Article Text

other Versions

Alterations to cell metabolism in connective tissues of the knee after ovariohysterectomy in a rabbit model: are there implications for the postmenopausal athlete?
  1. D A Hart,
  2. Y Achari
  1. McCaig Institute for Bone and Joint Health, University of Calgary, Calgary, Alberta, Canada
  1. Correspondence to Dr D A Hart, McCaig Institute for Bone and Joint Health, University of Calgary, 3280 Hospital Drive NW, Calgary, Alberta, Canada T2N 4Z6; hartd{at}


Background Participation in regular exercise and athletic activities across the lifespan is encouraged to maintain the cardiovascular and musculoskeletal systems and general wellbeing. Before the menopause there is an increased risk of anterior cruciate ligament (ACL) injuries in female athletes, whereas there is an increased risk of joint diseases such as knee osteoarthritis after the menopause. Although there are few data regarding alterations in individual connective tissues of the knee in humans either before, during or after the menopause, it is possible to assess changes in experimental models following surgical menopause.

Objective To assess changes in cell metabolism in the medial collateral ligament, ACL, patellar tendon, lateral and medial menisci, tibial plateau and femoral condyle articular cartilage and the synovium after surgical menopause in an experimental model system.

Methods Panels of rabbits were subjected to ovariohysterectomy or sham operations, and RNA from each tissue was assessed for collagen, proteoglycan, proteinase, growth factor, sex hormone receptor and inflammatory mediator messenger RNA levels by reverse transcribed PCR.

Results Unique alterations in cell metabolism were detected 2 months after surgical menopause and the pattern of significant changes was tissue specific (number of mRNA species altered, extent of changes, elevation/depression of changes).

Conclusions Changes in cell metabolism may alter the set point for the tissues of the knee and subsequently the functioning of the knee after the menopause. Such changes may contribute to an increased risk of injury and/or degenerative conditions. Further studies in pre and postmenopausal women athletes may also shed light on whether the present findings can be extrapolated to human populations.

Statistics from

Exercise through organised athletic activities or via recreational walking, running, swimming, etc has been encouraged as the population ages to maintain the cardiovascular system, as well as the musculoskeletal system, with emphasis on muscle tone and bone integrity. Many individuals in the fifth, sixth and seventh decade of life are participating in such activities. “Senior” level events are becoming more prevalent in many more arenas ranging from marathon and half marathons, 5 and 10 km races, tennis and other aerobic endeavours. Participation in such activities as the population ages, along with healthy lifestyle choices, has been suggested to contribute to the increased quality of cardiovascular tone, enhanced mobility and general wellbeing of many individuals.1 However, the potential downside risks of “enthusiastic” exercise on load-bearing joints has not been thoroughly explored, particularly for women as they proceed through the menopause.

It is well known that connective tissues of the knee express oestrogen and progesterone receptors,2 3 and are thus likely to be responsive to sex hormones. Female athletes competing at the collegiate and professional levels in many sports experience non-contact-mediated injuries to their anterior cruciate ligaments (ACL) of the knee at rates several times higher than those of their male counterparts.4,,7 Some studies have implicated fluctuations in sex hormones during the menstrual cycle as playing a role in such ACL injuries.5 Furthermore, some studies have indicated that knee joint laxity and function varies across the menstrual cycle for some women, but not others,8 implicating heterogeneity in response to hormones in women. However, although the mechanisms involved have not yet been identified, this body of evidence does imply that the connective tissues of the female knee are responsive to sex hormones.

Following menopause, sex hormone concentrations decline over a period of time and their regulatory role in connective tissue homeostasis is ultimately lost. The overt influence of sex hormones on such tissues, maintained for over 35 years (~13 to ~50 years of age), probably leads to a new “set point” for the tissues based on the mechanical environment without the modulating influence of sex hormones. It is possible that hormone replacement therapy (HRT) may overcome some of this hormonal loss, but many women now decline to use such therapies as a result of potential cancer-related issues. A recent paper by Cook et al9 has indicated that noninvasive assessment of the Achilles tendon using ultrasound indicated that active postmenopausal women who were taking HRT had fewer tendon abnormalities and less thick tendons than active women not taking HRT. Furthermore, the relationship between HRT and tendon abnormalities for active women were not observed for inactive women, possibly indicating an interaction between mechanical loading and hormones. Another recent study examined a cohort of asymptomatic postmenopausal women by magnetic resonance imaging for meniscal tears at baseline and 2 years later.10 In that study, the incidence of meniscal tears increased over the 2 years of the study, were related to the age of the women and an increased tibial plateau bone area. As postmenopausal women are at greater risk of the development of osteoarthritis than age-matched men or premenopausal women,3 11 this may indicate that meniscal degeneration in women plays a role in such an incidence of osteoarthritis.

