JOURNAL OF BONE AND MINERAL RESEARCH Volume 5, Number 6, 1990 Mary Ann Liebert, Inc., Publishers

Trabecular Bone Remodeling and Bone Mineral Density in the Adult Cat During Chronic Dietary Acidification with Ammonium Chloride SHELLEY V. CHING,'.' ROBERT W. NORRDIN,' MARTIN J. FETTMAN,' and RICHARD A. LECOUTEUR~

ABSTRACT Ammonium chloride (NH,CI) is used as a urinary acidifier in the treatment and prevention of feline urologic syndrome. It is reported to cause alterations in calcium and bone metabolism in humans, dogs, and rats. Adult cats with normal renal function were fed 1.5% NH,CI for 6 months to study the effects of chronic dietary acidification on trabecular bone remodeling of the iliac crest and bone mineral density (BMD) of lumbar vertebral trabecular bone and femoral cortex. Histomorphometric analyses of iliac crest biopsies were performed before and after treatment. Static and dynamic parameters of bone resorption and formation were determined. Single-energy quantitative computed tomography (SEQCT) was used to measure lumbar trabecular and femoral cortical BMD. There were no significant treatment effects in iliac crest trabecular bone remodeling or BMD of the vertebrae and femora. Bone remodeling activity decreased with time in both acidotic and control cats. Vertebral BMD increased with time in both groups of cats, whereas no change was seen in the femora. Thus, chronic dietary acidification for 6 months with therapeutic levels of NH,CI produced no significant changes in trabecular bone remodeling or bone mineral density in adult cats.

INTRODUCTION with ammonium chloride (NH4CI) produces metabolic acidosis and alterations in calcium and bone metabolism in humans and experimental animals. Total serum calcium levels are normal; however, blood ionized calcium concentrations are increased. Metabolic acidosis is also associated with increased urinary calcium excretion, normal to decreased intestinal absorption of calcium, and increased mobilization of calcium from bone. These combined alterations can result in a lower or net negative calcium balance, and osteopenia can develop with a decrease in bone mineral and r n a t r i ~ . ' ~ . ' ' ]

C

HRONIC DIETARY ACIDIFICATION

Bone serves as an extrarenal buffer source in chronic metabolic acidosis (CMA) since it has a large store of calcium carbonate, calcium phosphate, and other alkaline salts and large surface availability of sodium and potassium as readily exchangeable ions for H+.(5,6,9,11-'6) This large alkaline reserve functions slowly to stabilize the extracellular bicarbonate level by disposing of part of the endogenous acid load not excreted in the urine.~s-7~'o~'2~1~31 Current evidence indicates that there may be increased bone demineralization attributable to the dissolution of mineral in the acidic environment, loss of bone matrix due to increased cell-mediated bone resorption, or Changes in bone can be stimulated directly by the acid environment or indirectly through the actions

!Department of Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO

80523. *Present address: Department of Safety Assessment, Merck Sharp & Dohme Research Laboratories, WP45-213, West Point, PA 19486.

3Department of Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO 80523.

541

CHING ET AL.

548

of parathyroid hormone (PTH) and/or 1,25-dihydroxyvitamin D,.(4 l 9 2 1 - 2 3 ) Other studies have demonstrated that these effects are independent of PTH and dietary level of calcium and phosphorus. ( 4 * * 4 ) Histologically, cancellous bones in humans and experimental animals have decreased number and width of bony trabeculae in chronic acidosis." 2 5 2 6 ) H'istomorphometric data in humans with CMA and normal levels of blood PTH, 1,25-dihydroxyvitamin D3, calcium, and phosphorus have demonstrated a reduced mineral apposition rate and widened osteoid seams, although active mineralization was still present.(z7' Urinary hydroxyproline concentrations in acidosis are frequently increased, indicating an increase in bone turnover, particularly in bone r e s ~ r p t i o n .24' ~29 2 9 1 Ammonium chloride is commonly used in the treatment and prevention of feline urologic syndrome (FUS). Feline urologic syndrome is a multifactorial disorder of the feline lower urinary tract characterized by difficulty, straining, or increased frequency in urination, blood and/or crystals in the urine, cystitis, urethritis, and/or urethral obstruct i ~ n . ' The ~ ~ )incidence of struvite crystal formation and FUS increases when the urinary pH is greater than 6.4."O) Ammonium chloride has been demonstrated to be an efficacious urinary acidifier in the management of this disorder and in most cases must be given for long periods of time.'" The purpose of this study was to examine the impact of this therapeutic practice on several metabolic parameters(") and to evaluate the cat as a potential model to study the long-term effects of dietary acidification on bone metabolism.

MATERIALS AND METHODS

General and experimental design A total of 22 adult, healthy, specific pathogen-free (SPF) domestic cats were obtained for this study from Dr. E.A. Hoover (Department of Pathology, Colorado State University, Fort Collins, CO). Ages ranged from 1 to 6.5 years (mean = 2.2 * 1.4). Most were 1.5 years, with ages distributed equally in groups. There were 12 castrated male cats and 10 female cats, 2 of which were spayed. Castration had been performed at various ages and therefore the studies were not standardized to time of castration. All cats were removed from gang cages and individually housed in stainless steel cages in a light- (12 h cycle) and temperature- (22.2"C) controlled room. Each cat received daily exercise within the room for approximately 1 h during the cage-cleaning period. Experimental procedures were conducted in compliance with the guidelines in the "Guide for the Care and Use of Laboratory Animals" (USDHHS, PHS, NIH in Publication #85-23, 1985). All cats were acclimated to their cage environment for a minimum of 2 months before the beginning of the experiment. During the first month of acclimation, the cats were fed a commercial diet (Purina Cat Chow, Ralston-Purina Co., Checkerboard Square, St. Louis, MO) ad libitum. An experimental nutritionally balanced, neutral, dry basal diet (Tables 1 and 2) was then fed ad libitum during the second month. After the first month of acclimation, each cat was

evaluated for normal renal function by measuring glomerular filtration rate and effective renal plasma At the end of the acclimation period, the cats were paired for age and sex and randomly divided into control and experimental groups of 1 I cats each. The controls continued to consume the experimental basal diet, and the experimental group consumed the basal diet with 1.5% ammonium chloride (Tables 1 and 2). The groups were fed their respective diets for 6 months. Urine was collected anaerobically by cystocentesis once a month in the morning and urinary pH was measured immediately (Corning Digital 112 Research pH meter, Corning Medical, Medfield, MA). Venous blood samples (10

