Veterinary Immunology and Immunopathology, 34 ( 1992 ) 127-138
127
Elsevier Science Publishers B.V., Amsterdam
Identification and characterization of rat intestinal lamina propria cells: consequences of microbial colonization C h r i s t o p h e r J. W o o l v e r t o n a, L i s a C. H o l t b, D e b o r a h M i t c h e l l c a n d R . B a l f o u r S a r t o r a,b,d
aCenterfor Gastrointestinal Biology and Disease, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA bDepartment of Microbiology and Immunology, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA CBarrierIntact Facility, North Carolina State University, College of Veterinary Medicine, Raleigh, NC 27599, USA dDepartment of Medicine, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA (Accepted 15 January 1992)
ABSTRACT Woolverton, C.J., Holt, L.C., Mitchell, D. and Sartor, R.B., 1992. Identification and characterization of rat intestinal lamina propria cells: consequences of microbial colonization. Vet. Immunol. Immunopathol., 34: 127-138. Germ-free (GF) animals exhibit an abnormally diminished, cell-mediated immune response which can be rapidly normalized by bacterial colonization of the intestine. This conventionalization suggests that the development and/or regulation of the immune system is dependent upon intestinal bacteria or their products. Here we consider the ontogeny of gut-associated lymphoid tissue (GALT) immunocytes by isolating and characterizing the intestinal lamina propria cells (LPC) of GF rats responding to bacterial colonization or an irrelevant protein antigen, and compared to LPC of specific pathogen-free (SPF) rats which were conventionalized (CV) from birth. Isolation of cells was accomplished by successive EDTA washings of small intestine to remove the epithelium, and enzymatic digestion of the tissue generating single-cell suspensions. Resulting cell suspensions were characterized by monoclonal antibodies directed against leukocyte epitopes using flow cytometry. Functional characterization was measured by the tritiated thymidine proliferation assay with concanalin A (Con A) and lipopolysaccharide (LPS) as co-stimulators. Germ-free and SPF rats had fewer total LPC than CV rats. Antibody staining revealed that GF rats had fewer total leukocytes than CV and SPF rats, and that CV rats had a greater percentage of T-cells and cells positive for the C3 receptor than GF rats. Co-stimulation of LPC with mitogens only increased proliferation of cells from CV rats compared to GF and SPF rats. In addition, spleen cells from CV rats demonstrated significantly enhanced proliferative responses compared to spleen cells from GF rat and were more analogous to spleen cells from SPF rats in their ability to proliferate in vitro, with and without mitogens. We conclude that T-cells and CD35-positive ( C 3 B R + ) cells are recruited and/or proliferate in response to intestinal bacteria and/or their products, and that this results in the induction of immune competency.
Correspondence to: C.J. Woolverton, Suite 61563, Austin College, Sherman, T X 75091, USA.
© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-2427/92/$05.00
128
c.J. WOOLVERTONET AL.
ABBREVIATIONS
BSA, bovine serum albumin; cmf-HBSS, calcium magnesium-free Hank's balanced salts solution; Con A, concanavalin A; CPM, counts per minute; CV, conventionalized; DMEM, Dulbecco's minimum essential medium containing 10 mmol HEPES and 1000 U penicillin and 1000/~g m l - ' streptomycin; DTH, delayed-type hypersensitivity; FACS, fluorescence-activated cell sorter; GALT, gut-associated lymphoid tissue; GF, germ-free; HSA, human serum albumin; IEL, intraepithelial lymphocytes; JMOD, Joklick's modified MEM with 0.5% BSA, 20 mmol HEPES, 1 mol L-glutamine, 7.5% sodium bicarbonate and 1 mg ml - ' gentamicin; LP, lamina propria; LPC, lamina propria cells; LPS, lipopolysaccharide; Mabs, monoclonal antibodies: MEM, minimum essential medium; PBS, phosphate-buffered saline; SPF, specific pathogenfree.
