Testicular toxicity of boric acid (BA): Relationship of dose to lesion development and recovery in the F344 rat

Reproductive Toxicology, Vol. 7, pp. 305-319, 1993 0890-6238/93 $6.00 + .00 Printed in the U.S.A. All rights reserved. Copyright © 1993 Pergamon Press Ltd.

• Original Contribution

TESTICULAR TOXICITY OF BORIC ACID (BA): RELATIONSHIP OF DOSE TO LESION DEVELOPMENT A N D

RECOVERY IN THE F344 RAT

W A R R E N W . Ku,* ROBERT E. C H A P I N , * R O B E R T N. W I N E , * and B E T H C, GLADEN~"

*Developmental and Reproductive Toxicology Group, National Toxicology Program, and ¢Statistics and Biomathematics Branch, National Institute of Environmental Health Sciences;

Research Triangle Park, North Carolina

AbstractBHigh-dose boric acid (BA) produces testicular lesions in adult rats, characterized by inhibited spermiation followed by atrophy. The present study addressed whether inhibited spermiation can be separated from atrophy based on dose, compared testis boron (B) dosimetry to lesion development, determined how inhibited spermiation was reflected by common reproductive endpoints, and examined reversibility of the testicular lesions. Rats were fed 3000, 4500, 6000, or 9000 ppm BA for up to 9 weeks and examined. Recovery was assessed for up to 32 weeks post treatment. Inhibited spermiation could be separated from atrophy based on dose (inhibited spermiation: 3000/4500 ppm; atrophy: 6000/9000 ppm), with each lesion aspect expressed at different threshold testis B concentrations (inhibited spermiation: 5.6/xg B/g and atrophy: 11.9 pg B/g) with no B accumulation during the 9-week exposure. These data suggest that separate mechanisms may be operating for these lesion aspects based on testis B concentration and that B dose rate was important for testicular toxicity. Inhibited spermiation was most reliably reflected by informed testicular histology, with the more severe cases decreasing epididymal sperm count to levels that could affect fertility. After treatment, serum and testis B levels in all dose groups rapidly fell to background levels at the earliest time points evaluated (7 days and 8 weeks posttreatment, respectively). The severely inhibited spermiation at 4500 ppm was resolved by 16 weeks posttreatment, but areas of focal atrophy were detected that did not recover posttreatment. Also, no signs of recovery from atrophy were observed (6000 and 9000 ppm). Atrophic tubules contained a normal complement of spermatogonia (2.6 to 2.9 germ cells/100 Sertoli cells), with occasional dividing and degenerating germ cells. Elevations in serum FSH and LH levels suggested an intact hormonal response to the atrophy. In summary, 1) the different aspects of the BA-induced testicular lesion can be separated using different doses, 2) inhibited spermiation does not necessarily proceed to atrophy, and 3) there is no recovery from the atrophy despite the absence of testis B after treatment. The ability to separate inhibited spermiation from atrophy based on dose and testis B dosimetry will be useful in evaluating possible mechanisms. Furthermore, the presence of dividing spermatogonia during long-term BA-induced atrophy suggests that this model should be useful for identifying critical components involved in the reinitiation of spermatogenesis.

Key Words: boric acid; boron; testis; toxicity; rats.

INTRODUCTION

Boric acid (BA) (H3BO 3) is a water-soluble inorganic acid with widespread commercia l use. Thus, the potential exists for both consumer and occupational exposure and environmental contaminat ion by boron (B) (1-3). The extent and nature of the environmental health impact of B in humans has not been

Address correspondence to Warren W. Ku, National Insti- tute of Environmental Health Sciences (NIEHS), P.O. Box 12233 MD A2-02, Research Triangle Park, NC 27709.

Received 28 December 1992; Accepted 22 February 1993.

305

adequately studied (reviewed in 4), especially with respect to chronic low-level exposure . Limited studies in the Soviet Union suggested a relationship between high B levels in drinking water and an increased incidence of male infertility (5). Further- more, dis turbances in reproduct ive function and se- men quality have also been suggested in male work- ers involved in BA product ion (6).

In animals, test icular a t rophy was observed in dogs, rats, and mice on chronic dietary borax or BA at 1000 to 2000 ppm B equivalents (7-9). The National Toxicology Program ' s Reproduct ive As-

306 Reproductive Toxicology Volume 7, Number 4, 1993

sessment by Continuous Breeding study in mice exposed to BA showed multiple sites of action, with male fertility the most sensitive (10). The testicular lesion in adult rats fed 9000 ppm BA was characterized by an initial inhibition of spermiation followed by epithelial disorganization, germ cell loss, and atrophy (11).

The mechanism for the testicular toxicity of BA is unknown. One mechanism could be decreased testosterone (T). Rats fed BA exhibited slightly reduced basal serum T levels; this reduction appears to be CNS-mediated (11,12). However, it is unlikely that hormone changes can explain the atrophy, since it has been shown that spermatogenesis can be maintained in the presence of significantly reduced in- tratesticular T (13,14). Data on B tissue disposition suggested that neither the testicular toxicity nor the slight CNS hormonal effect can be explained on the basis of selective accumulation of B in the testis or brain/hypothalamus, respectively (15). Thus, these data suggest that other possible mechanisms should be considered.

The central tissue for any toxicant is to define its mechanism of action. However, a number of broader questions need to be addressed to aid in formulating a mechanistic hypothesis for the testicular toxicity of BA.

The previous approach used only a high dose to describe the BA testicular lesion (11). It is unknown if either or both lesion characteristics, the initial inhibited spermiation and atrophy, are exclusively a high-dose effect. The present study addressed whether inhibited spermiation can be separated from atrophy based on BA dose by characterizing and comparing testis lesion development in rats fed various doses of BA for up to 9 weeks. The ability to separate inhibited spermiation from atrophy by dose might suggest that separate mechanisms are responsible for these different lesions. This information would be essential for de- signing further mechanistic studies.

