JOURNAL OF BONE AND MINERAL RESEARCH Volume 5. Number 9, 1990 Mary Ann Liebert, Inc., Publishers

Vitamin D Metabolism in Rats with Adjuvant-Induced Arthritis CRAIG B. LANGMAN,' KATHY K. FORD,' LAUREN M. PACHMAN,' and FRANCIS GLORIEUX3

ABSTRACT Adjuvant-induced arthritis in rats shares many of the features of humans with rheumatoid arthritis, including the development of osteopenia in areas distal to erosive joint disease. We established adjuvant arthritis in male and female Sherman strain rats and then studied external calcium balances and vitamin D metabolism during the period of acute active clinical, serologic, and pathologic arthritis and osteopenia and in the preclinical period. While ingesting a calcium-sufficient vitamin D-replete diet (0.6% calcium, 0.65% phosphorus, and 2.2 IU D3 per g food), female rats with arthritis demonstrated reduced calcium balance (arthritic, 36 f 8 versus control, 169 f 13 mg per 6 days, p < 0.02) because of inefficient gastrointestinal absorption of calcium (arthritic 9.7% versus control 37%). This was associated with calcitriol deficiency (arthritic 52 f 7 versus control 70 f 10 pg/ml) and reduced osteocalcin levels. Male rats with arthritis demonstrated an inability to raise serum calcitriol levels to the same degree as control rats (200 f 30 versus 440 f 70, respectively) while ingesting a calcium-deficient diet (0.002% calcium, 0.34% phosphorus, and 2.2 IU D, per g food) and also had reduced balance (59 f 7 versus 85 f 10 mg per 6 days, respectively) due in part to decreased efficiency of absorption (55 versus 67%). No abnormalities in calcium balance or in serum calcitriol levels on the sufficient diet were present in the preclinical period. Physiologic calcitriol replacement to arthritic female rats increased osteoid available for mineralization and increased mineral apposition rates. Calcitriol deficiency in adjuvant-induced arthritis is responsible, in part, for reduced calcium balance and likely contributes to the development of osteopenia during the acute stage of the systemic inflammatory disease.

INTRODUCTION RACTURES AND BONY PAIN may

accompany the arthritis of patients with systemic rheumatoid disease. I ' ) Bony demineralization has also been noted in this group of patient~,"-~) although the mechanisms responsible for its occurrence are unknown. Previous studies in humans have suggested that abnormalities in calcium homeostasis, including hypercalciuria or diminished gastrointestinal absorption of calcium, may be present and perhaps responsible for altered mineralization.'4-11)The mechanisms that underlie these findings have not been determined.

F

Experimental arthritis may be established in susceptible strains of rats by the intradermal injection of modified Freund's adjuvant in oil, termed adjuvant-induced arthritis (AIA).(12)AIA shares some but not all of the clinical and biochemical features of patients with systemic rheumatoid d i ~ e a s e " ~ .and ' ~ ) has been proposed as an acceptable animal model of that disease.(") During the second month of AIA, osteopenia and bony demineralization have been demonstrated,"+") although no biochemical information was given that might determine an etiology. The purpose of this study was to create AIA in Sherman strain male and female rats and to then examine external

'Mineral Metabolism Laboratory, Division of Nephrology, Department of Pediatrics, Northwestern University and Children's Memorial Hospital, Chicago, IL. 'Division of Rheumatology/lmmunology, Department of Pediatrics, Northwestern University and Children's Memorial Hospital, Chicago, 1L. 'Departments of Surgery and Genetics, McGill University, Montreal, Quebec, Canada.

905

LANGMAN ET AL.

906 calcium balance, vitamin D metabolism, and bone histomorphometry to undersand the determinants of mineral metabolism that may result in osteopenia during systemic inflammation.

METHODS Induction of adjuvant-induced arthritis (AIA) To create AIA, 10 mg powdered Mycobacteriurn buryricum (Difco Laboratories, New York, NY) was mixed with 2.0 ml mineral oil and ground by hand with mortar and pestle for 150 s, according to previously reported metho d ~ . ( ~The ~ .resultant ~ ~ ’ adjuvant suspension (0.3 ml) or an equal volume of the oil alone was randomly injected intradermally into the base of the tail of 100 g (age 5-6 weeks, sexual maturity at 9-10 weeks of age) male and female Sherman rat littermates (Camm Research Institute, Wayne, NJ) to create arthritic (AIA) and control (C) rats of each gender, respectively. Beginning on day 5 after injection, C and AIA male rats had daily to every other day measurements made of fore- and hindlimb joint circumference using a constant-tension caliper (Lange Skinfold Caliper, Cambridge Scientific Industries). On days 5, 10, and 13 and at the conclusion of balance, male C and AIA rats had measurements of hematocrit and erythrocyte sedimentation rate using 0.25 ml blood obtained by tail vein puncture.

Dietary preparations Study A: At a time of active arthritis (see later), 10 days after the injection, the rats were placed in individual metabolic cages. During a 5 day cage adjustment period, the rats ate ad libitum a diet containing 0.6% calcium, 0.65% phosphorus, and 2.2 IU vitamin D3 per food (Teklad Test Diets, Madison, WI). Deionized distilled water was allowed ad libitum throughout the study. The diets were continued, and external calcium balance was performed on days 15-20 after injection of adjuvant. Study B: In a separate group of male AIA and C rats, external calcium balance was performed in the same manner as in study A, except that the diet from day 10 to the end of balance contained 0.002% calcium and 0.34% phosphorus (low-calcium diet, LCD). Study C: In a separate group of male and female AIA and C rats, external calcium balance was performed on days 5-10 inclusive, in the same manner and on the diet described for study A. Rats were randomly selected for this balance study from a cohort of male or female AIA or C rats. Animals not studied for balance served as time controls for disease expression. Study D: A separate group of AIA and C female rats ate ad libitum the diet used for study A, and had calcitriol or vehicle infused 2 days after injection (see later) over the same time period before sacrifice on day 16 for histomorphometric analysis of the femur (see later).