INTRODUCTION
The mammalian immune response is the result of environmental as well as genetic regulation. This is demonstrated by the differences between germ-free (GF) and conventional (CV) animals. Germ-free animals exhibit an underdeveloped immune system and are therefore analogous to neonatal animals. Their lymphoid tissues, including the spleen, thymus, lymph nodes, gut-associated lymphoid tissue (GALT), and intestinal lamina propria (LP), are hypocellular (Luckey, 1963; Hazenberg et al., 1981; Durkin et al., 1989; Bandeira et al., 1990). These animals have decreased immunological abilities, particularly of the cell-mediated limb as manifested by abnormal macrophage, T-cell and B-cell function, and by diminished resistance to infection (Luckey, 1963; Lee and Balish, 1981; Wade et al., 1986). This diminished immunological response is rapidly normalized as the GF animal is exposed to normal microbial flora (MacDonald and Carter, 1979; Lee and Balish, 1981; Durkin et al., 1989; Lefran~ois and Goodman, 1989; Bandeira et al., 1990), the process &conventionalization. It has been assumed that stimulation of the cell-mediated immunity is dependent upon viable bacteria (MacDonald and Carter, 1979). However, several studies have shown that purified bacterial cell wall polymers can change the cellular profile of GF animals to be more like CV animals (Wannemuehler et al., 1982; Durkin et al., 1989). Immunologic cells of GALT are found localized in discretely organized aggregates (e.g. Peyer's patches), disseminated throughout the intestinal LP (LPC), or between intestinal epithelial cells (e.g. intraepithelial lymphocytes, IEL). The Peyer's patch immunocytes, LPC and IEL then represent the interface of a closed, host immune system and the external, non-self, environment. Here, microbial and dietary antigen gains entry into the peripheral circulation leading to (systemic) immunologic sensitization (Strobel and Fer-
COLONIZATION OF RAT INTESTINAL LAMINA PROPRIA CELLS
129
guson, 1986). Further, uptake of macromolecules (Walker et al., 1972) and even whole bacteria (Owen et al., 1986 ) by specialized epithelial cells of Peyer's patches may be the normal mechanism of mucosal antigen uptake. Induction of the afferent GALT-mediated i m m u n e response probably arises in the organized lymphoid aggregates, while LPC and IEL are responsible for efferent, effector activities (James et al., 1988). However, little is known about the ontogeny of these GALT immunocytes in response to initial antigen stimulation. This study was therefore designed to examine the cellular changes in the GF rat intestinal LP in response to bacterial colonization by isolating and characterizing the responding LPC. METHODS AND MATERIALS
Animals Female, age-matched, GF Sprague-Dawley rats were obtained from the Barrier Intact Animal Facility (Core Center in Diarrheal Diseases, Raleigh, NC), housed no more than three per cage, and documented free of aerobic and anaerobic bacteria and viruses by gram stain and culture. Germ-free rats were maintained in a plastic Trexler isolator with free access to autoclaved lab chow and water throughout the experiment. Conventional, specific pathogen-free (SPF) rats were obtained as age- and sex-matched controls from Harlan Sprague-Dawley (Indianapolis, IN), and housed as above.
Experimental procedure Germ-free rats were gavage-fed a sterile solution of human serum albumin ( 100/~g g - 1 body weight) or viable SPF rat fecal flora, five times during 1 week. Specific pathogen-free rats were sham-treated by gavage feeding sterile distilled water (the h u m a n serum albumin (HSA) and bacterial suspension vehicle) according to the same schedule. Eighteen hours after the last feeding rats were killed and small intestines aseptically removed. Spleens were also removed and splenocytes included in assays as controls.
Isolation of intestinal lamina propria cells Lamina propria cells (LPC) were isolated using a modification of the method of Bull and Bookman (1979). After expelling fecal matter with cold calcium magnesium-free Hank's balanced salts solution (cmf-HBSS), intestines were placed in a plastic petri dish, rinsed with cold cmf-HBSS supplemented with 0.5% bovine serum albumin (BSA), 20 mmol HEPES, 1 mmol L-glutamine, 7.5% sodium bicarbonate, and 1 mg ml-~ gentamicin (wash buffer), and cut into 3-5 cm pieces. Intestinal tissue pieces were incubated
130
C.J, WOOLVERTON ET AL.