OF THE BASAL TABLE1. INGREDIENTCOMPOSITION EXPERIMENTAL DIET

Amount

(Yo of dry matter)

Ingredient

Ground and whole yellow corna Poultry meal Soybean meal Wheat Bone and meat meal Dicalcium phosphate Fish meal NaCl (iodized) Vitamin premixb Brewer's yeast Trace mineral mixb Choline chloride Spray-on fat digest

42.94 26.00 10.00 9.00 3.00 1 .00 1 .00 0.70 0.67 0.40 0.20 0.09 5.00

acornmeal (1.5%) was replaced by 1.5% ammonium chloride in the acidified experimental diet. bA standard feline mix, supplied by Ralston-Purina (St. Louis, MO). Composition is a trade secret.

TABLE2. CHEMICAL COMPOSITION OF THE BASAL AND ACIDIFIED EXPERIMENTAL DIETS Ingredient

Crude protein Ash Fat Calcium PhosPhorus Potassium Sodium Magnesium Chloride

Basal diet (% of dry matter)

Acidified diet (% of dry matter)

29.7 8.00 8.62 1.76 1.24 0.57 0.47

30.6a 8.00 7.76 1.68 1.22 0.55

0.15

0.51

0.51

0.15 1.61

acorrected for the amount of nitrogen contributed by 1.5% NH.CI.

BONE REMODELING IN AClDOSIS ml) were collected once a month from unanesthetized cats into lithium-heparinized syringes. Blood gas analyses (Corning 165/2 blood gas analyzer) were determined within 1 h.‘33) Bone biopsies and computed tomography (CT) were performed before treatment with NH,CI to obtain baseline values and again after 6 months. Thus, each cat served as its own control. All the cats were biopsied, but only six (three treated and three controls, age and sex matched) had C T scans. Urine samples were collected monthly during metabolic balance studies to measure urinary total hydroxyproline.

549 using a Leitz binocular microscope and Merz alternating hemispherical line eyepiece reticule. f 3 6 ) All measurem e n t ~ ‘began ~ ~ ) one optical field away (0.42 mm) from the endocortical surface, cartilage, and/or tissue edges to ensure that only trabecular bone surfaces were measured. The following static parameters were measured in trabecular bonec36,37) using a standardized nomenclature. ( 3 8 1 Trabecular bone volume (To), osteoid volume (Yo), osteoid surface (Vo), osteoblast surface (To), osteoclast surface (Yo), eroded surface (OC-, Vo), number of osteoclasts per square millimeter of surface area, osteoid thickness (pm), and surface density (mm’/mm”). Surface density

Procedure for bone labeling and biopsy All cats received injections for double labeling of bone for both pre- and posttreatment bone biopsies to apply specific fluorescent markers to actively mineralizing bone surfaces. The labeling schedule for both markers was 1-72-3 in which the marker was given 1 day, followed by a labeling-free interval of 7 days, and a second administration for 2 days. A 1% sterile calcein in buffer solution, 15 mg/ kg body weight per day, and oxytetracycline HC1, 25 mg/ kg body weight per day, were given intravenously as the pre- and posttreatment markers, respectively. Iliac crest bone biopsies were surgically removed using a Michele bone trephine 3 days after the second administration of the bone label as previously described. f 3 5 ) Pretreatment and posttreatment biopsies were taken from the right and left iliac crest, respectively. The cats recovered from the surgery well and resumed normal activity within 24 h.

Processing of bone biopsies Biopsy specimens were immediately fixed in chilled 10% phosphate-buffered formalin for 48 h. Processing consisted of gradual dehydration of undecalcified bone by increasing concentrations of ethanol. Specimens were embedded in methyl methacrylate. Histologic sections were cut o n a Jung Model K sledge-type microtome (R. Jung, Heidelburg, West Germany) at 5 pm, stained at pH 3.0 with 1.0% toluidine blue, and 15 pm thickness left unstained for ultraviolet microscopic quantitative evaluation of static and dynamic parameters of bone remodeling, respectively.

Histomorphometric measurements of bone biopsy sections Bone biopsy slides acceptable for quantitative evluation included those that contained both cortical and trabecular bone with a predominance of trabecular bone. Data from cats with biopsy samples containing primarily cortical or endocortical bone or having a large amount of processing, cutting, or staining artefact were eliminated. At least 20 fields were required to be acceptable for evaluation at a magnification of x 250 and 10 fields at x 100 for stained and unstained slides, respectively. Static parameters were measured with a manual counter

