Low Molecular Weight Silicones are Widely Distributed

[Guest guest]

American J of Pathology, Vol. 152 #3: 645-649

Short Communication Low Molecular Weight Silicones are

Widely Distributed after a Single Subcutaneous

Injection in Mice

Subbarao V. Kala* Ernest D. Lykissa* W. Neely*

and W. Lieberman*+

From the Depts of Pathology* and Cell Biology+. Baylor

College of Medicine, Houston Texas

To examine the distribution of low molecular weight

silicone sin body organs, separate groups of female

CD-1 mice were injected with either breast implant

distillate composed primarily of

hexamethylcyclotrisiloxane,

decamethylcyclotetrasiloxane,

decamethylcyclopentasiloxane, and

tetradecamethylecycloheptasiloxane or a

polydimethylsiloxane oil containing low molecular

weight linear siloxanes. Mice were injected

subcutaneously in the suprascapular area and killed at

different times. Levels of individual low molecular

weight silicones were measured in 10 different organs

(brain, heart, kidney, liver, lung, mesenteric lymph

nodes, ovaries, spleen, skeletal muscle and uterus).

In mice treated with the cyclosiloxane mixture and

killed at 3, 6, or 9 weeks, Highest levels of

cyclosiloxanes were found in the mesenteric lymph

nodes, ovaries, and uterus, but all organs examined

contained cyclosiloxanes. In a cohort killed at 1

year, most organs contained measurable cyclosiloxanes

with highest levels in mesenteric lymph nodes,

abdominal fat, and ovaries. Of the individual

cyclosiloxanes measured, selected retention of

decamethylcyclopentasiloxane and

dodecamethyclyclohexasiloxane relative to

octomethylcyclotetrasiloxane was seen in all organs at

the time points studied. Organs from animals receiving

the linear siloxane mixture were harvested at 9, 12,

and 15 weeks. We found maximum levels in the brain,

lungs, and mesenteric lymph nodes, but all other

organs contained measurable levels. These data are, to

the best of our knowledge, the first demonstration

that after a single subcutaneous injection silicones

are widely distributed throughout the body and can

persist over an extended period.

Silicone (polydimethylsiloxane) gels are the chief

component of breast implants. Because these gels are

composed largely of high molecular weight silicones1,

experimental analysis of silicone distribution and its

potential toxicity have been investigated after the

implantation of solid gels2-4. However, we and others

5-7, have demonstrated that 1 to 2% of the silicones

found in implanted gels are low molecular weight

silicones (LMWS) consisting of both cyclic and linear

compounds with repeating units of dimethylsiloxane

(n=3 to 20, Figure 1A). These studies indicate that

LMWS migrate out of intact implants along with the

platinum used as a catalyst in the polymerization

process of silicone gels.5 In addition, these

compounds would be released in the event of implant

rupture. However nothing is known about the

distribution of these LMWS in biological tissues. We

have recently developed a gas chromatographic/mass

spectrometric (GC/MS) detection method for both linear

and cyclic low molecular weight siloxanes in

biological tissues.8 This method is highly sensitive

and allows the examination of silicone-containing

compounds with a molecular mass less than 600 atomic

mass units. We have routinely been able to detect LMWS

in concetrations as low as 0.5mg/g tissue. To study

the distribution of LMWS released from breast implants

we have injected female CD-1 mice subcutaneously with

an enriched low molecular weight (LMW) cyclosiloxane

fraction obtained from explanted breast implants

(breast implant distillate) and followed their

distribution in different organs over the course of a

year. Similarly, injection of LMW linear siloxane

mixture (DMPS-V: Sigma) was used to follow the

distribution of linear siloxanes in biological tissues

over a 15 week period.