Study E; Female AIA and C rats ate ad libitum the diet used for study A and were placed in metabolic cages for adjustment, similar to study A, 3 days after injection. Complete 24 h urine specimens were collected from days 12-17 inclusive, before sacrifice for determination of chemistries.

External calcium balance The balance study was carried out from days 15-20 inclusive (studies A and B) or 5-10 days inclusive (study C) after adjuvant injection. During the balance experiment. complete individual 24 h urine specimens were collected in 0.25 ml 12 N hydrochloric acid, and individual 24 h stool specimens were obtained as previously described. (201 Calcium from each 24 h fecal collection was extracted by a solution of 35% nitric acid and 35% perchloric acid (volume/volume) at 95°C for 7-8 h. The resultant solution was brought up to 25 ml with deionized, distilled water before measurement of calcium concentrations. Daily food and water intake were recorded. Because in preliminary experiments we found that AIA rats of either gender weighed less and ate less food than C rats (data not shown), C and AIA rats were fed during the balance experiment. Data for C rats that ate less than 13 g food or that drank less than 15 ml water and data for AIA rats that ate less than 11 g food or drank less than 12 ml water were eliminated from the study. All data were analyzed individually for each day of balance, and the data for individual days for each rat were combined for further analysis. At the conclusion of balance on the morning of day 21 (studies A and B) or day 11 (study C) or on the morning of day 16 (study D) or day 17 (study E), rats were anesthetized with diethyl ether and exsanguinated through the carotid artery. In some studies, 0.3 ml blood was heparinized for measurement of ionized calcium. Serum from each rat was separated from cells within 20 minutes and frozen at -20°C for biochemical measurements; an aliquot of serum was stored at 4°C for determination of serum total calcium concentration.

Calcitriol infusion Alzette miniosmotic pumps (Aka Corporation, Palo Alto, CA) containing either 1,25-dihydroxyvitamin D, [ 1,25-(OH)2D3]in ethanol and propylene glycol (1 :1, vol/ vol) in a total dose calculated to deliver 35 pmol per 100 g body weight per day(21.221 for 14 days or vehicle alone were implanted subcutaneously in the dorsum of the back of female AIA rats beginning on day 2 (study D) or day 3 (study E) after adjuvant injection. Rats in study E received an intraperitoneal dose of tetracycline (HCl(20 mg) 7 days before sacrifice and an intraperitoneal dose of demeclocycline HCI (15 mg per 100 g body weight) 24 h before sacrifice.

Bone histomorphometry After fixation in formaldehyde-phosphate buffer (pH 7.1), the labeled rat femora were sectioned at midshaft

907

VITAMIN D METABOLISM IN RATS WITH ARTHRITIS

with a diamond saw and the distal ends processed. The specimens were dehydrated, cleaned, and embedded in methyl methacrylate. After polymerization, consecutive sections of 6 pm were cut on a Reichert-Jung Polycut S (Cambridge Instr-GmBH, West Germany), with the plane of section passing through the middle aspect of one condyle. One section was stained with Goldner’s trichrome, and the following section mounted unstained for fluorescence. Quantitative histomorphometric analysis was performed with a semiautomatic image analysis system (Videoplan, Kontron-Carl Zeiss, West Germany). osteoid volThe following parameters were ume/bone volume (OV/BV), percentage of cancellous bone volume occupied by osteoid; osteoid surface/bone surface (OS/BS), percentage of cancellous bone surface occupied by osteoid; osteoid thickness (O.Th, pm), average width of the osteoid layers covering the mineralized bone surface; mineral apposition rate (MAR, pm/day), thickness of the layer of new mineralized bone laid down per unit of time; and osteoid maturation time (Omt), mean time interval between the initiation of matrix synthesis and the onset of its mineralization.

Vitamin D metabolite measurements Serum (1.0 ml) was thawed to room temperature and 1500 cpm of [‘H]25-OHD3 and [3H]1,25-(OH),D, were added to monitor procedural l o s ~ e s The . ~ ~serum ~ ~ was extracted with an equal volume of acetonitrile and centrifuged at 1600 x g for 5 minutes. The supernatant was washed with an equal volume of 0.4 M potassium phosphate (KH,PO,), pH 10.5. The resulting supernatant was placed onto a C18 silica column (Waters Sep-Pak, Milford, MA), equilibrated in methanol, acetonitrile, and water (1: 1:1, vol/vol/vol), and eluted with 100% acetonitrile. The column fraction in which standard 25-OHD, and 1,25-(OH),D, eluted was collected, evaporated under nitrogen, and reconstituted in isopropanol and hexane, 1.5: 98.5 (vol/vol). The separate fractions corresponding to the elution of standard 25-OHD, and 1,25-(OH)2D, were collected from a silica column (Waters Sep-Pak, Milford, MA), equilibrated in isopropanol and hexane, 1.5:98.5, and eluted with a mobile-phase gradient of increasing polarity to a final eluate composition of 4050 isopropanol and hexane. 25-OHD3 in the silica column eluate was then measured in triplicate using a radioreceptor assay that utilizes dilute rat plasma as binding protein, as previously described.(z2) Overall recovery of sample was 68 3%. Least detectable dose ranged from 250 to 300 pg per assay tube; 50% displacement ranged from 6 to 8 x lo3 pg per assay tube. Fractions containing 1,25-(OH),D3 were measured in triplicate using the thymus cytosol radioreceptor assay, as previously described. (24) Overall recovery of sample was 53 * 2%. Least detectable dose ranged from 1 to 1.25 pg per assay tube. The 50% displacement ranged from 7 to 11 pg per assay tube.