=

c; 4 S,(,) = C h *d

where C i = total intersects on trabecular surfaces, C h = nP, = total number of grid points examined in all fields, and d = the distance between grid points in object space in millimeters determined using the calibrating stage micrometer. Dynamic parameters of bone formation and rates were measured using ultraviolet light microscopy, a microscope drawing tube, and a computerized digitizing board (Videoplan system, Carl Zeiss, Inc., New York, NY). Total trabecular surface and single- and double-labeled surfaces were traced at a magnification of x 100 using an optical field calibrated with a stage micrometer to equal 1 mm’. Mean distance between double labels was measured at a magnification of x 250. The distance between the centers of the double label lines was measured perpendicularly at points of intersection of the Merz grid lines with double labels. The following dynamic parameters were measured in trabecular bone: (36-38) Double- and single-labeled surfaces (Yo) and mean distance between double labels (pm). The derived parameters were as follows: Mineral apposition rate &m/day) was calculated as the ratio of the mean distance between double labels and the time between the midpoints of calcein or oxytetracycline administration in days multiplied by a stereologic correction factor of 7r/4 (distance between labels over 8.5 days for a 1-7-2-3 schedule x 0.79). Bone formation rate (pm”/pm2 per year) was calculated as appositional rate in pm multiplied by the fractional bone-forming surface, which is the fraction of trabecular label that is all of the double-labeled surface plus one-half the single-labeled surface, multiplied by 365 daydyear [M x (double + 112 single) x 3651. Adjusted apposition rate (prn’/pm* per year) was calculated using the equation for bone formation rate divided by the fraction of osteoid surface.

Computed tomography A total of six adult cats were anesthetized with thiopental sodium to prevent motion during the CT examination. C T was performed before and after treatment and within 1 week after bone biopsy.

CHING ET AL.

550

A Picker 1200 SX CT scanner was used to evaluate bone mineral density in the lumbar vertebrae and femoral cortex. Cats were placed in the supine position over calibration standards kindly provided by Dr. C.E. Cann (Dept. of Radiology, University of California, San Francisco, CA) of potassium phosphate (K2HP04) for bone (100 mg/ml mineral equivalent) and water. Simultaneous scanning of the cats and the standard was used to calibrate the CT number in the scan image and convert CT attenuation values to mineral equivalent densities. Single-energy CT scanning at 100 kVp was used for these studies. The scanning procedure for the lumbar vertebrae was similar to that performed in human^,(^^.^'^ with some modifications. The lateral computed radiograph (scout view) was used to localize the midpoint of the vertebral bodies, L1-L5 and the shaft of the femora. A 5 mm thick transverse scan was done at this point using factors of 100 kVp, 50 mA, field size of 14 cm, XR filter 1, image algorithm 3, compensator in, resolution high, and scan angle 398” (Dr. C.E. Cann, personal communication). An irregular elliptical region of interest (Fig. 1) was used to outline the medullary cavity of each of the vertebral bodies, excluding cortical bone.(4z) Point density measurements of the calibration standard and the anterior, posterior, medial, and lateral femoral cortex of both legs were performed. Because a computerized localization system was not available for this study,

the posttreatment positioning and scans were reproduced as closely as possible to the original scans by matching visual cues on the hard copy images.(421 To convert CT numbers to mineral equivalent (min eq) (mg/ml), the following equation was used(39,4’): CT number of vertebra CT number of K,HPO,

-

CT number of water K,HPO, mg/ml CT number of water

Urinary total hydroxyproline analysis Urine hydroxyproline concentration was assayed as described by E i n a r s ~ o n ( ~with ” ’ modifications in dilutions.

Statistical analysis Histomorphometric measurements were compared within treatment and sex groups with Student’s paired ttest. The unpaired t-test was used to compare the means between treatment groups. P 5 0.05 was considered statistically significant. A three-factor analysis of variance with repeated measures was used on CT data to compare differences due to site of measurement, between treatment groups, and time. A two-factor analysis of variance with repeated measures was used to compare differences within treatment, age, and sex groups and in the percentage change in mineral equivalence between the groups. It was also used to evaluate differences in urinary hydroxyproline concentrations within each group that may have occurred with time. Student’s unpaired f-test was used for urinary hydroxyproline to compare differences in the means between groups at each month. Fisher’s least significant difference (LSD) test was used to rank the means and identify those means that were significantly different (p < 0.05). Relationships between vertebral mineral density versus percentage TBV of the iliac crest, percentage change mineral density versus percentage change body weight, and femoral mineral density versus vertebral mineral density were determined by linear regression analyses. Data are expressed as means + standard deviations in text and tables. Fewer numbers of cats are included in the statistical analyses of paired samples compared to unpaired samples.

RESULTS The cats fed NH4C1 for 6 months developed metabolic acidosis and remained significantly acidotic throughout the study compared to control cats (data not shown).‘”) Blood pH and blood bicarbonate (month 6: blood pH treated; 7.29 + 0.06 versus control, 7.36 + 0.04; blood bicarbo2.61 nate treated, 17.90 + 2.69 versus control, 21.16 mmol/liter) were significantly lower in treated cats than in control cats.‘”) Urinary pH in treated cats remained at 6.4 or lower, which was significantly lower than in control cats, and relative to month 0.(331 Table 3 gives the histomorphometric data of the iliac crest comparing the acidotic cats with the control cats. Biopsy sample size was varible, and the amount of trabecular

*

FIG. 1. Transverse image of computed tomographic scan depicting elliptical region of interest used for measurement of average bone mineral density of trabecular bone in medullary cavity of lumbar vertebra.