The relationship between testis B concentration and the extent and severity of testis lesion development should be defined. Therefore, we addressed whether inhibited spermiation and atrophy are expressed at different testis B concentrations during the 9-week dosing period. Furthermore, these testis dosimetry data will be useful for setting relevant B concentrations to use in in vitro mechanistic studies. In addition, these data should determine if total BA dose or dose rate is important for testis lesion development.

Histologically, BA-induced inhibition of spermiation is characterized by varying degrees of

retention of step 19 spermatids in stage IX-XI tubules. We addressed the following questions: 1) To what extent can inhibited spermiation be reflected by an increase in either testis weight and/or testicular spermatid head count (TSHC)? and 2) How is inhibited spermiation reflected by epididymis weight and epididymal sperm count (ESC)?

Lastly, the recovery from BA-testicular lesions was evaluated for four complete cycles of spermatogenesis (32 weeks) following termination of exposure. The absence of recovery would suggest that BA exerts some permanent effects on the testis. To assess B "washout ," urinary, serum, and testis B levels were determined during the posttreatment period. Serum FSH and LH levels were also measured to determine if the gonadotropin response to the atrophy was intact.

MATERIALS A N D METHODS

Animals Adult male Fischer 344 (CDF (F344)/CrlBr®)

rats (60 to 70 days old, 200 to 220 g) were obtained from Charles River Breeding Laboratories (Raleigh, NC) and acclimated for 10 days to the NIEHS animal facility. Animals were housed three per polycarbonate cage with 12:12 h light/dark cycles, 50% -+ 10% humidity, and an ambient temperature of 20 + 1 °C.

Boric acid (BA) exposure Three hundred and sixty six rats were com-

puter-randomized by body weight and assigned to control (n = 6/group/week for 9 weeks) and 4 treatment groups (n = 6/treatment group/week). After a further 7-day acclimation to powdered NIH-07 certi- fied feed (Zeigler Bros., Inc., Gardners, PA), the treatment groups were fed powdered NIH-07 certi- fied feed containing BA (99.99%, Aldrich Chemical Company, Inc. Milwaukee, WI) at 3000 ppm (545 ppm B), 4500 ppm (788 ppm B), 6000 ppm (1050 ppm B), and 9000 ppm (1575 ppm B) (w/w) ad libitum for a period of up to 9 weeks. BA concentrations in the dosed feed were 96% to 110% of the target levels over the course of the study (Research Triangle In- stitute, Research Triangle Park, NC). The control groups received powdered feed with no additives (B levels <20 ppm, G. Rao, NTP, personal commu- nication). Deionized water was provided ad libitum. To estimate daily B intake, feed consumption was monitored gravimetrically during both weeks 6 and 7 of exposure. Feed spillage was negligible. A technical error in feeding was discovered at week 2: control and 3000 ppm BA feed were switched. Thus, during

Testicular toxicity of boric acid in rats • W. W. Ku ET AL. 307

week 2 only, both groups received dose feed some- where between 0 and 3000 ppm BA. This was veri- fied by B analysis of dose feed. Nonetheless, since no significant changes occurred at this time point for the low dose, the data were included for control and 3000 ppm at week 2. A pair-fed control group was not included, since a previous BA study showed that the testis histology of pair-fed rats was indistin- guishable from ad libitum controls (11), and the effects of body weight reduction by 30% on reproductive function in male rats are negligible (16).

At weekly intervals for 9 weeks, 6 rats from each group (control and 4 dose groups) were examined. Rats were weighed and briefly anesthetized with CO2, blood was collected by cardiac puncture, and then they were euthanatized by CO2 asphyxia- tion. The following tissues were removed: left testis for histology; right testis for weight and then for subdivision for B analysis and testicular spermatid head count (TSHC); right epididymis for weight and epididymal sperm count (ESC). Serum was separated from clotted whole blood, and sera and tissues were stored at -70 °C prior to processing and analysis.

Recovery Rats in control and 4500, 6000, and 9000 ppm

BA dose groups (n = 96, above) were placed on control NIH-31 pelleted feed after 9 weeks of exposure, and recovery was assessed at 8-week intervals for up to 32 weeks post treatment. Rats were given NIH-31 pelleted feed during the post-treatment period to avoid dental malocclusion problems. At each 8-week posttreatment interval, 6 rats from each group (control and 3 dose groups) were examined as described above.

Testis histology The left testis was fixed in 4% buffered para-

formaldehyde. Cut 2- to 3-mm transverse sections of testis were rinsed with phosphate-buffered saline, dehydrated through a graded series of ethanol, and embedded in glycol methacrylate (JB-4® Plus Em- bedding Kit, Polysciences, Inc., Warrington, PA). Sections (2- to 3-/zm) were cut and stained with peri- odic acid Schiff's reagent (PAS)/hematoxylin (Har- ris type). Seminiferous tubules were staged (17). A total of 200 to 300 tubules per animal was examined for lesions and recovery unblinded. The number of spermatogonia per 100 Sertoli cells was determined in atrophic tubules by counting a minimum of 1000 Sertoli cell nuclei and using the morphologic criteria for spermatogonia according to Clermont and Bustos-Obregon (18).

Quantitative sperm parameters (TSHC/ESC) Preweighed frozen samples of testis and cauda

epididymis were homogenized in ice-cold deionized water (dH20) (testis: 40% w/v; cauda: 4% w/v), di- luted 1:10 with 20% dimethyl sulfoxide (v/v in dH20), and mixed. Prefiltered trypan blue (0.4% w/v in dH20 ) was added, and spermatid or sperm heads were counted in a hemocytometer chamber.

Boron (B) analysis Serum, urine, and testis samples were prepared

for B analysis using the microwave acid digestion procedure (19). B levels were measured by induc- tively coupled plasma emission spectrometry (Re- search Triangle Institute, Research Triangle Park, NC) using appropriate matrix standard curves (cor- relation coefficients = 0.9999). The estimated detection limits for B in serum and testis were <4/.~g/ mL and 0.4 ~g/g wet weight, respectively. B recov- eries for all samples were greater than 90%.