with wash buffer supplemented with 0.75 m m o l EDTA for 20 min at 37°C, then copiously rinsed with wash buffer. This procedure was repeated twice to remove mucosal epithelial ceils. The intestinal tissue pieces were next digested by incubation in Joklick's modified minimal essential medium (MEM) supplemented with 0.5% BSA, 20 m m o l HEPES, 1 mmol L-glutamine, 7.5% sodium bicarbonate, and 1 mg ml-1 gentamicin (JMOD) containing 1.5 mg ml-~ dispase (Boehringer Mannheim, Indianapolis, IN) and 20 U m l - l collagenase (Sigma Chemicals, St. Louis, MO ). Tissue pieces were incubated on a Lab Line orbital shaker to keep the suspension constantly swirling for 30 rain at 37°C, 60 rain at 4°C, and finally 30 rain at 37°C. Cells were collected by passing the tissue digest through a 60 # m sterile, wire mesh, to exclude tissue. The cellular effluent was centrifuged at 600 × g for 5 min at 20 °C and washed twice in JMOD. Total cell counts were performed and corresponding viabilities were determined by trypan blue dye exclusion. Results are reported as mean cell counts.
Isolation of splenocytes Rat spleens were placed in sterile petri dishes containing 10 ml of Dulbecco's MEM containing 10 m m o l HEPES and a suspension of 1000 U penicillin and 1000 #g m1-1 streptomycin ( D M E M ) , pH 7.4. Spleen cells were obtained by making lateral incisions on both sides of the spleen with a sterile scalpel blade, and massaging the cells from within the reticular tissue of the spleen using sterile forceps. Cells were washed once in DMEM, counted, and resuspended as 1 × 107 m l - l in RPMI- 1640 m e d i u m supplemented with 10% heat-inactivated fetal calf serum, 2 m m o l L-glutamine, 0.18% sodium bicarbonate, 5 × 10- 5 mol 2-mercaptoethanol, and 1000 U penicillin and 1000 #g m l - 1 streptomycin (complete m e d i u m ).
Determination of lamina propria cell leukocyte subsets using monoclonal antibodies Germ-flee or conventional rat LP cells were characterized using monospecific antibodies to rat leukocyte surface markers. Monoclonal antibodies (Mabs) directed against rat leukocyte epitopes were obtained from the North Carolina State University fluorescence-activated cell sorter (FACS)/hybridoma facility (School of Veterinary Medicine, North Carolina State University, Raleigh, NC). These mouse anti-rat leukocyte Mabs recognized the following determinants: MRC OX- 1, m o n o m o r p h i c c o m m o n leukocyte antigen (Sunderland et al., 1979 ); MRC OX-8, most thymocytes, the cytotoxic/suppressor T-cell subset, and the majority of natural killer cells (Brideau et al., 1980; Cantrell et al., 1982 ); MRC OX-12, to-light chains (Hunt and Fowler, 1981 ); MRC OX-19, all thymocytes and peripheral T-cells, and a subset of
COLONIZATION OF RAT INTESTINAL LAMINA PROPRIA CELLS
131
B-cells (Dallman et al., 1984); MRC OX-42, a complement receptor Type-3 determinant on macrophages, granulocytes, and dendritic cells (Robinson et al., 1986); and W3/25, most thymocytes, the helper T-cell subset, and macrophages (White et al., 1978; Cantrell et al., 1982 ). Splenocytes were analyzed on a Coulter 753 FACS (Coulter, Hialeigha, FL), essentially as described by White et al. (1978). Briefly, 100 ~l aliquots of cells (106 cells) were suspended in RPMI-1640 containing no serum or antibiotics, and incubated with 100/A of each respective Mab for 30 min on ice. The cells were then washed twice in PBS and incubated with 50/d of FITC-conjugated goat anti-mouse IgG (Cappel Laboratories, West Chester, PA) on ice for 30 min. Cells were washed twice in PBS and examined by FACS for fluorescence. Propridium iodide was added to each tube to gate out non-viable cells. Controls included samples of cells incubated without the primary Mab and cells incubated with primary Mab but without the FITC-conjugated secondary antibody. Results are reported as the mean percentage of cells that are recognized by a given Mab, after subtracting the non-specific (background) binding to cells by the FITC-conjugated secondary antibody. This background was usually between 5% and 9%.