BONE REMODELING IN ACIDOSIS

551

TABLE3. HISTOMORPHOMETRIC PARAMETERS IN THE ILIAC CRESTOF ACIDOTIC AND CONTROL CATS^

Acidified cats (n

=

Control cats (n

9)

=

11)

Parameters

Pretreatment

Posttreatmen t

Pretreatment

Posttreatmen 1

Body weight, kg Trabecular bone volume, Vo Surface density, mmz/mm3

3.27 + 0.60 34.6 f 8.9 4.01 f 0.57

3.60 f 0.98 34.4 f 8.3 3.56 f 0.35b

3.44 f 0.89 30.1 f 7.3 3.67 f 0.91

3.67 f 1.05 31.2 f 4.8 3.36 f 0.93

Formation Osteoid surface, Vo Osteoid volume, Vo Osteoblast surface, 070 Osteoid thickness, pm Single-labeled surface, Vo Double-labeled surface, Vo Mineral apposition rate, pm/day Bone formation rate, pm3/pm2 per year Adjusted apposition rate, pm3/ pmZ per year Resorption Eroded surface, Vo (Oc') Eroded surface, Vo (Oc-) Number of osteoclasts, mm-' surface

19.0 3.5 16.6 7 3.44 7.19 1.11 0.042

0.18

*

16.4

f 3.5 f 16.3

* 2 f 1.96 f 6.00 f 0.39 f 0.041 f

0.06

1.4 f 1.5 4.6 f 2.9

0.05 f 0.04

7.2 1.0 4.2 7 3.54 3.88 0.74 0.018

f

6.9b

f 1.8b f 5

9 * 3 f 3.74 f 4.52 f 0.33b f 0.021

17.3 3.0 15.4 7 4.71 8.33 1.01 0.044

f 12.2 f 2.3 f 12.3

* 2 f 2.37

8.81 0.22 f 0.050 f

f

+

6.6 1.4 4.9 6 3.07 4.76 0.76 0.019

6 3 1.6 f 6.9~ f 2 f 2.57 f 3.61 f 0.34~ f 0.017c f

f

0.36

f

0.35

0.28

0.18

0.37

f

0.43

0.5 3.0 0.03

f

0.4 1.4 0.04

1.3 f 0.7 3.7 f 2.2 0.06 f 0.03

0.5 1.9 0.03

f

0.6~ 1.9b 0.03b

f f

f f

aData are expressed as means + SD. b O . l 2 P z 0.05 versus pretreatment. c P 5 0.05 versus pretreatment.

bone available for measurement was frequently small (mean = 7.49 f 2.51 mmz, range = 3.52-11.44 mm'). Significant decreases in percentage osteoid surface, percentage osteoblast surface, mineral apposition rate, bone formation rate, and percentage osteoclast resorption surface were seen in control cats with time. Acidotic cats demonstrated decreasing trends in some bone formation parameters with time, but no differences were seen between groups in these parameters or in the percentage change with time. Although not always statistically significant, bone-remodeling activity in all cats generally decreased during the experimental period. Measurements that were significantly different between sexes included surface density, percentage osteoid surface, percentage osteoblast surface, osteoid thickness, and actual perimeter length of total osteoid (Table 4). Only surface density of treated intact females was significantly different from controls. Castrated males, both treated and controls, had increased formation surfaces compared with intact females. The spayed females were excluded from these data because there were only one treated female and one control. The histomorphometric values from cats greater than 3 years of age are summarized in Table 5. These cats had significantly reduced posttreatment osteoid thickness, percentage double-labeled surface, and bone formation rate compared t o control cats, which was not seen in cats younger than 2.5-3 years of age.

Table 6 summarizes the effects of chronic NH,CI on urinary total hydroxyproline excretion. There were no significant differences in 24 h excretion between groups, but the urinary hydroxyproline/creatinine ratio was significantly lower in the acidotic cats than in control cats in month 5. The acidotic cats had significantly decreased urinary hydroxyproline excretion in months 3-6 relative to month 0. The percentage decrease with time measured in both control and acidotic cats was not significantly different between groups. Bone mineral densities derived from CT of the lumbar vertebrae and femoral cortices are given in Table 7. There is some variability between individual lumbar vertebrae, but none that is significant. No differences were seen as a result of the treatment, but significant increases in bone mineral density for each vertebra were measured with time in both groups. When the vertebrae were averaged together for a single mean value for lumbar mineral density per cat, there were no differences due to treatment, but differences with time were present. The percentage change with time was similar between groups. The bone mineral density of the femoral cortices did not change significantly with time and was unaffected by treatment. Linear regression analyses between lumbar bone mineral density versus percentage TBV of the iliac crest, percentage change in bone mineral density with time in the lumbar vertebrae versus percentage change in body weight, and

CHING ET AL. TABLE 4. SIGNIFICANT HISTOMORPHOMETRIC MEASUREMENTS IN THE ILIAC CREST OF MALEAND FEMALE CATS^

Acidified

Control

Measurement

Pretreatment

Post treatment

Pretreatment

Posttreatment

Surface density, mm2/mm3 Male Female

3.79 f 0.58 4.28 f 0.48

3.65 f 0.37 3.38 f 0.26b

3.74 f 0.76 3.73 f 1.29

3.69 f 1.19 3.17 f 0.48

Osteoid surface, Vo Male Female

28.3 f 16.9~ 7.5 f 3.8

10.4 f 6 . 1 ~ 0.6 f 0.4

23.5 10.0

Osteoblast surface, Vo Male Female

26.4 f 16.3~ 4.5 f 1.3

Osteoid thickness, pm Male Female Actual perimeter length total osteoid, mm/mm2 Male Female

8 f 1c 6 f 1

0.83 0.26

f f

0.51 0.15

6.3 0.0

f f

5.6~ 0.Ob

f f

11.0 11.1

7.6 f 4.0b~C 1.6 f 1.3

22.9 f 11.7~ 5.9 f 5.4

8 f 1c 3 f 4

8 f 2c 6 f 1

0.23 f 0.15~ 0.02 f 0.01

0.68 f 0.33 0.37 f 0.52

5.4 f 5.lb 0.5 f 0.5 7 f 3 5 f 3

0.21 f 0.09b.c 0.04 f 0.04

aData are expressed as means + SD; n = 5 average in each group (male, female, acidified, and control) for individual measurements. bP 5 0.05 paired 1-test between pre- and posttreatment. c P 5 0.05 between male and female.