B "Washout" Rats from control, 4500, 6000, and 9000 ppm

BA dose groups (total n = 24; n = 6/group) were housed individually in polycarbonate metabolic cages (Nalgene, Rochester, NY) and urine was collected daily for 14 days immediately following the last day of dosing. Pelleted NIH-31 food and water were provided ad libitum. On days 7 and 14, the animals were anesthetized with 70% CO2: 30% 02, blood was collected by retro-orbital sinus puncture and serum separated. Urine and sera were analyzed for B and urinary creatinine. Control urines from days 1, 7, and 14 days post treatment were analyzed for B. Urinary B concentrations were normalized to urinary creatinine to compensate for differences in urine volume. After the 14-day period, rats were returned to normal caging for further assessment of recovery.

Serum FSH/LH Serum LH and FSH were measured by double-

antibody radioimmunoassay (RIA). Purified LH and FSH standards and antisera were supplied by the Rat Pituitary Hormone Distribution Program, NIAMD. Rat LH and FSH were iodinated using the method of Greenwood and colleagues (20). The intra- and interassay coefficients of variation for LH were 4.8% and 14.5%, respectively. For FSH, the values were 8.2% and 5.5%, respectively.

Statistics For most endpoints, analysis of variance was

used to assess the effects of week, dose, and their


interaction. For terminal body weight, analysis of covariance was used instead; prestudy body weight was added as a covariate while studying the effects of week, dose, and their interaction. For both the analyses of variance and of covariance, F tests were used to assess overall effects, and t tests with pooled error terms were used to compare each dosed group to the control for that week.

Curves were also fit to the body and organ weight data to describe the changes over time. Body weight was considered to be a function of prestudy body weight as well as time and dose. Testis and epididymis weights were considered to be functions of only time and dose. Doses that did not differ significantly or differed only trivially were combined for presentation purposes. For serum gonadotropin levels, the data were analyzed by Dunnett's multiple comparison test. Differences were considered significant at P < 0.05.

RESULTS

Feed consumption and estimated daily boron (B) intake

An immediate and lasting decrease in body weight gain for the 9000 ppm dose group was observed (data not shown), and by week 9 this group weighed approximately 16% less than controls (controls = 323 ± 6 [SD] g; 9000 ppm = 270 ± 5 g). No changes in body weight gain were observed for the other dose groups. Mean (± SD) feed consumptions estimated during weeks 6 and 7 were 49.3 - 1.0, 50.2 ± 0.3, 49.2 ± 2.6, 49.2 -+ 1.6, and 44.0 ± 2.1 g/kg body weight/day for 0 (Control), 3000, 4500, 6000, and 9000 ppm BA dose groups, respectively, with the 9000 ppm dose group consuming approximately 11% less feed than controls. Based on feed consumption and body weights, the estimated daily intakes of B were <0.2, 26, 38, 52, and 68 mg B/kg body weight/day for control and the respective dose groups.

Testis lesion development Testes from control rats showed stages with

the normal cell associations consistent with intact spermatogenesis (reference 17; Figures 1A and 2A represent stages IX/X and XI, respectively). Rats fed various BA doses during the 9-week period de- veloped testicular lesions ranging from mildly inhibited spermiation to varying degrees of degenerative changes leading to atrophy. Atrophy was defined as the complete absence of postspermatogonial germ cells. The inhibited spermiation was characterized by the aberrant retention of step 19 spermatids in

stage 1X/X tubules (Figure IB), which, in more severe cases, involved stage XI/XII tubules (Figure 2B). Varying degrees of degenerative changes, ranging from epithelial disorganization and germ cell loss to eventual atrophy were also observed (Figures 3A-3C). Histologically intact spermatogonia (Fig- ure 3C inset) were present in atrophic tubules at a frequency of 2.6 to 2.9 spermatogonia per 100 Sertoli cell nuclei. The Leydig cells appeared histologically normal at the light microscopic level.

To assess testis lesion development over time for each dose group, lesions were assigned a numeric score between 0 and 6, depending on both the lesion characteristics and percentage of tubules affected (Table 1) and plotted (Figure 4):

Rats fed 3000 ppm BA showed only mildly inhibited spermiation (Grade 1) by week 5, which continued variably to week 9. Rats fed 4500 ppm BA showed severe and widespread inhibition of spermiation (Grade 2) by week 2 that was maintained out to week 9. At week 9, germ cell exfoliation (<5% of tubules) was also observed in this dose group, but overall, the lesion was relatively mild and limited. Rats fed 6000 ppm and 9000 ppm BA showed initially severe inhibition of spermiation by week 2, followed by progression to Grade 6 atrophy. The progression to atrophy was dose- and time-dependent, with 6000 ppm reaching atrophy by week 9 and 9000 ppm by week 6. This difference was due to the sustained presence of postspermatogonial germ cells in tubules of the 6000 ppm dose group between weeks 6 and 9 (Figure 3B).

Testis~Serum B concentration during lesion development

Control testis and serum B levels were at or near the limits of detection (0.4/~g/g testis; <4 t~g/ mL serum) throughout the 9-week period. Aside from the dosing error at week 2 (see Materials and Methods), there was a consistent relationship between BA dose and testis B levels (Figure 5). Mean (±SD) testis B levels over the 9-week period were 5.6 -+ 0.8 (minus week 2 data), 8.8 ± 0.7, 11.9 -+ 1.4, and 15.1 ± 1.9/~g/g for 3000, 4500, 6000, and 9000 ppm BA, respectively. Mean (± SD) serum B levels (weeks 1, 4, and 9) were 6.7 ± 1.0, 10.3 -+ 0.6, 13.3 ± 0.7, and 17.3 + 2.2 ~g/mL for these doses (data not shown). There was no B accumulation in testis over serum levels throughout the 9-week period, with mean testis/plasma ratios of less than one for all doses.

The relationship between estimated B intake, testis B levels, and the characteristics of testis lesions over the 9-week period is summarized in Table

Fig.