*

Biochemical measurements Rat parathyroid hormone was determined in triplicate in 0.2 ml serum by a radioimmunoassay previously described in which rat PTH-(1-34) (Calbiochem, LaJolla, CA) is used as standard.(24)Least detectable dose ranged from 25 to 27 pM; 50% displacement ranged from 320 to 350 pM. Rat osteocalcin was determined in quadruplicate in serum at a dilution of 150 by a double-antibody radioimmunoassay (Biomedical Technologies Inc., Stoughton, MA) that utilizes a goat antirat osteocalcin antibody and a donkey antigoat antibody. The least detectable dose was 30 pg/pl, and the 50% response was 220 pg/pl. Control rat serum used for the vitamin D metabolite assays served as an internal control serum for this assay; the mean SD value was 57 + 7 ng/ml for seven assays. Serum total calcium was measured by atomic absorption spectrophotometry and serum phosphorus by the colorimetric method of Fiske and Subbarow. Blood ionized calcium was measured with a calcium-selective electrode (ICA1, Radiometer, Copenhagen). Urine calcium was measured by automated murexide titration. In our previous studies, these values were not significantly different than calcium quantitated by atomic absorption spectrophotometry.

*

Statistical analysis Data for individual days of balance for control and AIA rats of each gender were compared with an unpaired I-test that did not assume equal variances within the groups. When data for individual days of balance were pooled for analysis, comparisons employed a repeated-measures analysis of variance; a Bonferroni correction factor was used to correct for the performance of multiple t-tests. All other analyses of two groups alone employed an unpaired t-test that did not assume equality of variance between the two groups. All analyses employed standard computer-assisted methodologies (BMDP, University of California, Berkeley, CA).

RESULTS Induction of arthritis In this experiment and in preliminary studies (data not shown), 94% of male and female rats injected with a suspension of modified, complete Freund’s adjuvant developed polyarticular arthritis by 13 days after injection. As seen in Table 1 in data reported for male rats, arthritis began in the hindlimb on day 10 and quickly became symmetrical in all four extremities. This arthritis persisted for the duration of calcium balance studies. In other experiments by us (data not shown), the joint edema disappeared by day 35 after injection, similar to previous reports.(”) Approximately 12% of the male rats given a suspension of adjuvant also developed an ulcerative lesion of the penis, similar to that described by other i n v e ~ t i g a t o r s . At ~ ~the ~~~~) time of the frank arthritis on day 13, serologic evidence of systemic inflammation was also present in male and female rats (Table 2, data for male rats) with an elevated erythro-

908

LANGMAN ET AL.

TABLE1. JOINTDIMENSIONS IN CONTROL AND ARTHRITIC MALER A T S ~ Days after injection

5

Limb

RF

Control

AIA

RR LR

4.3 f 4.5 f 5.2 f 5.4 f

0.1 0.1 0.1 0.1

4.2 f 4.6 f 5.3 f 5.4 f

0.1 0.1 0.1 0.1

7

RF LF RR LR

4.5 f 4.7 f 5.3 f 5.4 f

0.1 0.1 0.1 0.1

4.4 f 4.8 f 5.5 f 5.5 f

0.1 0.1 0.1 0.1

10

RF LF RR LR

4.6 f 4.8 f 5.5 f 5.5 f

0.1 0.1 0.1 0.1

4.7 f 4.9 f 5.7 f 7.2 f

0.1 0.1 0.1 0.lb

11

RF LF RR LR

4.9 f 4.8 f 5.4 f 5.5 f

0.1 0.1 0.1 0.1

LF

cyte sedimentation rate and anemia that persisted through the end of balance. Histologic examination of the left hip joint in male or female AIA rats demonstrated typical lymphocytic infiltration characteristic of AIA,(14)which was absent in C rats of either gender. Early pannus formation was also present in the AIA specimens. Plain-film whole-body radiographs taken on day 21 after adjuvant injection demonstrated the typical erosive arthritis,(*6)especially in the distal joints. Additionally, marked osteopenia in areas distant from the erosive joint destruction was also present in the female AIA rats fed the calcium-sufficient diet (study A) and in male AIA rats fed the LCD (study B). There were no radiographic changes in either gender given adjuvant when radiographed before the onset of clinical disease (study C).