TABLE 5. HISTOMORPHOMETRIC PARAMETERS OF

THE ILIAC CREST IN THAN 3 YEARSOF AGE^

Acidified cats (n Parameters

Pretreatment

Trabecular bone volume, Vo Surface density, mm2/mm3

34.7 4.27

Formation Osteoid surface, Vo Osteoid volume, Vo Osteoblast surface, Vo Osteoid thickness, pm Single-labeled surface, Vo Double-labeled surface, Vo Mineral apposition rate, pm/day Bone formation rate, pm3/~rn2 per year Adjusted apposition rate, pm3/ pm2 per year Resorption Eroded surface, Vo (Oc+) Eroded surface, Vo (Oc-) Number of osteoclasts, mm-' surface aData are expressed as means f SD. bO.l 2 P 2 0.05 versus pretreatment. CP 5 0.05 versus pretreatment. dP 5 0.05 between groups.

f f

4.1 0.59

6.6 f 4.1 1.5 f 1.4 3.9 f 0.7 6*1 1.76 f 1.88 0.57 f 0.65 0.90 f 0.28 0.005 f 0.001 0.11

f

0.4 3.1 0.05

f

0.03

0.4 f 0.7 f 0.04

=

ACIDIFIED AND

Control cats (n

3)

Post treatment 27.4 f 11.0 3.38 f 0.37b 0.4

f 0.2 0.0 0.oc O* 0.28 f 0.13 0.27 f 0.07d 0.45 f 0.64 0.001 f 0.001d

0.28

CONTROL CATS OF GREATER

f

0.40

0.4 f 0.2 2.9 f 1.5 0.01 f 0.001

Pretreatment 25.0 3.87

f

17.6 2.9 14.3 7 5.05 5.12 1.00 0.027

f

0.20

f

f

7.4 1.32

11.6 1.4 f 10.2 f l f 4.17 f 5.17 f 0.17 f 0.023 f

=

4)

Posttreatment 29.7 f 3.8 3.00 f 0.65

*

9.7 9.7 2.2 i2.3 8.5 & 10.1 8 & 1 1.61 i1.29 3.56 i1.39 0.88 & 0.12 0.014 & 0.006

0.12

0.17

i0.19

0.9 f 0.4 3.0 f 3.0 0.04 f 0.01

0.4 1.1 0.04

&

0.7 0.3 i0.04 &

BONE REMODELING IN ACIDOSIS

553

TABLE 6. EFFECTOF CHRONICDIETARYACIDIFICATION ON URINARY HYDROXYPROLINE EXCRETION IN ADULTCATS^

0

3

2

1

4

6

5

~~

Urine hydroxyproline/creatinine,a pg/mg NHdCI Control

69 f 40 60 f 40

55 51

24 h urinary hydroxyproline,a pmol/kg body weight NH4CI

22 f 11

Control

18 f 11

49 f 24 47 f 14

40 f 13b 37 f 5b 46 i 14 62 f 28

16 f 5

16 f 8

13

17 f 5

16 f 4

15 f 3

f f

24 20

f

5b

32 f 7b.c 53 f 19

15

f

2b

11

21

f

13

15 f 7

f Ib

29 47

f f

4b 41

8

f

2b

13 f 10

aData are expressed as mean f SD. Month 0 is baseline month and months 1-6 are treatment months. n = 5 NH,CI; n = 6 control. bP I 0.05 versus month 0. CP 5 0.05 between groups.

TABLE 7. EFFECTOF CHRONIC DIETARY ACIDIFICATION ON BONEMINERAL DENSITY IN ADULTCATSa Controlb (mg/ml)

Acidqiedb (mg/ml) Prefreatmenf Lumbar vertebrae L1 L2 L3 L4 L5 L 1-5 averaged 9'0 Change with time Femoral cortex Left femur Right femur

Posttreatmenf

204 f 70 178 f 82 173 f 70 153 f 89 142 f 64 170 f 75

208 f 192 f 198 i 196 f 194 f 198 f 15

726 730

f

67

+ 87

f

87c 94c lllc 109c lOlc 92c

Pretreatment 196 f 181 f 154 f 166 f 148 f 168 f

46 44 58 64 60 55

226 217 192 205 181 201 19

11

786 f 35 711 80

*

734 657

f f

Posttreatmenf

88 43

f

f f

f f f f

72c 81c 67c 84c 83c 76c

17 688 f 24 671 f 66

aData expressed as means f SD. Bone mineral density is expressed in mineral equivalents (mglml). b n = 3 in each group. CP 5 0.05 versus pretreatment.

femoral bone mineral density versus lumbar bone mineral density showed poor relationships, none of which were significant.

DISCUSS10N The results of this study demonstrated that chronic ingestion of therapeutic levels of NH4CI did not affect iliac crest trabecular bone remodeling, nor were there significant changes in vertebral or femora1 bone mineral density in cats. Thus, osteopenia did not develop as a result of CMA in this experimental model. The iliac crest was chosen to study trabecular bone remodeling because it is easily available without sacrificing cats and contains a sufficient amount of cancellous bone.(44)We have demonstrated in cats that this is indica-

tive of remodeling in vertebrae and bone remodeling values between right and left sides of the iliac crest are not significantly different.(35)Turnover rates for trabecular bone are four to eight times greater than cortical bone so that changes in bone metabolism due to systemic disease or stress are reflected more rapidly in areas that contain more cancellous bone. (45) Measurement of bone mineral density was performed by C T because it is an acceptable noninvasive method to clinically quantitate bone mass and bone mineral content in osteoporotic conditions. ( 3 9 , 4 6 ) Measurements were taken from sites in both the axial and appendicular skeleton to compare the relationship of mineral density. Data from CT have shown poor correlation in bone mineral density between skeletal sites, and the loss of bone mineral was detected earlier in the lumbar spine than in peripheral sites.'47) In metabolic studies in these cats, NH,CI induced chronic