1. I

nhib

ited

spe

rmia

tion

in

stag

e IX

/X s

emin

ifer

ous

tubu

les

duri

ng B

A e

xpo-

su

re.

A:

Con

trol

lat

e st

age

IX/e

arly

X t

ubul

e; B

: in

hibi

ted

sper

mia

tion

in

stag

e X

tu

bule

(90

00 p

pm B

A,

wee

k 2)

. N

ote

the

pres

ence

of

step

19

sper

mat

ids

(arr

ow-

head

s).

PA

S/h

emat

oxvl

in.

Bar

= 5

0 ~

m.

Fig

. 2.

In

hibi

ted

sper

mia

tion

in

stag

e X

I/X

II s

emin

ifer

ous

tubu

les

duri

ng B

A

expo

sure

. A

" C

ontr

ol s

tage

XI

tubu

le;

B:

Ret

aine

d st

ep 1

9 sp

erm

atid

s pr

esen

t in

st

age

XI/

XII

tubu

les

(900

0 pp

m B

A, w

eek

2). N

ote

basa

lly

loca

ted

step

19

sper

ma-

ti

ds (

arro

whe

ads)

. P

AS

/hem

atox

ylin

. B

ar =

50/

xm.


Fig. 3. Degenerative changes in seminiferous tubules during BA exposure. A: Exfoliative changes with epithelial disorganization. Some tubules show no postspermatogonial germ cells (asterisk) (6000 ppm BA, week 6); B: Atrophic changes, with some of the tubules showing significant numbers of residual spermatocytes, but few or no mature germ cells (6000 ppm BA, week 8); C: Complete atrophy (6000 ppm BA, week 9); C/Inset: Atrophic tubules exhibit histologically intact spermatogonia (arrowheads). PAS/hematoxylin. Bar = 50/xm.

2. Testis B levels of 5 to 6 /xg/g were associated with mildly inhibited spermiation appearing by week 5, which variably continued up to week 9. Levels of 8 to 9/zg/g were associated with severely inhibited spermiation by week 2, which continued up to week 9. Levels of 11 to 12 /xg/g were associated with severely inhibited spermiation (week 2) followed by progression to atrophy by week 9, and 15 to

16 ~g/g with inhibited spermiation (week 2) followed by atrophy by week 6.

Comparison of reproductive toxicity indices during lesion development

Testis and epididymis weights, TSHC, and ESC were measured over the 9-week exposure period and are depicted together in Figure 6. Mildy inhibited

Testicular toxicity of boric acid in rats • W. W. Ku ET AL.

Table 1. Histologic grading scheme for BA testicular lesions

311

Lesion grade 0 1 2 3 4 5 6

% Tubules at stages below the inhibited spermiation

Stages with retained spermatids

% Tubules with germ cell exfoliation

% Atrophic tubules

<5 25-50 >50 >50

IX X,XI XI,XII

<5 10-25 >75 <25 <5 <25 >75 >95

spermiation (3000 ppm BA) was detected histologically, but with no consistent changes in any parameter. Severe and widespread inhibition of spermiation (4500 ppm BA) was reflected by variable increases in TSHC (24% to 62%, detected initially and most significantly at week 2), with no significant changes in testis weight. This was followed by decreases

in epididymis weight (10% to 29%) and profound decreases in ESC (72% to 97%) during weeks 4 to 9. Severely inhibited spermiation and progression to atrophy (6000 and 9000 ppm BA) were represented initially by increased TSHC (31% to 51%), reflecting the inhibited spermiation at week 2, followed by progressive and profound decreases in testis weight

0 i m

0 , m

| m

m m

0 c- o

m m

, J

¢ -

2 5

e -

o

c -

-o._o ( D " - '

.-Q-E 1

- - -0- Control

--O-- 3000 ppm

4500 ppm

- - m - 6000 ppm

--D-- 9000 ppm

Testis Histopathology

0 1 2 3 4 5 6 7 8 9

Week

Fig. 4. Test is lesion deve lopment in BA dose groups ove r time. Sect ions of testis were examined for lesions, graded, and plotted. Values are the mean lesion score -- SE for 6 rats/group/week.


2 0 -

.~ 1 5 -

G) I - - 0 ' }

t - 1 0 - 0

0 a n

• "~ 5 -

0

•TTTTTk• ~"

I I I I 'i I '1 I 1 2 3 4 5 6 7 8 9

Week

I "-0- Control 3000 .pr I 4500 ppm I 6000 ppm 9000 ppm

Fig. 5. Testis B concentrations in BA dose groups over time. Values are the mean/zg B/g testis -+ SE for 6 rats/group/ week. Control testis B levels over the 9-week period were at or near the limits of detection (0.4 gg/g). The significance levels for comparing each week's dose to that week's control were all P < 0.05. Week 2 data were not analyzed as a result of dose feed error.

(12% to 68%), TSHC (16% to 99%), epididymis weight (12% to 57%), and ESC (78% to 99%), reflecting the progression to atrophy during weeks 3 to 9.

Assessment of recovery Upon termination of BA exposure, urinary B

in all dose groups was at control levels (mean 8.97 -+ 1.95 [SD] /zg B/mg creatinine) by 3 to 4 days post treatment (Figure 7). Serum and testis B levels were at or near estimated detection limits (<4 txg/mL serum and 0.4 txg/g testis) at the earliest post treatment times evaluated (7 days and 8 weeks, respectively, data not shown).