How the menopause affects homeostasis in individual connective tissues of the knee has not been well studied. Although not a study that can effectively be performed in humans for obvious reasons, it can be performed in experimental models to start to explore how tissues in different mechanical environments (compressive vs tensile, extra-articular vs intra-articular, high load vs low load) respond to the loss of hormones via surgical menopause. We report here that 2 months after ovariohysterectomy, during the “transition phase”, which may be analogous to a “perimenopausal” state in humans, analysis of tissues of the knee indicate that the response to the surgery is tissue specific when the responses of ligaments, menisci, cartilage surfaces, synovium and tendons of the knee are compared.

Materials and methods

Animals and animal surgery

Adult 1-year-old female rabbits were obtained from Riemen’s Rabbitry (St Agatha, Ontario, Canada). After acclimating for 10 days, randomly selected animals were subjected to ovariohysterectomy by Dr Morck (university veterinarian) as described previously12 with a protocol approved by the Faculty of Medicine Animal Care Committee. All animals tolerated the operations well and continued to maintain their weight postovariohysterectomy. At 2 months post-ovariohysterectomy, six sham and six ovariohysterectomy animals were killed and the tissues of the knee (medial collateral ligament (MCL), ACL, patellar tendon, medial menisci, lateral menisci, tibial plateau articular cartilage, femoral condyle cartilage and synovium) were immediately removed, weighed and frozen in liquid nitrogen. The frozen tissues were maintained at −80°C until processed.

RNA isolation

Total RNA was isolated by the TRIspin method, purified, and quantified.13 All samples for a given tissue were processed at the same time.

Reverse transcribed PCR

All RNA samples (1 μg) for a given tissue were reverse transcribed at the same time to avoid variability. Subsequently, all complementary DNA samples for a given tissue were assessed using validated rabbit-specific primers and PCR and were again assessed at the same time to estimate specific messenger RNA levels as described previously.2 12,,14 All non-reverse transcribed controls were negative for genomic DNA contamination, and values for individual molecules assessed were normalised to mRNA levels for beta-actin, which served as a housekeeping gene that was not affected by ovariohysterectomy (DA Hart, CR Reno and Y Achari, unpublished data). Values for the control animals were set at 100% and values for the ovariohysterectomy animals are presented as a percentage of control values (SEM).

Statistical analysis

Statistical analysis of the data was performed using analysis of variance, standard deviation, and standard error of the mean using Excel 5.0 software.


The results for the analysis of the eight knee tissues and 23 genes per tissue at 2 months post-surgical menopause are summarised in table 1. The panel of molecules assessed was composed of collagens, small proteoglycans, proteinases, growth factors and receptors, including sex hormone receptors and inflammatory mediators. All of these molecules are relevant to connective tissue maintenance, remodelling and degradation.

Table 1

Effect of ovariohysterectomy on adult rabbit knee tissue mRNA levels

If one focuses first on the molecules in the ovariohysterectomy group that had levels that were significantly different from control values (p<0.05), one can immediately see that four things are obvious. The first is that the number of molecules affected varied greatly between tissues, varying from 10 of 23 in the MCL and lateral menisci, to one of 23 in the medial menisci (table 1). It is of interest that such differences between the lateral menisci and the medial menisci were detected as the medial menisci is the first menisci to exhibit degeneration after ACL transection during the development of osteoarthritis in the rabbit model.14 15 Furthermore, the number of molecules affected by ovariohysterectomy in the MCL (10/23) exceeded the number of changes in the ACL (5/23). As the MCL is extraarticular and in a low load environment, whereas the ACL is intra-articular and in a high load environment, environment appears to play a role in the number of molecules affected.

Second, from table 1 it can be seen that the specific molecules affected by ovariohysterectomy vary between tissues. Ovariohysterectomy thus does not appear to affect mRNA levels for the same molecules in different tissues, implying that sex hormones are not contributing to tissue homeostasis equally. This is particularly evident in the MCL and ACL, where mRNA levels for collagen I, III and V were affected in the MCL, but not the ACL.

