI 1 :4ALONE PHA1:20SN 1:40 I 1 : 4 ~ 1 : 2 0 SN1:40

Figure 3. Biological activity of SNs from infected Xenopus. Twenty-four-h SNs from control (closed bars) and PHA-stimulated (open bars) infected J strain splenocytes were treated to remove mitogen and assayed for stimulatory activity on 4-day-old thymic blasts in a 3-day 3H-thymidine incorporation assay. The results are expressed as mean cpm _+ SE.

(3), this 16,000 Mr protein is most likely involved in this activity. Lanes C (control SN) and D (PHA SN) are supernatants from an infected frog. Lane C

shows that this individual is const i tu- tively secreting 14,000 to 16,000 M r peptides as well as IgM (H (heavy chain) = 73,000 M r, L (light chain) = 26,000

97"

67-

A B C

iii~i~i!iil;!i!i

31-

21-

D E F

14-

i[ ̧ , ,C,I,"

Figure 4. Labelled SN from uninfected and infected Xenopus. 3SS-Methionine-labelled splenocyte SN from one uninfected (lanes A and B) and two different infected (lanes C and D, and E and F) Xenopus were run on a 12% reducing SDS-PAGE and autoradiographed. Lanes A, C, and E are control SN (no mitogen); lanes B, D, and F are mitogen (PHA)-induced SN.


Mr) and IgY (Hv = 69,000 Mr doublet, L = 29,000 M r (1) as determined by immu- noprecipitation with the anti-Xenopus IgM mAb 6.16 (5) and the anti-Xenopus IgY mAb XY11.2.2 (6) (data not shown). St imula t ion with mitogen (lane D) slightly increases the level of secretion. Lanes E (control SN) and F (PHA SN) are supernatants from another infected frog. This individual is only constitutively secreting IgY (lane E), which is increased upon stimulation with mitogen (Lane F). Lane F also shows that stimulation with mitogen induces secretion of 16,000 and 18,000 Mr peptides, which are most likely cytokines involved in TCGF activity. The results from two different infected frogs are shown in order to dem- onstrate the individual variation in response to this infection.

Response to Mitogen-Induced Supernatant

As shown previously by Watkins and Cohen (3), freshly harvested Xenopus splenocytes from uninfected frogs do not respond to PHA SN. This is most likely due to the fact that these cells are in a resting state and, therefore, do not express growth factor receptors. Only after stimulation with antigen or mitogen are they capable of responding to the PHA SN. In representative experiments, splenocytes from an uninfected frog (Fig. 5A) did not proliferate in response to the PHA SN, whereas splenocytes from an infected frog (Fig. 5B) did proliferate. Additionally, both freshly harvested splenocytes and 3-day-old splenic blasts from infected frogs can absorb TCGF activity from PHA SN (data not shown). These results suggest that splenocytes from infected frogs are activated in vivo to express growth factor receptors. As mentioned in Materials and Methods, the assays involving uninfected animals were performed prior to March 1990, since all animals we have examined after

this time appear to be infected. These assays were performed with the same lot of FBS. Although different lots of PHA SN were used, this should have minimal effect on the results since all PHA SNs were prepared similarly. Stimulation in- dices (cpm with SN/cpm with medium) from experiments on fresh splenocytes before March 1990 ranged from 1.0 to 6.4 in 12 experiments assaying 25% PHA SN, whereas those performed after March 1990 ranged from I1.1 to 18.4 in five experiments.

Expression of Activation Markers

The mAb FJ17 recognizes a determinant on activated Xenopus lymhocytes (M. Flajnik, personal communication). Figure 6 is a representative experiment showing the time course of expression of the F J17 determinant on splenocytes from one uninfected and two infected frogs. Fresh splenocytes from the uninfected frog expressed low (<5%) levels of the FJI7 determinant and this expression was increased upon culture with mitogen up to a maximum of approximately 60% positive after 55 h. Freshly harvested splenocytes from infected animals were between 35 and 40% positive for the FJ17 determinant, and stimulation with mitogen for up to 70 h had little effect, indicating maximum expression. Uninfected and infected frogs exhibited no difference in class I or class II expression (data not shown).

Discussion

We have made several observations indicating a recent in vivo activation of lymphocytes from Xenopus raised in our laboratory and provided by four com- mercial suppliers. Freshly harvested splenocytes display a lymphoblastoid morphology, commonly associated with

460 L. Haynes et at.

10000

8000

6000

4000

2000

! i

NCt~ 1:4 1:8 1:16 SN CONCENTRATION

10000

8000

6000

4000

2000

I

w/'///,/////,41 , NCIklE 1:4 1:40 1:20

SN CONCENTRATION

Figure 5. Stimulation of splenocytes by PHA SN. Freshly harvested splenocytes from uninfected (A) and infected (B) Xenopus were assayed for response to PHA SN in a 3-day 3H-thymidine incorporation assay.

activation. These splenocytes exhibit constitutive secretion of immunoglobulins and cytokines. This in vivo upregu- lation of cytokine synthesis may be re- sponsible for the loss of MHC restriction in cytotoxic T cells from these frogs (8). These cells constitutively express TCGF receptors and are capable of responding to supernatant containing this cytokine. In addition, freshly harvested splenocytes express the activation marker detected by the monoclonal antibody FJ 17. They also express increased levels of the thymocyte and T-cell marker detected by the mAb AM22 (data not shown). Similar observat ions have also been

made by other researchers studying the Xenopus immune system (M. Flajnik and J. Horton, personal communications).