Histologically, evidence of mildly inhibited

spermiation was still present in the 4500 ppm BA dose group at 8 weeks post treatment (2/6 rats) (Fig- ure 8A); this effect resolved by 16 weeks post treatment (6/6 rats) (Figure 8B). Despite this improve- ment, atrophic tubules, which showed few or no postspermatogonial cells, were also seen (Figure 9A). This focal atrophy showed considerable vari- ability in the percentage of tubules affected (range: 5% to 100%; mean --~ 10% of tubules), with no evidence of recovery up to 32 weeks post treatment. In rats previously exposed to 6000 and 9000 ppm BA, no signs of recovery from atrophy were observed up to 32 weeks post treatment (Figure 9B). Atrophic tubules displayed histologically viable spermatogonia (2.6 to 2.9 germ cells/100 Sertoli cells). Most were

Table 2. Relationship between estimated boron (B) intake, testis B levels, and characteristics of testicular lesions

ppm

BA dose

approximate mg B/kg/day a Testis B b Lesion characteristics

(/~g/g) (week first detected)

3000 26 5.6 -+ 0.8 Mild IS ~ (5) 4500 38 8.8 -+ 0.7 Severe 1S (2) 6000 52 11.9 -+ 1.4 IS (2)/atrophy (9) 9000 68 15.1 _+ 1.9 IS (2)/atrophy (6)

a Calculations based on average feed consumption and body weight data during weeks 6 and 7. b Mean --- SD for each dose over the 9-week exposure period. c 1S = inhibited spermiation.


1.6 - A. Testis Weight

I:~ 1 . 2 -

z: ~ ) 0.8

0 . 4 - -

' o . . . . . . . . . • Fit to 0 -4500 ppm "D . . . . . . D- . . . . . D . . . . . "D

Fit to 6000 ppm

Fit to 9000 ppm

o.6o - B. Epid idymis Weight

I ~ 0.45

J: .~ 0.30

0.15

- - - 4 1 - - - -a,-

' 0 . . . . . . . . . . . I~ . . . . .

- - Fit to 0 - 3000 ppm - - - Fit to 4500 - 6000 ppm

Fit to 9000 ppm

C. " " Head Count 2o0

~ . 150

~ 1o- W~,= u

"1'- 50

0

12- -

" ~ 9 -

F : " 6 L X

0

D. Epid idymal Sperm Count

1 2 3 4 5 6 7 8 9

W e e k

I - - 0 - Control - - 0 - - 3000 ppm + 4500 ppm --B.-- 6000 ppm ~ 9000 ppm I

Fig. 6. Changes in commonly used indices of male reproductive toxicity in BA dose groups over time. Values are the mean weight (in grams) or count --- SE for 6 rats/group/week. For testis (A) and epididymis (B) weight, curves were fit to describe the changes over time. Doses that did not differ significantly or differed only trivially were combined. Testicular spermatid head count (TSHC) (C) and epididymal sperm count (ESC) (D) are expressed as spermatid heads/ testis and sperm/cauda, respectively. Isolated significant differences compared to respective control (P < 0.05) are listed for TSHC and ESC. TSHC: 3000 ppm: weeks 1, 8; 4500 ppm: weeks 2, 3, 6, 7; 6000 ppm: weeks 1 to 3, 5 to 9; 9000 ppm: weeks 2 to 9. ESC: 3000 ppm: weeks 6, 7; 4500 ppm: weeks 2, 4, 6 to 9; 6000 ppm: weeks 4 to 9; 9000 ppm: weeks 1, 3 t o9 .


G) c

°m

e- om

t~

O)

E e- O t , .

O

O~

800

700

600

500

400

300

200

100

0

• Control

A 4500 ppm

• 6000 ppm

a 9000 ppm

i I I ~ i I I I

0 2 4 6 8 1 0 1 2 1 4

D a y s P o s t - T r e a t m e n t

Fig. 7. Urinary boron levels post treatment. Values are the mean ~g B/mg creatinine -+ SE for 6 rats/group/day for 14 days. Mean control urinary B level (data combined from 1, 7, and 14 days post treatment) was 8.97 -+ 1.95 (SD) tzg B/mg creatinine.

basally located, occurred singly or occasionally in pairs, and lacked significant heterochromatin, sug- gesting that they were of the Type A class (18). Dividing spermatogonia were also observed (Figure 10A), with the occasional appearance of clusters of four or more cells connected by intercellular bridges (Figures 10B,C). Occasionally, degenerating spermatogonia were also seen (Figure 10D).

Representative serum FSH and LH levels during the posttreatment period for 24 weeks post treatment are presented in Table 3. Serum FSH was significantly increased (1.4 to 1.9-fold) in all BA dose groups. Serum LH showed apparent increases (1.8

Table 3. Serum FSH and LH levels in control and BA dose groups at 24 weeks post treatment

BA dose group (ppm) (ng/mL a) (pg/mL a)

FSH LH

0 (Control) b 10.8 -+ 1.0 723 +- 125 4500 15.5 -+ 1.5" 1330 + 404 6000 20.7 + 1.5" !492 -+ 222 9000 20.8 -+ 1.0" 2378 -+ 722*

aMean_+ SEM, N = 6 b Controls are same age animals given control feed during the 9- week BA exposure period. * Significant difference from control, P < 0.05.

to 3.3-fold) in all BA dose groups evaluated during posttreatment, but was significant only for the 9000 ppm dose group.

DISCUSSION

The present study demonstrated that it is possible to separate the different aspects of the BA testicular lesion. By using different BA doses, inhibited spermiation was maintained over a 9-week period without progression to atrophy. Comparison of testis pathology in this dose/time study led to the following conclusions. First, the inhibited spermiation is not exclusively a high-dose effect, since this was the only lesion seen in the two lower dose groups (3000 and 4500 ppm). Secondly, atrophy was associated only with the two higher doses (6000 and 9000 ppm) during the 9-week period, and the progression to atrophy was dose- and time-dependent. Atrophic tubules displayed histologically intact spermatogonia, with the frequency of total spermatogonia being 2.6 to 2.9 germ cells per 100 Sertoli cell nuclei. This estimate is quite similar to the observations for 2,5- hexanedione (2,5 HD)-induced atrophy (21).

The relationship between testis B dose and lesion development was assessed. There was a relationship between dietary and testis B levels, with

Fig.

8.

Rec

over

y fr

om i

nhib

ited

spe

rmia

tion

in

the

4500

ppm

BA

dos

e gr

oup.