Third, it can be seen that the effect of ovariohysterectomy on some tissues led to elevations in mRNA levels for a subset of molecules, whereas in other tissues significant depressions were observed (eg, cyclooxygenase 2 in the femoral condyle and MCL compared with the tibial plateau; progesterone receptor in the patellar tendon compared with the lateral menisci). The loss of sex hormones via surgical menopause thus did not lead to uniform effects on mRNA levels. As sex hormones plus ligands can interact with different DNA motifs (ERE, AP-1 sites) to yield different outcomes depending on the cell and the gene target,16 17 one cannot readily predict as yet how a specific tissue will respond to ovariohysterectomy.

Fourth, for some of the mRNA levels assessed considerable variation was noted between animals, thus influencing the statistical significance (table 1). Furthermore, for a number of the molecules assessed trends towards differences were noted (p = 0.05–0.10 or p = 0.1–0.2). Such results may be due to the relatively short time post-ovariohysterectomy that the analysis was performed (8 weeks); however, this time point was chosen as it probably represents a transition phase of the response to ovariohysterectomy and a time point of potential vulnerability for the tissues.


The results presented in this report clearly show that mRNA levels for a number of relevant molecules are affected by ovariohysterectomy in connective tissues of the rabbit knee. Table 1 Effect of ovariohysterectomy on adult rabbit knee tissue mRNA levels The tissues were influenced to differing degrees based on the number of molecules for which mRNA levels were affected, which molecules were affected and the direction of the change (elevations or depressions). Ovariohysterectomy thus does not affect all connective tissues of the knee the same, and the potential impact of the changes on the functioning of the tissue or the risk of injury may not be the same.

We deliberately chose a time post-ovariohysterectomy that was early after the onset of this model of surgical menopause as tissues in such a “transition stage” may potentially be more vulnerable to adverse sequalae associated with continued athletic activity at a high level or even a moderately high level. This time frame may be analogous to a “perimenopausal” state, which some authors have defined as a risk period for osteoarthritis development.18 However, with longer times post-ovariohysterectomy more significant differences may become evident, and possibly some of the variation for specific molecules between animals would disappear and the trends become significant. Whereas such a longitudinal study is feasible and may provide more insights into the post-ovariohysterectomy process, the results presented clearly show that even by 8 weeks post-surgical menopause in this model, significant tissue-specific alterations in mRNA levels were evident. Whereas mRNA changes are only potential protein changes, they do indicate indicative of that cellular metabolism has been affected by the “surgical menopause”. Relevant to this point are results of preliminary studies of the bones of the tibia and femur of these animals using micro computed tomography, which have indicated that early osteoporotic changes can be detected even by 8 weeks post-ovariohysterectomy (S Chiu, B Halgrimsson, Y Achari and DA Hart, unpublished data). However, in the future it will be important to determine whether significant protein changes are occurring in the connective tissues of the knee at longer times after the menopause, particularly for molecules such as the collagens and proteoglycans, which have a long half-life but are critically important for maintaining tissue integrity and repair processes.

The most critical questions raised by the results presented are related to the implications of the findings for the postmenopausal athlete engaged in moderate to rigorous levels of activity. It is known that a subset of postmenopausal women are at higher risk for the development of osteoarthritis than premenopausal women,11 but it is not known how that risk is manifested and there are no tests currently available to identify those at highest risk at the genetic or any other level. However, from twin studies and other longitudinal studies it is clear that genetic considerations contribute to some of this risk.19,,22

What is already known on this topic

There is a relative paucity of literature regarding this topic, although some recent papers9 10 have started to address specific aspects of the potential issues associated with the postmenopausal state with a focus on tendons and menisci of the knee, respectively.

What this study adds

The present study adds new insights into molecular changes in nearly all of the tissues of the knee following surgical menopause in an experimental model system. This study thus addresses the issue of coordinated alterations to individual tissues of the knee rather than one tissue, particularly at a critical “transitional” phase of the response to the “menopause” shortly after surgery when the potential risk of injury may be elevated. As this type of invasive study probably cannot be done in human populations, it will however, hopefully stimulate additional new investigations directed at postmenopausal populations to provide a further elaboration of risks.