These observations have coincided with the appearance of a cocobacilloid microorganism in the spleens of these animals. The microorganism has also been found in peripheral blood and other highly vascularized organs. We theorize that the observed in vivo activation is caused by this microorganism, which most likely came into the lab with a ship- ment of outbred Xenopus. Subsequently, this microorganism has been found in animals purchased from four different suppliers. We believe that it was transmitted


70

• 50 - - UNINFECTED

a, 40

30 ' INFECTED

20-

10-

0 I I 1 I I i I I

0 10 20 30 40 50 60 70 80

HOURS IN CULTURE WITH MI'rOGEN (PHA)

Figure 6. Express ion of the Xenopus T-cell act ivat ion marker detected by F J17. Splenocytes f rom infected and uninfected animals were cul tured with PHA for the indicated per iod of time, stained with the pr imary mAb F J17 and secondary goat ant i -mouse IgM-FITC, and analyzed on a f low cytometer• The percent F J17 posit ive is shown•

to all Xenopus in our laboratory, includ- ing inbred lines in three different rooms via contaminated water on gloves or nets. Splenomegaly appears to be the only pathological consequence of this infection thus far, with no increase in mortality.

This microorganism appears to thrive in in vitro cultures of Xenopus splenocytes, although it does not overgrow the cells and causes no obvious cytopathic effects. It also appears to be an obligate intracellular parasite, since it does not exhibit growth in cell-free media. The observed effects on the immune system are not unexpected since T-cell- mediated immunity, in the form of secreted lymphokines and cytotoxic T cells, is the main defense against this type of pathogen in mammals. In addition, intracellular parasites, in general, elicit an antibody response that is usually not pro-

tective, since the antibodies do not have access to the microorganism (9).

Although a definite diagnosis cannot be made at this time, the size (1 ixm diameter) of this microorganism suggests that it may belong to the family Rickett- siaceae and possibly the genra Coxiella or Ehrlichia, since arthropod vectors do not seem important in its transmission (10).

Acknowledgements - -We would like to thank R. Bauserman and R. Waugh for the video- enhanced microscopy and M. Flajnik for gen- erously supplying the monoclonal antibody FJ17 used in this paper. We would also like to thank the University of Rochester vivarium for the pathology report on the infected frogs. Finally, we would like to thank H. Holtfreter for the photomicrograph of the in vivo activated splenocytes and for her helpful advice. Supported by NIH grant HD-07901.

References

1. Du Pasquier, L.; Schwager, J.; Flajnik, M. E The immune system of Xenopus. Annu. Rev. Immunol. 7:251-275; 1985.

2. Watkins, D.; Parsons, S. C.; Cohen, N. A factor with interleukin-l-like activity is produced by peritoneal cells from the frog, Xenopus laevis. Immunology 62:669-673; 1987.

3. Watkins, D.; Cohen, N. Mitogen-activated Xenopus laevis lymphocytes produce a T- cell growth factor. Immunology 62:119-125; 1987.

4. Laemmli, W. K. Cleavage of structural pro- teins during the assembly of the head of bac- teriophage T4. Nature 227:680-685; 1970.


5. Bleicher, E A.; Cohen, N. Monoclonal anti- lgM can separate T cell from B cell prolifera- tive responses in the frog, Xenopus laevis. J. Immunol. 127:1549-1555; 1981.

6. Shoemaker, T. M.; Bleicher, P. A.; Cohen, N. Studies on the immunoglobulins of the clawed frog Xenopus. Dev. Comp. Immunol. Suppl. 3:156; 1984.

7. Watkins, D. T cell function in Xenopus. Roch- ester, NY: University of Rochester; 1985. Ph.D. thesis.

8. Harding, E A. Molecular and cellular aspects of the immune systems of lower vertebrates. Rochester, NY: University of Rochester; 1990. Ph.D. thesis.

9. Ryan, J. L. Bacterial diseases. In: Sites, D. R; Terr, A. 1., eds. Basic and clinical immunology. Norwalk, CT: Appleton & Lange; 1991: 637-645.

10. Krieg, N. R.; Holt, J. G., editors. Bergy's man- ual of systemic bacteriology, volume 1. Balti- more, MD: Williams & Wilkins; 1984: 687-709.

immune system activation associated with a naturally occurring infection in xenopus laevis

Documents