~_

: Sta

ge I

X t

ubul

e st

ill

show

ing

evid

ence

of

mil

dly

inhi

bite

d sp

erm

iati

on a

t 8

~ee

ks p

ost

trea

tmen

t (a

rrow

); B

: S

tage

IX

tub

ule

show

ing

no e

vide

nce

of in

hib-

te

d so

erm

iati

on a

t 16

wee

ks p

ost

trea

tmen

t. B

ar =

50

/xm

.

Fig

. 9.

Ap

pea

ran

ce a

nd p

ersi

sten

ce o

f at

roph

ic c

hang

es d

urin

g re

cov

ery

. A

: A

ppea

ranc

e of

foc

al a

trop

hy i

n th

e 45

00 p

pm B

A d

ose

grou

p at

8 w

eeks

pos

t L

reat

men

t; B

: P

ersi

sten

ce o

f to

tal

atro

phy

in t

he 6

000

ppm

BA

dos

e gr

oup

at

32 w

eeks

pos

t tr

eatm

ent.

Bar

=

50 p

~m.


Fig. 10. Characteristics of residual spermatogonia in atrophic tubules (6000 ppm BA dose group, 16 weeks post treatment). A: Dividing cells; B: Cluster of four cells connected by intercellular bridges (arrow); C: Larger cluster of connected cells (arrow); D: Degenerating spermatogonia (arrows). Dashed lines mark seminiferous tubule boundary, is = interstitium. Bar = 50 t~m.

no testis B accumulation over blood during the 9- week exposure period. These data are in agreement with, and further extend the results from, a previous 7-day B disposition study in which testis B appeared to reached steady-state levels by 3 to 4 days and showed no accumulation over plasma levels (15). The data also showed that the inhibited spermiation and atrophy were expressed at different testis B concentrations. A previous 9-week study identified 2000 ppm BA as the no-observed-adverse-effect level for testicular effects (unpublished data). Thus, the threshold testis B concentrations for inhibited spermiation and atrophy during the 9-week period

were approximately 5.6/xg B/g (3000 ppm BA) and 11.9/xg B/g (6000 ppm BA), respectively. Further- more, comparison of lesion development with testis B dose and time showed that dose rate, rather than total dose, is important for testicular toxicity. How- ever, there are limitations to this interpretation, due to the limited exposure period of 9 weeks. Lastly, the range of testis B concentrations achieved in this study was estimated. Assuming that the testis has a specific gravity of 1 g/cm 3 and is a homogeneous water-filled compartment, the calculated range of testis B concentrations was between 0.5 and 2 mM, with inhibited spermiation occurring at 0.5 to 1 mM,

Test icular toxicity of boric acid in rats • W. W. K u ET AL. 317

and atrophy between 1 and 2 mM B. This information will allow us to set relevant B concentrations for addressing mechanistic questions for these lesions in vitro.

Inhibited (or delayed) spermiation has been observed at various stages of lesion development for a number of testicular toxicants (22-28). However, few attempts have been made to assess the relationship between inhibited spermiation and changes in commonly used reproductive endpoints. Testis and epididymis weight, testicular spermatid head count (TSHC), and epididymal sperm count (ESC) were measured to assess their relative sensitivity in reflecting the BA-induced inhibition of spermiation. The results showed that, although epididymis weight and ESC could detect severely inhibited spermiation, of the testicular endpoints, only histology could identify the mildly inhibited spermiation at 3000 ppm BA. It is also worth noting how severely inhibited spermiation was depicted. Severe inhibition of spermiation in the 4500 ppm group was significant enough to decrease ESC to an extent similar to that observed for atrophy in the higher BA dose groups. Thus, although the histologic characteristics of inhibited spermiation and atrophy differ, both lesions produced profound decreases in ESC that could affect fertility. Overall, these inconsistencies reaffirm the importance of histologic evaluation and the need to assess multiple endpoints in characterizing the reproductive toxicity of a chemical (29-31).

Recovery from BA-induced inhibition of spermiation and from atrophy was also assessed for 32 weeks post treatment. Upon termination of BA exposure, urinary B in all dose groups was at control levels by 3 to 4 days posttreatment. Furthermore, serum and testis B levels in all dose groups were undetectable at the earliest time points evaluated (7 days and 8 weeks post treatment, respectively). This is consistent with observations in humans (32) and rabbits and guinea pigs (33), where nearly complete (>90%) excretion of BA was found within 48 to 72 h following intravenous infusion or oral administration. The absence of detectable serum and testis B after treatment is consistent with its anionic nature and known reactivity with biologic molecules, all of which are readily reversible (34).

Despite the efficient elimination of B, the reversibility of the testicular lesions varied. The inhibited spermiation seen in the 4500 ppm dose group was completely resolved by 16 weeks post treatment. This is consistent with data from Linder et al (35), which showed significant recovery from inhibited spermiation 43 days after its detection following

acute BA exposure in rats. Despite the resolution of inhibited spermiation in this dose group, atrophic changes were observed, showing considerable vari- ability in the percentage of tubules affected and no evidence of recovery up to 32 weeks posttreatment. It is reasonable to postulate that these atrophic changes seen posttreatment reflect further tubule degeneration resulting from effects initiated during the 9-week exposure. These atrophic changes in a previously nonatrophic dose group suggest either that the BA exposure period was insufficient (that is, a longer exposure would have eventually revealed atrophy in this group) or that a common mechanism is operating for both the inhibited spermiation and the atrophy. Future mechanistic studies should address this question.

No signs of recovery from atrophy were observed post treatment, with atrophic tubules dis- playing histologically viable spermatogonia and dividing cells. These characteristics are similar to the observations for 2,5-hexanedione-induced long- term testicular atrophy (21). Thus, although spermatogonia were present and showed evidence of occasional cell division, they failed to produce ade- quate regeneration and to progress. Increases in serum FSH and LH were also observed in rats dis- playing atrophy, consistent with previous findings (21,36). These data suggest that the gonadotropin response to the absence of germ cells was intact posttreatment.