Lymphocyte proliferation assays Rat splenocytes were seeded in 96-well microtiter plates (no. 3595, Coastar, Cambridge, MA) as 1 × 10 6 per well and cultured in 250/d complete medium (final vol. ). Cells were incubated with concanavalin A (Con A) ( l/lg per well, Pharmacia Fine Chemicals, Uppsala, Sweden), bacterial lipopolysaccharide ( l/~g ml-~, Escheria coli 0111 :B4 LPS, Ribi Immunochemicals, Hamilton, MT), or complete medium alone. Cells were cultured at 37 °C for 2 days and pulsed with 3H-thymidine (0.5 #Ci per well) during the last 6 h of culture. Cultures were harvested onto glass-fiber filters for measurement of 3H-thymidine incorporation by liquid scintillation spectrophotometry. Data are expressed as the arithmetic mean counts per minute (CPM) per l 0 6 cells from at least triplicate cultures per animal per group.
Statistics Mean values obtained for each experimental group were compared by nested analysis of variance with significance determinations made by the KruskalWallis or Student's t-tests (Conover, 1980). RESULTS
The enzymatic digestion of the LP tissue further resulted in the recovery of a significantly ( P = 0.004) larger population of LPC from CV rats than from
132
¢.J. WOOLVERTON ET AL.
control GF or SPF rats (Fig. 1 ). Analysis of the cellular component of the LP was made by screening the cell suspension with a battery of Mabs to rat leukocyte epitopes and subsequent identification by flow cytometry. Initial observations of two-parameter analyses showed marked differences between the forward angle and 90 ° light-scatter profiles between cells from GF and CV rats, indicating a shift from smaller, morphologically complex cells to larger, less complex cells (Fig. 2 ). These differences were elucidated by identifying the percentage of leukocytes staining positive for respective leukocyte markers in each group. The data of Table 1 characterize the respective LP populations of cells. Of note are the increases in the number of LPC from CV rats
LO
20
o
-
X v
t
I:C iii f:D
=E Z
Ill rJ Z iii
=E
GERM FREE
CONVENTIONAL ANIMAL
SPF
STATUS
Fig. 1. Effect o f microbial antigen in the intestine of GF, CV or SPF rats on m e a n LPC number. Cells were obtained by enzymatic digestion of intestines. N u m b e r s represent m e a n viable cells _+ 1 SD. * P < 0.001 by Student's t-test c o m p a r e d to the G F control.
FALS
FALS
. "-, .-'~.7.'-'.
GL
A GL
GL
B
,I90
,190
GL
Fig. 2. The effects of in vivo c o n v e n t i o n a l i z a t i o n on rat intestinal LPC as measured by flow cytometry. Forward-angle light-scatter ( F A L S ) is plotted against 90 ° light-scatter ( L I 9 0 ) to compare cell size and complexity. Panel A depicts data from one representative sham-treated G F rat fed h u m a n serum a l b u m i n five times during 1 week. Panel B depicts data from one representative experimental rat fed viable fecal flora five times during 1 week.
COLONIZATION OF RAT INTESTINAL LAMINA PROPRIA CELLS
133
TABLE 1
Percent monoclonal antibody staining of intestinal lamina propria cells from germ-free, conventionalized, and specific pathogen-free rats MAB
Percent staining
OX-I OX-8 OX-12 OX-19 OX-42 W3/25
GF
CV
SPF
1 8 . 7 + 1.3 9.7+3.8 7.5+3.2 7.7+5.0 15.6+ 1.8 12.4+4.1
24.6+3.2* 13.4+7.9 13.4+6.7 19.7+3.0" 29.4+5.1" 17.9+0.3 ~
33.0+7.1' 4.7+0.1 ND 14.1 + 1.2' 14.4+8.4 16.3+4.4
* P < 0.05 by Student's t-test compared to GF controls. ~P< 0.07 by Student's t-test compared to GF controls. Mean values + 1 SD are shown for 20 000 cells counted from which non-specific background fluores-
cence was subtracted. Monoclonal antibody specificity is described in Methods. ND, not determined. 20
~.) ¢a ...J
16
O
o. L)
8
o
4 o (31:
cv IN
VlVO
SPF
TREATMENT
Fig. 3. In vitro proliferation of intestinal lamina propria cells (LPC) from GF, CV and SPF rats. Triplicate wells containing 106 cells from each rat were cultured with medium alone (11), 1/tg per well Con A ( [] ), or 1/tg per well LPS ({~) for 48 h and analyzed for tritiated thymidine incorporation (mean+ 1 SD). *Significance between treated groups in vitro at P