Calcium balance

5.4 f 0 . l b Study A: Male and female AIA rats weighed the same 4.9 f 0.1 at the time of injection as C rats (male AIA 120 + 10 ver7.2 f 0.2b sus male C 123 f 1 1 g; female AIA 113 f 6 versus female 7.3 f 0.2b C 112 f 8 g; p = NS for both) but were smaller than C RF 13 4.9 f 0.1 5.8 f 0.lb rats of their respective gender at the beginning of the balLF 5.1 f 0.1 5.8 f 0.2b ance period (male AIA: 166 k l l versus male control 226 RR 5.6 f 0.1 8.1 f 0.4b f 4 g, p < O.OOO4; female AIA 154 f 6 versus female LR 5.8 f 0.1 8.1 f 0.4b control 170 f 5 g, p < 0.04). Daily weight gain and food intake, when expressed as a percentage of daily body 21 RF 6.2 f 0.1 9.2 f 0.4b weight, were significantly higher in male and female C rats LF 9.3 f 0.4b compared to AIA rats of the same gender (data not 6.3 f 0.1 RR 6.3 f 0.1 9.5 f 0.2b shown). LR 6.4 f 0.1 9.3 f 0.4b Daily intake of calcium was greater in C rats compared to AIA rats of the same gender on each balance day from aJoints were measured in control (C) and arthritic (AIA) rats using a constant-tension caliper, and values expressed in milli- the first, and therefore, cumulative intakes of calcium were meters represent the two-dimensional mean of seven to nine rats greater for the 6 days of balance in C rats than in AIA rats per group. RF, right forelimb; LF, left forelimb; RR, right hind- (Table 3) of each gender. Cumulative fecal calcium excrelimb; LR, left hindlimb. tion was also greater in C rats of each gender. The percenbp < 0.03, AIA greater than C. tage of ingested dietary calcium that was absorbed was similar in male C and AIA rats but was greater in female C rats than in female AIA rats (Table 3, p < 0.02). Daily urinary calcium excretion was greater in C rats compared TABLE2. HEMATOCRIT AND ERYTHROCYTE to AIA rats of each respective gender on each balance day AND SEDIMENTATION RATEIN CONTROL from the first. Therefore, cumulative urinary calcium exARTHRITIC MALER A T S ~ cretion was greater for C rats compared t o AIA rats of Control Day Parameter AIA each gender (Table 3). Daily net external calcium balance was greater for C rats compared to AIA rats of each re5 Hematocrit 46f2 43 f 2 spective gender from the first day of balance. Therefore, ESR 2f1 2fl cumulative net external calcium balances were greater for 44f2 44f2 10 Hematocrit C rats compared to AIA rats of each respective gender. As 2f1 14 f 4b ESR expected, female C rats absorbed a smaller percentage of dietary calcium and excreted a greater amount in the urine 43 f 2 32 f 2c 13 Hematocrit than male C rats.(*’) 1f1 ESR 16 f 3 b At the end of balance, blood ionized calcium and serum 45 f 2 21 Hematocrit 30 f 3 c total calcium were similar in each gender in the two groups If1 ESR 15 f 3 b (Table 4). However, serum phosphorus was significantly aValues of hematocrit (W) and erythrocyte sedimenreduced in the male AIA rats compared to controls (p < tation rate (ESR,mm/h) represent the mean f SD of 0.003)but there were no differences in female rats. This seven to nine animals per group on the day after adjuobservation was seen again in two similarly prepared series vant injection. of 6-8 male arthritic rats (data not shown). Serum levels of bAIA greater than control, p < 0.002. circulating parathyroid hormone were similar in C and CAIA less than controls, p < 0.03.

VITAMIN D METABOLISM IN RATS WITH ARTHRITIS TABLE3. CUMULATIVE CALCIUM BALANCE IN MALEAND FEMALE CONTROL AND ARTHRITIC RATSFEDA CALCIUM-SUFFICIENT DIET:STUDYAa

Female

Male Parameter Calcium intake Fecal calcium % Absorption Urine calcium Net balance

Control

A IA

Control

A IA

590 f 10 199 f 27 47.7 3.9 f 0.2 285 f 14

370 f 20b 182 f 36 49.6 2.7 f 0.2b 185 f l l b

430 f 20 291 f 50 37.2 5.3 f 0.3 169 f 13

331 f lob 266 f 46b 9.7b 2.5 f 0.lb 36 f 8b

aValues for six rats in each group ingesting a diet containing 0.6% calcium, 0.65% phosphorus, and 2.2 1U D, per g food from day 10 after adjuvant injection and studied on days 15-20 after adjuvant injection are mean f SEM. Calcium intake, excretions, and balance are expressed in mg per 6 days. bAIA less than control, same gender, p < 0.02.

TABLE4. BLOODMINERAL MEASUREMENTS IN MALEAND FEMALE CONTROL AND ARTHRITIC RATS FEDA CALCIUM-SUFFICIENT DIET:STUDYAa

Female

Male Parameter Calcium Ionized calcium Phosphorus Parathyroid hormone 25-OHD3 1,25-(OH)ZD3

Control 9.9 f 0.2 1.25 f 0.3 8.1 f 0.8 126 f 15 19 f 2 100 f 18

AZA

10.4 1.27 5.4 112 22 104

f f

0.4 0.3

f 0.P

10 3 f 12 f

f

Control

AIA

10.1 f 0.3 1.30 f 0.4 10.0 f 0.4 80 f 10 27 f 5 70 f 10

9.9 1.29 9.7 78 23 52 ~~~

~

0.3 0.4 f 0.4 f 8 f 2 f 7c f f

~

aValues for six rats in each group ingesting a diet containing 0.6% calcium, 0.65% phosphorus, and 2.2 IU D, per g food from day 10 after adjuvant injection and studied on days 15-20 after adjuvant injection are mean f SEM. Calcium is expressed in mg/dl, ionized calcium in mM, phosphorus in mg/dl, parathyroid hormone in pM, 25-OHD3 in ng/ml, and 1,25-(OH),D, in pg/ml. bAIA less than controls, same gender, p < 0.003. cAIA less than controls, same gender, p < 0.02.

AIA rats of each gender. Circulating levels of 25-hydroxyvitamin D3 were the same in C and AIA rats of each gender. Circulating levels of calcitriol were not different in male C and AIA rats but were reduced in female AIA rats compared to C rats. The observation of a reduction in calcitriol levels in female AIA rats compared to C rats was seen again in four similarly prepared series of 6-20 female rats (data not shown). As expected, male C rats when compared to female C rats, had higher circulating calcitriol levels and fractional calcium absorption rates.(Z7,28’ For all rats combined, there was a significant direct linear correlation between circulating levels of calcitriol and fractional calcium absorption ([calcitriol] = (0.06)(fractional calcium absorption) + 41; r = 0.748, p < lo-‘). Study B: Because there were no differences in fractional calcium absorption or calcitriol levels in male C and AIA rats on an adequate calcium intake but male AIA rats were hypophosphatemic compared to controls, external calcium balance was repeated in a separate group of C and AIA

male rats on the LCD. AIA rats weighed significantly less than C rats at the beginning of this balance (188 f 20 versus 225 + 19 g, p < 0.02). The difference in weights was similar to the differences observed during study A. Daily calcium intake was greater in C rats compared to AIA rats on each balance day from the first, and therefore, cumulative calcium intake was slightly less in AIA rats compared to C rats (Table 5). Cumulative fecal calcium excretion was similar in AIA rats compared to C rats, but fractional calcium absorption was reduced in AIA rats (Table 5). Cumulative urinary calcium excretion was significantly reduced in AIA rats compared to C rats. Cumulative net external calcium balance was reduced in AIA rats compared to C rats. As expected, the LCD was associated with secondary hyperparathyroidism in C rats and was not significantly different in AIA rats (Table 6). Circulating levels of 25-hydroxyvitamin D, were not different in AIA versus C rats. However, circulating levels of calcitriol were reduced in AIA rats compared to C rats (Table 6). Levels of blood

910

LANGMAN ET AL.