The reasons for the lack of recovery from BA- induced testicular atrophy are unknown. It appears that the gonadotropin response to the atrophy is intact. However, changes in autocrine and paracrine interactions cannot be ruled out. The impairment of recovery by covalent interactions is unlikely, since B shows no appreciable tissue accumulation (15), testis B levels were nondetectable during the posttreatment period, and B interactions with biologic molecules are reversible (34).

Adverse BA effects on Sertoli cells could impair recovery. The putative target cell for the testicular toxicants 2,5-HD (37,38) and TOCP (39) is the Sertoli cell. Interestingly, the testicular atrophy produced by TOCP is irreversible (40), and that produced by 2,5-HD is only partially reversible (21). BA may also exert permanent effects on the residual spermatogonia. Atrophic tubules showed histologically viable spermatogonia of the Type A class with occasional dividing and degenerating cells, but at reduced numbers. It was clear that the spermatogonia were capable of dividing, but failed to progress. Future studies are planned to address whether the levels of B that


produce atrophy (1 to 2 mM) induce permanent func- tional alterations in Sertoli cells and/or the residual spermatogonia.

Lastly, a number of paracine factors are produced within the testis and are postulated to influ- ence spermatogonial proliferation and maturation (reviewed in references 41 and 42). Currently, no data are available on paracrine responses to testicular atrophy or on their role in the reinitiation of spermatogenesis. Future studies are planned to eval- uate paracrine interactions during long-term BA- induced testicular atrophy.

In summary, the present BA dose/time study showed that inhibited spermiation is not exclusively a high-dose effect and can be separated within the 9-week time frame from atrophy, with each lesion aspect expressed at different threshold testis B concentrations. This study further revealed that B dose rate was important for the BA testicular toxicity; that inhibited spermiation was most reliably reflected by informed testicular histology, with decreases in sperm count only detected for more severe cases; and that severely inhibited spermiation by itself was capable of decreasing sperm count to an extent that could affect fertility. The ability to separate inhibited spermiation and atrophy based on BA dose and the relationship between testis B dosimetry and lesion expression will be useful in evaluating possible mechanisms in vivo and in vitro. This is the subject of future investigations. The present study also showed that high-dose BA produces nonrecoverable testicular atrophy with histologically viable spermatogonia and an intact gonadotropin response--and in the absence of detectable testis B after treatment. Future studies will utilize this atrophy model to identify critical components involved in the reinitiation of spermatogenesis.

A c k n o w l e d g m e n t s - - Dose feed preparation and tissue boron de- terminations were performed by Research Triangle Institute (RTI) under Contract NO 1-ES- 15307. The authors extend a special thanks to Mike Veselica for dose feed preparation; Dr. Robert W. Handy, Delores Brine, and Mary E. Lansdown for boron analyses; Brad Collins for chemistry support; Dr. Julius E. Thig- pen for microbial analysis of feed; Catharina Weaver of the Radio- immunoassay Core Facility, University of North Carolina, Chapel Hill, for serum hormone assays; and Dr. Mary F. Goelz, Clark Colgrove, and AI Caviness for veterinary support and coor- dination of animal care. The authors greatly appreciate the expert technical assistance provided by Martha Harris, Janet Allen, Eric Haskins, James Clark, Page Myers, and Joyce Snipe. The authors also extend a very special thanks to Drs. Michael Brabec, Jerrold Heindel, Bernard Schwetz, Phil Strong, and Ralph Linder for their critical review of this manuscript.

REFERENCES

1. Waggot A. An investigation of the potential problem of in- creasing boron concentrations in rivers and water courses. Water Res. 1969;3:749-65.

2. Newnham RE. Boron: its types of compounds and their toxicity; a review. In: Brown SS, Kodama Y, eds. Toxicology of metals. Chichester: Ellis Horwood; 1987:71-2.

3. Larsen LA. Boron. In: Seller HG, Sigel H, eds. Handbook on toxicity of inorganic compounds. New York: Marcel Dekker, 1988:129-41.

4. Minoia C, Gregotti C, Di Nucci A, et al. Toxicology and health impact of environmental exposure to boron. A review. G Ital Med Lay. 1987;9:119-24.

5. Khachatryan TS, Nazarenko AF. Hygienic evaluation of boron in drinking water. Zh Eksp Kiln Med (USSR). 1973;13:29-34.

6. Tarasenko NY, Kasparov AA, Strongina OM. The effect of boric acid on the generative function of the male organism. Gig Tr Prof Zabol. 1972;16:13-6.

7. Weir RJ Jr, Fisher RS. Toxicologic studies on borax and boric acid. Toxicol Appl Pharmacol. 1972;23:351-64.

8. Lee IP, Sherins RJ, Dixon RL. Evidence for induction of germinal aplasia in male rats by environmental exposure to boron. Toxicol Appl Pharmacol. 1978;45:577-90.

9. National Toxicology Program Report 1987. Toxicology and carcinogenesis studies of boric acid in B6C3F1 mice. Techni- cal Report 324, NIH Publ. no.: 88-2580. NTIS no. PB88213475. Available from: National Technical Information Service, Springfield, VA 22161.

10. Fail PA, George JD, Seely JC, et al. Reproductive toxicity of boric acid in Swiss (CD-1) mice: assessment using the continuous breeding protocol. Fundam Appl Toxicol. 1991 ;17:225-39.

11. Treinen KA, Chapin RE. Development of testicular lesions in F344 rats after treatment with boric acid. Toxicol Appl Pharmacol. 1991 ;107:325-35.

12. Anderson SA, Sauls HR, Pearce SW, et al. Endocrine responses after boric acid exposure for 2, 9, or 14 days in cannulated male CD rats. Biol Reprod. 1992;46(Suppl):124.

13. Rommerts FFG. How much androgen is required for maintenance of spermatogenesis? J Endocrinol. 1988;116:7-9.

14. Zirkin BR, Santulli R, Awoniyi CA, et al. Maintenance of advanced spermatogenic cells in the adult rat testis: quantitative relationship to testosterone concentration within the testis. Endocrinology. 1989;124;3043-9.