127
Elsevier Science Publishers B.V., Amsterdam
Identification and characterization of rat intestinal lamina propria cells: consequences of microbial colonization C h r i s t o p h e r J. W o o l v e r t o n a, L i s a C. H o l t b, D e b o r a h M i t c h e l l c a n d R . B a l f o u r S a r t o r a,b,d
aCenterfor Gastrointestinal Biology and Disease, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA bDepartment of Microbiology and Immunology, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA CBarrierIntact Facility, North Carolina State University, College of Veterinary Medicine, Raleigh, NC 27599, USA dDepartment of Medicine, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA (Accepted 15 January 1992)
ABSTRACT Woolverton, C.J., Holt, L.C., Mitchell, D. and Sartor, R.B., 1992. Identification and characterization of rat intestinal lamina propria cells: consequences of microbial colonization. Vet. Immunol. Immunopathol., 34: 127-138. Germ-free (GF) animals exhibit an abnormally diminished, cell-mediated immune response which can be rapidly normalized by bacterial colonization of the intestine. This conventionalization suggests that the development and/or regulation of the immune system is dependent upon intestinal bacteria or their products. Here we consider the ontogeny of gut-associated lymphoid tissue (GALT) immunocytes by isolating and characterizing the intestinal lamina propria cells (LPC) of GF rats responding to bacterial colonization or an irrelevant protein antigen, and compared to LPC of specific pathogen-free (SPF) rats which were conventionalized (CV) from birth. Isolation of cells was accomplished by successive EDTA washings of small intestine to remove the epithelium, and enzymatic digestion of the tissue generating single-cell suspensions. Resulting cell suspensions were characterized by monoclonal antibodies directed against leukocyte epitopes using flow cytometry. Functional characterization was measured by the tritiated thymidine proliferation assay with concanalin A (Con A) and lipopolysaccharide (LPS) as co-stimulators. Germ-free and SPF rats had fewer total LPC than CV rats. Antibody staining revealed that GF rats had fewer total leukocytes than CV and SPF rats, and that CV rats had a greater percentage of T-cells and cells positive for the C3 receptor than GF rats. Co-stimulation of LPC with mitogens only increased proliferation of cells from CV rats compared to GF and SPF rats. In addition, spleen cells from CV rats demonstrated significantly enhanced proliferative responses compared to spleen cells from GF rat and were more analogous to spleen cells from SPF rats in their ability to proliferate in vitro, with and without mitogens. We conclude that T-cells and CD35-positive ( C 3 B R + ) cells are recruited and/or proliferate in response to intestinal bacteria and/or their products, and that this results in the induction of immune competency.
Correspondence to: C.J. Woolverton, Suite 61563, Austin College, Sherman, T X 75091, USA.
© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-2427/92/$05.00
128
c.J. WOOLVERTONET AL.
ABBREVIATIONS
BSA, bovine serum albumin; cmf-HBSS, calcium magnesium-free Hank's balanced salts solution; Con A, concanavalin A; CPM, counts per minute; CV, conventionalized; DMEM, Dulbecco's minimum essential medium containing 10 mmol HEPES and 1000 U penicillin and 1000/~g m l - ' streptomycin; DTH, delayed-type hypersensitivity; FACS, fluorescence-activated cell sorter; GALT, gut-associated lymphoid tissue; GF, germ-free; HSA, human serum albumin; IEL, intraepithelial lymphocytes; JMOD, Joklick's modified MEM with 0.5% BSA, 20 mmol HEPES, 1 mol L-glutamine, 7.5% sodium bicarbonate and 1 mg ml - ' gentamicin; LP, lamina propria; LPC, lamina propria cells; LPS, lipopolysaccharide; Mabs, monoclonal antibodies: MEM, minimum essential medium; PBS, phosphate-buffered saline; SPF, specific pathogenfree.