MATERIALS AND METHODS

Animal Protocol

Female CD-1 mice (age 8 to 10 wks; 25 to 30 g) were

separated into two groups. Mice in the first group

received a single subcutaneous injection of 250 mg of

breast implant distillate (LMW cyclosiloxane mixture)

in the suprascapular area, and the control mice

received 250 mg of soy oil. Groups of six to eight

control and treated animals were killed at 3, 6, 9, or

52 wks after exposure to LMW cyclosiloxanes. Brain,

heart, kidney, liver, lung, mesenteric lymph nodes,

ovaries, spleen, skeletal muscle, and uterus were

dissected out for the analysis of silicones for 3, 6,

and 9 wk groups. For the 52 wk group, we also

collected adrenals, abdominal fat and perirenal fat.

Similarly, other mice received DMPS-V (low molecular

weight linear siloxane mixture) at a single

subcutaneous injection in the suprascapular area, and

the same were dissected out after 9, 12, 15 wks of

exposure. Preliminary studies have indicated that

linear siloxanes were not detectable in any organ

earlier than 9 weeks after injection. During the

dissection and separation of organs, precautions were

taken to eliminate any possible cross contamination

between the organs by cleaning the dissecting

instruments with ethyl acetate after the separation of

each organ. Harvested organs were weighed and washed

with saline before analysis. Ten or Twenty percent

homogenates of organs were prepared with deionized

water, and 0.1 to 1 ml was used for the extraction of

low molecular weight silicones with an equal volume of

ethyl acetate. No significant differences in body

weights were observed between control and the treated

mice. Food and water were provided ad libitum.

ANALYSIS OF LOW MOLECULAR WEIGHT SILICONES USING GC/MS

The detection of low molecular weight silicone in

mouse organs was carried out as previously described.8

Tissue extracts containing LMWS were injected (1 ml)

in to a gas chromatograph unit (Hewlett-Packard Model

6890) equipped with a low bleed column (J & W

Scientific, DB-XLB) and detected with mass

spectrometry (Hewlett-Packard Model 5972) using scan

mode operation. To quantify cyclosiloxanes, we used

external standard calibration curves obtained for

individual LMW cyclosiloxanes. Individual standard

Octomethylcyclotetrasiloxane (D4),

decamethylcyclopentasiloxane (D5), and

dodecamethylcyclohexasiloxane (D6) were purchased from

Ohio Valley Specialty Chemical (Marietta, OH).

Quantification was based on target ions: 281, 355, and

341 miz were selected for D4, D5, and D6,

respectively. As hexamethylcyclotrisiloxane (D3) is

present at very low levels in silicone breast implant

gels, we have not quantified its distribution.5

Individual components of DMPS-V in tissues were

quantified as described previously.8 The SCAN mode was

selected as opposed to SIM (Selected Ion Monitoring)

for the quantification. This procedure allowed us to

confirm the molecular structures of LMW cyclosiloxanes

in biological tissues by matching their spectra to

Wiley-library spectral data. In case of linear

siloxane determinations, the SIM mode was used for

quantification as described earlier.8 Ethyl acetate

blanks were run between samples to avoid any possible

carry over from one sample to another, and blank

values were subtracted from the sample values during

the data analysis. No detectable silicones were found

in control mice in any of the organs analyzed. The

limit of detection for cyclic and linear siloxanes by

GC/MS was 50 pg. Total siloxane (sum of D4, to D6 in

the case of cyclosiloxanes and sum of L5 to L11 in the

case of linear siloxanes) as well as individual

cyclosiloxane levels were expressed as mg/g wet

tissue.

Statistical analysis of the date were done using

Microsoft Excel Data analysis software package.

One-way analysis of variance was performed to

determine the statistical difference in means of total

or individual siloxanes among groups (3, 6, 9, and 52

weeks) or among D4, D5, and D6 within a group. All

data for the cyclosiloxane determinations (a total of

318 organs and tissues) with the exception of two

values (one of lung and one of spleen from 1-year

group showing exceptionally high values) were included

in our analysis.