Of relevance to this discussion is whether participation at the elite or a high level of athletics before the menopause alters the potential risk for postmenopausal osteoarthritis development in subsets of women who have genetic risk but no overt injuries associated with such participation. As tissues of the knee respond to loading, one could envision that participation could either protect against risk or contribute to the risk after the menopause. Whereas there is some literature that indicates that participation in athletics at the elite level can influence the risk of osteoarthritis later in life for male athletes23 depending partly on the type of sport, similar reports for female athletes were not found. However, with the expansion of participation by female athletes in elite level activities (Olympics, world championships, professional leagues, etc), this may be a topic for study in the future, particularly for women with a family history of disease.

It is also important to point out that at the present time it is not clear if the onset and progression of the menopause is accompanied by an increased risk of overt injury to knee tissues. A search of the literature has not revealed any definitive studies addressing this point. However, some studies such as that of Haapasalo et al24 have reported an age-related decline in injuries in a Finnish population. Whether this is due to a decline in intensity levels in the population studied or other factors is not known. However, the incidence of knee injuries in postmenopausal female athletes participating in more intense activities remains to be investigated. There are two main alternatives for this population, either there is no increased risk and it is not a significant problem, or there is a problem but it has not yet been appreciated or is attributed to other interpretations (ageing, previous unrecognised injuries, etc). If the situation is the latter, then it is possibly analogous to the increased incidence of ACL injuries in younger athletes, in which the problem existed but was unappreciated until attention was brought to bear on the issue.4 5 As the “baby boomers” age and continue their quest to maintain their youth, it may be possible to gain more evidence for or against the issue of athletic activity and the risk of adverse sequelae with regard to joint deterioration in the older female population.

It is known that muscles do change during ageing with the development of sarcopaenia in many individuals as they age, and there is an increased risk of muscle damage in older women as they engage and recover from certain types of exercises.25 26 There may be neural control considerations of joint stability in older women compared with men,27 so a number of factors could contribute to the postmenopausal state as metabolic changes noted functioning of the knee in the women age in addition to the in the present study. However, additional studies will have to be initiated in the future to determine whether there is a constellation of variables that could identify those at greatest risk of adverse events associated with continued athletic participation at a high level.

The above discussion should not be construed as advocating the avoidance of participation in exercise programmes in the postmenopausal population. As it is clear that maintenance of the cardiovascular, bone and muscle systems requires a reasonable exercise programme, the benefits of moderate exercise probably outweigh the risk for most individuals. As advances in “personalised medicine” lead to more accurate and comprehensive insights into disease predictions in the future, those individuals at higher risk can probably be cautioned against too much high-intensity athletic participation with increasing confidence, and others without such risk can still participate to maintain the biological systems indicated above as well as contributing to effective mobility during the ageing process.

In summary, the results presented indicate that the set point for specific connective tissues in the knee may be influenced by ovariohysterectomy/menopause, at least in this model system. Such results will have to be validated in some manner for human populations in the future. However, if we consider the knee as an organ system,28 then the new “set” point for the knee following the menopause is a composite of the individual tissue set points, and thus the risk of injury and/or degeneration may be reflected by new weaknesses. However, it should also be pointed out that tissues of the knee can adapt to an altered loading environment due to overt injury, and thus some of the observed changes could arise from the direct withdrawal of hormones after surgery, whereas other of the observed changes could arise secondarily in response to overt alterations in a subset of the tissues.28 Such an effect may become more apparent with time (and activity level), so this caveat should be kept in mind.

Therefore, this area of research should be further explored in more detail to determine whether such a new set point for the knee may contribute to degenerative processes in some of the women who engage in athletics and rigorous exercise after the menopause. As mentioned above, this is not meant to discourage exercise,29 but to raise a “flag” that exercise in the pre and postmenopausal state may impact connective tissues very differently, and one should be cautious in extrapolating findings from the study of a younger age bracket of women to an older postmenopausal population. In addition, unabashed endorsement of extreme exercise regimens and athletic endeavours1 may not be for everyone as they age and such thinking should be viewed with caution.


The authors would like to thank Carol Reno and Dr Doug Morck for their assistance with these studies.


View Abstract


  • Funding The studies were supported by grants from the CIHR Institute for Gender and Health and the Alberta Heritage Foundation for Medical Research via a team grant in osteoarthritis, as well as the Calgary Foundation–Grace Glaum Professorship to DAH. None of the sources of funding provided any conflict of interest.

  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.