554

metabolic acidosis, significantly increased blood ionized calcium concentrations, and significantly lowered calcium balance. (331 Lower calcium balance was most pronounced in months 1-4 and resulted from the combined effects of increased urinary calcium excretion and decreased intestinal calcium absorption.(.33)Plasma parathyroid hormone concentrations were unaffected; however, plasma 1,25-dihydroxyvitamin D, was significantly decreased in the acidotic cats.‘”) The lack of changes in parameters of bone metabolism despite CMA, lower calcium balance, and significant alterations in calcium metabolism in these cats is in contrast to previous studies in humans, dogs, and rats in which CMA induced by NH,CI resulted in increased bone resorption, increased bone mineral mobilization, and subsequent loss of bone mineral and matrix.(1,4.t1.2s1 Increased osteoclastic osteolysis has been associated with CMA but was not detected in the acidotic at^.(^.^^.^^^ Based upon previous studies in rats that demonstrated changes within 2 weeks to 6 months was considered adequate time for 300 bone changes to occur. Although the changes in acid-base and calcium metabolism in these cats were statistically significant, they may not have been quantitatively sufficient to induce changes detectable by histomorphometric or bone mineral density analyses. The diet composition and acidogenic potential of the diet used in this study compared to diets and acidifying supplements used in other studies in humans, rats, and dogs may differ. Although other studies reported the use of concentrations of 1.5% NH4CI or higher and subsequent development of osteopenia, the cation-anion balance of the diets was unknown. It is possible that in other studies the total amount of acid stress was greater than that induced in the cats of this study. Decreased remodeling activity with time in both groups could be attributable t o environmental and/or age-related effects in the cats. Although the cats were acclimated to their environment for a minimum of 2 months prior to the study, continued confinement, reduced physical activity, and other manipulations associated with the study may have contributed to the decrease in bone remodeling. Reduced bone remodeling activity may be attributable to skeletal aging and/or attainment of a steady state, which would account for some of the increased mineral density as bone matures. However, in the small group of acidotic cats (n = 3) older than 3 years of age, the pattern of remodeling changes characterized by a reduction in bone formation may represent an effect of chronic dietary acidification that more effectively causes bone changes in older individuals when active growth has ceased. Cats younger than 3 years old did not demonstrate significant differences in remodeling activity.(481 The decrease in urinary total hydroxyproline in both groups of cats supports the histomorphometric data that reduced bone remodeling occurred with time. This is in contrast t o data from other species in which hydroxyproline excretion increased in response to the administration of NH4C1.(4.24,28,29) Food consumption in both groups of cats also decreased with time, and this may have contributed to reduced urinary hydroxyproline excretion. (33L

CHING ET AL. The wide variation apparent in bone remodeling parameters and urinary hydroxyproline may be attributable t o several factors. Age, sex, hormonal status, length of time between castration and experiment, and other unknown factors from environmental conditions and experimental manipulation should be considered. There is little information on the quantitative parameters of feline bone remodeling, and it is unknown what effects these factors may have on feline calcium and bone metabolism. Further studies in a much larger group of older cats and cats of the same hormonal status may demonstrate that age and sex are important sources of variation in a mixed population of cats. The castrated male cats had generally more bone formation activity than the intact female cats both before and after treatment. It is undetermined whether age of castration or length of time between castration and experiment are influential because the cats were castrated at different ages. Bone mineral density of the lumbar spine and femora calculated from CT was not different in the acidotic cats compared to the control cats. Both groups had increases, rather than decreases, in vertebral bone mineral density with time. The precision of the scans with time determined using the calibration standard was 5-7% compared to 2-5% previously reported in To consider the possibility that bone mineral density changes were due to changes in weight gain and body size, mineral densities were standardized for individual body weight, but there were still no significant differences between groups. The values for bone mineral density in our cats are in contrast to previous studies in humans and experimental animals that have demonstrated increased loss of bone calcium to serve as an extrarenal buffer for chronic acidosis.iS 7 9 l l I2 2 0 ) R educed bone mineral content is seen in humans, dogs, and rats given NH4CI or who are chronically acidotic from metabolic disease. ‘ 2 . 2 5 49) Th e results of the present study agreee with those of Newell and Beau~ h e n e , (who ~ ’ demonstrated no changes in bone mineral in variably aged rats given NH4CI for 9 months. Detectable changes were seen in the lumbar vertebrae but not in the femoral cortices, which suggests that cancellous bone mineral accretion in the vertebrae was greater and more readily measured than in the cortical bone of the femora. These results in comparing appendicular and axial bone mineral densities in cats are consistent with several reports in humans.‘,’ The limitations of SEQCT and the potential for errors in measurement and interpretation of data due to bone marrow fat and positioning have been Three-dimensional positioning techniques and the use of dualenergy CT can reduce this error, but these were not available for this Our results indicated that measurably significant values were obtained and suggest that CT is a useful noninvasive method to measure BMD in animals. It is unclear how much the skeleton contributed to extracellular buffering in CMA or to the increase in blood ionized calcium measured in the acidotic cats. Perhaps there was continued release of surface-available calcium ions and buffers by a passive physicochemical equilibrium ”)

BONE REMODELING IN ACIDOSIS mechanism between the extracellular fluid and the more readily available amorphous soluble apatite. ( 1 s . s 2 - s 4 ) A recent in vitro study demonstrated that proton-induced calcium efflux is due to dissolution of bone calcium carbonate, not the result of an alteration in bone cell function. ( 5 5 ) The equilibrating buffer ion-exchange rate of bone has been reported to be 10-50 times greater than net mineral turnover from bone remodeling. ( 5 4 . 5 6 ) The cat has an uniquely higher dietary protein requirement than many other species, including the human. As a result of this additional acid stress, we can only speculate that perhaps the feline skeleton may have evolved to be more resistant to acid-base changes compared to other species and may be able to provide buffering with bone carbonates and phosphates without stimulation of cellmediated bone matrix turnover or significant losses of bone mineral content.