RESULTS:

The Molecular structures of low molecular weight

cyclic and linear siloxanes are presented in Figure

1A. We prepared breast implant distillate and analyzed

its compositions by GC/MS.5,8 We Found as expected,

the relative proportion of D3, D4, D5, D6, and

tetradecamethylcycloheptasiloxane (D7) within the

distillate to be ~30, ~45, ~15, ~8, and ~2%,

respectively5,8, GC/MS analysis of DMPS-V revealed

that the mixture consists of low molecular linear

siloxanes L5 to L16. Approximately 80% of this mixture

is L6 to L13.

We obtained gas chromatographic profiles for

cyclosiloxanes from individual organs. The approach is

illustrated for ovary at 9 weeks. The spectral matches

obtained using Wiley-Library mass spectral data

confirm the presence of D4 to D6. We used these data

to analyze the total siloxane content and the

abundance of individual cyclosiloxanes (D4, D5, and D6

) in various organs at different times after

subcutaneous injection of breast implant distillate.

Of the individual cyclic components measured in

organs, only D7 was not detectable.

We found that a 3, 6, and 9 weeks we could detect

cyclosiloxanes in every organ examined. Changes in the

levels of cyclosiloxanes (sum of D4, D5, and D6) in

various organs of mice injected with breast implant

distillate with time are presented in Fig. 2B.

Mesenteric lymph nodes, ovaries and uterus exhibit the

highest levels of cyclosiloxanes among the organs

studied. From 3 to 6 weeks, levels of total

cyclosiloxanes increase in heart, kidney, lung,

mesenteric lymph nodes, ovaries, and uterus with a

slight drop in these levels at 9 weeks. In an entirely

independent experiment we repeated the 3-week and 6

week cyclosiloxane protocol. For each time point we

used nine mice injected with 250 mg of breast implant

distillate and five mice injected with 250 mg of soy

oil. In the distillate-treated mice, we found similar

levels of total as well as individual cyclosiloxanes

in different organs at both time points, indicating

the reproducibility of our results (data not shown).

We also found a large variation in the levels of these

low molecular weight cyclosiloxanes in individual

mice. This variation is illustrated for the levels of

total cyclosiloxane in the organs of 8 mice at 3

weeks. (Figure 2C) Note also that the relative

distribution from organ to organ of these

cyclosiloxanes varies from mouse to mouse for example,

mouse number 4, shows very high levels in spleen

compared with other mice and relatively low levels in

uterus compared with other mice. At present we do not

understand the basis for this idiosyncratic

distribution.

The relative proportions of individual components of

breast implant distillate (D4, D5, and D6) in various

organs for mice exposed for 3, 6, 9, weeks were also

determined (figures 3, A to C). D4, D5, and D6 were

found in all organs. Organs from the 3 week group

exhibited proportions of D4, D5 and D6 similar to that

found in the starting material (breast implant

distillate) (Figures 1B and 3A). In the distillate the

ratios of D4:D5 and D5:D6 were approximately 3 and 2.

In a similar fashion in mesenteric lymph nodes (which

show the highest level of cyclosiloxanes) the ratio of

D4:D5 and D5:D6 were approximately 3 and 2. At 6

weeks, the levels of D4 were similar to those of 3

weeks; however, levels of D5 and D6 increased at 6 and

9 weeks over the 3-week values (Figure 3, AtoC). These

data suggest that there may be a selective retention

of D5 and D6 relative to D4.

Because we found significant retention of

cyclosiloxanes in all organs over a 9 week period, we

were interested in knowing if there was long term

retention of these compounds. Therefore, we killed

another group of mice l year after injection. We also

evaluated retention of cyclosiloxanes in abdominal

fat, perirenal fat and adrenals (Figure 3D). We found

that even after l year most organs have measurable

levels of these compounds. Highest levels were seen in

mesenteric lymph nodes, abdominal fat, and ovaries. In

mesenteric lymph nodes, cyclosiloxane levels at 1 year

are similar to the 9 week levels, whereas in ovaries

and uterus they approach 50% of the 9 week values. As

with the earlier times, D5 and D6 levels are

relatively higher than the D4 levels.