INTRODUCTION
The mammalian immune response is the result of environmental as well as genetic regulation. This is demonstrated by the differences between germ-free (GF) and conventional (CV) animals. Germ-free animals exhibit an underdeveloped immune system and are therefore analogous to neonatal animals. Their lymphoid tissues, including the spleen, thymus, lymph nodes, gut-associated lymphoid tissue (GALT), and intestinal lamina propria (LP), are hypocellular (Luckey, 1963; Hazenberg et al., 1981; Durkin et al., 1989; Bandeira et al., 1990). These animals have decreased immunological abilities, particularly of the cell-mediated limb as manifested by abnormal macrophage, T-cell and B-cell function, and by diminished resistance to infection (Luckey, 1963; Lee and Balish, 1981; Wade et al., 1986). This diminished immunological response is rapidly normalized as the GF animal is exposed to normal microbial flora (MacDonald and Carter, 1979; Lee and Balish, 1981; Durkin et al., 1989; Lefran~ois and Goodman, 1989; Bandeira et al., 1990), the process &conventionalization. It has been assumed that stimulation of the cell-mediated immunity is dependent upon viable bacteria (MacDonald and Carter, 1979). However, several studies have shown that purified bacterial cell wall polymers can change the cellular profile of GF animals to be more like CV animals (Wannemuehler et al., 1982; Durkin et al., 1989). Immunologic cells of GALT are found localized in discretely organized aggregates (e.g. Peyer's patches), disseminated throughout the intestinal LP (LPC), or between intestinal epithelial cells (e.g. intraepithelial lymphocytes, IEL). The Peyer's patch immunocytes, LPC and IEL then represent the interface of a closed, host immune system and the external, non-self, environment. Here, microbial and dietary antigen gains entry into the peripheral circulation leading to (systemic) immunologic sensitization (Strobel and Fer-
COLONIZATION OF RAT INTESTINAL LAMINA PROPRIA CELLS
129
guson, 1986). Further, uptake of macromolecules (Walker et al., 1972) and even whole bacteria (Owen et al., 1986 ) by specialized epithelial cells of Peyer's patches may be the normal mechanism of mucosal antigen uptake. Induction of the afferent GALT-mediated i m m u n e response probably arises in the organized lymphoid aggregates, while LPC and IEL are responsible for efferent, effector activities (James et al., 1988). However, little is known about the ontogeny of these GALT immunocytes in response to initial antigen stimulation. This study was therefore designed to examine the cellular changes in the GF rat intestinal LP in response to bacterial colonization by isolating and characterizing the responding LPC. METHODS AND MATERIALS
Animals Female, age-matched, GF Sprague-Dawley rats were obtained from the Barrier Intact Animal Facility (Core Center in Diarrheal Diseases, Raleigh, NC), housed no more than three per cage, and documented free of aerobic and anaerobic bacteria and viruses by gram stain and culture. Germ-free rats were maintained in a plastic Trexler isolator with free access to autoclaved lab chow and water throughout the experiment. Conventional, specific pathogen-free (SPF) rats were obtained as age- and sex-matched controls from Harlan Sprague-Dawley (Indianapolis, IN), and housed as above.
Experimental procedure Germ-free rats were gavage-fed a sterile solution of human serum albumin ( 100/~g g - 1 body weight) or viable SPF rat fecal flora, five times during 1 week. Specific pathogen-free rats were sham-treated by gavage feeding sterile distilled water (the h u m a n serum albumin (HSA) and bacterial suspension vehicle) according to the same schedule. Eighteen hours after the last feeding rats were killed and small intestines aseptically removed. Spleens were also removed and splenocytes included in assays as controls.
Isolation of intestinal lamina propria cells Lamina propria cells (LPC) were isolated using a modification of the method of Bull and Bookman (1979). After expelling fecal matter with cold calcium magnesium-free Hank's balanced salts solution (cmf-HBSS), intestines were placed in a plastic petri dish, rinsed with cold cmf-HBSS supplemented with 0.5% bovine serum albumin (BSA), 20 mmol HEPES, 1 mmol L-glutamine, 7.5% sodium bicarbonate, and 1 mg ml-~ gentamicin (wash buffer), and cut into 3-5 cm pieces. Intestinal tissue pieces were incubated
130
C.J, WOOLVERTON ET AL.