ACKNOWLEDGMENTS Financial support for this research was provided by the Ralston Purina Company, Robert H. Winn Foundation, and the Morris Animal Foundation. The authors acknowledge technical assistance from Kay Smith and Maggie Voorhees, the histopathology laboratory, and Sue Ridgel for CT. We thank Dr. Christopher E. Cann for technical advice for CT and review of this manuscript.

REFERENCES 1 Jaffe HL, Bodansky A, Chandler J P 1932 NH,CI decalcification, as modified by calcium intake: The relation between generalized osteoporosis and osteitis fibrosa. J Exp Med 56: 823-834. 2. Gafter U, Kraut JA, Lee DBN, Silis V, Walling MW, Kurokawa K, Haussler MR, Coburn JW 1980 Effect of metabolic acidosis on intestinal absorption of calcium and phosphorus. Am J Physiol 239:G480-484. 3. Keiser BA, Leitz H, Moore ES, Favus MJ, Coe FL 1982 The effect of chronic metabolic acidosis on calcium and phosphorus balance (abstract). Pediatr Res 16:323A. 4. Kraut JA, Mishler DR, Singer FR, Goodman WG 1986 The effects of metabolic acidosis on bone formation and bone resorption in the rat. Kidney Int 30:694-700. 5 . Lemann J, Lennon EJ 1972 Role of diet, gastrointestinal tract and bone in acid base homeostasis. Kidney lnt 1:275279. 6. Lemann J, Lennon EJ, Goodman AD, Litzow JR, Relman AS 1965 Net balance of acid in subjects given large loads of acid or alkali. J Clin Invest 44507-517. 7. Lemann J, Litzow JR, Lennon EJ 1966 The effects of chronic acid loads in normal man: Further evidence for the participation of bone mineral in the defense against chronic metabolic acidosis. J Clin Invest 451608-1614. 8. Lemann J , Litzow JR, Lennon EJ 1967 Studies of the mechanism by which chronic metabolic acidosis augments urinary calcium excretion in man. J Clin Invest 46:1318-1328. 9. Newell GK, Beauchene RE 1975 Effects of dietary calcium level, acid stress and age on renal, serum, and bone responses

555 of rats. J Nutr 105:1039-1047. 10. Petito SL, Evan J L 1984 Calcium status of growing rat as affected by diet acidity from NH,CI, phosphate and protein. J Nutr 114:1049-1059. 11. Barzel US, Jowsey J 1969 The effects of chronic acid and alkali administration on bone turnover in adult rats. Clin Sci 36:517-524. 12. Goodman AD. Lemann J. Lennon EJ. Relman AS 1965 Production, excretion and net balance of fixed acid in patients with renal acidosis. J Clin Invest 44:495-506. 13 Litzow JR, Lemann J , Lennon EJ 1967 The effect of treatment of acidosis on calcium balance in patients with chronic azotemic renal disease. J Clin Invest 46:280-286. 14. Bettice JA 1984 Skeletal CO, stores during metabolic acidosis. Am J Physiol 247:F326-330. 15. Burnell JM 1971 Changes in bone sodium and carbonate in metabolic acidosis and alkalosis in the dog. J Clin Invest 50: 327-33 1. 16. Bushinsky DA, Levi-Setti R, Coe FL 1986 Ion microprobe determination of bone surface elements: Effects of reduced medium pH. Am J Physiol 250:F1090-1097. 17. Bailey RR 1985 Chronic acidosis with metabolic bone disease. NZ Med J 98:483-484. 18. Bushinsky DA, Goldring JM, Coe FL 1985 Cellular contribution to pH mediated calcium flux in neonatal mouse calvariae. Am J Physiol 248:F785-789. 19. Kraut JA, Mishler DR, Kurokawa K 1984 Effect of colchicine and calcitonin on the calcemic response to metabolic acidosis. Kidney Int 29608-612. 20. Kunkel ME, Roughead ZK, Nichter EA, Navia JM 1986 The effects of dietary acid stress on bone metabolism in young ovarietomized and intact rats. Br J Nutr 5579-86. 21. Chan Y-L, Savdie E, Mason RS, Posen S 1985 The effect of metabolic acidosis on vitamin D metabolites and bone histology in uremic rats. Calcif Tissue Int 37:158-164. 22. Cunningham J, Fraher LJ, Clemens TL, Revell PA, Papapoulos SE 1982 Chronic acidosis with metabolic bone disease. Am J Med 73:199-204. 23. Kraut JA, Misher DR, Kurokawa K 1984 Effect of metabolic acidosis on bone resorption and responsiveness to PTH. In: Cohn DV, Fujita T, Potts JT, Talmadge RV (eds) Endocrine Control of Bone and Calcium Metabolism. Elsevier Science, Amsterdam, pp 272-275. 24. Wachman A, Bernstein DS 1970 Parathyroid hormone in metabolic acidosis. Clin Orthop Relat Res 69:252-263. 25. Barzel US 1969 The effect of excessive acid feeding on bone. Calcif Tissue Res 4:94-100. 26. Barzel U S 1975 Studies in osteoporosis: The long term effect of oophorectomy and of NH,CI ingestion on the bone of mature rats. Endocrinology %:1304-1306. 27. Phelps KR, Einhorn TA, Vigorita VJ, Lieberman RL, Uribarri J 1986 Acidosis-induced osteomalacia: Metabolic studies and skeletal histomorphometry. Bone 7:171-179. 28. Goulding A, Campbell DR 1984 Hypocalciuric effects of hydrochlorothiazide in the rat during NaHCO,, NaCI, and NH,CI loading. Renal Physiol 7:185-191. 29. Sutton RAL, Walker VR 1986 Bone resorption and hypercalciuria in calcium stoneformers. Metabolism 35485-488. 30. Lewis LD, Morris ML, Hand MS 1987 Feline urologic syndrome (FUS). In: Small Animal Clinical Nutrition 111. Mark Morris Associates, Topeka, Kansas, Chap. 9, pp 2-32. 31. Taton GF, Hamar DW, Lewis LD 1984 Evaluation of ammonium chloride as a urinary acidifier in the cat. J Am Vet Med Assoc 184:433. 32. Taton GF, Hamar DW, Lewis LD 1984 Urinary acidification in the prevention and treatment of feline struvite urolithiasis.