15. Ku WW, Chapin RE, Moseman RF, et al. Tissue disposition of boron in male Fischer rats. Toxicol Appl Pharmacol. 1991;111:145-51.

16. Chapin RE, Gulati DK, Barnes LH, Teague JL. The effects of feed restriction on reproductive function in Sprague-Dawley rats. Fundam Appl Toxicol. [In press].

17. Leblond CP, Clermont Y. Definition of the stages of the cycle of the seminiferous epithelium in the rat. Ann NY Acad Sci. 1952;55:548-73.

18. Clermont Y, Bustos-Obregon E. Re-examination ofsperma- togonial renewal in the rat by means of seminiferous tubules mounted "in toto". Am J Anat. 1968;122:237-47.

19. Moseman RF, Brink RE, Jameson CW, et al. Method development and validation for the determination of ppm levels of boron in rat tissues. Toxicologist. 1991;11:278.

20. Greenwood FC, Hunter FW, Glover JS. The preparation of 131I-labelled human growth hormone of high specific radioac- tivity. Biochem J. 1963;89:114-23.

21. Boekelheide K, Hall SJ. 2,5-Hexanedione exposure in the rat results in long-term testicular atrophy despite the presence of residual spermatogonia. J Androl. 1991 ;12:18-26.

22. Chapin RE, White RD, Morgan KT, et al. Studies of lesions induced in the testis and epididymis of F-344 rats by inhaled methyl chloride. Toxicol Appl Pharmacol. 1984;76:328-43.

23. Chapin RE, Dutton SL, Ross MD, et al. Development of reproductive tract lesions in male F344 rats after treatment with dimethyl methylphosphonate. Exp Mol Pathol. 1984;41:126-40.

24. Chapin RE, Dutton SL, Ross MD, et al. The effects of ethyl- ene glycol monomethyl ether on testicular histology in F344 rats. J Androl. 1984;5:369-80.


25. Anderson RA, Berryman SH, Phillips JF, Biochemical and structural evidence for ethanol-induced impairment of testicular development: Apparent lack of Leydig cell involvement. Toxicol Appl Pharmacol. 1989;100:62-85.

26. Parvinen L-M. Early effects of procarbazine (N-isopropyl- L-(2-methylhydrazino)-p-toluamide hydrochloride) on rat spermatogenesis. Exp Mol Pathol. 1979;30:1-11.

27. Hess RA, Linder RE, Strader LF, et al. Acute effects and long-term sequelae of 1,3-dinitrobenzene on male reproduc- tion in the rat; 2: quantitative and qualitative histopathology of the testis. J Androl. 1988;9:327-42.

28. Creasy DM, Ford GR, Gray TJB. The morphogenesis of cyclohexylamine-induced testicular atrophy in the rat: in vivo and in vitro studies. Exp Mol Pathol. 1990;52:155-69.

29. Perreault SD, Linder RE, Strader LF, et al. The value of multiple endpoint data in male reproductive toxicology: reve- lations in the rat. In: Burger EJ Jr, Tardiff RG, Scialli AR, Zenick H, eds. Sperm measures and reproductive success. New York: Alan R. Liss; 1989:179-92.

30. Working PK, Chellman GJ. The use of multiple endpoints to define the mechanism of action of reproductive toxicants and germ cell mutagens. In: Burger EJ Jr, Tardiff RG, Scialli AR, Zenick H, eds. Sperm measures and reproductive success. New York: Alan R. Liss, 1989:211-27.

31. Blazak WF, Significance of cellular end points in assessment of male reproductive toxicity. In: Working PK, ed. Toxicol- ogy of the male and female reproductive systems. New York: Hemisphere Publishing; 1989:157-72.

32. Jansen JA, Andersen J, Schou JS. Boric acid single dose pharmacokinetics after intravenous administration to man. Arch Toxicol. 1984;55:64-7.

33. Akagi M, Misawa T, Kaneshima H. Studies on the metabo-

lism of borate; 1: excretion of the boron into the urine and distribution in some organs. Yakugaku Zasshi. 1962;82:934-6.

34. Zittle CA. Reaction of borate with substances of biological interest. Adv Enzymol. 1951 ; 12:493-527.

35. Linder RE, Strader LF, Rehnberg GL. Effect of acute exposure to boric acid on the male reproductive system of the rat. J Toxicol Environ Health. 1990;31:133-46.

36. Rich KA, De Kretser DM. Effect of differing degrees of destruction of the rat seminiferous epithelium on levels of serum follicle stimulating hormone and androgen binding pro- tein. Endocrinology. 1977;101:959-68.

37. Chapin RE, Morgan KT, Bus JS. The morphogenesis of testicular degeneration induced in rats by orally administered 2,5-hexanedione. Exp Mol Pathol. 1983;38:149-69.

38. Boekelheide K. Rat testis during 2,5-hexanedione intoxi- cation and recovery; 1: dose response and the reversibility of germ cell loss. Toxicol Appl Pharmacol. 1988;92: 18-27.

39. Somkuti SG, Lapadula DM, Chapin RE, et al. Light and electron microscopic evidence of tri-o-cresyl phosphate (TOCP)-mediated testicular toxicity in Fischer 344 rats. Tox- icol Appl Pharmacol. 1991;107:35-46.

40. Somkuti SG, Lapadula DM, Chapin RE, et al. Time course of the tri-o-cresyl phosphate-induced testicular lesion in F- 344 rats: enzymatic, hormonal, and sperm parameter studies. Toxicol Appl Pharmacol. 1987;89:64-72.

41. Heindel J J, Treinen KA. Physiology of the male reproductive system: endocrine, paracrine and autocrine regulation. Tox- icol Pathol. 1989;17:411-45.

42. Skinner MK. Cell-cell interactions in the testis. Endocrine Rev. 1991;12:45-77.

Documents

Testicular toxicity of boric acid (BA): Relationship of dose to lesion development and recovery in the F344 rat