with wash buffer supplemented with 0.75 m m o l EDTA for 20 min at 37°C, then copiously rinsed with wash buffer. This procedure was repeated twice to remove mucosal epithelial ceils. The intestinal tissue pieces were next digested by incubation in Joklick's modified minimal essential medium (MEM) supplemented with 0.5% BSA, 20 m m o l HEPES, 1 mmol L-glutamine, 7.5% sodium bicarbonate, and 1 mg ml-1 gentamicin (JMOD) containing 1.5 mg ml-~ dispase (Boehringer Mannheim, Indianapolis, IN) and 20 U m l - l collagenase (Sigma Chemicals, St. Louis, MO ). Tissue pieces were incubated on a Lab Line orbital shaker to keep the suspension constantly swirling for 30 rain at 37°C, 60 rain at 4°C, and finally 30 rain at 37°C. Cells were collected by passing the tissue digest through a 60 # m sterile, wire mesh, to exclude tissue. The cellular effluent was centrifuged at 600 × g for 5 min at 20 °C and washed twice in JMOD. Total cell counts were performed and corresponding viabilities were determined by trypan blue dye exclusion. Results are reported as mean cell counts.
Isolation of splenocytes Rat spleens were placed in sterile petri dishes containing 10 ml of Dulbecco's MEM containing 10 m m o l HEPES and a suspension of 1000 U penicillin and 1000 #g m1-1 streptomycin ( D M E M ) , pH 7.4. Spleen cells were obtained by making lateral incisions on both sides of the spleen with a sterile scalpel blade, and massaging the cells from within the reticular tissue of the spleen using sterile forceps. Cells were washed once in DMEM, counted, and resuspended as 1 × 107 m l - l in RPMI- 1640 m e d i u m supplemented with 10% heat-inactivated fetal calf serum, 2 m m o l L-glutamine, 0.18% sodium bicarbonate, 5 × 10- 5 mol 2-mercaptoethanol, and 1000 U penicillin and 1000 #g m l - 1 streptomycin (complete m e d i u m ).
Determination of lamina propria cell leukocyte subsets using monoclonal antibodies Germ-flee or conventional rat LP cells were characterized using monospecific antibodies to rat leukocyte surface markers. Monoclonal antibodies (Mabs) directed against rat leukocyte epitopes were obtained from the North Carolina State University fluorescence-activated cell sorter (FACS)/hybridoma facility (School of Veterinary Medicine, North Carolina State University, Raleigh, NC). These mouse anti-rat leukocyte Mabs recognized the following determinants: MRC OX- 1, m o n o m o r p h i c c o m m o n leukocyte antigen (Sunderland et al., 1979 ); MRC OX-8, most thymocytes, the cytotoxic/suppressor T-cell subset, and the majority of natural killer cells (Brideau et al., 1980; Cantrell et al., 1982 ); MRC OX-12, to-light chains (Hunt and Fowler, 1981 ); MRC OX-19, all thymocytes and peripheral T-cells, and a subset of
COLONIZATION OF RAT INTESTINAL LAMINA PROPRIA CELLS
131
B-cells (Dallman et al., 1984); MRC OX-42, a complement receptor Type-3 determinant on macrophages, granulocytes, and dendritic cells (Robinson et al., 1986); and W3/25, most thymocytes, the helper T-cell subset, and macrophages (White et al., 1978; Cantrell et al., 1982 ). Splenocytes were analyzed on a Coulter 753 FACS (Coulter, Hialeigha, FL), essentially as described by White et al. (1978). Briefly, 100 ~l aliquots of cells (106 cells) were suspended in RPMI-1640 containing no serum or antibiotics, and incubated with 100/A of each respective Mab for 30 min on ice. The cells were then washed twice in PBS and incubated with 50/d of FITC-conjugated goat anti-mouse IgG (Cappel Laboratories, West Chester, PA) on ice for 30 min. Cells were washed twice in PBS and examined by FACS for fluorescence. Propridium iodide was added to each tube to gate out non-viable cells. Controls included samples of cells incubated without the primary Mab and cells incubated with primary Mab but without the FITC-conjugated secondary antibody. Results are reported as the mean percentage of cells that are recognized by a given Mab, after subtracting the non-specific (background) binding to cells by the FITC-conjugated secondary antibody. This background was usually between 5% and 9%.