We used a similar approach to analyze the distribution

and abundance of linear LMWS. A representative gas

chromatogram obtained for ethyl acetate extracts of

brain from a mouse injected with DMPS-V and killed at

12 weeks is presented in Figure 4A. Several components

of DMPS-V (L6 to L12) were readily identifiable. The

data representing the changes in the levels of total

linear siloxanes in various tissues of mice exposed to

DMPS-V are presented in Figure 4B. No detectable

levels of linear siloxane were found in any organs of

mice injected with DMPS-V and killed at 3 or 6 weeks.

However, by 9 weeks we detected linear siloxanes, and

with the exception of lung, organ levels of these

siloxanes remained relatively constant at 12 and 15

weeks. In contrast to the cyclosiloxanes, brain and

lung accumulate the maximum levels of linear

siloxanes.

DISCUSSION

Our findings clearly demonstrate that low molecular

weight silicones persist in the organs of mice for at

least 1 year after a single subcutaneous injection.

Additionally, every organ examined accumulated

silicones. We have focused on the LMW cyclosiloxanes

(D4 to D7) because these are known to be the major

components of Breast Implants.5 Individual

cyclosiloxanes show differential retention in tissues.

D5 and D6 appear to persist longer than D4. The

explanation for the observation is unclear, but the

release of individual silicones from individual organs

all contribute to the observed " kinetics " . The

hydrophobicity/lipophilicity of these compounds with

increasing chain length may also contribute to the

selective distribution and retention in various

organs. The substantial interanimal variation seen

from organ to organ is perplexing, it is unclear why

mice vary so greatly in the amount of cyclosiloxane

taken up by individual organs and in the relative

uptake of these organs.

Surprisingly, we found that levels of cyclosiloxanes

were very high in ovary and moderately high in uterus

and that the high levels persisted for 1 year in these

organs. It is unknown whether the presence of LMW

cyclosiloxanes has reproductive implications, but it

is worth noting that other have reported an affinity

of cyclosiloxanes for estrogen receptors. 9 Similarly

our finding that linear siloxanes accumulate

preferentially in brain warrants the need for

additional investigation.

To the best of our knowledge, this is the first

comprehensive analysis of the distribution and

persistence of low molecular weight silicones in a

mammal. Whether these compounds persist indefinitely

and to what extent is an important area for additional

study. Also of interest is the question of whether the

presence of these compounds have any adverse

biological effects. We caution that following

distribution of LMWS injected subcutaneously may not

mimic precisely what might happen with transmigration

of LMWS from a subcutaneously placed implant or its

rupture. However, this approach provides a guide for

additional study. The fact remains that implants

contain LMWS that can migrate through the capsule

underscores the importence of the present study.5 The

wide spread distribution of low molecular weight

silicones and their persistence raises the issue of

possible untoward consequences.

References

1. Lane, et al. Silica, Silicon, and

Silicones...unraveling the mystery. Immunology of

Silicones. Edited by M. Potter, NR Rose, New York,

Springer, 1996, pp3-12

2. Nakamura A, et al. J Biomed Mater Res 1992, 26:

631-650

3. Bradley, SG. et al. Drug Chem Toxicol 1994, 17:

175-220

4. Patter, M., et al, J Nat'l Cancer Inst. 1994, 86:

297-304

5. Lykissa ED, et al, Anal Chem 1997 , 69: 4912-4916

6 Yu L, et al, PRS 1995, 97: 756-764

7. Garrido L, et al, Magn Reson Med 1994, 31: 328-330

8. Kala SV, et al, Anal Chem 1997, 69: 1267-1271

9. Levier RR, et al, Biochemistry of Silicon and

Related Problems, Edited by G. Bendz, I Lindquist, New

York, Plenum, 1978, pp 473-513