Calcitriol infusion

TABLE5. CUMULATWE CALCIUM BALANCE IN MALECONTROL AND ARTHRITIC RATSFED A CALCIUM-RESTRICTED DIET: STUDYBa

Parameter Calcium intake Fecal calcium To Absorption Urine calcium Net balance

Control 152

f

50 f

AIA

20 6

113 51

llb f 8 55b

1 10

3.4 59

f lb

67 6.8 85

f f

Studies D and E: Calcitriol infusion did not increase the body weight of the AIA rats compared to the vehicle-infused group or alter the clinical expression of the arthritis.

f

7b

f

Walues for five rats in each group ingesting a diet containing 0.002% calcium, 0.34% phosphorus, and 2.2 IU D, per g food from day 10 after adjuvant injection and studied on days 15-20 after adjuvant injection are mean f SEM. Calcium intake, excretions, and balance are expressed in mg per 6 days. bAIA less than control, p < 0.02.

Biochemical Studies: Rats with AIA receiving vehicle alone had biochemistries similar to those in study A (Table 8). Rats with AIA receiving calcitriol by infusion had a small increase in serum total calcium (0.4 mg/dl). Additionally, when compared to AIA rats receiving vehicle, calcitriol-infused rats significantly increased urinary calcium excretion (6.9 f 1.9 versus 23.5 f 3.1 mg per 5 days, p < 0.02) and had increased circulating levels of osteocalcin (Table 8).

Bone Histomorphometry: Rats with AIA receiving vehicle had significantly less OV/BV and OS/BS compared to rats with AIA that received calcitriol by chronic infusion (Table 9). Mineral apposition rates were increased with the infusion of calcitriol, although differences in the TABLE6. BLOODMINERAL MEASUREMENTS IN MALE time for osteoid maturation were not different (AIA 1.72 RATS FEDA CALCIUM-RESTRICTED CONTROL AND ARTHRITIC f 0.37 versus AIA + 1,25-(OH),D, 1.62 f 0.39 (mean f DIET: STUDYBa SD) days, p = NS].

Parameter Calcium intake Ionized calcium Phosphorus Parathyroid hormone 25-OHDJ 1,25-(OH)ZD,

Control 10 f 1.27 f 11 f 223 + 13.5 f 440 f

0.2 0.1 0.3 23 1 70

AIA

10.2 1.28 10.9 184 14 200

f f f f f

f

0.2 0.1 0.3 23 1.3 30b

Walues for five rats in each group ingesting a diet containing 0.002% calcium, 0.34% phosphorus, and 2.2 IU D, per g food from day 10 after adjuvant injection and studied on days 15-20 after adjuvant injection are mean f SEM. Calcium is expressed in mg/dl, ionized calcium in mM, phosphorus in mg/dl, parathyroid hormone in pM, 25-OHD3 in ng/ml, and 1,25-(OH),D3 in pg/ml. bAIA less than control, p < 0.02.

ionized calcium, serum total calcium, and phosphorus were the same in AIA and C rats (Table 6).

Study C: Male and female rats weighed the same at the time of injection as C rats and at the time of inception of balance (data not shown). Unlike the previous balance studies, there was no significant difference in daily weight gain between AIA and C rats of the same gender (final body weights: male AIA 182 f 4 versus male C 183 f 8 g; female AIA 192 f 9 versus female C 194 f 6 g; p = NS for each comparison within gender). There were no differences among AIA and C rats of the same gender in daily food intake, and therefore, cumulative calcium intake was the same in AIA and C rats of each gender (Table 7). There were no differences between AIA and C rats of each gender with respsect to serum total calcium, phosphorus, 25-OHD3, or 1,25-(OH),D, (data not shown). There were no differences in net external calcium balance between C and AIA rats of each gender (Table 7).

DISCUSSION The Sherman strain rat has been shown in this study to be susceptible to the induction of the development of adjuvant-induced arthritis, similar to several other strains of Many of the serologic and clinical abnormalities that occur in humans with rheumatoid disease also occurred in our rats, including an elevation of the erythrocyte sedimentation rate, anemia, and progressive, migrating, polyarticular arthritis of acute and chronic nature. However, as in other strains of rats with AIA previously studied, rheumatoid factor was absent from the sera of our rats. Histologic examination of the hindlimb joints of our rats revealed typical lymphocytic infiltrates previously noted in other strains of rats and similar to those that occur in humans. Some of the rats developed ocular manifestations (uveitis and scleritis), which are not uncommon in humans. Additionally, some of the rats also developed genital lesions, which is not a common feature of the human disease. Alterations in vitamin D metabolism in rats with AIA may be responsible for the expression of osteopenia. Calcitriol, the active form of vitamin D, was reduced in female rats with AIA and in male rats with AIA while the latter were ingesting a calcium-deficient diet. The reduction in calcitriol levels was associated with a diminution of its major biologic effect, in that there was decreased efficiency of gastrointestinal absorption of ingested calcium and reduced calcium balance. Calcitriol infusions during the period of acute arthritis increased urinary calcium excretion and raised serum osteocalcin levels, suggesting that bone formation rates were increased. This was demonstrated by dynamic histomorphometry in which the surface area available for mineralization increased with calcitriol