Lymphocyte proliferation assays Rat splenocytes were seeded in 96-well microtiter plates (no. 3595, Coastar, Cambridge, MA) as 1 × 10 6 per well and cultured in 250/d complete medium (final vol. ). Cells were incubated with concanavalin A (Con A) ( l/lg per well, Pharmacia Fine Chemicals, Uppsala, Sweden), bacterial lipopolysaccharide ( l/~g ml-~, Escheria coli 0111 :B4 LPS, Ribi Immunochemicals, Hamilton, MT), or complete medium alone. Cells were cultured at 37 °C for 2 days and pulsed with 3H-thymidine (0.5 #Ci per well) during the last 6 h of culture. Cultures were harvested onto glass-fiber filters for measurement of 3H-thymidine incorporation by liquid scintillation spectrophotometry. Data are expressed as the arithmetic mean counts per minute (CPM) per l 0 6 cells from at least triplicate cultures per animal per group.
Statistics Mean values obtained for each experimental group were compared by nested analysis of variance with significance determinations made by the KruskalWallis or Student's t-tests (Conover, 1980). RESULTS
The enzymatic digestion of the LP tissue further resulted in the recovery of a significantly ( P = 0.004) larger population of LPC from CV rats than from
132
¢.J. WOOLVERTON ET AL.
control GF or SPF rats (Fig. 1 ). Analysis of the cellular component of the LP was made by screening the cell suspension with a battery of Mabs to rat leukocyte epitopes and subsequent identification by flow cytometry. Initial observations of two-parameter analyses showed marked differences between the forward angle and 90 ° light-scatter profiles between cells from GF and CV rats, indicating a shift from smaller, morphologically complex cells to larger, less complex cells (Fig. 2 ). These differences were elucidated by identifying the percentage of leukocytes staining positive for respective leukocyte markers in each group. The data of Table 1 characterize the respective LP populations of cells. Of note are the increases in the number of LPC from CV rats
LO
20
o
-
X v
t
I:C iii f:D
=E Z
Ill rJ Z iii
=E
GERM FREE
CONVENTIONAL ANIMAL
SPF
STATUS
Fig. 1. Effect o f microbial antigen in the intestine of GF, CV or SPF rats on m e a n LPC number. Cells were obtained by enzymatic digestion of intestines. N u m b e r s represent m e a n viable cells _+ 1 SD. * P < 0.001 by Student's t-test c o m p a r e d to the G F control.
FALS
FALS
. "-, .-'~.7.'-'.
GL
A GL
GL
B
,I90
,190
GL
Fig. 2. The effects of in vivo c o n v e n t i o n a l i z a t i o n on rat intestinal LPC as measured by flow cytometry. Forward-angle light-scatter ( F A L S ) is plotted against 90 ° light-scatter ( L I 9 0 ) to compare cell size and complexity. Panel A depicts data from one representative sham-treated G F rat fed h u m a n serum a l b u m i n five times during 1 week. Panel B depicts data from one representative experimental rat fed viable fecal flora five times during 1 week.
COLONIZATION OF RAT INTESTINAL LAMINA PROPRIA CELLS
133
TABLE 1
Percent monoclonal antibody staining of intestinal lamina propria cells from germ-free, conventionalized, and specific pathogen-free rats MAB
Percent staining
OX-I OX-8 OX-12 OX-19 OX-42 W3/25
GF
CV
SPF
1 8 . 7 + 1.3 9.7+3.8 7.5+3.2 7.7+5.0 15.6+ 1.8 12.4+4.1
24.6+3.2* 13.4+7.9 13.4+6.7 19.7+3.0" 29.4+5.1" 17.9+0.3 ~
33.0+7.1' 4.7+0.1 ND 14.1 + 1.2' 14.4+8.4 16.3+4.4
* P < 0.05 by Student's t-test compared to GF controls. ~P< 0.07 by Student's t-test compared to GF controls. Mean values + 1 SD are shown for 20 000 cells counted from which non-specific background fluores-
cence was subtracted. Monoclonal antibody specificity is described in Methods. ND, not determined. 20
~.) ¢a ...J
16
O
o. L)
8
o
4 o (31:
cv IN
VlVO
SPF
TREATMENT
Fig. 3. In vitro proliferation of intestinal lamina propria cells (LPC) from GF, CV and SPF rats. Triplicate wells containing 106 cells from each rat were cultured with medium alone (11), 1/tg per well Con A ( [] ), or 1/tg per well LPS ({~) for 48 h and analyzed for tritiated thymidine incorporation (mean+ 1 SD). *Significance between treated groups in vitro at P