CHING ET AL.

556 J Am Vet Med Assoc 184:437. 33. Ching SV, Fettman MJ, Hamar DW, Nagode LA, Smith KR 1989 The effect of chronic dietary acidification using ammonium chloride on acid-base and mineral metabolism in the adult cat. J Nutr 119902-915. 34. Fettman MJ, Allen TA, Wilke WL, Radin MJ, Eubank MC 1985 Single-injection method for evaluation of renal function with ''C-inulin and 3H-tetraethylammonium bromide in dogs and cats. Am J Vet Res 46:482-485. 35. Ching SV, Norrdin RW 1990 Histomorphometric comparison of measurements of trabecular bone remodeling in iliac crest biopsy sites and lumbar vertebrae in cats. Am J Vet Res 51:447-450. 36. Merz WA, Schenk RK 1970 Quantitative structural analysis of human cancellous bone. Acta Anat (Basel) 7554-66. 37. Parfitt AM 1983 Stereologic basis of bone histomorphometry; theory of quantitative microscopy and reconstruction of the third dimension. In: Recker RR (ed) Bone Histomorphometry: Techniques and Interpretation. CRC Press, Boca Raton, FL, pp 54-85. 38. 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 2595-610. 39. Cann CE, Genant HK 1980 Precise measurement of vertebral mineral content using computer tomography. J Comput Assist Tomogr 4:493-500. 40. Genant HK, Cann CE, Ettinger B, Gordan GS 1982 Quantitative computed tomography of vertebral spongiosa: A sensitive method for detecting early bone loss after oophorectorny. Ann Intern Med 97:699-705. 41. Genant HK, Turski PA, Moss AA 1983 Advances in CT assessment of metabolic and endocrine disorders. In: Stollerman GH (ed) Advances in Internal Medicine, Vol. 28. Year Book Medical Publishers, Chicago, pp 409-447. 42. Cann CE, Genant HK, Young DR 1980 Comparison of vertebral and peripheral mineral losses in disuse osteoporosis in monkeys. Radiology 134525-529. 43. Einarsson S 1985 Selective determination of secondary amino acids using precolumn derivatization with 9-fluorenylmethylchloroformate and reversed-phase high-performance liquid chromatography. J Chromatogr 348:213-220. 44. Nilsson BE, Wiklund PE 1983 Iliac crest biopsy in the diag-

nosis of metabolic bone disease. Acta Med Scand 213:151-

155. 45. Frost HM 1969 Tetracycline-based histological analysis of bone remodeling. Calcif Tissue Res 3:211-237. 46. Genant HK 1987 Quantitative computed tomography: Update 1987. Calcif Tissue Int 41:179-186. 47. Cann CE, Genant HK, Ettinger B, Gordan GS 1980 Spinal mineral loss in oophorectomized women. JAMA 244:20562059. 48. Ching SV 1988 The effect of chronic dietary acidification on acid-base, mineral and bone metabolism in adult cats. Ph.D. Dissertation, Colorado State University, Fort Collins. 49. Cunningham J, Avioli LV 1982 Effects of systemic pH on calcium regulating hormones and bone. Adv Exp Med Biol 151:333-339. 50. Dalen N, Jacobson B 1974 Bone mineral assay: Choice of measuring sites. Invest Radio1 9174. 51. Pacifici R, Susman N, Carr PL, Birge SJ, Avioli LV 1987 Single and dual energy tomographic analysis of spinal trabecular bone: A comparative study in normal and osteoporotic women. J Clin Endocrinol Metab 64:209-214. 52. Biltz RM, Pelligrino ED, Letteri JM 1981 Skeletal carbonates and acid base regulation. Min Electrolyte Metab 5:l-7. 53. Neuman MW 1982 Blood:bone equilibrium. Calcif Tissue Int 34: I 17- 120. 54. Parfitt AM 1987 Bone and plasma calcium homeostasis. Bone 8(Suppl. l):Sl-S8. 5 5 . Bushinsky DA, Lechleider RJ 1987 Mechanism of proton-induced bone calcium release: Calcium carbonate dissolution. Am J Physiol 253:F998-1005. 56. Neuman MW, Imae K , Kawase T, Saito S 1987 The calciumbuffering phase of bone mineral: Some clues to its form and formation. J Bone Min Res 2:171-181.

Address reprint requests to: Dr. Martin J . Fettman Department of Pathology Colorado State University Fort Coffins,C O 80523 Received for publication June 7, 1989; in revised form December 18, 1989; accepted January 9, 1990.