911

VITAMIN D METABOLISM IN RATS WITH ARTHRITIS

TABLE7. CUMULATIVE CALCIUM BALANCE IN MALEAND FEMALE CONTROL AND ARTHRITIC RATSFEDA CALCIUM-SUFFICIENT DIET DURING PREARTHRITIS: STUDYCa

Male Parameter

Female

Control

AIA

Control

AIA

529 f 12 303 f 12 50.5 4.6 f 0.8

509 f 13 256 f 12 45.3 4.6 f 0.3

451 f 16 311 f 17 30.7 7.6 f 0.7

469 f 8 293 f 6 38.3 6.3 f 0.7

~~

Calcium intake Fecal calcium Yo Absorption Urine calcium

aValues for five rats in each group ingesting a diet containing 0.6% calcium, 0.65% phosphorus, and 2.2 IU D, per g food from days 5-10 after adjuvant injection are mean f SEM. Calcium intake, excretions, and balance are expressed in mg per 6 days. There were no significant differences between C and AIA rats of each gender.

infusions and the mineralization rate of newly formed osteoid was increased when compared to vehicle-infused arthritic animals. There was no defect in the absorption of the parent Parameter + Calcitriol + Vehicle compound, vitamin D, from the diet of AIA rats or in its hydroxylation by the P450mixed-function oxidase of the 10.1 f 0.1 9.7 f 0.lb Calcium liver microsomes to 25-OHD3, as levels of the latter com10 f 0.3 10.6 f 0.3 Phosphorus pound were similar in control and experimental rats of 15.7 f 1.3 22.3 f 1 . 1 ~ each gender. However, female rats with AIA had a lower 25-OHD3 1,25-(OH)zD, 130 f 7 66 f 7d circulating level of calcitriol compared to female control Osteocalcin 82 f 5 69 f 2b rats ingesting a normal calcium diet and male AIA and control rats on this diet had similar values. However, in aValues (mean f SEM) for 12 rats in each group ingesting a diet male rats with AIA ingesting a calcium-sufficient diet, containing 0.6%calcium, 0.65% phosphorus, and 2.2 IU D, per g food from day 2 after adjuvant injection and studied on day 17 modest hypophosphatemia developed in comparison to and infused with calcitriol (35 pmol per 100 g per day) or vehicle controls and could have increased calcitriol levels to confor the same time. Calcium is expressed in mg/dl, phosphorus in trol values. 13') Restriction of dietary calcium demonstrated mg/dl, 25-OHD, in ng/ml, 1,25-(OH),D, in pg/ml, and osteocalthe inability of male rats with AIA to increase their levels cin in ng/ml. of calcitriol to the same degree as controls. The reduction bVehicle less than infusion, p < 0.05. of calcitriol was not secondary to substrate deficiency or to CVehicle less than infusion, p < 0.001. dvehicle less than infusion, p < 10'. altered levels of parathyroid hormone or blood ionized calcium, two important regulators of the renal 25-OHD3-1hydroxylase The use of the restricted calcium diet did not produce hypophosphatemia and was associated with reduced calcitriol levels in male rats with AIA. Similar to the inflammatory polyarthritis induced in rats TABLE9. HISTOMORPHOMETRIC PARAMETERS IN by type I1 collagen i m m ~ n i z a t i o n , ~rats ~ ~with . ~ ~adjuvant) FEMALE ARTHRITIC RATS INFUSED WITH CALCITRIOL OR VEHICLE:STUDYEa induced arthritis had a reduction in circulating levels of osteocalcin during the acute phase of the disease. HowParameter + Calcitriol + Vehicle ever, in the latter study, no biochemical or histomorphometric studies were performed to elucidate the cause or the 7.2 f 0.5 5.3 f 0.7 OV/BV, Yo effect of the change in osteocalcin. In our study, the bioO.Th, Fm 4.70 f 0.3 3.99 f 0 . 9 chemical findings of a reduction in calcitriol and osteocal2.93 f 0.34 2.34 f 0.42b MAR, pm/day cin levels in acute arthritis and their increase with calcitriol 1.62 f 0.39 1.72 f 0.37 OMT, day infusions correlated with the mineralization process. Vitamin D metabolism in adjuvant-induced arthritis difaValues (mean f SEM) for 10-12 rats in each group infers from that in several other forms of inflammation-assogesting a diet containing 0.6% calcium, 0.65% phosphate, ciated or independent osteopenia in rodents that are not and 2.2 IU D, per g food from day 3 after adjuvant infection and studied on day 17 and infused with calcitriol (35 associated with calcitriol deficiency. Inflammation-medipmol per 100 g per day) or vehicle for the same time. ated osteopenia created by the intraperitoneal injection of b p < 0.05, vehicle group versus calcitriol infusion group. talcum powder in rats was not associated with altered levels of 25-OHD3 or 1,25-(OH)zD3.'34.35) Serum calcium was increased in that study, unlike the normal blood ionTABLE8. BLOODMINERAL MEASUREMENTS IN FEMALE ARTHRITIC RATS INFUSED WITH CALCITRIOL OR VEHICLE:STUDYDa

912

ized calcium in rats with adjuvant-induced arthritis in our study. Additionally, in those studies, the biologic effects of calcitriol were not evaluated with respect to external calcium balance. Osteopenia, which occurs in immobilizati or^'^^) and is simulated in rodents by suspension of the hindlimbs,(37)is associated with hypercalciuria, an increase in blood ionized calcium levels, and suppression of serum levels of parathyroid hormone and calcitriol. These features of immobilization osteopenia were absent in our rats with AIA. Also, we did not observe a reduction of physical activity in rats with AIA. The mechanism whereby adjuvant-induced arthritis causes calcitriol deficiency remains unknown at present. We have demonstrated that the temporal sequence of calcitriol deficiency is synchronous with the onset of clinical and serologic expression of disease activity, but not beforehand. Many of the clinical and serologic manifestations of AIA are related to increased macrophage activity, with increased synthesis of prostaglandin El and release of interleukin-1 and tumor necrosis f a ~ t o r . ' ~In~ other . ~ ~ ) inflammatory disorders, such as sarcoidosis, activated macrophages have been shown to contain a 25-OHD-la-hydroxylase enzyme and produce c a l ~ i t r i o l . ~Perhaps ~ ~ ~ ~ ' it) is also possible that macrophages activated in adjuvant arthritis in other ways and expressing other cytokines could alter the metabolism of calcitriol. Calcitriol levels have been demonstrated previously to be reduced in female rats compared to male secondary to reduced renal synthesis of calcitriol.(28)The reduction in calcitriol has been shown to be associated with reduced external balance and decreased gastrointestinal uptake of calcium.i27)In AIA, we also demonstrated that gender plays a role in the response of calcitriol levels in that male rats with AIA did not have the reduction of serum levels of calcitriol that female rats exhibited on a calcium-sufficient diet, although the modest hypophosphatemia in males may have been responsible for normalization of the calcitriol levels. Additionally, female rats were hypercalciuric and had reduced external calcium balance compared to male rats with AIA. This difference in gender metabolism of vitamin D and in expression of calcitriol deficiency in AIA has an important clinical corollary in that the overwhelming preponderance of bone disease in children's) and adults'42)with rheumatoid disease occurs in women. Perhaps the subtle gender differences in vitamin D metabolism, which are generally obscured in human studies and are manifest in our rat study, may explain, in part, this clinical observation. In summary, we have studied a model of systemic inflammation associated with acute arthritis and osteopenia. In the arthritic rat, calcium homeostasis is altered with reduced absorption of dietary calcium associated with reduced levels of calcitriol. It is likely that the reduced calcium balance is the consequence of calcitriol deficiency and that both together contribute in part to the development of osteopenia. Our study provides evidence for the concept that during acute inflammation there is systemic regulation of bone metabolism that may act in part through changes in calcitriol metabolism.

LANGMAN ET AL.

ACKNOWLEDGMENTS The authors wish to thank Cindy Blifeld, M.D., Jacob Radde (deceased), Lori Aaronson, and Neil Knudsen for excellent technical assistance and Ms. Janet Bojan for expert editorial preparation. This work was supported in part by grants (Langman) from the National Institutes of Health (AR30692 and DK33949) and from the Otto Sprague Memorial Fund. Presented in part at the Midwest Section, American Federation of Clinical Research, November 1986, Chicago, IL, and the Society for Pediatric Research, May 1989, Washington, D.C.

REFERENCES 1. Robinson DR, Tashijian A, Levine L 1975 Prostaglandin stimulated bone resorption by rheumatoid synovia. A possible mechanism for bone destruction in rheumatoid arthritis. J Clin Invest 56:1181-1188. 2. Tannenbaum H 1984 Osteopenia in rheumatology practice: Pathogenesis and therapy. Semin Arthritis Rheum 13:337-348. 3. Duncan H 1972 Osteoporosis in rheumatoid arthritis and corticosteroid-induced osteoporosis. Orthop Clin North Am 3:571-583. 4. Maddison PJ, Bacon PA 1974 Vitamin D deficiency, spontaneous fractures and osteopenia in rheumatoid arthritis. Br Med J 4(5942):433-435. 5 . Reed A, Haugen M, Pachman LM, Langman CB 1990 Abnormalities in serum osteocalcin in children with chronic rheumatic diseases. J Pediatr 116574-580. 6. Kennedy AC, Lindsay LR, Buchanan WW, Allam BF 1976 Bone resorbing activity in the sera of patients with rheumatoid arthritis. Clin Sci Mol Med 51:205-207. 7. Olhagen B, Reizenstein P, Sjoberg HE 1969 Retention of oral "'Ca in patients with rheumatoid arthritis and osteopenia. Acta Rheum Scand 15:145-151. 8. Kennedy AC, Allam BF, Rooney PJ, Watson ME, Fairney A, Cuchanan KD, Hillyard CJ 1979 Hypercalcemia in rheumatoid arthritis: Investigation of its causes and implications. Ann Rheum Dis 38:40-412. 9. Stapleton FB, Hanissian HS, Miller LA 1985 Hypercalciuria in children with juvenile rheumatoid arthritis: Association with hematuria. J Pediatr 102235-239. 10. Lund B, Storm TL, Lund B, Melsen F, Mosekidle L, Anderson RB, Egmose C, Sorensen CH 1985 Bone mineral loss, bone histomorphometry an vitamin D metabolism in patients with rheumatoid arthritis on long-term glucocorticoid therapy. Clin Rheum 4:143-149. 11. Bird HA, Peacock M, Storeer JH, Wright V 1980 Comparison of serum 25(OH)D concentration in rheumatoid arthritis and osteoarthrosis. Br Med J 280:1416. 12. Pearson CM, Wood FD 1959 Studies of polyarthritis and other lesions induced in rats by injection of mycobacterial adjuvant. Arthritis Rheum 2:440-459. 13. Pearson CM 1963 Experimental joint disease: Observations on adjuvant-induced arthritis. J Chron Dis 162363474. 14. Weichman BM 1989 Rat adjuvant arthritis: A model of chronic inflammation. In: Chang JY, Lewis AJ (eds) Pharmacological Methods in the Control of Inflammation. Alan R. Liss, New York, pp. 363-380. 15. Pearson CM, Wood FD 1963 Studies of arthritis and other lesions induced in rats by the injection of mycobacterial ad-

913

VITAMIN D METABOLISM IN RATS WITH ARTHRITIS

16.

17.

18.

19.

20.

21.

22.

23.

24,

25

I

26.

27.

28.

29.

30.

31.

32.

juvant. 111. Pathologic details of the arthritis and spondylitis. Am J Pathol 42:73-95. Martel RR, Klicius K, Metcalf G 1984 Effect of etodolac on articular and bone pathology associated with adjuvant arthritis in rats. Agents Actions 14257-264. Martel RR, Klicius K, Metcalf G, Rona GA 1984 Comparative effects of long term therapy with etodolac, naproxen and ibuprofen on articular and bony changes associated with adjuvant arthritis in rats. Agents Actions 15:403-412. Best R, Christian R, Lewis DA 1984 Effects of particle size of dried mycobacteria on adjuvant induced arthritis in the rat. Agents Actions 14:265-268. Liyange SP, Currey HLF, Roberts BV 1975 Influence of tubercle aggregate size on severity of adjuvant arthritis in the rat. Ann Rheum Dis 34:49-53. Bushinsky DA, Favus MJ, Langman CB, Coe FL 1986 Mechanism of chronic hypercalciuria with furosemide: Increased calcium absorption. Am J Physiol 251:F17-F24. Rizzoli R, Fleisch H, Bonjour J-P 1977 Effects of thyroparathyroidectomy on calcium metabolism in rats: Role of 1,25-dihydroxyvitamin D. Am J Physiol 233:E160-E164. Favus MJ, Langman CB 1984 Effects of 1,25-dihydroxyvitamin D, on colonic calcium transport in vitamin D-deficient and normal rats. Am J Physiol 246:G268-G273. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR 1987 Bone histomorphometry: Standardization of nomenclature, symbols and units. J Bone Min Res 6595-610. Langman CB, Bushinsky DA, Favus MJ, Coe FL 1986 Ca and P regulation of 1,25(OH)>D, synthesis by D replete rat tubules during acidosis. Am J Physiol 251:F911-F918. Favus MJ, Langman CB 1986 Evidence for calcium-dependent control of 1,25-dihydroxyvitamin D, production by rat kidney proximal tubules. J Biol Chem 261:11224-11229. Blackham A, Burns JW, Farner JB, Radziwonik H, Westwick J 1977 An x-ray analysis of adjuvant arthritis in the rat. Agents Actions 7:145-151. Bushinsky DA, Favus MJ, Coe FL 1984 Elevated 1,25(OH),D,, intestinal absorption and renal mineral conservation in male rats. Am J Physiol 246:F140-F146. Langman CB, Favus MJ, Bushinsky DA, Coe FL 1985 Effects of dietary calcium restriction on 1,25-dihydroxyvitamin D, net synthesis by rat proximal tubules. J Lab Clin Med 106~286-292. Swingle KF, Jaques LW, Kram DC 1969 Differences in the severity of adjuvant arthritis in four strains of rats. Proc SOC Exp Biol Med 132:608-612. Mackenzie AR, Sibley PR, White BP 1979 Resistance and susceptibility to the induction of rat adjuvant disease. Br J Exp Pathol 60507-512. Tanaka Y, DeLuca HF 1975 The control of 25-hydroxyvitamin D, metabolism by inorganic phosphorus. Arch Biochem Biophys 154566-572. Trentham DE, Townes AS, King AH 1977 Autoimmunity to

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

type I1 collagen: An experimental model of arthritis. J Exp Med 146:857-868. Tanaka H, Shirota H, Kuwada M, Terato K, Katayama K 1988 Plasma and bone osteocalcin levels in rats with type 11 collagen-induced arthritis. Arthritis Rheum 31: 1413- 1420. Mime HW, Pfeilschifter J, Scharla S, Mutschelknauss S, Schwarz A, Krempien B, Ziegler R 1984 Inflammation-mediated osteopenia in the rat: A new animal model for pathological loss of bone mass. Endocrinology 115:50-54. Pfeilschifter J, Mime HW, Enzmann E, Krempien B, Ziegler R 1985 Inflammation-mediated osteopenia in the rat: The effects of artificial granuloma and sham operation on cortical and trabecular bone. Bone 6:461-465. Donaldson CL, Hulley SB, Vogel JM, Hattner RS, Bayers JH, McMillan DE 1970 Effect of prolonged bed rest on bone mineral. Metabolism 19:1071-1084. Halloran BP, Bikle DD, Wronski TJ, Globus RK, Levens MJ, Morey-Holton E 1986 The role of 1,25-dihydroxyvitamin D, in the inhibition of bone formation induced by skeletal unloading. Endocrinology 118:948-954. Connolly KM, Stecher VJ, Kent L 1988 Examination of interleukin-1 activity, the acute phase response and leucocyte subpopulations in rats with adjuvant-induced arthritis. J Lab Clin Med 111:341-347. Johnson WJ, Muirhead KA, Meunier PC 1986 Macrophage activation in rat models of inflammation and arthritis. Arthritis Rheum 29:1122-1130. Reichel H, Koeffler HP, Bishop JE, Norman AW 1987 25Hydroxyvitamin D, metabolism by lipopolysaccharide-stimulated normal human macrophages. J Clin Endocrinol Metab 64:l-9. Reichel H, Koeffler HP, Barbers R, Norman AW 1987 Regulation of 1,25-dihydroxyvitamin D, production by cultured alveolar macrophages from normal human donors and from patients with sarcoidosis. J Clin Endocrinol Metab 65:12011209. van Soesbergen RM, Lips S, van den Ende A, van der Korst JK 1986 Bone metabolism in rheumatoid arthritis compared with postmenopausal osteoporosis. Ann Rheum Dis 45: 149155.

Address reprint requests to: Craig B. Langman, M.D. Mineral Metabolism Laboratory Nephrology Division Children’s Memorial Hospital 2300 Children’s Plaza Box #37 Chicago, IL 60614 Received for publication July 7, 1989; in revised form May 4, 1990; accepted May 7, 1990.